diff options
Diffstat (limited to 'drivers/gpu/drm/amd/display')
198 files changed, 26844 insertions, 3242 deletions
diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index f954bf61af28..71991a28a775 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -26,6 +26,14 @@ config DRM_AMD_DC_DCN2_0 Choose this option if you want to have Navi support for display engine +config DRM_AMD_DC_DCN2_1 + bool "DCN 2.1 family" + depends on DRM_AMD_DC && X86 + depends on DRM_AMD_DC_DCN2_0 + help + Choose this option if you want to have + Renoir support for display engine + config DRM_AMD_DC_DSC_SUPPORT bool "DSC support" default y diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 4a29f72334d0..8cab6da512a0 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -688,12 +688,15 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) */ if (adev->flags & AMD_IS_APU && adev->asic_type >= CHIP_CARRIZO && - adev->asic_type < CHIP_RAVEN) + adev->asic_type <= CHIP_RAVEN) init_data.flags.gpu_vm_support = true; if (amdgpu_dc_feature_mask & DC_FBC_MASK) init_data.flags.fbc_support = true; + if (amdgpu_dc_feature_mask & DC_MULTI_MON_PP_MCLK_SWITCH_MASK) + init_data.flags.multi_mon_pp_mclk_switch = true; + init_data.flags.power_down_display_on_boot = true; #ifdef CONFIG_DRM_AMD_DC_DCN2_0 @@ -809,6 +812,9 @@ static int load_dmcu_fw(struct amdgpu_device *adev) case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_NAVI10: + case CHIP_NAVI14: + case CHIP_NAVI12: + case CHIP_RENOIR: return 0; case CHIP_RAVEN: if (ASICREV_IS_PICASSO(adev->external_rev_id)) @@ -2107,6 +2113,7 @@ static int amdgpu_dm_backlight_get_brightness(struct backlight_device *bd) } static const struct backlight_ops amdgpu_dm_backlight_ops = { + .options = BL_CORE_SUSPENDRESUME, .get_brightness = amdgpu_dm_backlight_get_brightness, .update_status = amdgpu_dm_backlight_update_status, }; @@ -2358,7 +2365,12 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) #if defined(CONFIG_DRM_AMD_DC_DCN1_0) case CHIP_RAVEN: #if defined(CONFIG_DRM_AMD_DC_DCN2_0) + case CHIP_NAVI12: case CHIP_NAVI10: + case CHIP_NAVI14: +#endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + case CHIP_RENOIR: #endif if (dcn10_register_irq_handlers(dm->adev)) { DRM_ERROR("DM: Failed to initialize IRQ\n"); @@ -2373,6 +2385,8 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) if (adev->asic_type != CHIP_CARRIZO && adev->asic_type != CHIP_STONEY) dm->dc->debug.disable_stutter = amdgpu_pp_feature_mask & PP_STUTTER_MODE ? false : true; + if (adev->asic_type == CHIP_RENOIR) + dm->dc->debug.disable_stutter = true; return 0; fail: @@ -2428,8 +2442,7 @@ static ssize_t s3_debug_store(struct device *device, { int ret; int s3_state; - struct pci_dev *pdev = to_pci_dev(device); - struct drm_device *drm_dev = pci_get_drvdata(pdev); + struct drm_device *drm_dev = dev_get_drvdata(device); struct amdgpu_device *adev = drm_dev->dev_private; ret = kstrtoint(buf, 0, &s3_state); @@ -2515,10 +2528,23 @@ static int dm_early_init(void *handle) #endif #if defined(CONFIG_DRM_AMD_DC_DCN2_0) case CHIP_NAVI10: + case CHIP_NAVI12: adev->mode_info.num_crtc = 6; adev->mode_info.num_hpd = 6; adev->mode_info.num_dig = 6; break; + case CHIP_NAVI14: + adev->mode_info.num_crtc = 5; + adev->mode_info.num_hpd = 5; + adev->mode_info.num_dig = 5; + break; +#endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + case CHIP_RENOIR: + adev->mode_info.num_crtc = 4; + adev->mode_info.num_hpd = 4; + adev->mode_info.num_dig = 4; + break; #endif default: DRM_ERROR("Unsupported ASIC type: 0x%X\n", adev->asic_type); @@ -2665,7 +2691,7 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, const struct amdgpu_framebuffer *afb, const enum surface_pixel_format format, const enum dc_rotation_angle rotation, - const union plane_size *plane_size, + const struct plane_size *plane_size, const union dc_tiling_info *tiling_info, const uint64_t info, struct dc_plane_dcc_param *dcc, @@ -2691,8 +2717,8 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, return -EINVAL; input.format = format; - input.surface_size.width = plane_size->grph.surface_size.width; - input.surface_size.height = plane_size->grph.surface_size.height; + input.surface_size.width = plane_size->surface_size.width; + input.surface_size.height = plane_size->surface_size.height; input.swizzle_mode = tiling_info->gfx9.swizzle; if (rotation == ROTATION_ANGLE_0 || rotation == ROTATION_ANGLE_180) @@ -2710,9 +2736,9 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, return -EINVAL; dcc->enable = 1; - dcc->grph.meta_pitch = + dcc->meta_pitch = AMDGPU_TILING_GET(info, DCC_PITCH_MAX) + 1; - dcc->grph.independent_64b_blks = i64b; + dcc->independent_64b_blks = i64b; dcc_address = get_dcc_address(afb->address, info); address->grph.meta_addr.low_part = lower_32_bits(dcc_address); @@ -2728,7 +2754,7 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, const enum dc_rotation_angle rotation, const uint64_t tiling_flags, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, struct dc_plane_dcc_param *dcc, struct dc_plane_address *address) { @@ -2741,11 +2767,11 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, memset(address, 0, sizeof(*address)); if (format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) { - plane_size->grph.surface_size.x = 0; - plane_size->grph.surface_size.y = 0; - plane_size->grph.surface_size.width = fb->width; - plane_size->grph.surface_size.height = fb->height; - plane_size->grph.surface_pitch = + plane_size->surface_size.x = 0; + plane_size->surface_size.y = 0; + plane_size->surface_size.width = fb->width; + plane_size->surface_size.height = fb->height; + plane_size->surface_pitch = fb->pitches[0] / fb->format->cpp[0]; address->type = PLN_ADDR_TYPE_GRAPHICS; @@ -2754,20 +2780,20 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, } else if (format < SURFACE_PIXEL_FORMAT_INVALID) { uint64_t chroma_addr = afb->address + fb->offsets[1]; - plane_size->video.luma_size.x = 0; - plane_size->video.luma_size.y = 0; - plane_size->video.luma_size.width = fb->width; - plane_size->video.luma_size.height = fb->height; - plane_size->video.luma_pitch = + plane_size->surface_size.x = 0; + plane_size->surface_size.y = 0; + plane_size->surface_size.width = fb->width; + plane_size->surface_size.height = fb->height; + plane_size->surface_pitch = fb->pitches[0] / fb->format->cpp[0]; - plane_size->video.chroma_size.x = 0; - plane_size->video.chroma_size.y = 0; + plane_size->chroma_size.x = 0; + plane_size->chroma_size.y = 0; /* TODO: set these based on surface format */ - plane_size->video.chroma_size.width = fb->width / 2; - plane_size->video.chroma_size.height = fb->height / 2; + plane_size->chroma_size.width = fb->width / 2; + plane_size->chroma_size.height = fb->height / 2; - plane_size->video.chroma_pitch = + plane_size->chroma_pitch = fb->pitches[1] / fb->format->cpp[1]; address->type = PLN_ADDR_TYPE_VIDEO_PROGRESSIVE; @@ -2814,6 +2840,11 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, adev->asic_type == CHIP_VEGA20 || #if defined(CONFIG_DRM_AMD_DC_DCN2_0) adev->asic_type == CHIP_NAVI10 || + adev->asic_type == CHIP_NAVI14 || + adev->asic_type == CHIP_NAVI12 || +#endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + adev->asic_type == CHIP_RENOIR || #endif adev->asic_type == CHIP_RAVEN) { /* Fill GFX9 params */ @@ -2995,6 +3026,8 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, plane_info->visible = true; plane_info->stereo_format = PLANE_STEREO_FORMAT_NONE; + plane_info->layer_index = 0; + ret = fill_plane_color_attributes(plane_state, plane_info->format, &plane_info->color_space); if (ret) @@ -3060,6 +3093,7 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev, dc_plane_state->global_alpha = plane_info.global_alpha; dc_plane_state->global_alpha_value = plane_info.global_alpha_value; dc_plane_state->dcc = plane_info.dcc; + dc_plane_state->layer_index = plane_info.layer_index; // Always returns 0 /* * Always set input transfer function, since plane state is refreshed @@ -3131,13 +3165,25 @@ static enum dc_color_depth convert_color_depth_from_display_info(const struct drm_connector *connector, const struct drm_connector_state *state) { - uint32_t bpc = connector->display_info.bpc; + uint8_t bpc = (uint8_t)connector->display_info.bpc; + + /* Assume 8 bpc by default if no bpc is specified. */ + bpc = bpc ? bpc : 8; if (!state) state = connector->state; if (state) { - bpc = state->max_bpc; + /* + * Cap display bpc based on the user requested value. + * + * The value for state->max_bpc may not correctly updated + * depending on when the connector gets added to the state + * or if this was called outside of atomic check, so it + * can't be used directly. + */ + bpc = min(bpc, state->max_requested_bpc); + /* Round down to the nearest even number. */ bpc = bpc - (bpc & 1); } @@ -3491,6 +3537,10 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, bool scale = dm_state ? (dm_state->scaling != RMX_OFF) : false; int mode_refresh; int preferred_refresh = 0; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + struct dsc_dec_dpcd_caps dsc_caps; + uint32_t link_bandwidth_kbps; +#endif struct dc_sink *sink = NULL; if (aconnector == NULL) { @@ -3563,17 +3613,23 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, &mode, &aconnector->base, con_state, old_stream); #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - /* stream->timing.flags.DSC = 0; */ - /* */ - /* if (aconnector->dc_link && */ - /* aconnector->dc_link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT #<{(|&& */ - /* aconnector->dc_link->dpcd_caps.dsc_caps.dsc_basic_caps.is_dsc_supported|)}>#) */ - /* if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc, */ - /* &aconnector->dc_link->dpcd_caps.dsc_caps, */ - /* dc_link_bandwidth_kbps(aconnector->dc_link, dc_link_get_link_cap(aconnector->dc_link)), */ - /* &stream->timing, */ - /* &stream->timing.dsc_cfg)) */ - /* stream->timing.flags.DSC = 1; */ + stream->timing.flags.DSC = 0; + + if (aconnector->dc_link && sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT) { + dc_dsc_parse_dsc_dpcd(aconnector->dc_link->dpcd_caps.dsc_caps.dsc_basic_caps.raw, + aconnector->dc_link->dpcd_caps.dsc_caps.dsc_ext_caps.raw, + &dsc_caps); + link_bandwidth_kbps = dc_link_bandwidth_kbps(aconnector->dc_link, + dc_link_get_link_cap(aconnector->dc_link)); + + if (dsc_caps.is_dsc_supported) + if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc, + &dsc_caps, + link_bandwidth_kbps, + &stream->timing, + &stream->timing.dsc_cfg)) + stream->timing.flags.DSC = 1; + } #endif update_stream_scaling_settings(&mode, dm_state, stream); @@ -3657,7 +3713,7 @@ dm_crtc_duplicate_state(struct drm_crtc *crtc) state->abm_level = cur->abm_level; state->vrr_supported = cur->vrr_supported; state->freesync_config = cur->freesync_config; - state->crc_enabled = cur->crc_enabled; + state->crc_src = cur->crc_src; state->cm_has_degamma = cur->cm_has_degamma; state->cm_is_degamma_srgb = cur->cm_is_degamma_srgb; @@ -3727,6 +3783,7 @@ static const struct drm_crtc_funcs amdgpu_dm_crtc_funcs = { .atomic_destroy_state = dm_crtc_destroy_state, .set_crc_source = amdgpu_dm_crtc_set_crc_source, .verify_crc_source = amdgpu_dm_crtc_verify_crc_source, + .get_crc_sources = amdgpu_dm_crtc_get_crc_sources, .enable_vblank = dm_enable_vblank, .disable_vblank = dm_disable_vblank, }; @@ -4446,7 +4503,7 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane, } if (plane->type != DRM_PLANE_TYPE_CURSOR) - domain = amdgpu_display_supported_domains(adev); + domain = amdgpu_display_supported_domains(adev, rbo->flags); else domain = AMDGPU_GEM_DOMAIN_VRAM; @@ -4536,20 +4593,10 @@ static int dm_plane_atomic_check(struct drm_plane *plane, static int dm_plane_atomic_async_check(struct drm_plane *plane, struct drm_plane_state *new_plane_state) { - struct drm_plane_state *old_plane_state = - drm_atomic_get_old_plane_state(new_plane_state->state, plane); - /* Only support async updates on cursor planes. */ if (plane->type != DRM_PLANE_TYPE_CURSOR) return -EINVAL; - /* - * DRM calls prepare_fb and cleanup_fb on new_plane_state for - * async commits so don't allow fb changes. - */ - if (old_plane_state->fb != new_plane_state->fb) - return -EINVAL; - return 0; } @@ -5693,11 +5740,11 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, * deadlock during GPU reset when this fence will not signal * but we hold reservation lock for the BO. */ - r = reservation_object_wait_timeout_rcu(abo->tbo.resv, true, + r = dma_resv_wait_timeout_rcu(abo->tbo.base.resv, true, false, msecs_to_jiffies(5000)); if (unlikely(r <= 0)) - DRM_ERROR("Waiting for fences timed out or interrupted!"); + DRM_ERROR("Waiting for fences timed out!"); /* * TODO This might fail and hence better not used, wait @@ -5721,8 +5768,13 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, bundle->surface_updates[planes_count].plane_info = &bundle->plane_infos[planes_count]; + /* + * Only allow immediate flips for fast updates that don't + * change FB pitch, DCC state, rotation or mirroing. + */ bundle->flip_addrs[planes_count].flip_immediate = - (crtc->state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) != 0; + crtc->state->async_flip && + acrtc_state->update_type == UPDATE_TYPE_FAST; timestamp_ns = ktime_get_ns(); bundle->flip_addrs[planes_count].flip_timestamp_in_us = div_u64(timestamp_ns, 1000); @@ -5967,6 +6019,7 @@ static void amdgpu_dm_enable_crtc_interrupts(struct drm_device *dev, struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state, *new_crtc_state; int i; + enum amdgpu_dm_pipe_crc_source source; for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { @@ -5992,9 +6045,11 @@ static void amdgpu_dm_enable_crtc_interrupts(struct drm_device *dev, #ifdef CONFIG_DEBUG_FS /* The stream has changed so CRC capture needs to re-enabled. */ - if (dm_new_crtc_state->crc_enabled) { - dm_new_crtc_state->crc_enabled = false; - amdgpu_dm_crtc_set_crc_source(crtc, "auto"); + source = dm_new_crtc_state->crc_src; + if (amdgpu_dm_is_valid_crc_source(source)) { + amdgpu_dm_crtc_configure_crc_source( + crtc, dm_new_crtc_state, + dm_new_crtc_state->crc_src); } #endif } @@ -6045,23 +6100,8 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev, if (dm_old_crtc_state->interrupts_enabled && (!dm_new_crtc_state->interrupts_enabled || - drm_atomic_crtc_needs_modeset(new_crtc_state))) { - /* - * Drop the extra vblank reference added by CRC - * capture if applicable. - */ - if (dm_new_crtc_state->crc_enabled) - drm_crtc_vblank_put(crtc); - - /* - * Only keep CRC capture enabled if there's - * still a stream for the CRTC. - */ - if (!dm_new_crtc_state->stream) - dm_new_crtc_state->crc_enabled = false; - + drm_atomic_crtc_needs_modeset(new_crtc_state))) manage_dm_interrupts(adev, acrtc, false); - } } /* * Add check here for SoC's that support hardware cursor plane, to @@ -6310,7 +6350,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) amdgpu_dm_enable_crtc_interrupts(dev, state, true); for_each_new_crtc_in_state(state, crtc, new_crtc_state, j) - if (new_crtc_state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) + if (new_crtc_state->async_flip) wait_for_vblank = false; /* update planes when needed per crtc*/ @@ -7033,6 +7073,12 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, continue; for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, j) { + const struct amdgpu_framebuffer *amdgpu_fb = + to_amdgpu_framebuffer(new_plane_state->fb); + struct dc_plane_info plane_info; + struct dc_flip_addrs flip_addr; + uint64_t tiling_flags; + new_plane_crtc = new_plane_state->crtc; old_plane_crtc = old_plane_state->crtc; new_dm_plane_state = to_dm_plane_state(new_plane_state); @@ -7076,6 +7122,24 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, updates[num_plane].scaling_info = &scaling_info; + if (amdgpu_fb) { + ret = get_fb_info(amdgpu_fb, &tiling_flags); + if (ret) + goto cleanup; + + memset(&flip_addr, 0, sizeof(flip_addr)); + + ret = fill_dc_plane_info_and_addr( + dm->adev, new_plane_state, tiling_flags, + &plane_info, + &flip_addr.address); + if (ret) + goto cleanup; + + updates[num_plane].plane_info = &plane_info; + updates[num_plane].flip_addr = &flip_addr; + } + num_plane++; } @@ -7272,6 +7336,26 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, if (ret) goto fail; + if (state->legacy_cursor_update) { + /* + * This is a fast cursor update coming from the plane update + * helper, check if it can be done asynchronously for better + * performance. + */ + state->async_update = + !drm_atomic_helper_async_check(dev, state); + + /* + * Skip the remaining global validation if this is an async + * update. Cursor updates can be done without affecting + * state or bandwidth calcs and this avoids the performance + * penalty of locking the private state object and + * allocating a new dc_state. + */ + if (state->async_update) + return 0; + } + /* Check scaling and underscan changes*/ /* TODO Removed scaling changes validation due to inability to commit * new stream into context w\o causing full reset. Need to @@ -7324,13 +7408,37 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, ret = -EINVAL; goto fail; } - } else if (state->legacy_cursor_update) { + } else { /* - * This is a fast cursor update coming from the plane update - * helper, check if it can be done asynchronously for better - * performance. + * The commit is a fast update. Fast updates shouldn't change + * the DC context, affect global validation, and can have their + * commit work done in parallel with other commits not touching + * the same resource. If we have a new DC context as part of + * the DM atomic state from validation we need to free it and + * retain the existing one instead. */ - state->async_update = !drm_atomic_helper_async_check(dev, state); + struct dm_atomic_state *new_dm_state, *old_dm_state; + + new_dm_state = dm_atomic_get_new_state(state); + old_dm_state = dm_atomic_get_old_state(state); + + if (new_dm_state && old_dm_state) { + if (new_dm_state->context) + dc_release_state(new_dm_state->context); + + new_dm_state->context = old_dm_state->context; + + if (old_dm_state->context) + dc_retain_state(old_dm_state->context); + } + } + + /* Store the overall update type for use later in atomic check. */ + for_each_new_crtc_in_state (state, crtc, new_crtc_state, i) { + struct dm_crtc_state *dm_new_crtc_state = + to_dm_crtc_state(new_crtc_state); + + dm_new_crtc_state->update_type = (int)overall_update_type; } /* Must be success */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index b89cbbfcc0e9..c8c525a2b505 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -50,6 +50,7 @@ #include "irq_types.h" #include "signal_types.h" +#include "amdgpu_dm_crc.h" /* Forward declarations */ struct amdgpu_device; @@ -309,11 +310,12 @@ struct dm_crtc_state { bool cm_has_degamma; bool cm_is_degamma_srgb; + int update_type; int active_planes; bool interrupts_enabled; int crc_skip_count; - bool crc_enabled; + enum amdgpu_dm_pipe_crc_source crc_src; bool freesync_timing_changed; bool freesync_vrr_info_changed; @@ -380,19 +382,6 @@ void dm_restore_drm_connector_state(struct drm_device *dev, void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, struct edid *edid); -/* amdgpu_dm_crc.c */ -#ifdef CONFIG_DEBUG_FS -int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name); -int amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, - const char *src_name, - size_t *values_cnt); -void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc); -#else -#define amdgpu_dm_crtc_set_crc_source NULL -#define amdgpu_dm_crtc_verify_crc_source NULL -#define amdgpu_dm_crtc_handle_crc_irq(x) -#endif - #define MAX_COLOR_LUT_ENTRIES 4096 /* Legacy gamm LUT users such as X doesn't like large LUT sizes */ #define MAX_COLOR_LEGACY_LUT_ENTRIES 256 diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c index bc67e6502733..a549c7c717dd 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c @@ -30,23 +30,57 @@ #include "amdgpu_dm.h" #include "dc.h" -enum amdgpu_dm_pipe_crc_source { - AMDGPU_DM_PIPE_CRC_SOURCE_NONE = 0, - AMDGPU_DM_PIPE_CRC_SOURCE_AUTO, - AMDGPU_DM_PIPE_CRC_SOURCE_MAX, - AMDGPU_DM_PIPE_CRC_SOURCE_INVALID = -1, +static const char *const pipe_crc_sources[] = { + "none", + "crtc", + "crtc dither", + "dprx", + "dprx dither", + "auto", }; static enum amdgpu_dm_pipe_crc_source dm_parse_crc_source(const char *source) { if (!source || !strcmp(source, "none")) return AMDGPU_DM_PIPE_CRC_SOURCE_NONE; - if (!strcmp(source, "auto")) - return AMDGPU_DM_PIPE_CRC_SOURCE_AUTO; + if (!strcmp(source, "auto") || !strcmp(source, "crtc")) + return AMDGPU_DM_PIPE_CRC_SOURCE_CRTC; + if (!strcmp(source, "dprx")) + return AMDGPU_DM_PIPE_CRC_SOURCE_DPRX; + if (!strcmp(source, "crtc dither")) + return AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER; + if (!strcmp(source, "dprx dither")) + return AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER; return AMDGPU_DM_PIPE_CRC_SOURCE_INVALID; } +static bool dm_is_crc_source_crtc(enum amdgpu_dm_pipe_crc_source src) +{ + return (src == AMDGPU_DM_PIPE_CRC_SOURCE_CRTC) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER); +} + +static bool dm_is_crc_source_dprx(enum amdgpu_dm_pipe_crc_source src) +{ + return (src == AMDGPU_DM_PIPE_CRC_SOURCE_DPRX) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER); +} + +static bool dm_need_crc_dither(enum amdgpu_dm_pipe_crc_source src) +{ + return (src == AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_NONE); +} + +const char *const *amdgpu_dm_crtc_get_crc_sources(struct drm_crtc *crtc, + size_t *count) +{ + *count = ARRAY_SIZE(pipe_crc_sources); + return pipe_crc_sources; +} + int amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, const char *src_name, size_t *values_cnt) @@ -63,14 +97,52 @@ amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, const char *src_name, return 0; } -int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) +int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc, + struct dm_crtc_state *dm_crtc_state, + enum amdgpu_dm_pipe_crc_source source) { struct amdgpu_device *adev = crtc->dev->dev_private; - struct dm_crtc_state *crtc_state = to_dm_crtc_state(crtc->state); - struct dc_stream_state *stream_state = crtc_state->stream; - bool enable; + struct dc_stream_state *stream_state = dm_crtc_state->stream; + bool enable = amdgpu_dm_is_valid_crc_source(source); + int ret = 0; + /* Configuration will be deferred to stream enable. */ + if (!stream_state) + return 0; + + mutex_lock(&adev->dm.dc_lock); + + /* Enable CRTC CRC generation if necessary. */ + if (dm_is_crc_source_crtc(source)) { + if (!dc_stream_configure_crc(stream_state->ctx->dc, + stream_state, enable, enable)) { + ret = -EINVAL; + goto unlock; + } + } + + /* Configure dithering */ + if (!dm_need_crc_dither(source)) + dc_stream_set_dither_option(stream_state, DITHER_OPTION_TRUN8); + else + dc_stream_set_dither_option(stream_state, + DITHER_OPTION_DEFAULT); + +unlock: + mutex_unlock(&adev->dm.dc_lock); + + return ret; +} + +int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) +{ enum amdgpu_dm_pipe_crc_source source = dm_parse_crc_source(src_name); + struct drm_crtc_commit *commit; + struct dm_crtc_state *crtc_state; + struct drm_dp_aux *aux = NULL; + bool enable = false; + bool enabled = false; + int ret = 0; if (source < 0) { DRM_DEBUG_DRIVER("Unknown CRC source %s for CRTC%d\n", @@ -78,41 +150,124 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) return -EINVAL; } - if (!stream_state) { - DRM_ERROR("No stream state for CRTC%d\n", crtc->index); - return -EINVAL; - } + ret = drm_modeset_lock(&crtc->mutex, NULL); + if (ret) + return ret; - enable = (source == AMDGPU_DM_PIPE_CRC_SOURCE_AUTO); + spin_lock(&crtc->commit_lock); + commit = list_first_entry_or_null(&crtc->commit_list, + struct drm_crtc_commit, commit_entry); + if (commit) + drm_crtc_commit_get(commit); + spin_unlock(&crtc->commit_lock); - mutex_lock(&adev->dm.dc_lock); - if (!dc_stream_configure_crc(stream_state->ctx->dc, stream_state, - enable, enable)) { - mutex_unlock(&adev->dm.dc_lock); - return -EINVAL; + if (commit) { + /* + * Need to wait for all outstanding programming to complete + * in commit tail since it can modify CRC related fields and + * hardware state. Since we're holding the CRTC lock we're + * guaranteed that no other commit work can be queued off + * before we modify the state below. + */ + ret = wait_for_completion_interruptible_timeout( + &commit->hw_done, 10 * HZ); + if (ret) + goto cleanup; } - /* When enabling CRC, we should also disable dithering. */ - dc_stream_set_dither_option(stream_state, - enable ? DITHER_OPTION_TRUN8 - : DITHER_OPTION_DEFAULT); + enable = amdgpu_dm_is_valid_crc_source(source); + crtc_state = to_dm_crtc_state(crtc->state); - mutex_unlock(&adev->dm.dc_lock); + /* + * USER REQ SRC | CURRENT SRC | BEHAVIOR + * ----------------------------- + * None | None | Do nothing + * None | CRTC | Disable CRTC CRC, set default to dither + * None | DPRX | Disable DPRX CRC, need 'aux', set default to dither + * None | CRTC DITHER | Disable CRTC CRC + * None | DPRX DITHER | Disable DPRX CRC, need 'aux' + * CRTC | XXXX | Enable CRTC CRC, no dither + * DPRX | XXXX | Enable DPRX CRC, need 'aux', no dither + * CRTC DITHER | XXXX | Enable CRTC CRC, set dither + * DPRX DITHER | XXXX | Enable DPRX CRC, need 'aux', set dither + */ + if (dm_is_crc_source_dprx(source) || + (source == AMDGPU_DM_PIPE_CRC_SOURCE_NONE && + dm_is_crc_source_dprx(crtc_state->crc_src))) { + struct amdgpu_dm_connector *aconn = NULL; + struct drm_connector *connector; + struct drm_connector_list_iter conn_iter; + + drm_connector_list_iter_begin(crtc->dev, &conn_iter); + drm_for_each_connector_iter(connector, &conn_iter) { + if (!connector->state || connector->state->crtc != crtc) + continue; + + aconn = to_amdgpu_dm_connector(connector); + break; + } + drm_connector_list_iter_end(&conn_iter); + + if (!aconn) { + DRM_DEBUG_DRIVER("No amd connector matching CRTC-%d\n", crtc->index); + ret = -EINVAL; + goto cleanup; + } + + aux = &aconn->dm_dp_aux.aux; + + if (!aux) { + DRM_DEBUG_DRIVER("No dp aux for amd connector\n"); + ret = -EINVAL; + goto cleanup; + } + } + + if (amdgpu_dm_crtc_configure_crc_source(crtc, crtc_state, source)) { + ret = -EINVAL; + goto cleanup; + } /* * Reading the CRC requires the vblank interrupt handler to be * enabled. Keep a reference until CRC capture stops. */ - if (!crtc_state->crc_enabled && enable) - drm_crtc_vblank_get(crtc); - else if (crtc_state->crc_enabled && !enable) + enabled = amdgpu_dm_is_valid_crc_source(crtc_state->crc_src); + if (!enabled && enable) { + ret = drm_crtc_vblank_get(crtc); + if (ret) + goto cleanup; + + if (dm_is_crc_source_dprx(source)) { + if (drm_dp_start_crc(aux, crtc)) { + DRM_DEBUG_DRIVER("dp start crc failed\n"); + ret = -EINVAL; + goto cleanup; + } + } + } else if (enabled && !enable) { drm_crtc_vblank_put(crtc); + if (dm_is_crc_source_dprx(source)) { + if (drm_dp_stop_crc(aux)) { + DRM_DEBUG_DRIVER("dp stop crc failed\n"); + ret = -EINVAL; + goto cleanup; + } + } + } - crtc_state->crc_enabled = enable; + crtc_state->crc_src = source; /* Reset crc_skipped on dm state */ crtc_state->crc_skip_count = 0; - return 0; + +cleanup: + if (commit) + drm_crtc_commit_put(commit); + + drm_modeset_unlock(&crtc->mutex); + + return ret; } /** @@ -135,7 +290,7 @@ void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc) stream_state = crtc_state->stream; /* Early return if CRC capture is not enabled. */ - if (!crtc_state->crc_enabled) + if (!amdgpu_dm_is_valid_crc_source(crtc_state->crc_src)) return; /* @@ -149,10 +304,12 @@ void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc) return; } - if (!dc_stream_get_crc(stream_state->ctx->dc, stream_state, - &crcs[0], &crcs[1], &crcs[2])) - return; + if (dm_is_crc_source_crtc(crtc_state->crc_src)) { + if (!dc_stream_get_crc(stream_state->ctx->dc, stream_state, + &crcs[0], &crcs[1], &crcs[2])) + return; - drm_crtc_add_crc_entry(crtc, true, - drm_crtc_accurate_vblank_count(crtc), crcs); + drm_crtc_add_crc_entry(crtc, true, + drm_crtc_accurate_vblank_count(crtc), crcs); + } } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h new file mode 100644 index 000000000000..f7d731797d3f --- /dev/null +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h @@ -0,0 +1,67 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ +#define AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ + +struct drm_crtc; +struct dm_crtc_state; + +enum amdgpu_dm_pipe_crc_source { + AMDGPU_DM_PIPE_CRC_SOURCE_NONE = 0, + AMDGPU_DM_PIPE_CRC_SOURCE_CRTC, + AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER, + AMDGPU_DM_PIPE_CRC_SOURCE_DPRX, + AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER, + AMDGPU_DM_PIPE_CRC_SOURCE_MAX, + AMDGPU_DM_PIPE_CRC_SOURCE_INVALID = -1, +}; + +static inline bool amdgpu_dm_is_valid_crc_source(enum amdgpu_dm_pipe_crc_source source) +{ + return (source > AMDGPU_DM_PIPE_CRC_SOURCE_NONE) && + (source < AMDGPU_DM_PIPE_CRC_SOURCE_MAX); +} + +/* amdgpu_dm_crc.c */ +#ifdef CONFIG_DEBUG_FS +int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc, + struct dm_crtc_state *dm_crtc_state, + enum amdgpu_dm_pipe_crc_source source); +int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name); +int amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, + const char *src_name, + size_t *values_cnt); +const char *const *amdgpu_dm_crtc_get_crc_sources(struct drm_crtc *crtc, + size_t *count); +void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc); +#else +#define amdgpu_dm_crtc_set_crc_source NULL +#define amdgpu_dm_crtc_verify_crc_source NULL +#define amdgpu_dm_crtc_get_crc_sources NULL +#define amdgpu_dm_crtc_handle_crc_irq(x) +#endif + +#endif /* AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 36a1d794b4af..f3dfb2887ae0 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -1053,9 +1053,33 @@ static int target_backlight_read(struct seq_file *m, void *data) return 0; } +static int mst_topo(struct seq_file *m, void *unused) +{ + struct drm_info_node *node = (struct drm_info_node *)m->private; + struct drm_device *dev = node->minor->dev; + struct drm_connector *connector; + struct drm_connector_list_iter conn_iter; + struct amdgpu_dm_connector *aconnector; + + drm_connector_list_iter_begin(dev, &conn_iter); + drm_for_each_connector_iter(connector, &conn_iter) { + if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort) + continue; + + aconnector = to_amdgpu_dm_connector(connector); + + seq_printf(m, "\nMST topology for connector %d\n", aconnector->connector_id); + drm_dp_mst_dump_topology(m, &aconnector->mst_mgr); + } + drm_connector_list_iter_end(&conn_iter); + + return 0; +} + static const struct drm_info_list amdgpu_dm_debugfs_list[] = { {"amdgpu_current_backlight_pwm", ¤t_backlight_read}, {"amdgpu_target_backlight_pwm", &target_backlight_read}, + {"amdgpu_mst_topology", &mst_topo}, }; /* diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index a0ed0154a9f0..ee1dc75f5ddc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -548,7 +548,9 @@ bool dm_helpers_dp_write_dsc_enable( bool enable ) { - return false; + uint8_t enable_dsc = enable ? 1 : 0; + + return dm_helpers_dp_write_dpcd(ctx, stream->sink->link, DP_DSC_ENABLE, &enable_dsc, 1); } #endif diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 6e205ee36ac3..16218a202b59 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -156,6 +156,26 @@ dm_dp_mst_connector_destroy(struct drm_connector *connector) kfree(amdgpu_dm_connector); } +static int +amdgpu_dm_mst_connector_late_register(struct drm_connector *connector) +{ + struct amdgpu_dm_connector *amdgpu_dm_connector = + to_amdgpu_dm_connector(connector); + struct drm_dp_mst_port *port = amdgpu_dm_connector->port; + + return drm_dp_mst_connector_late_register(connector, port); +} + +static void +amdgpu_dm_mst_connector_early_unregister(struct drm_connector *connector) +{ + struct amdgpu_dm_connector *amdgpu_dm_connector = + to_amdgpu_dm_connector(connector); + struct drm_dp_mst_port *port = amdgpu_dm_connector->port; + + drm_dp_mst_connector_early_unregister(connector, port); +} + static const struct drm_connector_funcs dm_dp_mst_connector_funcs = { .detect = dm_dp_mst_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -164,7 +184,9 @@ static const struct drm_connector_funcs dm_dp_mst_connector_funcs = { .atomic_duplicate_state = amdgpu_dm_connector_atomic_duplicate_state, .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, .atomic_set_property = amdgpu_dm_connector_atomic_set_property, - .atomic_get_property = amdgpu_dm_connector_atomic_get_property + .atomic_get_property = amdgpu_dm_connector_atomic_get_property, + .late_register = amdgpu_dm_mst_connector_late_register, + .early_unregister = amdgpu_dm_mst_connector_early_unregister, }; static int dm_dp_mst_get_modes(struct drm_connector *connector) @@ -388,7 +410,7 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, struct amdgpu_dm_connector *aconnector) { aconnector->dm_dp_aux.aux.name = "dmdc"; - aconnector->dm_dp_aux.aux.dev = dm->adev->dev; + aconnector->dm_dp_aux.aux.dev = aconnector->base.kdev; aconnector->dm_dp_aux.aux.transfer = dm_dp_aux_transfer; aconnector->dm_dp_aux.ddc_service = aconnector->dc_link->ddc; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c index 592fa499c9f8..f4cfa0caeba8 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c @@ -151,18 +151,31 @@ static void get_default_clock_levels( static enum smu_clk_type dc_to_smu_clock_type( enum dm_pp_clock_type dm_pp_clk_type) { -#define DCCLK_MAP_SMUCLK(dcclk, smuclk) \ - [dcclk] = smuclk + enum smu_clk_type smu_clk_type = SMU_CLK_COUNT; - static int dc_clk_type_map[] = { - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_DISPLAY_CLK, SMU_DISPCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_ENGINE_CLK, SMU_GFXCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_MEMORY_CLK, SMU_MCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_DCEFCLK, SMU_DCEFCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_SOCCLK, SMU_SOCCLK), - }; + switch (dm_pp_clk_type) { + case DM_PP_CLOCK_TYPE_DISPLAY_CLK: + smu_clk_type = SMU_DISPCLK; + break; + case DM_PP_CLOCK_TYPE_ENGINE_CLK: + smu_clk_type = SMU_GFXCLK; + break; + case DM_PP_CLOCK_TYPE_MEMORY_CLK: + smu_clk_type = SMU_MCLK; + break; + case DM_PP_CLOCK_TYPE_DCEFCLK: + smu_clk_type = SMU_DCEFCLK; + break; + case DM_PP_CLOCK_TYPE_SOCCLK: + smu_clk_type = SMU_SOCCLK; + break; + default: + DRM_ERROR("DM_PPLIB: invalid clock type: %d!\n", + dm_pp_clk_type); + break; + } - return dc_clk_type_map[dm_pp_clk_type]; + return smu_clk_type; } static enum amd_pp_clock_type dc_to_pp_clock_type( @@ -334,7 +347,7 @@ bool dm_pp_get_clock_levels_by_type( } } else if (adev->smu.funcs && adev->smu.funcs->get_clock_by_type) { if (smu_get_clock_by_type(&adev->smu, - dc_to_smu_clock_type(clk_type), + dc_to_pp_clock_type(clk_type), &pp_clks)) { get_default_clock_levels(clk_type, dc_clks); return true; @@ -419,7 +432,7 @@ bool dm_pp_get_clock_levels_by_type_with_latency( return false; } else if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->get_clock_by_type_with_latency) { if (smu_get_clock_by_type_with_latency(&adev->smu, - dc_to_pp_clock_type(clk_type), + dc_to_smu_clock_type(clk_type), &pp_clks)) return false; } @@ -801,6 +814,19 @@ enum pp_smu_status pp_nv_set_hard_min_uclk_by_freq(struct pp_smu *pp, int mhz) return PP_SMU_RESULT_OK; } +enum pp_smu_status pp_nv_set_pstate_handshake_support( + struct pp_smu *pp, BOOLEAN pstate_handshake_supported) +{ + const struct dc_context *ctx = pp->dm; + struct amdgpu_device *adev = ctx->driver_context; + struct smu_context *smu = &adev->smu; + + if (smu_display_disable_memory_clock_switch(smu, !pstate_handshake_supported)) + return PP_SMU_RESULT_FAIL; + + return PP_SMU_RESULT_OK; +} + enum pp_smu_status pp_nv_set_voltage_by_freq(struct pp_smu *pp, enum pp_smu_nv_clock_id clock_id, int mhz) { @@ -916,6 +942,7 @@ void dm_pp_get_funcs( funcs->nv_funcs.get_maximum_sustainable_clocks = pp_nv_get_maximum_sustainable_clocks; /*todo compare data with window driver */ funcs->nv_funcs.get_uclk_dpm_states = pp_nv_get_uclk_dpm_states; + funcs->nv_funcs.set_pstate_handshake_support = pp_nv_set_pstate_handshake_support; break; #endif default: diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile index 55ce5b657390..627982cb15d2 100644 --- a/drivers/gpu/drm/amd/display/dc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/Makefile @@ -37,6 +37,9 @@ endif ifdef CONFIG_DRM_AMD_DC_DCN1_0 DC_LIBS += dcn10 dml endif +ifdef CONFIG_DRM_AMD_DC_DCN2_1 +DC_LIBS += dcn21 +endif DC_LIBS += dce120 diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c index 461eef1de124..221e0f56389f 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c @@ -2796,8 +2796,6 @@ static const struct dc_vbios_funcs vbios_funcs = { .get_device_tag = bios_parser_get_device_tag, - .get_firmware_info = bios_parser_get_firmware_info, - .get_spread_spectrum_info = bios_parser_get_spread_spectrum_info, .get_ss_entry_number = bios_parser_get_ss_entry_number, @@ -2922,6 +2920,7 @@ static bool bios_parser_construct( dal_bios_parser_init_cmd_tbl_helper(&bp->cmd_helper, dce_version); bp->base.integrated_info = bios_parser_create_integrated_info(&bp->base); + bp->base.fw_info_valid = bios_parser_get_firmware_info(&bp->base, &bp->base.fw_info) == BP_RESULT_OK; return true; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 6aa2e56dfb67..dff65c0fe82f 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -1881,8 +1881,6 @@ static const struct dc_vbios_funcs vbios_funcs = { .get_device_tag = bios_parser_get_device_tag, - .get_firmware_info = bios_parser_get_firmware_info, - .get_spread_spectrum_info = bios_parser_get_spread_spectrum_info, .get_ss_entry_number = bios_parser_get_ss_entry_number, @@ -1998,6 +1996,7 @@ static bool bios_parser_construct( dal_bios_parser_init_cmd_tbl_helper2(&bp->cmd_helper, dce_version); bp->base.integrated_info = bios_parser_create_integrated_info(&bp->base); + bp->base.fw_info_valid = bios_parser_get_firmware_info(&bp->base, &bp->base.fw_info) == BP_RESULT_OK; return true; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c index f9439dfc7b75..db153ddf0fee 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c @@ -67,6 +67,11 @@ bool dal_bios_parser_init_cmd_tbl_helper2( *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + case DCN_VERSION_2_1: + *h = dal_cmd_tbl_helper_dce112_get_table2(); + return true; +#endif case DCE_VERSION_12_0: case DCE_VERSION_12_1: *h = dal_cmd_tbl_helper_dce112_get_table2(); diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c b/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c index ca24154468c7..11bf247bb180 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c +++ b/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c @@ -153,38 +153,10 @@ static uint8_t hpd_sel_to_atom(enum hpd_source_id id) static uint8_t dig_encoder_sel_to_atom(enum engine_id id) { - uint8_t atom_dig_encoder_sel = 0; - - switch (id) { - case ENGINE_ID_DIGA: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGA_SEL; - break; - case ENGINE_ID_DIGB: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGB_SEL; - break; - case ENGINE_ID_DIGC: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGC_SEL; - break; - case ENGINE_ID_DIGD: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGD_SEL; - break; - case ENGINE_ID_DIGE: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGE_SEL; - break; - case ENGINE_ID_DIGF: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGF_SEL; - break; - case ENGINE_ID_DIGG: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGG_SEL; - break; - case ENGINE_ID_UNKNOWN: - /* No DIG_FRONT is associated to DIG_BACKEND */ - atom_dig_encoder_sel = 0; - break; - default: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGA_SEL; - break; - } + /* On any ASIC after DCE80, we manually program the DIG_FE + * selection (see connect_dig_be_to_fe function of the link + * encoder), so translation should always return 0 (no FE). + */ return 0; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c index 0237ae575068..755b6e33140a 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c +++ b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c @@ -150,38 +150,10 @@ static uint8_t hpd_sel_to_atom(enum hpd_source_id id) static uint8_t dig_encoder_sel_to_atom(enum engine_id id) { - uint8_t atom_dig_encoder_sel = 0; - - switch (id) { - case ENGINE_ID_DIGA: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - case ENGINE_ID_DIGB: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGB_SEL; - break; - case ENGINE_ID_DIGC: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGC_SEL; - break; - case ENGINE_ID_DIGD: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGD_SEL; - break; - case ENGINE_ID_DIGE: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGE_SEL; - break; - case ENGINE_ID_DIGF: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGF_SEL; - break; - case ENGINE_ID_DIGG: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGG_SEL; - break; - case ENGINE_ID_UNKNOWN: - /* No DIG_FRONT is associated to DIG_BACKEND */ - atom_dig_encoder_sel = 0; - break; - default: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - } + /* On any ASIC after DCE80, we manually program the DIG_FE + * selection (see connect_dig_be_to_fe function of the link + * encoder), so translation should always return 0 (no FE). + */ return 0; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c index 452034f83e4c..06b4f7fa4a50 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c +++ b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c @@ -150,38 +150,10 @@ static uint8_t hpd_sel_to_atom(enum hpd_source_id id) static uint8_t dig_encoder_sel_to_atom(enum engine_id id) { - uint8_t atom_dig_encoder_sel = 0; - - switch (id) { - case ENGINE_ID_DIGA: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - case ENGINE_ID_DIGB: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGB_SEL; - break; - case ENGINE_ID_DIGC: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGC_SEL; - break; - case ENGINE_ID_DIGD: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGD_SEL; - break; - case ENGINE_ID_DIGE: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGE_SEL; - break; - case ENGINE_ID_DIGF: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGF_SEL; - break; - case ENGINE_ID_DIGG: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGG_SEL; - break; - case ENGINE_ID_UNKNOWN: - /* No DIG_FRONT is associated to DIG_BACKEND */ - atom_dig_encoder_sel = 0; - break; - default: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - } + /* On any ASIC after DCE80, we manually program the DIG_FE + * selection (see connect_dig_be_to_fe function of the link + * encoder), so translation should always return 0 (no FE). + */ return 0; } diff --git a/drivers/gpu/drm/amd/display/dc/calcs/Makefile b/drivers/gpu/drm/amd/display/dc/calcs/Makefile index 95f332ee3e7e..985633c08a26 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile +++ b/drivers/gpu/drm/amd/display/dc/calcs/Makefile @@ -32,9 +32,13 @@ endif calcs_ccflags := -mhard-float -msse $(cc_stack_align) -CFLAGS_dcn_calcs.o := $(calcs_ccflags) -CFLAGS_dcn_calc_auto.o := $(calcs_ccflags) -CFLAGS_dcn_calc_math.o := $(calcs_ccflags) -Wno-tautological-compare +ifdef CONFIG_CC_IS_CLANG +calcs_ccflags += -msse2 +endif + +CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_ccflags) +CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_ccflags) +CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_math.o := $(calcs_ccflags) -Wno-tautological-compare BW_CALCS = dce_calcs.o bw_fixed.o custom_float.o diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c index 7108d51a9c5b..a1d49256fab7 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c @@ -25,6 +25,7 @@ #include <linux/slab.h> +#include "resource.h" #include "dm_services.h" #include "dce_calcs.h" #include "dc.h" @@ -2852,7 +2853,7 @@ static void populate_initial_data( data->src_height[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.viewport.height); data->src_width[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.viewport.width); data->pitch_in_pixels[num_displays * 2 + j] = bw_int_to_fixed( - pipe[i].bottom_pipe->plane_state->plane_size.grph.surface_pitch); + pipe[i].bottom_pipe->plane_state->plane_size.surface_pitch); data->h_taps[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.taps.h_taps); data->v_taps[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.taps.v_taps); data->h_scale_ratio[num_displays * 2 + j] = fixed31_32_to_bw_fixed( @@ -2977,6 +2978,32 @@ static void populate_initial_data( data->number_of_displays = num_displays; } +static bool all_displays_in_sync(const struct pipe_ctx pipe[], + int pipe_count) +{ + const struct pipe_ctx *active_pipes[MAX_PIPES]; + int i, num_active_pipes = 0; + + for (i = 0; i < pipe_count; i++) { + if (!pipe[i].stream || pipe[i].top_pipe) + continue; + + active_pipes[num_active_pipes++] = &pipe[i]; + } + + if (!num_active_pipes) + return false; + + for (i = 1; i < num_active_pipes; ++i) { + if (!resource_are_streams_timing_synchronizable( + active_pipes[0]->stream, active_pipes[i]->stream)) { + return false; + } + } + + return true; +} + /** * Return: * true - Display(s) configuration supported. @@ -2998,8 +3025,10 @@ bool bw_calcs(struct dc_context *ctx, populate_initial_data(pipe, pipe_count, data); - /*TODO: this should be taken out calcs output and assigned during timing sync for pplib use*/ - calcs_output->all_displays_in_sync = false; + if (ctx->dc->config.multi_mon_pp_mclk_switch) + calcs_output->all_displays_in_sync = all_displays_in_sync(pipe, pipe_count); + else + calcs_output->all_displays_in_sync = false; if (data->number_of_displays != 0) { uint8_t yclk_lvl, sclk_lvl; diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c index 38365dd911a3..9b2cb57bf2ba 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -329,7 +329,7 @@ static void pipe_ctx_to_e2e_pipe_params ( dcc_support_pixel_format(pipe->plane_state->format, &bpe) ? 1 : 0; } input->src.dcc_rate = 1; - input->src.meta_pitch = pipe->plane_state->dcc.grph.meta_pitch; + input->src.meta_pitch = pipe->plane_state->dcc.meta_pitch; input->src.source_scan = dm_horz; input->src.sw_mode = pipe->plane_state->tiling_info.gfx9.swizzle; @@ -705,6 +705,17 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v, hack_force_pipe_split(v, context->streams[0]->timing.pix_clk_100hz); } + +unsigned int get_highest_allowed_voltage_level(uint32_t hw_internal_rev) +{ + /* for dali, the highest voltage level we want is 0 */ + if (ASICREV_IS_DALI(hw_internal_rev)) + return 0; + + /* we are ok with all levels */ + return 4; +} + bool dcn_validate_bandwidth( struct dc *dc, struct dc_state *context, @@ -732,6 +743,7 @@ bool dcn_validate_bandwidth( memset(v, 0, sizeof(*v)); kernel_fpu_begin(); + v->sr_exit_time = dc->dcn_soc->sr_exit_time; v->sr_enter_plus_exit_time = dc->dcn_soc->sr_enter_plus_exit_time; v->urgent_latency = dc->dcn_soc->urgent_latency; @@ -1268,7 +1280,7 @@ bool dcn_validate_bandwidth( PERFORMANCE_TRACE_END(); BW_VAL_TRACE_FINISH(); - if (bw_limit_pass && v->voltage_level != 5) + if (bw_limit_pass && v->voltage_level <= get_highest_allowed_voltage_level(dc->ctx->asic_id.hw_internal_rev)) return true; else return false; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile b/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile index 003c27767e9c..b864869cc7e3 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile @@ -85,3 +85,13 @@ AMD_DAL_CLK_MGR_DCN20 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn20/,$(CLK_MGR_DC AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN20) endif +ifdef CONFIG_DRM_AMD_DC_DCN2_1 +############################################################################### +# DCN21 +############################################################################### +CLK_MGR_DCN21 = rn_clk_mgr.o rn_clk_mgr_vbios_smu.o + +AMD_DAL_CLK_MGR_DCN21 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn21/,$(CLK_MGR_DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN21) +endif diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c index 6b8fc5cbabb8..c43797bea413 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c @@ -37,6 +37,9 @@ #include "dcn10/rv1_clk_mgr.h" #include "dcn10/rv2_clk_mgr.h" #include "dcn20/dcn20_clk_mgr.h" +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#include "dcn21/rn_clk_mgr.h" +#endif int clk_mgr_helper_get_active_display_cnt( @@ -108,6 +111,12 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p #if defined(CONFIG_DRM_AMD_DC_DCN1_0) case FAMILY_RV: +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + if (ASICREV_IS_RENOIR(asic_id.hw_internal_rev)) { + rn_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); + break; + } +#endif /* DCN2_1 */ if (ASICREV_IS_RAVEN2(asic_id.hw_internal_rev)) { rv2_clk_mgr_construct(ctx, clk_mgr, pp_smu); break; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c index 814450fefffa..c5c8c4901eed 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c @@ -273,18 +273,12 @@ static void dce_clock_read_integrated_info(struct clk_mgr_internal *clk_mgr_dce) { struct dc_debug_options *debug = &clk_mgr_dce->base.ctx->dc->debug; struct dc_bios *bp = clk_mgr_dce->base.ctx->dc_bios; - struct integrated_info info = { { { 0 } } }; - struct dc_firmware_info fw_info = { { 0 } }; int i; if (bp->integrated_info) - info = *bp->integrated_info; - - clk_mgr_dce->dentist_vco_freq_khz = info.dentist_vco_freq; + clk_mgr_dce->dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq; if (clk_mgr_dce->dentist_vco_freq_khz == 0) { - bp->funcs->get_firmware_info(bp, &fw_info); - clk_mgr_dce->dentist_vco_freq_khz = - fw_info.smu_gpu_pll_output_freq; + clk_mgr_dce->dentist_vco_freq_khz = bp->fw_info.smu_gpu_pll_output_freq; if (clk_mgr_dce->dentist_vco_freq_khz == 0) clk_mgr_dce->dentist_vco_freq_khz = 3600000; } @@ -317,9 +311,10 @@ static void dce_clock_read_integrated_info(struct clk_mgr_internal *clk_mgr_dce) /*Do not allow bad VBIOS/SBIOS to override with invalid values, * check for > 100MHz*/ - if (info.disp_clk_voltage[i].max_supported_clk >= 100000) - clk_mgr_dce->max_clks_by_state[clk_state].display_clk_khz = - info.disp_clk_voltage[i].max_supported_clk; + if (bp->integrated_info) + if (bp->integrated_info->disp_clk_voltage[i].max_supported_clk >= 100000) + clk_mgr_dce->max_clks_by_state[clk_state].display_clk_khz = + bp->integrated_info->disp_clk_voltage[i].max_supported_clk; } if (!debug->disable_dfs_bypass && bp->integrated_info) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c index 5cc3acccda2a..b1e657e137a9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c @@ -98,11 +98,14 @@ uint32_t dce110_get_min_vblank_time_us(const struct dc_state *context) struct dc_stream_state *stream = context->streams[j]; uint32_t vertical_blank_in_pixels = 0; uint32_t vertical_blank_time = 0; + uint32_t vertical_total_min = stream->timing.v_total; + struct dc_crtc_timing_adjust adjust = stream->adjust; + if (adjust.v_total_max != adjust.v_total_min) + vertical_total_min = adjust.v_total_min; vertical_blank_in_pixels = stream->timing.h_total * - (stream->timing.v_total + (vertical_total_min - stream->timing.v_addressable); - vertical_blank_time = vertical_blank_in_pixels * 10000 / stream->timing.pix_clk_100hz; @@ -171,6 +174,10 @@ void dce11_pplib_apply_display_requirements( struct dc_state *context) { struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg; + int memory_type_multiplier = MEMORY_TYPE_MULTIPLIER_CZ; + + if (dc->bw_vbios && dc->bw_vbios->memory_type == bw_def_hbm) + memory_type_multiplier = MEMORY_TYPE_HBM; pp_display_cfg->all_displays_in_sync = context->bw_ctx.bw.dce.all_displays_in_sync; @@ -183,8 +190,20 @@ void dce11_pplib_apply_display_requirements( pp_display_cfg->cpu_pstate_separation_time = context->bw_ctx.bw.dce.blackout_recovery_time_us; - pp_display_cfg->min_memory_clock_khz = context->bw_ctx.bw.dce.yclk_khz - / MEMORY_TYPE_MULTIPLIER_CZ; + /* + * TODO: determine whether the bandwidth has reached memory's limitation + * , then change minimum memory clock based on real-time bandwidth + * limitation. + */ + if (ASICREV_IS_VEGA20_P(dc->ctx->asic_id.hw_internal_rev) && (context->stream_count >= 2)) { + pp_display_cfg->min_memory_clock_khz = max(pp_display_cfg->min_memory_clock_khz, + (uint32_t) div64_s64( + div64_s64(dc->bw_vbios->high_yclk.value, + memory_type_multiplier), 10000)); + } else { + pp_display_cfg->min_memory_clock_khz = context->bw_ctx.bw.dce.yclk_khz + / memory_type_multiplier; + } pp_display_cfg->min_engine_clock_khz = determine_sclk_from_bounding_box( dc, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c index caf8a4a4e442..47f529ce280a 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c @@ -34,6 +34,11 @@ #include "rv1_clk_mgr_vbios_smu.h" #include "rv1_clk_mgr_clk.h" +void rv1_init_clocks(struct clk_mgr *clk_mgr) +{ + memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); +} + static int rv1_determine_dppclk_threshold(struct clk_mgr_internal *clk_mgr, struct dc_clocks *new_clocks) { bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz; @@ -232,6 +237,7 @@ static void rv1_enable_pme_wa(struct clk_mgr *clk_mgr_base) } static struct clk_mgr_funcs rv1_clk_funcs = { + .init_clocks = rv1_init_clocks, .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, .update_clocks = rv1_update_clocks, .enable_pme_wa = rv1_enable_pme_wa, @@ -246,7 +252,6 @@ void rv1_clk_mgr_construct(struct dc_context *ctx, struct clk_mgr_internal *clk_ { struct dc_debug_options *debug = &ctx->dc->debug; struct dc_bios *bp = ctx->dc_bios; - struct dc_firmware_info fw_info = { { 0 } }; clk_mgr->base.ctx = ctx; clk_mgr->pp_smu = pp_smu; @@ -262,9 +267,8 @@ void rv1_clk_mgr_construct(struct dc_context *ctx, struct clk_mgr_internal *clk_ if (bp->integrated_info) clk_mgr->dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq; - if (clk_mgr->dentist_vco_freq_khz == 0) { - bp->funcs->get_firmware_info(bp, &fw_info); - clk_mgr->dentist_vco_freq_khz = fw_info.smu_gpu_pll_output_freq; + if (bp->fw_info_valid && clk_mgr->dentist_vco_freq_khz == 0) { + clk_mgr->dentist_vco_freq_khz = bp->fw_info.smu_gpu_pll_output_freq; if (clk_mgr->dentist_vco_freq_khz == 0) clk_mgr->dentist_vco_freq_khz = 3600000; } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c index 50bfb5921de0..3e8ac303bd52 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c @@ -26,8 +26,6 @@ #include "dccg.h" #include "clk_mgr_internal.h" - -#include "dcn20/dcn20_clk_mgr.h" #include "dce100/dce_clk_mgr.h" #include "reg_helper.h" #include "core_types.h" @@ -106,7 +104,6 @@ void dcn20_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, { int i; - clk_mgr->dccg->ref_dppclk = clk_mgr->base.clks.dppclk_khz; for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) { int dpp_inst, dppclk_khz; @@ -116,28 +113,75 @@ void dcn20_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, dpp_inst = context->res_ctx.pipe_ctx[i].plane_res.dpp->inst; dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; clk_mgr->dccg->funcs->update_dpp_dto( - clk_mgr->dccg, dpp_inst, dppclk_khz); + clk_mgr->dccg, dpp_inst, dppclk_khz, false); } } -void dcn20_update_clocks_update_dentist(struct clk_mgr_internal *clk_mgr) +static void update_global_dpp_clk(struct clk_mgr_internal *clk_mgr, unsigned int khz) { int dpp_divider = DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->dentist_vco_freq_khz / clk_mgr->base.clks.dppclk_khz; - int disp_divider = DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->dentist_vco_freq_khz / clk_mgr->base.clks.dispclk_khz; + * clk_mgr->dentist_vco_freq_khz / khz; uint32_t dppclk_wdivider = dentist_get_did_from_divider(dpp_divider); - uint32_t dispclk_wdivider = dentist_get_did_from_divider(disp_divider); REG_UPDATE(DENTIST_DISPCLK_CNTL, - DENTIST_DISPCLK_WDIVIDER, dispclk_wdivider); -// REG_WAIT(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, 1, 5, 100); - REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DPPCLK_WDIVIDER, dppclk_wdivider); REG_WAIT(DENTIST_DISPCLK_CNTL, DENTIST_DPPCLK_CHG_DONE, 1, 5, 100); } +static void update_display_clk(struct clk_mgr_internal *clk_mgr, unsigned int khz) +{ + int disp_divider = DENTIST_DIVIDER_RANGE_SCALE_FACTOR + * clk_mgr->dentist_vco_freq_khz / khz; + + uint32_t dispclk_wdivider = dentist_get_did_from_divider(disp_divider); + + REG_UPDATE(DENTIST_DISPCLK_CNTL, + DENTIST_DISPCLK_WDIVIDER, dispclk_wdivider); +} + +static void request_voltage_and_program_disp_clk(struct clk_mgr *clk_mgr_base, unsigned int khz) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + struct dc *dc = clk_mgr_base->ctx->dc; + struct pp_smu_funcs_nv *pp_smu = NULL; + bool going_up = clk_mgr->base.clks.dispclk_khz < khz; + + if (dc->res_pool->pp_smu) + pp_smu = &dc->res_pool->pp_smu->nv_funcs; + + clk_mgr->base.clks.dispclk_khz = khz; + + if (going_up && pp_smu && pp_smu->set_voltage_by_freq) + pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_DISPCLK, clk_mgr_base->clks.dispclk_khz / 1000); + + update_display_clk(clk_mgr, khz); + + if (!going_up && pp_smu && pp_smu->set_voltage_by_freq) + pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_DISPCLK, clk_mgr_base->clks.dispclk_khz / 1000); +} + +static void request_voltage_and_program_global_dpp_clk(struct clk_mgr *clk_mgr_base, unsigned int khz) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + struct dc *dc = clk_mgr_base->ctx->dc; + struct pp_smu_funcs_nv *pp_smu = NULL; + bool going_up = clk_mgr->base.clks.dppclk_khz < khz; + + if (dc->res_pool->pp_smu) + pp_smu = &dc->res_pool->pp_smu->nv_funcs; + + clk_mgr->base.clks.dppclk_khz = khz; + clk_mgr->dccg->ref_dppclk = khz; + + if (going_up && pp_smu && pp_smu->set_voltage_by_freq) + pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_PIXELCLK, clk_mgr_base->clks.dppclk_khz / 1000); + + update_global_dpp_clk(clk_mgr, khz); + + if (!going_up && pp_smu && pp_smu->set_voltage_by_freq) + pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_PIXELCLK, clk_mgr_base->clks.dppclk_khz / 1000); +} void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, struct dc_state *context, @@ -148,12 +192,21 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, struct dc *dc = clk_mgr_base->ctx->dc; struct pp_smu_funcs_nv *pp_smu = NULL; int display_count; - bool update_dppclk = false; bool update_dispclk = false; bool enter_display_off = false; - bool dpp_clock_lowered = false; struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu; + bool force_reset = false; + int i; + if (dc->work_arounds.skip_clock_update) + return; + + if (clk_mgr_base->clks.dispclk_khz == 0 || + dc->debug.force_clock_mode & 0x1) { + //this is from resume or boot up, if forced_clock cfg option used, we bypass program dispclk and DPPCLK, but need set them for S3. + force_reset = true; + //force_clock_mode 0x1: force reset the clock even it is the same clock as long as it is in Passive level. + } display_count = clk_mgr_helper_get_active_display_cnt(dc, context); if (dc->res_pool->pp_smu) pp_smu = &dc->res_pool->pp_smu->nv_funcs; @@ -172,6 +225,7 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_PHYCLK, clk_mgr_base->clks.phyclk_khz / 1000); } + if (dc->debug.force_min_dcfclk_mhz > 0) new_clocks->dcfclk_khz = (new_clocks->dcfclk_khz > (dc->debug.force_min_dcfclk_mhz * 1000)) ? new_clocks->dcfclk_khz : (dc->debug.force_min_dcfclk_mhz * 1000); @@ -196,10 +250,13 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, } if (should_update_pstate_support(safe_to_lower, new_clocks->p_state_change_support, clk_mgr_base->clks.p_state_change_support)) { + clk_mgr_base->clks.prev_p_state_change_support = clk_mgr_base->clks.p_state_change_support; + clk_mgr_base->clks.p_state_change_support = new_clocks->p_state_change_support; if (pp_smu && pp_smu->set_pstate_handshake_support) pp_smu->set_pstate_handshake_support(&pp_smu->pp_smu, clk_mgr_base->clks.p_state_change_support); } + clk_mgr_base->clks.prev_p_state_change_support = clk_mgr_base->clks.p_state_change_support; if (should_set_clock(safe_to_lower, new_clocks->dramclk_khz, clk_mgr_base->clks.dramclk_khz)) { clk_mgr_base->clks.dramclk_khz = new_clocks->dramclk_khz; @@ -207,35 +264,48 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, pp_smu->set_hard_min_uclk_by_freq(&pp_smu->pp_smu, clk_mgr_base->clks.dramclk_khz / 1000); } - if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { - if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) - dpp_clock_lowered = true; - clk_mgr->base.clks.dppclk_khz = new_clocks->dppclk_khz; + if (dc->config.forced_clocks == false) { + // First update display clock + if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) + request_voltage_and_program_disp_clk(clk_mgr_base, new_clocks->dispclk_khz); - if (pp_smu && pp_smu->set_voltage_by_freq) - pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_PIXELCLK, clk_mgr_base->clks.dppclk_khz / 1000); + // Updating DPP clock requires some more logic + if (!safe_to_lower) { + // For pre-programming, we need to make sure any DPP clock that will go up has to go up - update_dppclk = true; - } + // First raise the global reference if needed + if (new_clocks->dppclk_khz > clk_mgr_base->clks.dppclk_khz) + request_voltage_and_program_global_dpp_clk(clk_mgr_base, new_clocks->dppclk_khz); - if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) { - clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz; - if (pp_smu && pp_smu->set_voltage_by_freq) - pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_DISPCLK, clk_mgr_base->clks.dispclk_khz / 1000); + // Then raise any dividers that need raising + for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) { + int dpp_inst, dppclk_khz; - update_dispclk = true; - } - if (dc->config.forced_clocks == false) { - if (dpp_clock_lowered) { - // if clock is being lowered, increase DTO before lowering refclk - dcn20_update_clocks_update_dpp_dto(clk_mgr, context); - dcn20_update_clocks_update_dentist(clk_mgr); + if (!context->res_ctx.pipe_ctx[i].plane_state) + continue; + + dpp_inst = context->res_ctx.pipe_ctx[i].plane_res.dpp->inst; + dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; + + clk_mgr->dccg->funcs->update_dpp_dto(clk_mgr->dccg, dpp_inst, dppclk_khz, true); + } } else { - // if clock is being raised, increase refclk before lowering DTO - if (update_dppclk || update_dispclk) - dcn20_update_clocks_update_dentist(clk_mgr); - if (update_dppclk) - dcn20_update_clocks_update_dpp_dto(clk_mgr, context); + // For post-programming, we can lower ref clk if needed, and unconditionally set all the DTOs + + if (new_clocks->dppclk_khz < clk_mgr_base->clks.dppclk_khz) + request_voltage_and_program_global_dpp_clk(clk_mgr_base, new_clocks->dppclk_khz); + + for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) { + int dpp_inst, dppclk_khz; + + if (!context->res_ctx.pipe_ctx[i].plane_state) + continue; + + dpp_inst = context->res_ctx.pipe_ctx[i].plane_res.dpp->inst; + dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; + + clk_mgr->dccg->funcs->update_dpp_dto(clk_mgr->dccg, dpp_inst, dppclk_khz, false); + } } } if (update_dispclk && @@ -303,6 +373,7 @@ void dcn2_init_clocks(struct clk_mgr *clk_mgr) memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); // Assumption is that boot state always supports pstate clk_mgr->clks.p_state_change_support = true; + clk_mgr->clks.prev_p_state_change_support = true; } void dcn2_enable_pme_wa(struct clk_mgr *clk_mgr_base) @@ -318,11 +389,32 @@ void dcn2_enable_pme_wa(struct clk_mgr *clk_mgr_base) } } +void dcn2_get_clock(struct clk_mgr *clk_mgr, + struct dc_state *context, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg) +{ + + if (clock_type == DC_CLOCK_TYPE_DISPCLK) { + clock_cfg->max_clock_khz = context->bw_ctx.bw.dcn.clk.max_supported_dispclk_khz; + clock_cfg->min_clock_khz = DCN_MINIMUM_DISPCLK_Khz; + clock_cfg->current_clock_khz = clk_mgr->clks.dispclk_khz; + clock_cfg->bw_requirequired_clock_khz = context->bw_ctx.bw.dcn.clk.bw_dispclk_khz; + } + if (clock_type == DC_CLOCK_TYPE_DPPCLK) { + clock_cfg->max_clock_khz = context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz; + clock_cfg->min_clock_khz = DCN_MINIMUM_DPPCLK_Khz; + clock_cfg->current_clock_khz = clk_mgr->clks.dppclk_khz; + clock_cfg->bw_requirequired_clock_khz = context->bw_ctx.bw.dcn.clk.bw_dppclk_khz; + } +} + static struct clk_mgr_funcs dcn2_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, .update_clocks = dcn2_update_clocks, .init_clocks = dcn2_init_clocks, - .enable_pme_wa = dcn2_enable_pme_wa + .enable_pme_wa = dcn2_enable_pme_wa, + .get_clock = dcn2_get_clock, }; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h index 5661a5a89847..ac31a9787305 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h @@ -45,4 +45,9 @@ void dcn20_clk_mgr_construct(struct dc_context *ctx, uint32_t dentist_get_did_from_divider(int divider); +void dcn2_get_clock(struct clk_mgr *clk_mgr, + struct dc_state *context, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg); + #endif //__DCN20_CLK_MGR_H__ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c new file mode 100644 index 000000000000..787f94d815f4 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -0,0 +1,590 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dccg.h" +#include "clk_mgr_internal.h" + + +#include "dcn20/dcn20_clk_mgr.h" +#include "rn_clk_mgr.h" + + +#include "dce100/dce_clk_mgr.h" +#include "rn_clk_mgr_vbios_smu.h" +#include "reg_helper.h" +#include "core_types.h" +#include "dm_helpers.h" + +#include "atomfirmware.h" +#include "clk/clk_10_0_2_offset.h" +#include "clk/clk_10_0_2_sh_mask.h" +#include "renoir_ip_offset.h" + + +/* Constants */ + +#define LPDDR_MEM_RETRAIN_LATENCY 4.977 /* Number obtained from LPDDR4 Training Counter Requirement doc */ + +/* Macros */ + +#define REG(reg_name) \ + (CLK_BASE.instance[0].segment[mm ## reg_name ## _BASE_IDX] + mm ## reg_name) + +void rn_update_clocks(struct clk_mgr *clk_mgr_base, + struct dc_state *context, + bool safe_to_lower) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk; + struct dc *dc = clk_mgr_base->ctx->dc; + int display_count; + bool update_dppclk = false; + bool update_dispclk = false; + bool enter_display_off = false; + bool dpp_clock_lowered = false; + struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu; + + display_count = clk_mgr_helper_get_active_display_cnt(dc, context); + + if (display_count == 0) + enter_display_off = true; + + if (enter_display_off == safe_to_lower) { + rn_vbios_smu_set_display_count(clk_mgr, display_count); + } + + if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr_base->clks.phyclk_khz)) { + clk_mgr_base->clks.phyclk_khz = new_clocks->phyclk_khz; + rn_vbios_smu_set_phyclk(clk_mgr, clk_mgr_base->clks.phyclk_khz); + } + + if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr_base->clks.dcfclk_khz)) { + clk_mgr_base->clks.dcfclk_khz = new_clocks->dcfclk_khz; + rn_vbios_smu_set_hard_min_dcfclk(clk_mgr, clk_mgr_base->clks.dcfclk_khz); + } + + if (should_set_clock(safe_to_lower, + new_clocks->dcfclk_deep_sleep_khz, clk_mgr_base->clks.dcfclk_deep_sleep_khz)) { + clk_mgr_base->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz; + rn_vbios_smu_set_min_deep_sleep_dcfclk(clk_mgr, clk_mgr_base->clks.dcfclk_deep_sleep_khz); + } + + if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { + if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) + dpp_clock_lowered = true; + clk_mgr_base->clks.dppclk_khz = new_clocks->dppclk_khz; + update_dppclk = true; + } + + if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) { + clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz; + rn_vbios_smu_set_dispclk(clk_mgr, clk_mgr_base->clks.dispclk_khz); + + update_dispclk = true; + } + + if (dpp_clock_lowered) { + // if clock is being lowered, increase DTO before lowering refclk + dcn20_update_clocks_update_dpp_dto(clk_mgr, context); + rn_vbios_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz); + } else { + // if clock is being raised, increase refclk before lowering DTO + if (update_dppclk || update_dispclk) + rn_vbios_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz); + if (update_dppclk) + dcn20_update_clocks_update_dpp_dto(clk_mgr, context); + } + + if (update_dispclk && + dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + /*update dmcu for wait_loop count*/ + dmcu->funcs->set_psr_wait_loop(dmcu, + clk_mgr_base->clks.dispclk_khz / 1000 / 7); + } +} + + +static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) +{ + /* get FbMult value */ + struct fixed31_32 pll_req; + unsigned int fbmult_frac_val = 0; + unsigned int fbmult_int_val = 0; + + + /* + * Register value of fbmult is in 8.16 format, we are converting to 31.32 + * to leverage the fix point operations available in driver + */ + + REG_GET(CLK1_CLK_PLL_REQ, FbMult_frac, &fbmult_frac_val); /* 16 bit fractional part*/ + REG_GET(CLK1_CLK_PLL_REQ, FbMult_int, &fbmult_int_val); /* 8 bit integer part */ + + pll_req = dc_fixpt_from_int(fbmult_int_val); + + /* + * since fractional part is only 16 bit in register definition but is 32 bit + * in our fix point definiton, need to shift left by 16 to obtain correct value + */ + pll_req.value |= fbmult_frac_val << 16; + + /* multiply by REFCLK period */ + pll_req = dc_fixpt_mul_int(pll_req, clk_mgr->dfs_ref_freq_khz); + + /* integer part is now VCO frequency in kHz */ + return dc_fixpt_floor(pll_req); +} + +static void rn_dump_clk_registers_internal(struct rn_clk_internal *internal, struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + internal->CLK1_CLK3_CURRENT_CNT = REG_READ(CLK1_CLK3_CURRENT_CNT); + internal->CLK1_CLK3_BYPASS_CNTL = REG_READ(CLK1_CLK3_BYPASS_CNTL); + + internal->CLK1_CLK3_DS_CNTL = REG_READ(CLK1_CLK3_DS_CNTL); //dcf deep sleep divider + internal->CLK1_CLK3_ALLOW_DS = REG_READ(CLK1_CLK3_ALLOW_DS); + + internal->CLK1_CLK1_CURRENT_CNT = REG_READ(CLK1_CLK1_CURRENT_CNT); + internal->CLK1_CLK1_BYPASS_CNTL = REG_READ(CLK1_CLK1_BYPASS_CNTL); + + internal->CLK1_CLK2_CURRENT_CNT = REG_READ(CLK1_CLK2_CURRENT_CNT); + internal->CLK1_CLK2_BYPASS_CNTL = REG_READ(CLK1_CLK2_BYPASS_CNTL); + + internal->CLK1_CLK0_CURRENT_CNT = REG_READ(CLK1_CLK0_CURRENT_CNT); + internal->CLK1_CLK0_BYPASS_CNTL = REG_READ(CLK1_CLK0_BYPASS_CNTL); +} + +/* This function collect raw clk register values */ +static void rn_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, + struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info) +{ + struct rn_clk_internal internal = {0}; + char *bypass_clks[5] = {"0x0 DFS", "0x1 REFCLK", "0x2 ERROR", "0x3 400 FCH", "0x4 600 FCH"}; + unsigned int chars_printed = 0; + unsigned int remaining_buffer = log_info->bufSize; + + rn_dump_clk_registers_internal(&internal, clk_mgr_base); + + regs_and_bypass->dcfclk = internal.CLK1_CLK3_CURRENT_CNT / 10; + regs_and_bypass->dcf_deep_sleep_divider = internal.CLK1_CLK3_DS_CNTL / 10; + regs_and_bypass->dcf_deep_sleep_allow = internal.CLK1_CLK3_ALLOW_DS; + regs_and_bypass->dprefclk = internal.CLK1_CLK2_CURRENT_CNT / 10; + regs_and_bypass->dispclk = internal.CLK1_CLK0_CURRENT_CNT / 10; + regs_and_bypass->dppclk = internal.CLK1_CLK1_CURRENT_CNT / 10; + + regs_and_bypass->dppclk_bypass = internal.CLK1_CLK1_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dppclk_bypass < 0 || regs_and_bypass->dppclk_bypass > 4) + regs_and_bypass->dppclk_bypass = 0; + regs_and_bypass->dcfclk_bypass = internal.CLK1_CLK3_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dcfclk_bypass < 0 || regs_and_bypass->dcfclk_bypass > 4) + regs_and_bypass->dcfclk_bypass = 0; + regs_and_bypass->dispclk_bypass = internal.CLK1_CLK0_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dispclk_bypass < 0 || regs_and_bypass->dispclk_bypass > 4) + regs_and_bypass->dispclk_bypass = 0; + regs_and_bypass->dprefclk_bypass = internal.CLK1_CLK2_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dprefclk_bypass < 0 || regs_and_bypass->dprefclk_bypass > 4) + regs_and_bypass->dprefclk_bypass = 0; + + if (log_info->enabled) { + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "clk_type,clk_value,deepsleep_cntl,deepsleep_allow,bypass\n"); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "dcfclk,%d,%d,%d,%s\n", + regs_and_bypass->dcfclk, + regs_and_bypass->dcf_deep_sleep_divider, + regs_and_bypass->dcf_deep_sleep_allow, + bypass_clks[(int) regs_and_bypass->dcfclk_bypass]); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "dprefclk,%d,N/A,N/A,%s\n", + regs_and_bypass->dprefclk, + bypass_clks[(int) regs_and_bypass->dprefclk_bypass]); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "dispclk,%d,N/A,N/A,%s\n", + regs_and_bypass->dispclk, + bypass_clks[(int) regs_and_bypass->dispclk_bypass]); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + //split + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "SPLIT\n"); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + // REGISTER VALUES + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "reg_name,value,clk_type\n"); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_CURRENT_CNT,%d,dcfclk\n", + internal.CLK1_CLK3_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_DS_CNTL,%d,dcf_deep_sleep_divider\n", + internal.CLK1_CLK3_DS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_ALLOW_DS,%d,dcf_deep_sleep_allow\n", + internal.CLK1_CLK3_ALLOW_DS); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK2_CURRENT_CNT,%d,dprefclk\n", + internal.CLK1_CLK2_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK0_CURRENT_CNT,%d,dispclk\n", + internal.CLK1_CLK0_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK1_CURRENT_CNT,%d,dppclk\n", + internal.CLK1_CLK1_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_BYPASS_CNTL,%d,dcfclk_bypass\n", + internal.CLK1_CLK3_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK2_BYPASS_CNTL,%d,dprefclk_bypass\n", + internal.CLK1_CLK2_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK0_BYPASS_CNTL,%d,dispclk_bypass\n", + internal.CLK1_CLK0_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK1_BYPASS_CNTL,%d,dppclk_bypass\n", + internal.CLK1_CLK1_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + } +} + +/* This function produce translated logical clk state values*/ +void rn_get_clk_states(struct clk_mgr *clk_mgr_base, struct clk_states *s) +{ + struct clk_state_registers_and_bypass sb = { 0 }; + struct clk_log_info log_info = { 0 }; + + rn_dump_clk_registers(&sb, clk_mgr_base, &log_info); + + s->dprefclk_khz = sb.dprefclk; +} + +void rn_enable_pme_wa(struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + rn_vbios_smu_enable_pme_wa(clk_mgr); +} + +static struct clk_mgr_funcs dcn21_funcs = { + .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, + .update_clocks = rn_update_clocks, + .init_clocks = dcn2_init_clocks, + .enable_pme_wa = rn_enable_pme_wa, + /* .dump_clk_registers = rn_dump_clk_registers */ +}; + +struct clk_bw_params rn_bw_params = { + .vram_type = Ddr4MemType, + .num_channels = 1, + .clk_table = { + .entries = { + { + .voltage = 0, + .dcfclk_mhz = 400, + .fclk_mhz = 400, + .memclk_mhz = 800, + .socclk_mhz = 0, + }, + { + .voltage = 0, + .dcfclk_mhz = 483, + .fclk_mhz = 800, + .memclk_mhz = 1600, + .socclk_mhz = 0, + }, + { + .voltage = 0, + .dcfclk_mhz = 602, + .fclk_mhz = 1067, + .memclk_mhz = 1067, + .socclk_mhz = 0, + }, + { + .voltage = 0, + .dcfclk_mhz = 738, + .fclk_mhz = 1333, + .memclk_mhz = 1600, + .socclk_mhz = 0, + }, + }, + + .num_entries = 4, + }, + + .wm_table = { + .entries = { + { + .wm_inst = WM_A, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .valid = true, + }, + { + .wm_inst = WM_B, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .valid = true, + }, + { + .wm_inst = WM_C, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .valid = true, + }, + { + .wm_inst = WM_D, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .valid = true, + }, + }, + } +}; + +void build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_range_sets *ranges) +{ + int i, num_valid_sets; + + num_valid_sets = 0; + + for (i = 0; i < WM_SET_COUNT; i++) { + /* skip empty entries, the smu array has no holes*/ + if (!bw_params->wm_table.entries[i].valid) + continue; + + ranges->reader_wm_sets[num_valid_sets].wm_inst = bw_params->wm_table.entries[i].wm_inst; + ranges->reader_wm_sets[num_valid_sets].wm_type = bw_params->wm_table.entries[i].wm_type;; + /* We will not select WM based on dcfclk, so leave it as unconstrained */ + ranges->reader_wm_sets[num_valid_sets].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[num_valid_sets].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + /* fclk wil be used to select WM*/ + + if (ranges->reader_wm_sets[num_valid_sets].wm_type == WM_TYPE_PSTATE_CHG) { + if (i == 0) + ranges->reader_wm_sets[num_valid_sets].min_fill_clk_mhz = 0; + else { + /* add 1 to make it non-overlapping with next lvl */ + ranges->reader_wm_sets[num_valid_sets].min_fill_clk_mhz = bw_params->clk_table.entries[i - 1].fclk_mhz + 1; + } + ranges->reader_wm_sets[num_valid_sets].max_fill_clk_mhz = bw_params->clk_table.entries[i].fclk_mhz; + + } else { + /* unconstrained for memory retraining */ + ranges->reader_wm_sets[num_valid_sets].min_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[num_valid_sets].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + + /* Modify previous watermark range to cover up to max */ + ranges->reader_wm_sets[num_valid_sets - 1].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + } + num_valid_sets++; + } + + ASSERT(num_valid_sets != 0); /* Must have at least one set of valid watermarks */ + ranges->num_reader_wm_sets = num_valid_sets; + + /* modify the min and max to make sure we cover the whole range*/ + ranges->reader_wm_sets[0].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[0].min_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[ranges->num_reader_wm_sets - 1].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + ranges->reader_wm_sets[ranges->num_reader_wm_sets - 1].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + + /* This is for writeback only, does not matter currently as no writeback support*/ + ranges->num_writer_wm_sets = 1; + ranges->writer_wm_sets[0].wm_inst = WM_A; + ranges->writer_wm_sets[0].min_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->writer_wm_sets[0].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + ranges->writer_wm_sets[0].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->writer_wm_sets[0].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + +} + +void clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params, struct dpm_clocks *clock_table, struct hw_asic_id *asic_id) +{ + int i; + + ASSERT(PP_SMU_NUM_FCLK_DPM_LEVELS <= MAX_NUM_DPM_LVL); + + for (i = 0; i < PP_SMU_NUM_FCLK_DPM_LEVELS; i++) { + if (clock_table->FClocks[i].Freq == 0) + break; + + bw_params->clk_table.entries[i].dcfclk_mhz = clock_table->DcfClocks[i].Freq; + bw_params->clk_table.entries[i].fclk_mhz = clock_table->FClocks[i].Freq; + bw_params->clk_table.entries[i].memclk_mhz = clock_table->MemClocks[i].Freq; + bw_params->clk_table.entries[i].socclk_mhz = clock_table->SocClocks[i].Freq; + bw_params->clk_table.entries[i].voltage = clock_table->FClocks[i].Vol; + } + bw_params->clk_table.num_entries = i; + + bw_params->vram_type = asic_id->vram_type; + bw_params->num_channels = asic_id->vram_width / DDR4_DRAM_WIDTH; + + for (i = 0; i < WM_SET_COUNT; i++) { + bw_params->wm_table.entries[i].wm_inst = i; + + if (clock_table->FClocks[i].Freq == 0) { + bw_params->wm_table.entries[i].valid = false; + continue; + } + + bw_params->wm_table.entries[i].wm_type = WM_TYPE_PSTATE_CHG; + bw_params->wm_table.entries[i].valid = true; + } + + if (bw_params->vram_type == LpDdr4MemType) { + /* + * WM set D will be re-purposed for memory retraining + */ + bw_params->wm_table.entries[WM_D].pstate_latency_us = LPDDR_MEM_RETRAIN_LATENCY; + bw_params->wm_table.entries[WM_D].wm_inst = WM_D; + bw_params->wm_table.entries[WM_D].wm_type = WM_TYPE_RETRAINING; + bw_params->wm_table.entries[WM_D].valid = true; + } + +} + +void rn_clk_mgr_construct( + struct dc_context *ctx, + struct clk_mgr_internal *clk_mgr, + struct pp_smu_funcs *pp_smu, + struct dccg *dccg) +{ + struct dc_debug_options *debug = &ctx->dc->debug; + struct dpm_clocks clock_table = { 0 }; + struct clk_state_registers_and_bypass s = { 0 }; + + clk_mgr->base.ctx = ctx; + clk_mgr->base.funcs = &dcn21_funcs; + + clk_mgr->pp_smu = pp_smu; + + clk_mgr->dccg = dccg; + clk_mgr->dfs_bypass_disp_clk = 0; + + clk_mgr->dprefclk_ss_percentage = 0; + clk_mgr->dprefclk_ss_divider = 1000; + clk_mgr->ss_on_dprefclk = false; + clk_mgr->dfs_ref_freq_khz = 48000; + + clk_mgr->smu_ver = rn_vbios_smu_get_smu_version(clk_mgr); + + if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { + dcn21_funcs.update_clocks = dcn2_update_clocks_fpga; + clk_mgr->dentist_vco_freq_khz = 3600000; + clk_mgr->base.dprefclk_khz = 600000; + } else { + struct clk_log_info log_info = {0}; + + /* TODO: Check we get what we expect during bringup */ + clk_mgr->dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr); + + /* in case we don't get a value from the register, use default */ + if (clk_mgr->dentist_vco_freq_khz == 0) + clk_mgr->dentist_vco_freq_khz = 3600000; + + rn_dump_clk_registers(&s, &clk_mgr->base, &log_info); + clk_mgr->base.dprefclk_khz = s.dprefclk; + + if (clk_mgr->base.dprefclk_khz != 600000) { + clk_mgr->base.dprefclk_khz = 600000; + ASSERT(1); //TODO: Renoir follow up. + } + + /* in case we don't get a value from the register, use default */ + if (clk_mgr->base.dprefclk_khz == 0) + clk_mgr->base.dprefclk_khz = 600000; + } + + dce_clock_read_ss_info(clk_mgr); + + clk_mgr->base.bw_params = &rn_bw_params; + + if (pp_smu) { + pp_smu->rn_funcs.get_dpm_clock_table(&pp_smu->rn_funcs.pp_smu, &clock_table); + clk_mgr_helper_populate_bw_params(clk_mgr->base.bw_params, &clock_table, &ctx->asic_id); + } + + /* + * Notify SMU which set of WM should be selected for different ranges of fclk + * On Renoir there is a maximumum of 4 DF pstates supported, could be less + * depending on DDR speed and fused maximum fclk. + */ + if (!debug->disable_pplib_wm_range) { + struct pp_smu_wm_range_sets ranges = {0}; + + build_watermark_ranges(clk_mgr->base.bw_params, &ranges); + + /* Notify PP Lib/SMU which Watermarks to use for which clock ranges */ + if (pp_smu && pp_smu->rn_funcs.set_wm_ranges) + pp_smu->rn_funcs.set_wm_ranges(&pp_smu->rn_funcs.pp_smu, &ranges); + } + + /* enable powerfeatures when displaycount goes to 0 */ + if (!debug->disable_48mhz_pwrdwn) + rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(clk_mgr); +} + diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h new file mode 100644 index 000000000000..aadec06fde10 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h @@ -0,0 +1,39 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __RN_CLK_MGR_H__ +#define __RN_CLK_MGR_H__ + +struct rn_clk_registers { + uint32_t CLK1_CLK0_CURRENT_CNT; /* DPREFCLK */ +}; + + +void rn_clk_mgr_construct(struct dc_context *ctx, + struct clk_mgr_internal *clk_mgr, + struct pp_smu_funcs *pp_smu, + struct dccg *dccg); + +#endif //__RN_CLK_MGR_H__ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c new file mode 100644 index 000000000000..50984c1811bb --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c @@ -0,0 +1,200 @@ +/* + * Copyright 2012-16 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "core_types.h" +#include "clk_mgr_internal.h" +#include "reg_helper.h" + +#include "renoir_ip_offset.h" + +#include "mp/mp_12_0_0_offset.h" +#include "mp/mp_12_0_0_sh_mask.h" + +#define REG(reg_name) \ + (MP1_BASE.instance[0].segment[mm ## reg_name ## _BASE_IDX] + mm ## reg_name) + +#define FN(reg_name, field) \ + FD(reg_name##__##field) + +#define VBIOSSMC_MSG_TestMessage 0x1 +#define VBIOSSMC_MSG_GetSmuVersion 0x2 +#define VBIOSSMC_MSG_PowerUpGfx 0x3 +#define VBIOSSMC_MSG_SetDispclkFreq 0x4 +#define VBIOSSMC_MSG_SetDprefclkFreq 0x5 +#define VBIOSSMC_MSG_PowerDownGfx 0x6 +#define VBIOSSMC_MSG_SetDppclkFreq 0x7 +#define VBIOSSMC_MSG_SetHardMinDcfclkByFreq 0x8 +#define VBIOSSMC_MSG_SetMinDeepSleepDcfclk 0x9 +#define VBIOSSMC_MSG_SetPhyclkVoltageByFreq 0xA +#define VBIOSSMC_MSG_GetFclkFrequency 0xB +#define VBIOSSMC_MSG_SetDisplayCount 0xC +#define VBIOSSMC_MSG_EnableTmdp48MHzRefclkPwrDown 0xD +#define VBIOSSMC_MSG_UpdatePmeRestore 0xE + +int rn_vbios_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, unsigned int msg_id, unsigned int param) +{ + /* First clear response register */ + REG_WRITE(MP1_SMN_C2PMSG_91, 0); + + /* Set the parameter register for the SMU message, unit is Mhz */ + REG_WRITE(MP1_SMN_C2PMSG_83, param); + + /* Trigger the message transaction by writing the message ID */ + REG_WRITE(MP1_SMN_C2PMSG_67, msg_id); + + REG_WAIT(MP1_SMN_C2PMSG_91, CONTENT, 1, 10, 200000); + + /* Actual dispclk set is returned in the parameter register */ + return REG_READ(MP1_SMN_C2PMSG_83); +} + +int rn_vbios_smu_get_smu_version(struct clk_mgr_internal *clk_mgr) +{ + return rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_GetSmuVersion, + 0); +} + + +int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dispclk_khz) +{ + int actual_dispclk_set_mhz = -1; + struct dc *core_dc = clk_mgr->base.ctx->dc; + struct dmcu *dmcu = core_dc->res_pool->dmcu; + uint32_t clk = requested_dispclk_khz / 1000; + + if (clk <= 100) + clk = 101; + + /* Unit of SMU msg parameter is Mhz */ + actual_dispclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDispclkFreq, + clk); + + if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) + dmcu->funcs->set_psr_wait_loop(dmcu, + actual_dispclk_set_mhz / 7); + } + } + + return actual_dispclk_set_mhz * 1000; +} + +int rn_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr) +{ + int actual_dprefclk_set_mhz = -1; + + actual_dprefclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDprefclkFreq, + clk_mgr->base.dprefclk_khz / 1000); + + /* TODO: add code for programing DP DTO, currently this is down by command table */ + + return actual_dprefclk_set_mhz * 1000; +} + +int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_dcfclk_khz) +{ + int actual_dcfclk_set_mhz = -1; + + if (clk_mgr->smu_ver < 0xFFFFFFFF) + return actual_dcfclk_set_mhz; + + actual_dcfclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetHardMinDcfclkByFreq, + requested_dcfclk_khz / 1000); + + return actual_dcfclk_set_mhz * 1000; +} + +int rn_vbios_smu_set_min_deep_sleep_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_min_ds_dcfclk_khz) +{ + int actual_min_ds_dcfclk_mhz = -1; + + if (clk_mgr->smu_ver < 0xFFFFFFFF) + return actual_min_ds_dcfclk_mhz; + + actual_min_ds_dcfclk_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetMinDeepSleepDcfclk, + requested_min_ds_dcfclk_khz / 1000); + + return actual_min_ds_dcfclk_mhz * 1000; +} + +void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz) +{ + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetPhyclkVoltageByFreq, + requested_phyclk_khz / 1000); +} + +int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_khz) +{ + int actual_dppclk_set_mhz = -1; + + uint32_t clk = requested_dpp_khz / 1000; + + if (clk <= 100) + clk = 101; + + actual_dppclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDppclkFreq, + clk); + + return actual_dppclk_set_mhz * 1000; +} + +void rn_vbios_smu_set_display_count(struct clk_mgr_internal *clk_mgr, int display_count) +{ + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDisplayCount, + display_count); +} + +void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr) +{ + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_EnableTmdp48MHzRefclkPwrDown, + 0); +} + +void rn_vbios_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr) +{ + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_UpdatePmeRestore, + 0); +} diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h new file mode 100644 index 000000000000..da3a49487c6d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h @@ -0,0 +1,40 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef DAL_DC_RN_CLK_MGR_VBIOS_SMU_H_ +#define DAL_DC_RN_CLK_MGR_VBIOS_SMU_H_ + +int rn_vbios_smu_get_smu_version(struct clk_mgr_internal *clk_mgr); +int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dispclk_khz); +int rn_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr); +int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_dcfclk_khz); +int rn_vbios_smu_set_min_deep_sleep_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_min_ds_dcfclk_khz); +void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz); +int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_khz); +void rn_vbios_smu_set_display_count(struct clk_mgr_internal *clk_mgr, int display_count); +void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr); +void rn_vbios_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr); + +#endif /* DAL_DC_DCN10_RV1_CLK_MGR_VBIOS_SMU_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index fa20201eef3a..5d1adeda4d90 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -23,6 +23,7 @@ */ #include <linux/slab.h> +#include <linux/mm.h> #include "dm_services.h" @@ -180,13 +181,25 @@ static bool create_links( link = link_create(&link_init_params); if (link) { - if (dc->config.edp_not_connected && - link->connector_signal == SIGNAL_TYPE_EDP) { - link_destroy(&link); - } else { + bool should_destory_link = false; + + if (link->connector_signal == SIGNAL_TYPE_EDP) { + if (dc->config.edp_not_connected) + should_destory_link = true; + else if (dc->debug.remove_disconnect_edp) { + enum dc_connection_type type; + dc_link_detect_sink(link, &type); + if (type == dc_connection_none) + should_destory_link = true; + } + } + + if (!should_destory_link) { dc->links[dc->link_count] = link; link->dc = dc; ++dc->link_count; + } else { + link_destroy(&link); } } } @@ -278,7 +291,9 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc, dc->hwss.set_drr(&pipe, 1, adjust->v_total_min, - adjust->v_total_max); + adjust->v_total_max, + adjust->v_total_mid, + adjust->v_total_mid_frame_num); ret = true; } @@ -674,6 +689,11 @@ static bool construct(struct dc *dc, if (!dc->clk_mgr) goto fail; +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 + if (dc->res_pool->funcs->update_bw_bounding_box) + dc->res_pool->funcs->update_bw_bounding_box(dc, dc->clk_mgr->bw_params); +#endif + /* Creation of current_state must occur after dc->dml * is initialized in dc_create_resource_pool because * on creation it copies the contents of dc->dml @@ -947,7 +967,7 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc, { struct timing_generator *tg; struct dc_link *link = sink->link; - unsigned int inst; + unsigned int enc_inst, tg_inst; /* Check for enabled DIG to identify enabled display */ if (!link->link_enc->funcs->is_dig_enabled(link->link_enc)) @@ -959,13 +979,22 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc, * current implementation always map 1-to-1, so this code makes * the same assumption and doesn't check OTG source. */ - inst = link->link_enc->funcs->get_dig_frontend(link->link_enc) - 1; + enc_inst = link->link_enc->funcs->get_dig_frontend(link->link_enc); /* Instance should be within the range of the pool */ - if (inst >= dc->res_pool->pipe_count) + if (enc_inst >= dc->res_pool->pipe_count) + return false; + + if (enc_inst >= dc->res_pool->stream_enc_count) + return false; + + tg_inst = dc->res_pool->stream_enc[enc_inst]->funcs->dig_source_otg( + dc->res_pool->stream_enc[enc_inst]); + + if (tg_inst >= dc->res_pool->timing_generator_count) return false; - tg = dc->res_pool->timing_generators[inst]; + tg = dc->res_pool->timing_generators[tg_inst]; if (!tg->funcs->is_matching_timing) return false; @@ -978,10 +1007,11 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc, dc->res_pool->dp_clock_source->funcs->get_pixel_clk_frequency_100hz( dc->res_pool->dp_clock_source, - inst, &pix_clk_100hz); + tg_inst, &pix_clk_100hz); if (crtc_timing->pix_clk_100hz != pix_clk_100hz) return false; + } return true; @@ -1064,7 +1094,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c if (result != DC_OK) return result; - if (context->stream_count > 1) { + if (context->stream_count > 1 && !dc->debug.disable_timing_sync) { enable_timing_multisync(dc, context); program_timing_sync(dc, context); } @@ -1171,8 +1201,8 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc) struct dc_state *dc_create_state(struct dc *dc) { - struct dc_state *context = kzalloc(sizeof(struct dc_state), - GFP_KERNEL); + struct dc_state *context = kvzalloc(sizeof(struct dc_state), + GFP_KERNEL); if (!context) return NULL; @@ -1192,11 +1222,11 @@ struct dc_state *dc_create_state(struct dc *dc) struct dc_state *dc_copy_state(struct dc_state *src_ctx) { int i, j; - struct dc_state *new_ctx = kmemdup(src_ctx, - sizeof(struct dc_state), GFP_KERNEL); + struct dc_state *new_ctx = kvmalloc(sizeof(struct dc_state), GFP_KERNEL); if (!new_ctx) return NULL; + memcpy(new_ctx, src_ctx, sizeof(struct dc_state)); for (i = 0; i < MAX_PIPES; i++) { struct pipe_ctx *cur_pipe = &new_ctx->res_ctx.pipe_ctx[i]; @@ -1207,6 +1237,12 @@ struct dc_state *dc_copy_state(struct dc_state *src_ctx) if (cur_pipe->bottom_pipe) cur_pipe->bottom_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx]; + if (cur_pipe->prev_odm_pipe) + cur_pipe->prev_odm_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->prev_odm_pipe->pipe_idx]; + + if (cur_pipe->next_odm_pipe) + cur_pipe->next_odm_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->next_odm_pipe->pipe_idx]; + } for (i = 0; i < new_ctx->stream_count; i++) { @@ -1230,7 +1266,7 @@ static void dc_state_free(struct kref *kref) { struct dc_state *context = container_of(kref, struct dc_state, refcount); dc_resource_state_destruct(context); - kfree(context); + kvfree(context); } void dc_release_state(struct dc_state *context) @@ -1238,6 +1274,55 @@ void dc_release_state(struct dc_state *context) kref_put(&context->refcount, dc_state_free); } +bool dc_set_generic_gpio_for_stereo(bool enable, + struct gpio_service *gpio_service) +{ + enum gpio_result gpio_result = GPIO_RESULT_NON_SPECIFIC_ERROR; + struct gpio_pin_info pin_info; + struct gpio *generic; + struct gpio_generic_mux_config *config = kzalloc(sizeof(struct gpio_generic_mux_config), + GFP_KERNEL); + + if (!config) + return false; + pin_info = dal_gpio_get_generic_pin_info(gpio_service, GPIO_ID_GENERIC, 0); + + if (pin_info.mask == 0xFFFFFFFF || pin_info.offset == 0xFFFFFFFF) { + kfree(config); + return false; + } else { + generic = dal_gpio_service_create_generic_mux( + gpio_service, + pin_info.offset, + pin_info.mask); + } + + if (!generic) { + kfree(config); + return false; + } + + gpio_result = dal_gpio_open(generic, GPIO_MODE_OUTPUT); + + config->enable_output_from_mux = enable; + config->mux_select = GPIO_SIGNAL_SOURCE_PASS_THROUGH_STEREO_SYNC; + + if (gpio_result == GPIO_RESULT_OK) + gpio_result = dal_mux_setup_config(generic, config); + + if (gpio_result == GPIO_RESULT_OK) { + dal_gpio_close(generic); + dal_gpio_destroy_generic_mux(&generic); + kfree(config); + return true; + } else { + dal_gpio_close(generic); + dal_gpio_destroy_generic_mux(&generic); + kfree(config); + return false; + } +} + static bool is_surface_in_context( const struct dc_state *context, const struct dc_plane_state *plane_state) @@ -1304,8 +1389,8 @@ static enum surface_update_type get_plane_info_update_type(const struct dc_surfa } if (u->plane_info->dcc.enable != u->surface->dcc.enable - || u->plane_info->dcc.grph.independent_64b_blks != u->surface->dcc.grph.independent_64b_blks - || u->plane_info->dcc.grph.meta_pitch != u->surface->dcc.grph.meta_pitch) { + || u->plane_info->dcc.independent_64b_blks != u->surface->dcc.independent_64b_blks + || u->plane_info->dcc.meta_pitch != u->surface->dcc.meta_pitch) { update_flags->bits.dcc_change = 1; elevate_update_type(&update_type, UPDATE_TYPE_MED); } @@ -1319,9 +1404,9 @@ static enum surface_update_type get_plane_info_update_type(const struct dc_surfa elevate_update_type(&update_type, UPDATE_TYPE_FULL); } - if (u->plane_info->plane_size.grph.surface_pitch != u->surface->plane_size.grph.surface_pitch - || u->plane_info->plane_size.video.luma_pitch != u->surface->plane_size.video.luma_pitch - || u->plane_info->plane_size.video.chroma_pitch != u->surface->plane_size.video.chroma_pitch) { + if (u->plane_info->plane_size.surface_pitch != u->surface->plane_size.surface_pitch + || u->plane_info->plane_size.surface_pitch != u->surface->plane_size.surface_pitch + || u->plane_info->plane_size.chroma_pitch != u->surface->plane_size.chroma_pitch) { update_flags->bits.plane_size_change = 1; elevate_update_type(&update_type, UPDATE_TYPE_MED); } @@ -1541,6 +1626,9 @@ enum surface_update_type dc_check_update_surfaces_for_stream( for (i = 0; i < surface_count; i++) updates[i].surface->update_flags.raw = 0xFFFFFFFF; + if (type == UPDATE_TYPE_FAST && memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0) + dc->optimized_required = true; + return type; } @@ -1617,6 +1705,8 @@ static void copy_surface_update_to_plane( srf_update->plane_info->dcc; surface->sdr_white_level = srf_update->plane_info->sdr_white_level; + surface->layer_index = + srf_update->plane_info->layer_index; } if (srf_update->gamma && @@ -1783,9 +1873,7 @@ static void commit_planes_do_stream_update(struct dc *dc, for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - if (!pipe_ctx->top_pipe && - pipe_ctx->stream && - pipe_ctx->stream == stream) { + if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe && pipe_ctx->stream == stream) { if (stream_update->periodic_interrupt0 && dc->hwss.setup_periodic_interrupt) @@ -1811,7 +1899,7 @@ static void commit_planes_do_stream_update(struct dc *dc, if (stream_update->dither_option) { #if defined(CONFIG_DRM_AMD_DC_DCN2_0) - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; #endif resource_build_bit_depth_reduction_params(pipe_ctx->stream, &pipe_ctx->stream->bit_depth_params); @@ -1819,10 +1907,12 @@ static void commit_planes_do_stream_update(struct dc *dc, &stream->bit_depth_params, &stream->clamping); #if defined(CONFIG_DRM_AMD_DC_DCN2_0) - if (odm_pipe) + while (odm_pipe) { odm_pipe->stream_res.opp->funcs->opp_program_fmt(odm_pipe->stream_res.opp, &stream->bit_depth_params, &stream->clamping); + odm_pipe = odm_pipe->next_odm_pipe; + } #endif } @@ -1839,13 +1929,21 @@ static void commit_planes_do_stream_update(struct dc *dc, if (stream_update->dpms_off) { dc->hwss.pipe_control_lock(dc, pipe_ctx, true); + if (*stream_update->dpms_off) { - core_link_disable_stream(pipe_ctx, KEEP_ACQUIRED_RESOURCE); + core_link_disable_stream(pipe_ctx); + /* for dpms, keep acquired resources*/ + if (pipe_ctx->stream_res.audio && !dc->debug.az_endpoint_mute_only) + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + dc->hwss.optimize_bandwidth(dc, dc->current_state); } else { - dc->hwss.prepare_bandwidth(dc, dc->current_state); + if (!dc->optimize_seamless_boot) + dc->hwss.prepare_bandwidth(dc, dc->current_state); + core_link_enable_stream(dc->current_state, pipe_ctx); } + dc->hwss.pipe_control_lock(dc, pipe_ctx, false); } @@ -1935,6 +2033,7 @@ static void commit_planes_for_stream(struct dc *dc, struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; if (!pipe_ctx->top_pipe && + !pipe_ctx->prev_odm_pipe && pipe_ctx->stream && pipe_ctx->stream == stream) { struct dc_stream_status *stream_status = NULL; @@ -2049,7 +2148,7 @@ void dc_commit_updates_for_stream(struct dc *dc, enum surface_update_type update_type; struct dc_state *context; struct dc_context *dc_ctx = dc->ctx; - int i, j; + int i; stream_status = dc_stream_get_status(stream); context = dc->current_state; @@ -2087,16 +2186,6 @@ void dc_commit_updates_for_stream(struct dc *dc, copy_surface_update_to_plane(surface, &srf_updates[i]); - if (update_type >= UPDATE_TYPE_MED) { - for (j = 0; j < dc->res_pool->pipe_count; j++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - - if (pipe_ctx->plane_state != surface) - continue; - - resource_build_scaling_params(pipe_ctx); - } - } } copy_stream_update_to_stream(dc, context, stream, stream_update); @@ -2186,6 +2275,14 @@ void dc_set_power_state( dc_resource_state_construct(dc, dc->current_state); dc->hwss.init_hw(dc); + +#ifdef CONFIG_DRM_AMD_DC_DCN2_0 + if (dc->hwss.init_sys_ctx != NULL && + dc->vm_pa_config.valid) { + dc->hwss.init_sys_ctx(dc->hwseq, dc, &dc->vm_pa_config); + } +#endif + break; default: ASSERT(dc->current_state->stream_count == 0); @@ -2386,3 +2483,14 @@ void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx info->fClock = (unsigned int)state->bw_ctx.bw.dcn.clk.fclk_khz; info->phyClock = (unsigned int)state->bw_ctx.bw.dcn.clk.phyclk_khz; } +enum dc_status dc_set_clock(struct dc *dc, enum dc_clock_type clock_type, uint32_t clk_khz, uint32_t stepping) +{ + if (dc->hwss.set_clock) + return dc->hwss.set_clock(dc, clock_type, clk_khz, stepping); + return DC_ERROR_UNEXPECTED; +} +void dc_get_clock(struct dc *dc, enum dc_clock_type clock_type, struct dc_clock_config *clock_cfg) +{ + if (dc->hwss.get_clock) + dc->hwss.get_clock(dc, clock_type, clock_cfg); +} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c index 5903e7822f98..b9227d5de3a3 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c @@ -115,16 +115,16 @@ void pre_surface_trace( plane_state->clip_rect.height); SURFACE_TRACE( - "plane_state->plane_size.grph.surface_size.x = %d;\n" - "plane_state->plane_size.grph.surface_size.y = %d;\n" - "plane_state->plane_size.grph.surface_size.width = %d;\n" - "plane_state->plane_size.grph.surface_size.height = %d;\n" - "plane_state->plane_size.grph.surface_pitch = %d;\n", - plane_state->plane_size.grph.surface_size.x, - plane_state->plane_size.grph.surface_size.y, - plane_state->plane_size.grph.surface_size.width, - plane_state->plane_size.grph.surface_size.height, - plane_state->plane_size.grph.surface_pitch); + "plane_state->plane_size.surface_size.x = %d;\n" + "plane_state->plane_size.surface_size.y = %d;\n" + "plane_state->plane_size.surface_size.width = %d;\n" + "plane_state->plane_size.surface_size.height = %d;\n" + "plane_state->plane_size.surface_pitch = %d;\n", + plane_state->plane_size.surface_size.x, + plane_state->plane_size.surface_size.y, + plane_state->plane_size.surface_size.width, + plane_state->plane_size.surface_size.height, + plane_state->plane_size.surface_pitch); SURFACE_TRACE( @@ -202,20 +202,20 @@ void update_surface_trace( SURFACE_TRACE( "plane_info->color_space = %d;\n" "plane_info->format = %d;\n" - "plane_info->plane_size.grph.surface_pitch = %d;\n" - "plane_info->plane_size.grph.surface_size.height = %d;\n" - "plane_info->plane_size.grph.surface_size.width = %d;\n" - "plane_info->plane_size.grph.surface_size.x = %d;\n" - "plane_info->plane_size.grph.surface_size.y = %d;\n" + "plane_info->plane_size.surface_pitch = %d;\n" + "plane_info->plane_size.surface_size.height = %d;\n" + "plane_info->plane_size.surface_size.width = %d;\n" + "plane_info->plane_size.surface_size.x = %d;\n" + "plane_info->plane_size.surface_size.y = %d;\n" "plane_info->rotation = %d;\n" "plane_info->stereo_format = %d;\n", update->plane_info->color_space, update->plane_info->format, - update->plane_info->plane_size.grph.surface_pitch, - update->plane_info->plane_size.grph.surface_size.height, - update->plane_info->plane_size.grph.surface_size.width, - update->plane_info->plane_size.grph.surface_size.x, - update->plane_info->plane_size.grph.surface_size.y, + update->plane_info->plane_size.surface_pitch, + update->plane_info->plane_size.surface_size.height, + update->plane_info->plane_size.surface_size.width, + update->plane_info->plane_size.surface_size.x, + update->plane_info->plane_size.surface_size.y, update->plane_info->rotation, update->plane_info->stereo_format); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 355b4ba12796..ca20b150afcc 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -45,10 +45,6 @@ #include "dpcd_defs.h" #include "dmcu.h" #include "hw/clk_mgr.h" -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) -#include "resource.h" -#endif -#include "hw/clk_mgr.h" #define DC_LOGGER_INIT(logger) @@ -684,6 +680,56 @@ static bool is_same_edid(struct dc_edid *old_edid, struct dc_edid *new_edid) return (memcmp(old_edid->raw_edid, new_edid->raw_edid, new_edid->length) == 0); } +bool wait_for_alt_mode(struct dc_link *link) +{ + + /** + * something is terribly wrong if time out is > 200ms. (5Hz) + * 500 microseconds * 400 tries us 200 ms + **/ + unsigned int sleep_time_in_microseconds = 500; + unsigned int tries_allowed = 400; + bool is_in_alt_mode; + unsigned long long enter_timestamp; + unsigned long long finish_timestamp; + unsigned long long time_taken_in_ns; + int tries_taken; + + DC_LOGGER_INIT(link->ctx->logger); + + if (link->link_enc->funcs->is_in_alt_mode == NULL) + return true; + + is_in_alt_mode = link->link_enc->funcs->is_in_alt_mode(link->link_enc); + DC_LOG_WARNING("DP Alt mode state on HPD: %d\n", is_in_alt_mode); + + if (is_in_alt_mode) + return true; + + enter_timestamp = dm_get_timestamp(link->ctx); + + for (tries_taken = 0; tries_taken < tries_allowed; tries_taken++) { + udelay(sleep_time_in_microseconds); + /* ask the link if alt mode is enabled, if so return ok */ + if (link->link_enc->funcs->is_in_alt_mode(link->link_enc)) { + + finish_timestamp = dm_get_timestamp(link->ctx); + time_taken_in_ns = dm_get_elapse_time_in_ns( + link->ctx, finish_timestamp, enter_timestamp); + DC_LOG_WARNING("Alt mode entered finished after %llu ms\n", + div_u64(time_taken_in_ns, 1000000)); + return true; + } + + } + finish_timestamp = dm_get_timestamp(link->ctx); + time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, + enter_timestamp); + DC_LOG_WARNING("Alt mode has timed out after %llu ms\n", + div_u64(time_taken_in_ns, 1000000)); + return false; +} + /** * dc_link_detect() - Detect if a sink is attached to a given link * @@ -772,6 +818,15 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) } case SIGNAL_TYPE_DISPLAY_PORT: { + /* wa HPD high coming too early*/ + if (link->link_enc->features.flags.bits.DP_IS_USB_C == 1) { + + /* if alt mode times out, return false */ + if (wait_for_alt_mode(link) == false) { + return false; + } + } + if (!detect_dp( link, &sink_caps, @@ -795,16 +850,9 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) dc_sink_release(prev_sink); } else { /* Empty dongle plug in */ - for (i = 0; i < LINK_TRAINING_MAX_VERIFY_RETRY; i++) { - int fail_count = 0; - - dp_verify_link_cap(link, - &link->reported_link_cap, - &fail_count); - - if (fail_count == 0) - break; - } + dp_verify_link_cap_with_retries(link, + &link->reported_link_cap, + LINK_TRAINING_MAX_VERIFY_RETRY); } return true; } @@ -908,17 +956,9 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) */ /* deal with non-mst cases */ - for (i = 0; i < LINK_TRAINING_MAX_VERIFY_RETRY; i++) { - int fail_count = 0; - - dp_verify_link_cap(link, - &link->reported_link_cap, - &fail_count); - - if (fail_count == 0) - break; - } - + dp_verify_link_cap_with_retries(link, + &link->reported_link_cap, + LINK_TRAINING_MAX_VERIFY_RETRY); } else { // If edid is the same, then discard new sink and revert back to original sink if (same_edid) { @@ -1188,6 +1228,9 @@ static bool construct( link->ctx = dc_ctx; link->link_index = init_params->link_index; + memset(&link->preferred_training_settings, 0, sizeof(struct dc_link_training_overrides)); + memset(&link->preferred_link_setting, 0, sizeof(struct dc_link_settings)); + link->link_id = bios->funcs->get_connector_id(bios, init_params->connector_index); if (link->link_id.type != OBJECT_TYPE_CONNECTOR) { @@ -1384,57 +1427,6 @@ void link_destroy(struct dc_link **link) *link = NULL; } -static void dpcd_configure_panel_mode( - struct dc_link *link, - enum dp_panel_mode panel_mode) -{ - union dpcd_edp_config edp_config_set; - bool panel_mode_edp = false; - DC_LOGGER_INIT(link->ctx->logger); - - memset(&edp_config_set, '\0', sizeof(union dpcd_edp_config)); - - if (DP_PANEL_MODE_DEFAULT != panel_mode) { - - switch (panel_mode) { - case DP_PANEL_MODE_EDP: - case DP_PANEL_MODE_SPECIAL: - panel_mode_edp = true; - break; - - default: - break; - } - - /*set edp panel mode in receiver*/ - core_link_read_dpcd( - link, - DP_EDP_CONFIGURATION_SET, - &edp_config_set.raw, - sizeof(edp_config_set.raw)); - - if (edp_config_set.bits.PANEL_MODE_EDP - != panel_mode_edp) { - enum ddc_result result = DDC_RESULT_UNKNOWN; - - edp_config_set.bits.PANEL_MODE_EDP = - panel_mode_edp; - result = core_link_write_dpcd( - link, - DP_EDP_CONFIGURATION_SET, - &edp_config_set.raw, - sizeof(edp_config_set.raw)); - - ASSERT(result == DDC_RESULT_SUCESSFULL); - } - } - DC_LOG_DETECTION_DP_CAPS("Link: %d eDP panel mode supported: %d " - "eDP panel mode enabled: %d \n", - link->link_index, - link->dpcd_caps.panel_mode_edp, - panel_mode_edp); -} - static void enable_stream_features(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; @@ -1466,6 +1458,19 @@ static enum dc_status enable_link_dp( struct dc_link *link = stream->link; struct dc_link_settings link_settings = {0}; enum dp_panel_mode panel_mode; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + bool fec_enable; +#endif + int i; + bool apply_seamless_boot_optimization = false; + + // check for seamless boot + for (i = 0; i < state->stream_count; i++) { + if (state->streams[i]->apply_seamless_boot_optimization) { + apply_seamless_boot_optimization = true; + break; + } + } /* get link settings for video mode timing */ decide_link_settings(stream, &link_settings); @@ -1487,7 +1492,8 @@ static enum dc_status enable_link_dp( pipe_ctx->stream_res.pix_clk_params.requested_sym_clk = link_settings.link_rate * LINK_RATE_REF_FREQ_IN_KHZ; - state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false); + if (!apply_seamless_boot_optimization) + state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false); dp_enable_link_phy( link, @@ -1502,18 +1508,19 @@ static enum dc_status enable_link_dp( } panel_mode = dp_get_panel_mode(link); - dpcd_configure_panel_mode(link, panel_mode); + dp_set_panel_mode(link, panel_mode); skip_video_pattern = true; if (link_settings.link_rate == LINK_RATE_LOW) skip_video_pattern = false; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - dp_set_fec_ready(link, true); -#endif + if (link->aux_access_disabled) { + dc_link_dp_perform_link_training_skip_aux(link, &link_settings); - if (perform_link_training_with_retries( + link->cur_link_settings = link_settings; + status = DC_OK; + } else if (perform_link_training_with_retries( link, &link_settings, skip_video_pattern, @@ -1525,7 +1532,12 @@ static enum dc_status enable_link_dp( status = DC_FAIL_DP_LINK_TRAINING; #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - dp_set_fec_enable(link, true); + if (link->preferred_training_settings.fec_enable != NULL) + fec_enable = *link->preferred_training_settings.fec_enable; + else + fec_enable = true; + + dp_set_fec_enable(link, fec_enable); #endif return status; } @@ -2755,21 +2767,27 @@ void core_link_enable_stream( CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, COLOR_DEPTH_UNDEFINED); +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal) || + dc_is_virtual_signal(pipe_ctx->stream->signal)) + dp_set_dsc_enable(pipe_ctx, true); + } +#endif core_dc->hwss.enable_stream(pipe_ctx); - if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) - allocate_mst_payload(pipe_ctx); - #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (pipe_ctx->stream->timing.flags.DSC && - (dc_is_dp_signal(pipe_ctx->stream->signal) || - dc_is_virtual_signal(pipe_ctx->stream->signal))) { - dp_set_dsc_enable(pipe_ctx, true); - pipe_ctx->stream_res.tg->funcs->wait_for_state( - pipe_ctx->stream_res.tg, - CRTC_STATE_VBLANK); + /* Set DPS PPS SDP (AKA "info frames") */ + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal) || + dc_is_virtual_signal(pipe_ctx->stream->signal)) + dp_set_dsc_pps_sdp(pipe_ctx, true); } #endif + + if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) + allocate_mst_payload(pipe_ctx); + core_dc->hwss.unblank_stream(pipe_ctx, &pipe_ctx->stream->link->cur_link_settings); @@ -2786,7 +2804,7 @@ void core_link_enable_stream( #endif } -void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option) +void core_link_disable_stream(struct pipe_ctx *pipe_ctx) { struct dc *core_dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; @@ -2821,13 +2839,13 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option) write_i2c_redriver_setting(pipe_ctx, false); } } - core_dc->hwss.disable_stream(pipe_ctx, option); + core_dc->hwss.disable_stream(pipe_ctx); disable_link(pipe_ctx->stream->link, pipe_ctx->stream->signal); #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (pipe_ctx->stream->timing.flags.DSC && - dc_is_dp_signal(pipe_ctx->stream->signal)) { - dp_set_dsc_enable(pipe_ctx, false); + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + dp_set_dsc_enable(pipe_ctx, false); } #endif } @@ -2836,7 +2854,7 @@ void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable) { struct dc *core_dc = pipe_ctx->stream->ctx->dc; - if (pipe_ctx->stream->signal != SIGNAL_TYPE_HDMI_TYPE_A) + if (!dc_is_hdmi_signal(pipe_ctx->stream->signal)) return; core_dc->hwss.set_avmute(pipe_ctx, enable); @@ -2999,8 +3017,10 @@ void dc_link_set_preferred_link_settings(struct dc *dc, for (i = 0; i < MAX_PIPES; i++) { pipe = &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe->stream && pipe->stream->link) { - if (pipe->stream->link == link) + if (pipe->stream->link == link) { + link_stream = pipe->stream; break; + } } } @@ -3008,20 +3028,40 @@ void dc_link_set_preferred_link_settings(struct dc *dc, if (i == MAX_PIPES) return; - link_stream = link->dc->current_state->res_ctx.pipe_ctx[i].stream; - /* Cannot retrain link if backend is off */ if (link_stream->dpms_off) return; - if (link_stream) - decide_link_settings(link_stream, &store_settings); + decide_link_settings(link_stream, &store_settings); if ((store_settings.lane_count != LANE_COUNT_UNKNOWN) && (store_settings.link_rate != LINK_RATE_UNKNOWN)) dp_retrain_link_dp_test(link, &store_settings, false); } +void dc_link_set_preferred_training_settings(struct dc *dc, + struct dc_link_settings *link_setting, + struct dc_link_training_overrides *lt_overrides, + struct dc_link *link, + bool skip_immediate_retrain) +{ + if (lt_overrides != NULL) + link->preferred_training_settings = *lt_overrides; + else + memset(&link->preferred_training_settings, 0, sizeof(link->preferred_training_settings)); + + if (link_setting != NULL) { + link->preferred_link_setting = *link_setting; + } else { + link->preferred_link_setting.lane_count = LANE_COUNT_UNKNOWN; + link->preferred_link_setting.link_rate = LINK_RATE_UNKNOWN; + } + + /* Retrain now, or wait until next stream update to apply */ + if (skip_immediate_retrain == false) + dc_link_set_preferred_link_settings(dc, &link->preferred_link_setting, link); +} + void dc_link_enable_hpd(const struct dc_link *link) { dc_link_dp_enable_hpd(link); @@ -3032,7 +3072,6 @@ void dc_link_disable_hpd(const struct dc_link *link) dc_link_dp_disable_hpd(link); } - void dc_link_set_test_pattern(struct dc_link *link, enum dp_test_pattern test_pattern, const struct link_training_settings *p_link_settings, diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c index e6da8506128b..505967b48e14 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c @@ -294,7 +294,7 @@ static uint32_t defer_delay_converter_wa( { struct dc_link *link = ddc->link; - if (link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_4 && + if (link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_0080E1 && !memcmp(link->dpcd_caps.branch_dev_name, DP_DVI_CONVERTER_ID_4, sizeof(link->dpcd_caps.branch_dev_name))) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 2c7aaed907b9..f5742719b5d9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -49,7 +49,7 @@ static struct dc_link_settings get_common_supported_link_settings( struct dc_link_settings link_setting_a, struct dc_link_settings link_setting_b); -static void wait_for_training_aux_rd_interval( +static uint32_t get_training_aux_rd_interval( struct dc_link *link, uint32_t default_wait_in_micro_secs) { @@ -68,15 +68,21 @@ static void wait_for_training_aux_rd_interval( sizeof(training_rd_interval)); if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL) - default_wait_in_micro_secs = - training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000; + default_wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000; } - udelay(default_wait_in_micro_secs); + return default_wait_in_micro_secs; +} + +static void wait_for_training_aux_rd_interval( + struct dc_link *link, + uint32_t wait_in_micro_secs) +{ + udelay(wait_in_micro_secs); DC_LOG_HW_LINK_TRAINING("%s:\n wait = %d\n", __func__, - default_wait_in_micro_secs); + wait_in_micro_secs); } static void dpcd_set_training_pattern( @@ -95,27 +101,27 @@ static void dpcd_set_training_pattern( dpcd_pattern.v1_4.TRAINING_PATTERN_SET); } -static enum hw_dp_training_pattern get_supported_tp(struct dc_link *link) +static enum dc_dp_training_pattern get_supported_tp(struct dc_link *link) { - enum hw_dp_training_pattern highest_tp = HW_DP_TRAINING_PATTERN_2; + enum dc_dp_training_pattern highest_tp = DP_TRAINING_PATTERN_SEQUENCE_2; struct encoder_feature_support *features = &link->link_enc->features; struct dpcd_caps *dpcd_caps = &link->dpcd_caps; if (features->flags.bits.IS_TPS3_CAPABLE) - highest_tp = HW_DP_TRAINING_PATTERN_3; + highest_tp = DP_TRAINING_PATTERN_SEQUENCE_3; if (features->flags.bits.IS_TPS4_CAPABLE) - highest_tp = HW_DP_TRAINING_PATTERN_4; + highest_tp = DP_TRAINING_PATTERN_SEQUENCE_4; if (dpcd_caps->max_down_spread.bits.TPS4_SUPPORTED && - highest_tp >= HW_DP_TRAINING_PATTERN_4) - return HW_DP_TRAINING_PATTERN_4; + highest_tp >= DP_TRAINING_PATTERN_SEQUENCE_4) + return DP_TRAINING_PATTERN_SEQUENCE_4; if (dpcd_caps->max_ln_count.bits.TPS3_SUPPORTED && - highest_tp >= HW_DP_TRAINING_PATTERN_3) - return HW_DP_TRAINING_PATTERN_3; + highest_tp >= DP_TRAINING_PATTERN_SEQUENCE_3) + return DP_TRAINING_PATTERN_SEQUENCE_3; - return HW_DP_TRAINING_PATTERN_2; + return DP_TRAINING_PATTERN_SEQUENCE_2; } static void dpcd_set_link_settings( @@ -126,7 +132,7 @@ static void dpcd_set_link_settings( union down_spread_ctrl downspread = { {0} }; union lane_count_set lane_count_set = { {0} }; - enum hw_dp_training_pattern hw_tr_pattern; + enum dc_dp_training_pattern dp_tr_pattern; downspread.raw = (uint8_t) (lt_settings->link_settings.link_spread); @@ -134,21 +140,21 @@ static void dpcd_set_link_settings( lane_count_set.bits.LANE_COUNT_SET = lt_settings->link_settings.lane_count; - lane_count_set.bits.ENHANCED_FRAMING = 1; - + lane_count_set.bits.ENHANCED_FRAMING = lt_settings->enhanced_framing; lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = 0; - hw_tr_pattern = get_supported_tp(link); - if (hw_tr_pattern != HW_DP_TRAINING_PATTERN_4) { + dp_tr_pattern = get_supported_tp(link); + + if (dp_tr_pattern != DP_TRAINING_PATTERN_SEQUENCE_4) { lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED; } core_link_write_dpcd(link, DP_DOWNSPREAD_CTRL, - &downspread.raw, sizeof(downspread)); + &downspread.raw, sizeof(downspread)); core_link_write_dpcd(link, DP_LANE_COUNT_SET, - &lane_count_set.raw, 1); + &lane_count_set.raw, 1); if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14 && lt_settings->link_settings.use_link_rate_set == true) { @@ -162,46 +168,47 @@ static void dpcd_set_link_settings( } if (rate) { - DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x\n %x spread = %x\n", + DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x framing = %x\n %x spread = %x\n", __func__, DP_LINK_BW_SET, lt_settings->link_settings.link_rate, DP_LANE_COUNT_SET, lt_settings->link_settings.lane_count, + lt_settings->enhanced_framing, DP_DOWNSPREAD_CTRL, lt_settings->link_settings.link_spread); } else { - DC_LOG_HW_LINK_TRAINING("%s\n %x rate set = %x\n %x lane = %x\n %x spread = %x\n", + DC_LOG_HW_LINK_TRAINING("%s\n %x rate set = %x\n %x lane = %x framing = %x\n %x spread = %x\n", __func__, DP_LINK_RATE_SET, lt_settings->link_settings.link_rate_set, DP_LANE_COUNT_SET, lt_settings->link_settings.lane_count, + lt_settings->enhanced_framing, DP_DOWNSPREAD_CTRL, lt_settings->link_settings.link_spread); } - } static enum dpcd_training_patterns - hw_training_pattern_to_dpcd_training_pattern( + dc_dp_training_pattern_to_dpcd_training_pattern( struct dc_link *link, - enum hw_dp_training_pattern pattern) + enum dc_dp_training_pattern pattern) { enum dpcd_training_patterns dpcd_tr_pattern = DPCD_TRAINING_PATTERN_VIDEOIDLE; switch (pattern) { - case HW_DP_TRAINING_PATTERN_1: + case DP_TRAINING_PATTERN_SEQUENCE_1: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_1; break; - case HW_DP_TRAINING_PATTERN_2: + case DP_TRAINING_PATTERN_SEQUENCE_2: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_2; break; - case HW_DP_TRAINING_PATTERN_3: + case DP_TRAINING_PATTERN_SEQUENCE_3: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_3; break; - case HW_DP_TRAINING_PATTERN_4: + case DP_TRAINING_PATTERN_SEQUENCE_4: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_4; break; default: @@ -212,13 +219,12 @@ static enum dpcd_training_patterns } return dpcd_tr_pattern; - } static void dpcd_set_lt_pattern_and_lane_settings( struct dc_link *link, const struct link_training_settings *lt_settings, - enum hw_dp_training_pattern pattern) + enum dc_dp_training_pattern pattern) { union dpcd_training_lane dpcd_lane[LANE_COUNT_DP_MAX] = { { {0} } }; const uint32_t dpcd_base_lt_offset = @@ -233,7 +239,7 @@ static void dpcd_set_lt_pattern_and_lane_settings( * DpcdAddress_TrainingPatternSet *****************************************************************/ dpcd_pattern.v1_4.TRAINING_PATTERN_SET = - hw_training_pattern_to_dpcd_training_pattern(link, pattern); + dc_dp_training_pattern_to_dpcd_training_pattern(link, pattern); dpcd_lt_buffer[DP_TRAINING_PATTERN_SET - dpcd_base_lt_offset] = dpcd_pattern.raw; @@ -346,12 +352,20 @@ static void update_drive_settings( { uint32_t lane; for (lane = 0; lane < src.link_settings.lane_count; lane++) { - dest->lane_settings[lane].VOLTAGE_SWING = - src.lane_settings[lane].VOLTAGE_SWING; - dest->lane_settings[lane].PRE_EMPHASIS = - src.lane_settings[lane].PRE_EMPHASIS; - dest->lane_settings[lane].POST_CURSOR2 = - src.lane_settings[lane].POST_CURSOR2; + if (dest->voltage_swing == NULL) + dest->lane_settings[lane].VOLTAGE_SWING = src.lane_settings[lane].VOLTAGE_SWING; + else + dest->lane_settings[lane].VOLTAGE_SWING = *dest->voltage_swing; + + if (dest->pre_emphasis == NULL) + dest->lane_settings[lane].PRE_EMPHASIS = src.lane_settings[lane].PRE_EMPHASIS; + else + dest->lane_settings[lane].PRE_EMPHASIS = *dest->pre_emphasis; + + if (dest->post_cursor2 == NULL) + dest->lane_settings[lane].POST_CURSOR2 = src.lane_settings[lane].POST_CURSOR2; + else + dest->lane_settings[lane].POST_CURSOR2 = *dest->post_cursor2; } } @@ -754,15 +768,15 @@ static enum link_training_result perform_channel_equalization_sequence( struct link_training_settings *lt_settings) { struct link_training_settings req_settings; - enum hw_dp_training_pattern hw_tr_pattern; + enum dc_dp_training_pattern tr_pattern; uint32_t retries_ch_eq; enum dc_lane_count lane_count = lt_settings->link_settings.lane_count; union lane_align_status_updated dpcd_lane_status_updated = { {0} }; union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = { { {0} } }; - hw_tr_pattern = get_supported_tp(link); + tr_pattern = lt_settings->pattern_for_eq; - dp_set_hw_training_pattern(link, hw_tr_pattern); + dp_set_hw_training_pattern(link, tr_pattern); for (retries_ch_eq = 0; retries_ch_eq <= LINK_TRAINING_MAX_RETRY_COUNT; retries_ch_eq++) { @@ -776,12 +790,12 @@ static enum link_training_result perform_channel_equalization_sequence( dpcd_set_lt_pattern_and_lane_settings( link, lt_settings, - hw_tr_pattern); + tr_pattern); else dpcd_set_lane_settings(link, lt_settings); /* 3. wait for receiver to lock-on*/ - wait_for_training_aux_rd_interval(link, 400); + wait_for_training_aux_rd_interval(link, lt_settings->eq_pattern_time); /* 4. Read lane status and requested * drive settings as set by the sink*/ @@ -817,27 +831,16 @@ static enum link_training_result perform_clock_recovery_sequence( { uint32_t retries_cr; uint32_t retry_count; - uint32_t lane; struct link_training_settings req_settings; - enum dc_lane_count lane_count = - lt_settings->link_settings.lane_count; - enum hw_dp_training_pattern hw_tr_pattern = HW_DP_TRAINING_PATTERN_1; + enum dc_lane_count lane_count = lt_settings->link_settings.lane_count; + enum dc_dp_training_pattern tr_pattern = DP_TRAINING_PATTERN_SEQUENCE_1; union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX]; union lane_align_status_updated dpcd_lane_status_updated; retries_cr = 0; retry_count = 0; - /* initial drive setting (VS/PE/PC2)*/ - for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { - lt_settings->lane_settings[lane].VOLTAGE_SWING = - VOLTAGE_SWING_LEVEL0; - lt_settings->lane_settings[lane].PRE_EMPHASIS = - PRE_EMPHASIS_DISABLED; - lt_settings->lane_settings[lane].POST_CURSOR2 = - POST_CURSOR2_DISABLED; - } - dp_set_hw_training_pattern(link, hw_tr_pattern); + dp_set_hw_training_pattern(link, tr_pattern); /* najeeb - The synaptics MST hub can put the LT in * infinite loop by switching the VS @@ -845,7 +848,7 @@ static enum link_training_result perform_clock_recovery_sequence( /* between level 0 and level 1 continuously, here * we try for CR lock for LinkTrainingMaxCRRetry count*/ while ((retries_cr < LINK_TRAINING_MAX_RETRY_COUNT) && - (retry_count < LINK_TRAINING_MAX_CR_RETRY)) { + (retry_count < LINK_TRAINING_MAX_CR_RETRY)) { memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status)); memset(&dpcd_lane_status_updated, '\0', @@ -863,7 +866,7 @@ static enum link_training_result perform_clock_recovery_sequence( dpcd_set_lt_pattern_and_lane_settings( link, lt_settings, - hw_tr_pattern); + tr_pattern); else dpcd_set_lane_settings( link, @@ -872,7 +875,7 @@ static enum link_training_result perform_clock_recovery_sequence( /* 3. wait receiver to lock-on*/ wait_for_training_aux_rd_interval( link, - 100); + lt_settings->cr_pattern_time); /* 4. Read lane status and requested drive * settings as set by the sink @@ -939,7 +942,7 @@ static inline enum link_training_result perform_link_training_int( * TPS4 must be used instead of POST_LT_ADJ_REQ. */ if (link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED != 1 || - get_supported_tp(link) == HW_DP_TRAINING_PATTERN_4) + get_supported_tp(link) == DP_TRAINING_PATTERN_SEQUENCE_4) return status; if (status == LINK_TRAINING_SUCCESS && @@ -947,7 +950,7 @@ static inline enum link_training_result perform_link_training_int( status = LINK_TRAINING_LQA_FAIL; lane_count_set.bits.LANE_COUNT_SET = lt_settings->link_settings.lane_count; - lane_count_set.bits.ENHANCED_FRAMING = 1; + lane_count_set.bits.ENHANCED_FRAMING = lt_settings->enhanced_framing; lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = 0; core_link_write_dpcd( @@ -959,24 +962,29 @@ static inline enum link_training_result perform_link_training_int( return status; } -enum link_training_result dc_link_dp_perform_link_training( - struct dc_link *link, +static void initialize_training_settings( + struct dc_link *link, const struct dc_link_settings *link_setting, - bool skip_video_pattern) + const struct dc_link_training_overrides *overrides, + struct link_training_settings *lt_settings) { - enum link_training_result status = LINK_TRAINING_SUCCESS; + uint32_t lane; - char *link_rate = "Unknown"; - char *lt_result = "Unknown"; + memset(lt_settings, '\0', sizeof(struct link_training_settings)); - struct link_training_settings lt_settings; + /* Initialize link settings */ + lt_settings->link_settings.use_link_rate_set = link_setting->use_link_rate_set; + lt_settings->link_settings.link_rate_set = link_setting->link_rate_set; - memset(<_settings, '\0', sizeof(lt_settings)); + if (link->preferred_link_setting.link_rate != LINK_RATE_UNKNOWN) + lt_settings->link_settings.link_rate = link->preferred_link_setting.link_rate; + else + lt_settings->link_settings.link_rate = link_setting->link_rate; - lt_settings.link_settings.link_rate = link_setting->link_rate; - lt_settings.link_settings.lane_count = link_setting->lane_count; - lt_settings.link_settings.use_link_rate_set = link_setting->use_link_rate_set; - lt_settings.link_settings.link_rate_set = link_setting->link_rate_set; + if (link->preferred_link_setting.lane_count != LANE_COUNT_UNKNOWN) + lt_settings->link_settings.lane_count = link->preferred_link_setting.lane_count; + else + lt_settings->link_settings.lane_count = link_setting->lane_count; /*@todo[vdevulap] move SS to LS, should not be handled by displaypath*/ @@ -987,31 +995,75 @@ enum link_training_result dc_link_dp_perform_link_training( * LINK_SPREAD_05_DOWNSPREAD_30KHZ : * LINK_SPREAD_DISABLED; */ + /* Initialize link spread */ if (link->dp_ss_off) - lt_settings.link_settings.link_spread = LINK_SPREAD_DISABLED; + lt_settings->link_settings.link_spread = LINK_SPREAD_DISABLED; + else if (overrides->downspread != NULL) + lt_settings->link_settings.link_spread + = *overrides->downspread + ? LINK_SPREAD_05_DOWNSPREAD_30KHZ + : LINK_SPREAD_DISABLED; else - lt_settings.link_settings.link_spread = LINK_SPREAD_05_DOWNSPREAD_30KHZ; + lt_settings->link_settings.link_spread = LINK_SPREAD_05_DOWNSPREAD_30KHZ; - /* 1. set link rate, lane count and spread*/ - dpcd_set_link_settings(link, <_settings); + /* Initialize lane settings overrides */ + if (overrides->voltage_swing != NULL) + lt_settings->voltage_swing = overrides->voltage_swing; - /* 2. perform link training (set link training done - * to false is done as well)*/ - status = perform_clock_recovery_sequence(link, <_settings); - if (status == LINK_TRAINING_SUCCESS) { - status = perform_channel_equalization_sequence(link, - <_settings); - } + if (overrides->pre_emphasis != NULL) + lt_settings->pre_emphasis = overrides->pre_emphasis; - if ((status == LINK_TRAINING_SUCCESS) || !skip_video_pattern) { - status = perform_link_training_int(link, - <_settings, - status); + if (overrides->post_cursor2 != NULL) + lt_settings->post_cursor2 = overrides->post_cursor2; + + /* Initialize lane settings (VS/PE/PC2) */ + for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { + lt_settings->lane_settings[lane].VOLTAGE_SWING = + lt_settings->voltage_swing != NULL ? + *lt_settings->voltage_swing : + VOLTAGE_SWING_LEVEL0; + lt_settings->lane_settings[lane].PRE_EMPHASIS = + lt_settings->pre_emphasis != NULL ? + *lt_settings->pre_emphasis + : PRE_EMPHASIS_DISABLED; + lt_settings->lane_settings[lane].POST_CURSOR2 = + lt_settings->post_cursor2 != NULL ? + *lt_settings->post_cursor2 + : POST_CURSOR2_DISABLED; } - /* 6. print status message*/ - switch (lt_settings.link_settings.link_rate) { + /* Initialize training timings */ + if (overrides->cr_pattern_time != NULL) + lt_settings->cr_pattern_time = *overrides->cr_pattern_time; + else + lt_settings->cr_pattern_time = get_training_aux_rd_interval(link, 100); + + if (overrides->eq_pattern_time != NULL) + lt_settings->eq_pattern_time = *overrides->eq_pattern_time; + else + lt_settings->eq_pattern_time = get_training_aux_rd_interval(link, 400); + + if (overrides->pattern_for_eq != NULL) + lt_settings->pattern_for_eq = *overrides->pattern_for_eq; + else + lt_settings->pattern_for_eq = get_supported_tp(link); + + if (overrides->enhanced_framing != NULL) + lt_settings->enhanced_framing = *overrides->enhanced_framing; + else + lt_settings->enhanced_framing = 1; +} + +static void print_status_message( + struct dc_link *link, + const struct link_training_settings *lt_settings, + enum link_training_result status) +{ + char *link_rate = "Unknown"; + char *lt_result = "Unknown"; + char *lt_spread = "Disabled"; + switch (lt_settings->link_settings.link_rate) { case LINK_RATE_LOW: link_rate = "RBR"; break; @@ -1057,13 +1109,122 @@ enum link_training_result dc_link_dp_perform_link_training( break; } + switch (lt_settings->link_settings.link_spread) { + case LINK_SPREAD_DISABLED: + lt_spread = "Disabled"; + break; + case LINK_SPREAD_05_DOWNSPREAD_30KHZ: + lt_spread = "0.5% 30KHz"; + break; + case LINK_SPREAD_05_DOWNSPREAD_33KHZ: + lt_spread = "0.5% 33KHz"; + break; + default: + break; + } + /* Connectivity log: link training */ - CONN_MSG_LT(link, "%sx%d %s VS=%d, PE=%d", - link_rate, - lt_settings.link_settings.lane_count, - lt_result, - lt_settings.lane_settings[0].VOLTAGE_SWING, - lt_settings.lane_settings[0].PRE_EMPHASIS); + CONN_MSG_LT(link, "%sx%d %s VS=%d, PE=%d, DS=%s", + link_rate, + lt_settings->link_settings.lane_count, + lt_result, + lt_settings->lane_settings[0].VOLTAGE_SWING, + lt_settings->lane_settings[0].PRE_EMPHASIS, + lt_spread); +} + +bool dc_link_dp_perform_link_training_skip_aux( + struct dc_link *link, + const struct dc_link_settings *link_setting) +{ + struct link_training_settings lt_settings; + enum dc_dp_training_pattern pattern_for_cr = DP_TRAINING_PATTERN_SEQUENCE_1; + + initialize_training_settings( + link, + link_setting, + &link->preferred_training_settings, + <_settings); + + /* 1. Perform_clock_recovery_sequence. */ + + /* transmit training pattern for clock recovery */ + dp_set_hw_training_pattern(link, pattern_for_cr); + + /* call HWSS to set lane settings*/ + dp_set_hw_lane_settings(link, <_settings); + + /* wait receiver to lock-on*/ + wait_for_training_aux_rd_interval(link, lt_settings.cr_pattern_time); + + /* 2. Perform_channel_equalization_sequence. */ + + /* transmit training pattern for channel equalization. */ + dp_set_hw_training_pattern(link, lt_settings.pattern_for_eq); + + /* call HWSS to set lane settings*/ + dp_set_hw_lane_settings(link, <_settings); + + /* wait receiver to lock-on. */ + wait_for_training_aux_rd_interval(link, lt_settings.eq_pattern_time); + + /* 3. Perform_link_training_int. */ + + /* Mainlink output idle pattern. */ + dp_set_hw_test_pattern(link, DP_TEST_PATTERN_VIDEO_MODE, NULL, 0); + + print_status_message(link, <_settings, LINK_TRAINING_SUCCESS); + + return true; +} + +enum link_training_result dc_link_dp_perform_link_training( + struct dc_link *link, + const struct dc_link_settings *link_setting, + bool skip_video_pattern) +{ + enum link_training_result status = LINK_TRAINING_SUCCESS; + struct link_training_settings lt_settings; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + bool fec_enable; +#endif + + initialize_training_settings( + link, + link_setting, + &link->preferred_training_settings, + <_settings); + + /* 1. set link rate, lane count and spread. */ + dpcd_set_link_settings(link, <_settings); + +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + if (link->preferred_training_settings.fec_enable != NULL) + fec_enable = *link->preferred_training_settings.fec_enable; + else + fec_enable = true; + + dp_set_fec_ready(link, fec_enable); +#endif + + + /* 2. perform link training (set link training done + * to false is done as well) + */ + status = perform_clock_recovery_sequence(link, <_settings); + if (status == LINK_TRAINING_SUCCESS) { + status = perform_channel_equalization_sequence(link, + <_settings); + } + + if ((status == LINK_TRAINING_SUCCESS) || !skip_video_pattern) { + status = perform_link_training_int(link, + <_settings, + status); + } + + /* 6. print status message*/ + print_status_message(link, <_settings, status); if (status != LINK_TRAINING_SUCCESS) link->ctx->dc->debug_data.ltFailCount++; @@ -1071,7 +1232,6 @@ enum link_training_result dc_link_dp_perform_link_training( return status; } - bool perform_link_training_with_retries( struct dc_link *link, const struct dc_link_settings *link_setting, @@ -1096,6 +1256,146 @@ bool perform_link_training_with_retries( return false; } +static enum clock_source_id get_clock_source_id(struct dc_link *link) +{ + enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_UNDEFINED; + struct clock_source *dp_cs = link->dc->res_pool->dp_clock_source; + + if (dp_cs != NULL) { + dp_cs_id = dp_cs->id; + } else { + /* + * dp clock source is not initialized for some reason. + * Should not happen, CLOCK_SOURCE_ID_EXTERNAL will be used + */ + ASSERT(dp_cs); + } + + return dp_cs_id; +} + +static void set_dp_mst_mode(struct dc_link *link, bool mst_enable) +{ + if (mst_enable == false && + link->type == dc_connection_mst_branch) { + /* Disable MST on link. Use only local sink. */ + dp_disable_link_phy_mst(link, link->connector_signal); + + link->type = dc_connection_single; + link->local_sink = link->remote_sinks[0]; + link->local_sink->sink_signal = SIGNAL_TYPE_DISPLAY_PORT; + } else if (mst_enable == true && + link->type == dc_connection_single && + link->remote_sinks[0] != NULL) { + /* Re-enable MST on link. */ + dp_disable_link_phy(link, link->connector_signal); + dp_enable_mst_on_sink(link, true); + + link->type = dc_connection_mst_branch; + link->local_sink->sink_signal = SIGNAL_TYPE_DISPLAY_PORT_MST; + } +} + +bool dc_link_dp_sync_lt_begin(struct dc_link *link) +{ + /* Begin Sync LT. During this time, + * DPCD:600h must not be powered down. + */ + link->sync_lt_in_progress = true; + + /*Clear any existing preferred settings.*/ + memset(&link->preferred_training_settings, 0, + sizeof(struct dc_link_training_overrides)); + memset(&link->preferred_link_setting, 0, + sizeof(struct dc_link_settings)); + + return true; +} + +enum link_training_result dc_link_dp_sync_lt_attempt( + struct dc_link *link, + struct dc_link_settings *link_settings, + struct dc_link_training_overrides *lt_overrides) +{ + struct link_training_settings lt_settings; + enum link_training_result lt_status = LINK_TRAINING_SUCCESS; + enum dp_panel_mode panel_mode = DP_PANEL_MODE_DEFAULT; + enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_EXTERNAL; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + bool fec_enable = false; +#endif + + initialize_training_settings( + link, + link_settings, + lt_overrides, + <_settings); + + /* Setup MST Mode */ + if (lt_overrides->mst_enable) + set_dp_mst_mode(link, *lt_overrides->mst_enable); + + /* Disable link */ + dp_disable_link_phy(link, link->connector_signal); + + /* Enable link */ + dp_cs_id = get_clock_source_id(link); + dp_enable_link_phy(link, link->connector_signal, + dp_cs_id, link_settings); + +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + /* Set FEC enable */ + fec_enable = lt_overrides->fec_enable && *lt_overrides->fec_enable; + dp_set_fec_ready(link, fec_enable); +#endif + + if (lt_overrides->alternate_scrambler_reset) { + if (*lt_overrides->alternate_scrambler_reset) + panel_mode = DP_PANEL_MODE_EDP; + else + panel_mode = DP_PANEL_MODE_DEFAULT; + } else + panel_mode = dp_get_panel_mode(link); + + dp_set_panel_mode(link, panel_mode); + + /* Attempt to train with given link training settings */ + + /* Set link rate, lane count and spread. */ + dpcd_set_link_settings(link, <_settings); + + /* 2. perform link training (set link training done + * to false is done as well) + */ + lt_status = perform_clock_recovery_sequence(link, <_settings); + if (lt_status == LINK_TRAINING_SUCCESS) { + lt_status = perform_channel_equalization_sequence(link, + <_settings); + } + + /* 3. Sync LT must skip TRAINING_PATTERN_SET:0 (video pattern)*/ + /* 4. print status message*/ + print_status_message(link, <_settings, lt_status); + + return lt_status; +} + +bool dc_link_dp_sync_lt_end(struct dc_link *link, bool link_down) +{ + /* If input parameter is set, shut down phy. + * Still shouldn't turn off dp_receiver (DPCD:600h) + */ + if (link_down == true) { + dp_disable_link_phy(link, link->connector_signal); +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + dp_set_fec_ready(link, false); +#endif + } + + link->sync_lt_in_progress = false; + return true; +} + static struct dc_link_settings get_max_link_cap(struct dc_link *link) { /* Set Default link settings */ @@ -1250,7 +1550,6 @@ bool dp_verify_link_cap( bool success; bool skip_link_training; bool skip_video_pattern; - struct clock_source *dp_cs; enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_EXTERNAL; enum link_training_result status; union hpd_irq_data irq_data; @@ -1274,17 +1573,7 @@ bool dp_verify_link_cap( /* disable PHY done possible by BIOS, will be done by driver itself */ dp_disable_link_phy(link, link->connector_signal); - dp_cs = link->dc->res_pool->dp_clock_source; - - if (dp_cs) - dp_cs_id = dp_cs->id; - else { - /* - * dp clock source is not initialized for some reason. - * Should not happen, CLOCK_SOURCE_ID_EXTERNAL will be used - */ - ASSERT(dp_cs); - } + dp_cs_id = get_clock_source_id(link); /* link training starts with the maximum common settings * supported by both sink and ASIC. @@ -1354,6 +1643,33 @@ bool dp_verify_link_cap( return success; } +bool dp_verify_link_cap_with_retries( + struct dc_link *link, + struct dc_link_settings *known_limit_link_setting, + int attempts) +{ + uint8_t i = 0; + bool success = false; + + for (i = 0; i < attempts; i++) { + int fail_count = 0; + enum dc_connection_type type; + + memset(&link->verified_link_cap, 0, + sizeof(struct dc_link_settings)); + if (!dc_link_detect_sink(link, &type)) { + break; + } else if (dp_verify_link_cap(link, + &link->reported_link_cap, + &fail_count) && fail_count == 0) { + success = true; + break; + } + msleep(10); + } + return success; +} + static struct dc_link_settings get_common_supported_link_settings( struct dc_link_settings link_setting_a, struct dc_link_settings link_setting_b) @@ -2156,6 +2472,11 @@ bool is_mst_supported(struct dc_link *link) union dpcd_rev rev; union mstm_cap cap; + if (link->preferred_training_settings.mst_enable && + *link->preferred_training_settings.mst_enable == false) { + return false; + } + rev.raw = 0; cap.raw = 0; @@ -2363,13 +2684,13 @@ static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data, if (link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_VGA_CONVERTER) { switch (link->dpcd_caps.branch_dev_id) { - /* Some active dongles (DP-VGA, DP-DLDVI converters) power down + /* 0010FA active dongles (DP-VGA, DP-DLDVI converters) power down * all internal circuits including AUX communication preventing * reading DPCD table and EDID (spec violation). * Encoder will skip DP RX power down on disable_output to * keep receiver powered all the time.*/ - case DP_BRANCH_DEVICE_ID_1: - case DP_BRANCH_DEVICE_ID_4: + case DP_BRANCH_DEVICE_ID_0010FA: + case DP_BRANCH_DEVICE_ID_0080E1: link->wa_flags.dp_keep_receiver_powered = true; break; @@ -2774,14 +3095,19 @@ static void set_crtc_test_pattern(struct dc_link *link, controller_test_pattern, color_depth); #if defined(CONFIG_DRM_AMD_DC_DCN2_0) else if (opp->funcs->opp_set_disp_pattern_generator) { - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; - if (bot_odm_pipe) { - struct output_pixel_processor *bot_opp = bot_odm_pipe->stream_res.opp; + width /= opp_cnt; - bot_opp->funcs->opp_program_bit_depth_reduction(bot_opp, ¶ms); - width /= 2; - bot_opp->funcs->opp_set_disp_pattern_generator(bot_opp, + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; + + odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); + odm_opp->funcs->opp_set_disp_pattern_generator(odm_opp, controller_test_pattern, color_depth, NULL, @@ -2810,14 +3136,18 @@ static void set_crtc_test_pattern(struct dc_link *link, color_depth); #if defined(CONFIG_DRM_AMD_DC_DCN2_0) else if (opp->funcs->opp_set_disp_pattern_generator) { - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; - if (bot_odm_pipe) { - struct output_pixel_processor *bot_opp = bot_odm_pipe->stream_res.opp; + width /= opp_cnt; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; - bot_opp->funcs->opp_program_bit_depth_reduction(bot_opp, ¶ms); - width /= 2; - bot_opp->funcs->opp_set_disp_pattern_generator(bot_opp, + odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); + odm_opp->funcs->opp_set_disp_pattern_generator(odm_opp, CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, color_depth, NULL, @@ -2858,7 +3188,7 @@ bool dc_link_dp_set_test_pattern( memset(&training_pattern, 0, sizeof(training_pattern)); for (i = 0; i < MAX_PIPES; i++) { - if (pipes[i].stream->link == link) { + if (pipes[i].stream->link == link && !pipes[i].top_pipe && !pipes[i].prev_odm_pipe) { pipe_ctx = &pipes[i]; break; } @@ -3007,6 +3337,105 @@ void dp_enable_mst_on_sink(struct dc_link *link, bool enable) core_link_write_dpcd(link, DP_MSTM_CTRL, &mstmCntl, 1); } +void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode) +{ + union dpcd_edp_config edp_config_set; + bool panel_mode_edp = false; + + memset(&edp_config_set, '\0', sizeof(union dpcd_edp_config)); + + if (panel_mode != DP_PANEL_MODE_DEFAULT) { + + switch (panel_mode) { + case DP_PANEL_MODE_EDP: + case DP_PANEL_MODE_SPECIAL: + panel_mode_edp = true; + break; + + default: + break; + } + + /*set edp panel mode in receiver*/ + core_link_read_dpcd( + link, + DP_EDP_CONFIGURATION_SET, + &edp_config_set.raw, + sizeof(edp_config_set.raw)); + + if (edp_config_set.bits.PANEL_MODE_EDP + != panel_mode_edp) { + enum ddc_result result = DDC_RESULT_UNKNOWN; + + edp_config_set.bits.PANEL_MODE_EDP = + panel_mode_edp; + result = core_link_write_dpcd( + link, + DP_EDP_CONFIGURATION_SET, + &edp_config_set.raw, + sizeof(edp_config_set.raw)); + + ASSERT(result == DDC_RESULT_SUCESSFULL); + } + } + DC_LOG_DETECTION_DP_CAPS("Link: %d eDP panel mode supported: %d " + "eDP panel mode enabled: %d \n", + link->link_index, + link->dpcd_caps.panel_mode_edp, + panel_mode_edp); +} + +enum dp_panel_mode dp_get_panel_mode(struct dc_link *link) +{ + /* We need to explicitly check that connector + * is not DP. Some Travis_VGA get reported + * by video bios as DP. + */ + if (link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT) { + + switch (link->dpcd_caps.branch_dev_id) { + case DP_BRANCH_DEVICE_ID_0022B9: + /* alternate scrambler reset is required for Travis + * for the case when external chip does not + * provide sink device id, alternate scrambler + * scheme will be overriden later by querying + * Encoder features + */ + if (strncmp( + link->dpcd_caps.branch_dev_name, + DP_VGA_LVDS_CONVERTER_ID_2, + sizeof( + link->dpcd_caps. + branch_dev_name)) == 0) { + return DP_PANEL_MODE_SPECIAL; + } + break; + case DP_BRANCH_DEVICE_ID_00001A: + /* alternate scrambler reset is required for Travis + * for the case when external chip does not provide + * sink device id, alternate scrambler scheme will + * be overriden later by querying Encoder feature + */ + if (strncmp(link->dpcd_caps.branch_dev_name, + DP_VGA_LVDS_CONVERTER_ID_3, + sizeof( + link->dpcd_caps. + branch_dev_name)) == 0) { + return DP_PANEL_MODE_SPECIAL; + } + break; + default: + break; + } + } + + if (link->dpcd_caps.panel_mode_edp) { + return DP_PANEL_MODE_EDP; + } + + return DP_PANEL_MODE_DEFAULT; +} + #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT void dp_set_fec_ready(struct dc_link *link, bool ready) { @@ -3024,7 +3453,7 @@ void dp_set_fec_ready(struct dc_link *link, bool ready) if (link_enc->funcs->fec_set_ready && link->dpcd_caps.fec_cap.bits.FEC_CAPABLE) { - if (link->fec_state == dc_link_fec_not_ready && ready) { + if (ready) { fec_config = 1; if (core_link_write_dpcd(link, DP_FEC_CONFIGURATION, @@ -3033,9 +3462,11 @@ void dp_set_fec_ready(struct dc_link *link, bool ready) link_enc->funcs->fec_set_ready(link_enc, true); link->fec_state = dc_link_fec_ready; } else { + link->link_enc->funcs->fec_set_ready(link->link_enc, false); + link->fec_state = dc_link_fec_not_ready; dm_error("dpcd write failed to set fec_ready"); } - } else if (link->fec_state == dc_link_fec_ready && !ready) { + } else if (link->fec_state == dc_link_fec_ready) { fec_config = 0; core_link_write_dpcd(link, DP_FEC_CONFIGURATION, diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c index 2d019e1f6135..79438c4f1e20 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c @@ -25,10 +25,11 @@ enum dc_status core_link_read_dpcd( uint8_t *data, uint32_t size) { - if (!dm_helpers_dp_read_dpcd(link->ctx, - link, - address, data, size)) - return DC_ERROR_UNEXPECTED; + if (!link->aux_access_disabled && + !dm_helpers_dp_read_dpcd(link->ctx, + link, address, data, size)) { + return DC_ERROR_UNEXPECTED; + } return DC_OK; } @@ -39,10 +40,11 @@ enum dc_status core_link_write_dpcd( const uint8_t *data, uint32_t size) { - if (!dm_helpers_dp_write_dpcd(link->ctx, - link, - address, data, size)) - return DC_ERROR_UNEXPECTED; + if (!link->aux_access_disabled && + !dm_helpers_dp_write_dpcd(link->ctx, + link, address, data, size)) { + return DC_ERROR_UNEXPECTED; + } return DC_OK; } @@ -53,6 +55,9 @@ void dp_receiver_power_ctrl(struct dc_link *link, bool on) state = on ? DP_POWER_STATE_D0 : DP_POWER_STATE_D3; + if (link->sync_lt_in_progress) + return; + core_link_write_dpcd(link, DP_SET_POWER, &state, sizeof(state)); } @@ -160,6 +165,10 @@ bool edp_receiver_ready_T7(struct dc_link *link) break; udelay(25); //MAx T7 is 50ms } while (++tries < 300); + + if (link->local_sink->edid_caps.panel_patch.extra_t7_ms > 0) + udelay(link->local_sink->edid_caps.panel_patch.extra_t7_ms * 1000); + return result; } @@ -203,21 +212,21 @@ void dp_disable_link_phy_mst(struct dc_link *link, enum signal_type signal) bool dp_set_hw_training_pattern( struct dc_link *link, - enum hw_dp_training_pattern pattern) + enum dc_dp_training_pattern pattern) { enum dp_test_pattern test_pattern = DP_TEST_PATTERN_UNSUPPORTED; switch (pattern) { - case HW_DP_TRAINING_PATTERN_1: + case DP_TRAINING_PATTERN_SEQUENCE_1: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN1; break; - case HW_DP_TRAINING_PATTERN_2: + case DP_TRAINING_PATTERN_SEQUENCE_2: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN2; break; - case HW_DP_TRAINING_PATTERN_3: + case DP_TRAINING_PATTERN_SEQUENCE_3: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN3; break; - case HW_DP_TRAINING_PATTERN_4: + case DP_TRAINING_PATTERN_SEQUENCE_4: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN4; break; default: @@ -239,46 +248,6 @@ void dp_set_hw_lane_settings( encoder->funcs->dp_set_lane_settings(encoder, link_settings); } -enum dp_panel_mode dp_get_panel_mode(struct dc_link *link) -{ - /* We need to explicitly check that connector - * is not DP. Some Travis_VGA get reported - * by video bios as DP. - */ - if (link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT) { - - switch (link->dpcd_caps.branch_dev_id) { - case DP_BRANCH_DEVICE_ID_2: - if (strncmp( - link->dpcd_caps.branch_dev_name, - DP_VGA_LVDS_CONVERTER_ID_2, - sizeof( - link->dpcd_caps. - branch_dev_name)) == 0) { - return DP_PANEL_MODE_SPECIAL; - } - break; - case DP_BRANCH_DEVICE_ID_3: - if (strncmp(link->dpcd_caps.branch_dev_name, - DP_VGA_LVDS_CONVERTER_ID_3, - sizeof( - link->dpcd_caps. - branch_dev_name)) == 0) { - return DP_PANEL_MODE_SPECIAL; - } - break; - default: - break; - } - } - - if (link->dpcd_caps.panel_mode_edp) { - return DP_PANEL_MODE_EDP; - } - - return DP_PANEL_MODE_DEFAULT; -} - void dp_set_hw_test_pattern( struct dc_link *link, enum dp_test_pattern test_pattern, @@ -306,7 +275,7 @@ void dp_retrain_link_dp_test(struct dc_link *link, for (i = 0; i < MAX_PIPES; i++) { if (pipes[i].stream != NULL && - !pipes[i].top_pipe && + !pipes[i].top_pipe && !pipes[i].prev_odm_pipe && pipes[i].stream->link != NULL && pipes[i].stream_res.stream_enc != NULL) { udelay(100); @@ -320,7 +289,9 @@ void dp_retrain_link_dp_test(struct dc_link *link, dp_receiver_power_ctrl(link, false); - link->dc->hwss.disable_stream(&pipes[i], KEEP_ACQUIRED_RESOURCE); + link->dc->hwss.disable_stream(&pipes[i]); + if ((&pipes[i])->stream_res.audio && !link->dc->debug.az_endpoint_mute_only) + (&pipes[i])->stream_res.audio->funcs->az_disable((&pipes[i])->stream_res.audio); link->link_enc->funcs->disable_output( link->link_enc, @@ -373,10 +344,22 @@ void dp_retrain_link_dp_test(struct dc_link *link, static void dsc_optc_config_log(struct display_stream_compressor *dsc, struct dsc_optc_config *config) { - DC_LOG_DSC("Setting optc DSC config at DSC inst %d", dsc->inst); - DC_LOG_DSC("\n\tbytes_per_pixel %d\n\tis_pixel_format_444 %d\n\tslice_width %d", - config->bytes_per_pixel, - config->is_pixel_format_444, config->slice_width); + uint32_t precision = 1 << 28; + uint32_t bytes_per_pixel_int = config->bytes_per_pixel / precision; + uint32_t bytes_per_pixel_mod = config->bytes_per_pixel % precision; + uint64_t ll_bytes_per_pix_fraq = bytes_per_pixel_mod; + + /* 7 fractional digits decimal precision for bytes per pixel is enough because DSC + * bits per pixel precision is 1/16th of a pixel, which means bytes per pixel precision is + * 1/16/8 = 1/128 of a byte, or 0.0078125 decimal + */ + ll_bytes_per_pix_fraq *= 10000000; + ll_bytes_per_pix_fraq /= precision; + + DC_LOG_DSC("\tbytes_per_pixel 0x%08x (%d.%07d)", + config->bytes_per_pixel, bytes_per_pixel_int, (uint32_t)ll_bytes_per_pix_fraq); + DC_LOG_DSC("\tis_pixel_format_444 %d", config->is_pixel_format_444); + DC_LOG_DSC("\tslice_width %d", config->slice_width); } static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) @@ -392,55 +375,62 @@ static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) return result; } -/* This has to be done after DSC was enabled on RX first, i.e. after dp_enable_dsc_on_rx() had been called +/* The stream with these settings can be sent (unblanked) only after DSC was enabled on RX first, + * i.e. after dp_enable_dsc_on_rx() had been called */ -static void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) +void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) { struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; struct dc *core_dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; if (enable) { - /* TODO proper function */ struct dsc_config dsc_cfg; struct dsc_optc_config dsc_optc_cfg; enum optc_dsc_mode optc_dsc_mode; - uint8_t dsc_packed_pps[128]; /* Enable DSC hw block */ - dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; + dsc_cfg.pic_width = (stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right) / opp_cnt; dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % opp_cnt == 0); + dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; - dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg, &dsc_packed_pps[0]); - if (odm_pipe) { - struct display_stream_compressor *bot_dsc = odm_pipe->stream_res.dsc; - uint8_t dsc_packed_pps_odm[128]; - - dsc_cfg.pic_width /= 2; - ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % 2 == 0); - dsc_cfg.dc_dsc_cfg.num_slices_h /= 2; - dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg, &dsc_packed_pps_odm[0]); - bot_dsc->funcs->dsc_set_config(bot_dsc, &dsc_cfg, &dsc_optc_cfg, &dsc_packed_pps_odm[0]); - bot_dsc->funcs->dsc_enable(bot_dsc, odm_pipe->stream_res.opp->inst); - } + dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; + + odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); + odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); + } + dsc_cfg.dc_dsc_cfg.num_slices_h *= opp_cnt; + dsc_cfg.pic_width *= opp_cnt; optc_dsc_mode = dsc_optc_cfg.is_pixel_format_444 ? OPTC_DSC_ENABLED_444 : OPTC_DSC_ENABLED_NATIVE_SUBSAMPLED; - dsc_optc_config_log(dsc, &dsc_optc_cfg); /* Enable DSC in encoder */ - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment) && pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config) + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + DC_LOG_DSC("Setting stream encoder DSC config for engine %d:", (int)pipe_ctx->stream_res.stream_enc->id); + dsc_optc_config_log(dsc, &dsc_optc_cfg); pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(pipe_ctx->stream_res.stream_enc, optc_dsc_mode, dsc_optc_cfg.bytes_per_pixel, - dsc_optc_cfg.slice_width, - &dsc_packed_pps[0]); + dsc_optc_cfg.slice_width); + + /* PPS SDP is set elsewhere because it has to be done after DIG FE is connected to DIG BE */ + } /* Enable DSC in OPTC */ + DC_LOG_DSC("Setting optc DSC config for tg instance %d:", pipe_ctx->stream_res.tg->inst); + dsc_optc_config_log(dsc, &dsc_optc_cfg); pipe_ctx->stream_res.tg->funcs->set_dsc_config(pipe_ctx->stream_res.tg, optc_dsc_mode, dsc_optc_cfg.bytes_per_pixel, @@ -452,15 +442,18 @@ static void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) OPTC_DSC_DISABLED, 0, 0); /* disable DSC in stream encoder */ - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config( pipe_ctx->stream_res.stream_enc, - OPTC_DSC_DISABLED, 0, 0, NULL); + OPTC_DSC_DISABLED, 0, 0); + + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, false, NULL); } /* disable DSC block */ pipe_ctx->stream_res.dsc->funcs->dsc_disable(pipe_ctx->stream_res.dsc); - if (odm_pipe) + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) odm_pipe->stream_res.dsc->funcs->dsc_disable(odm_pipe->stream_res.dsc); } } @@ -489,6 +482,47 @@ out: return result; } +bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable) +{ + struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; + struct dc *core_dc = pipe_ctx->stream->ctx->dc; + struct dc_stream_state *stream = pipe_ctx->stream; + + if (!pipe_ctx->stream->timing.flags.DSC || !dsc) + return false; + + if (enable) { + struct dsc_config dsc_cfg; + uint8_t dsc_packed_pps[128]; + + /* Enable DSC hw block */ + dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; + dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; + dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; + dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + + DC_LOG_DSC(" "); + dsc->funcs->dsc_get_packed_pps(dsc, &dsc_cfg, &dsc_packed_pps[0]); + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + DC_LOG_DSC("Setting stream encoder DSC PPS SDP for engine %d\n", (int)pipe_ctx->stream_res.stream_enc->id); + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, + true, + &dsc_packed_pps[0]); + } + } else { + /* disable DSC PPS in stream encoder */ + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, false, NULL); + } + } + + return true; +} + + bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx) { struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; @@ -499,8 +533,8 @@ bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx) return false; dp_set_dsc_on_stream(pipe_ctx, true); + dp_set_dsc_pps_sdp(pipe_ctx, true); return true; } - #endif diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 2ceaab4fb5de..8f70295179ff 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -52,6 +52,9 @@ #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/dcn20_resource.h" #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#include "dcn21/dcn21_resource.h" +#endif #include "dce120/dce120_resource.h" #define DC_LOGGER_INIT(logger) @@ -101,6 +104,10 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) dc_version = DCN_VERSION_1_0; if (ASICREV_IS_RAVEN2(asic_id.hw_internal_rev)) dc_version = DCN_VERSION_1_01; +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + if (ASICREV_IS_RENOIR(asic_id.hw_internal_rev)) + dc_version = DCN_VERSION_2_1; +#endif break; #endif @@ -168,17 +175,20 @@ struct resource_pool *dc_create_resource_pool(struct dc *dc, res_pool = dcn20_create_resource_pool(init_data, dc); break; #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + case DCN_VERSION_2_1: + res_pool = dcn21_create_resource_pool(init_data, dc); + break; +#endif default: break; } - if (res_pool != NULL) { - struct dc_firmware_info fw_info = { { 0 } }; - if (dc->ctx->dc_bios->funcs->get_firmware_info(dc->ctx->dc_bios, - &fw_info) == BP_RESULT_OK) { + if (res_pool != NULL) { + if (dc->ctx->dc_bios->fw_info_valid) { res_pool->ref_clocks.xtalin_clock_inKhz = - fw_info.pll_info.crystal_frequency; + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; /* initialize with firmware data first, no all * ASIC have DCCG SW component. FPGA or * simulation need initialization of @@ -265,12 +275,10 @@ bool resource_construct( DC_ERR("DC: failed to create audio!\n"); return false; } - if (!aud->funcs->endpoint_valid(aud)) { aud->funcs->destroy(&aud); break; } - pool->audios[i] = aud; pool->audio_count++; } @@ -940,7 +948,14 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx) data->inits.v_c_bot = dc_fixpt_add(data->inits.v_c, data->ratios.vert_c); } +static bool are_rect_integer_multiples(struct rect src, struct rect dest) +{ + if (dest.width >= src.width && dest.width % src.width == 0 && + dest.height >= src.height && dest.height % src.height == 0) + return true; + return false; +} bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; @@ -983,6 +998,15 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) if (pipe_ctx->plane_res.dpp != NULL) res = pipe_ctx->plane_res.dpp->funcs->dpp_get_optimal_number_of_taps( pipe_ctx->plane_res.dpp, &pipe_ctx->plane_res.scl_data, &plane_state->scaling_quality); + + if (res && + plane_state->scaling_quality.integer_scaling && + are_rect_integer_multiples(pipe_ctx->plane_res.scl_data.viewport, + pipe_ctx->plane_res.scl_data.recout)) { + pipe_ctx->plane_res.scl_data.taps.v_taps = 1; + pipe_ctx->plane_res.scl_data.taps.h_taps = 1; + } + if (!res) { /* Try 24 bpp linebuffer */ pipe_ctx->plane_res.scl_data.lb_params.depth = LB_PIXEL_DEPTH_24BPP; @@ -1103,25 +1127,21 @@ struct pipe_ctx *resource_get_head_pipe_for_stream( struct dc_stream_state *stream) { int i; + for (i = 0; i < MAX_PIPES; i++) { - if (res_ctx->pipe_ctx[i].stream == stream && - !res_ctx->pipe_ctx[i].top_pipe) { + if (res_ctx->pipe_ctx[i].stream == stream + && !res_ctx->pipe_ctx[i].top_pipe + && !res_ctx->pipe_ctx[i].prev_odm_pipe) return &res_ctx->pipe_ctx[i]; - break; - } } return NULL; } -static struct pipe_ctx *resource_get_tail_pipe_for_stream( +static struct pipe_ctx *resource_get_tail_pipe( struct resource_context *res_ctx, - struct dc_stream_state *stream) + struct pipe_ctx *head_pipe) { - struct pipe_ctx *head_pipe, *tail_pipe; - head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream); - - if (!head_pipe) - return NULL; + struct pipe_ctx *tail_pipe; tail_pipe = head_pipe->bottom_pipe; @@ -1137,31 +1157,20 @@ static struct pipe_ctx *resource_get_tail_pipe_for_stream( * A free_pipe for a stream is defined here as a pipe * that has no surface attached yet */ -static struct pipe_ctx *acquire_free_pipe_for_stream( +static struct pipe_ctx *acquire_free_pipe_for_head( struct dc_state *context, const struct resource_pool *pool, - struct dc_stream_state *stream) + struct pipe_ctx *head_pipe) { int i; struct resource_context *res_ctx = &context->res_ctx; - struct pipe_ctx *head_pipe = NULL; - - /* Find head pipe, which has the back end set up*/ - - head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream); - - if (!head_pipe) { - ASSERT(0); - return NULL; - } - if (!head_pipe->plane_state) return head_pipe; /* Re-use pipe already acquired for this stream if available*/ for (i = pool->pipe_count - 1; i >= 0; i--) { - if (res_ctx->pipe_ctx[i].stream == stream && + if (res_ctx->pipe_ctx[i].stream == head_pipe->stream && !res_ctx->pipe_ctx[i].plane_state) { return &res_ctx->pipe_ctx[i]; } @@ -1175,8 +1184,7 @@ static struct pipe_ctx *acquire_free_pipe_for_stream( if (!pool->funcs->acquire_idle_pipe_for_layer) return NULL; - return pool->funcs->acquire_idle_pipe_for_layer(context, pool, stream); - + return pool->funcs->acquire_idle_pipe_for_layer(context, pool, head_pipe->stream); } #if defined(CONFIG_DRM_AMD_DC_DCN1_0) @@ -1190,7 +1198,7 @@ static int acquire_first_split_pipe( for (i = 0; i < pool->pipe_count; i++) { struct pipe_ctx *split_pipe = &res_ctx->pipe_ctx[i]; - if (split_pipe->top_pipe && !dc_res_is_odm_head_pipe(split_pipe) && + if (split_pipe->top_pipe && split_pipe->top_pipe->plane_state == split_pipe->plane_state) { split_pipe->top_pipe->bottom_pipe = split_pipe->bottom_pipe; if (split_pipe->bottom_pipe) @@ -1251,39 +1259,41 @@ bool dc_add_plane_to_context( return false; } - tail_pipe = resource_get_tail_pipe_for_stream(&context->res_ctx, stream); - ASSERT(tail_pipe); + /* retain new surface, but only once per stream */ + dc_plane_state_retain(plane_state); - free_pipe = acquire_free_pipe_for_stream(context, pool, stream); + while (head_pipe) { + tail_pipe = resource_get_tail_pipe(&context->res_ctx, head_pipe); + ASSERT(tail_pipe); -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) - if (!free_pipe) { - int pipe_idx = acquire_first_split_pipe(&context->res_ctx, pool, stream); - if (pipe_idx >= 0) - free_pipe = &context->res_ctx.pipe_ctx[pipe_idx]; - } -#endif - if (!free_pipe) - return false; + free_pipe = acquire_free_pipe_for_head(context, pool, head_pipe); - /* retain new surfaces */ - dc_plane_state_retain(plane_state); - free_pipe->plane_state = plane_state; - - if (head_pipe != free_pipe) { - free_pipe->stream_res.tg = tail_pipe->stream_res.tg; - free_pipe->stream_res.abm = tail_pipe->stream_res.abm; - free_pipe->stream_res.opp = tail_pipe->stream_res.opp; - free_pipe->stream_res.stream_enc = tail_pipe->stream_res.stream_enc; - free_pipe->stream_res.audio = tail_pipe->stream_res.audio; - free_pipe->clock_source = tail_pipe->clock_source; - free_pipe->top_pipe = tail_pipe; - tail_pipe->bottom_pipe = free_pipe; - } else if (free_pipe->bottom_pipe && free_pipe->bottom_pipe->plane_state == NULL) { - ASSERT(free_pipe->bottom_pipe->stream_res.opp != free_pipe->stream_res.opp); - free_pipe->bottom_pipe->plane_state = plane_state; - } + #if defined(CONFIG_DRM_AMD_DC_DCN1_0) + if (!free_pipe) { + int pipe_idx = acquire_first_split_pipe(&context->res_ctx, pool, stream); + if (pipe_idx >= 0) + free_pipe = &context->res_ctx.pipe_ctx[pipe_idx]; + } + #endif + if (!free_pipe) { + dc_plane_state_release(plane_state); + return false; + } + free_pipe->plane_state = plane_state; + + if (head_pipe != free_pipe) { + free_pipe->stream_res.tg = tail_pipe->stream_res.tg; + free_pipe->stream_res.abm = tail_pipe->stream_res.abm; + free_pipe->stream_res.opp = tail_pipe->stream_res.opp; + free_pipe->stream_res.stream_enc = tail_pipe->stream_res.stream_enc; + free_pipe->stream_res.audio = tail_pipe->stream_res.audio; + free_pipe->clock_source = tail_pipe->clock_source; + free_pipe->top_pipe = tail_pipe; + tail_pipe->bottom_pipe = free_pipe; + } + head_pipe = head_pipe->next_odm_pipe; + } /* assign new surfaces*/ stream_status->plane_states[stream_status->plane_count] = plane_state; @@ -1292,35 +1302,6 @@ bool dc_add_plane_to_context( return true; } -struct pipe_ctx *dc_res_get_odm_bottom_pipe(struct pipe_ctx *pipe_ctx) -{ - struct pipe_ctx *bottom_pipe = pipe_ctx->bottom_pipe; - - /* ODM should only be updated once per otg */ - if (pipe_ctx->top_pipe) - return NULL; - - while (bottom_pipe) { - if (bottom_pipe->stream_res.opp != pipe_ctx->stream_res.opp) - break; - bottom_pipe = bottom_pipe->bottom_pipe; - } - - return bottom_pipe; -} - -bool dc_res_is_odm_head_pipe(struct pipe_ctx *pipe_ctx) -{ - struct pipe_ctx *top_pipe = pipe_ctx->top_pipe; - - if (!top_pipe) - return false; - if (top_pipe && top_pipe->stream_res.opp == pipe_ctx->stream_res.opp) - return false; - - return true; -} - bool dc_remove_plane_from_context( const struct dc *dc, struct dc_stream_state *stream, @@ -1347,12 +1328,6 @@ bool dc_remove_plane_from_context( struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; if (pipe_ctx->plane_state == plane_state) { - if (dc_res_is_odm_head_pipe(pipe_ctx)) { - pipe_ctx->plane_state = NULL; - pipe_ctx->bottom_pipe = NULL; - continue; - } - if (pipe_ctx->top_pipe) pipe_ctx->top_pipe->bottom_pipe = pipe_ctx->bottom_pipe; @@ -1367,13 +1342,10 @@ bool dc_remove_plane_from_context( * For head pipe detach surfaces from pipe for tail * pipe just zero it out */ - if (!pipe_ctx->top_pipe) { + if (!pipe_ctx->top_pipe) pipe_ctx->plane_state = NULL; - if (!dc_res_get_odm_bottom_pipe(pipe_ctx)) - pipe_ctx->bottom_pipe = NULL; - } else { + else memset(pipe_ctx, 0, sizeof(*pipe_ctx)); - } } } @@ -1659,24 +1631,25 @@ static struct audio *find_first_free_audio( const struct resource_pool *pool, enum engine_id id) { - int i; - for (i = 0; i < pool->audio_count; i++) { + int i, available_audio_count; + + available_audio_count = pool->audio_count; + + for (i = 0; i < available_audio_count; i++) { if ((res_ctx->is_audio_acquired[i] == false) && (res_ctx->is_stream_enc_acquired[i] == true)) { /*we have enough audio endpoint, find the matching inst*/ if (id != i) continue; - return pool->audios[i]; } } - /* use engine id to find free audio */ - if ((id < pool->audio_count) && (res_ctx->is_audio_acquired[id] == false)) { + /* use engine id to find free audio */ + if ((id < available_audio_count) && (res_ctx->is_audio_acquired[id] == false)) { return pool->audios[id]; } - /*not found the matching one, first come first serve*/ - for (i = 0; i < pool->audio_count; i++) { + for (i = 0; i < available_audio_count; i++) { if (res_ctx->is_audio_acquired[i] == false) { return pool->audios[i]; } @@ -1736,50 +1709,45 @@ enum dc_status dc_remove_stream_from_ctx( { int i; struct dc_context *dc_ctx = dc->ctx; - struct pipe_ctx *del_pipe = NULL; + struct pipe_ctx *del_pipe = resource_get_head_pipe_for_stream(&new_ctx->res_ctx, stream); + struct pipe_ctx *odm_pipe; - /* Release primary pipe */ - for (i = 0; i < MAX_PIPES; i++) { - if (new_ctx->res_ctx.pipe_ctx[i].stream == stream && - !new_ctx->res_ctx.pipe_ctx[i].top_pipe) { - struct pipe_ctx *odm_pipe = - dc_res_get_odm_bottom_pipe(&new_ctx->res_ctx.pipe_ctx[i]); + if (!del_pipe) { + DC_ERROR("Pipe not found for stream %p !\n", stream); + return DC_ERROR_UNEXPECTED; + } - del_pipe = &new_ctx->res_ctx.pipe_ctx[i]; + odm_pipe = del_pipe->next_odm_pipe; - ASSERT(del_pipe->stream_res.stream_enc); - update_stream_engine_usage( - &new_ctx->res_ctx, - dc->res_pool, - del_pipe->stream_res.stream_enc, - false); + /* Release primary pipe */ + ASSERT(del_pipe->stream_res.stream_enc); + update_stream_engine_usage( + &new_ctx->res_ctx, + dc->res_pool, + del_pipe->stream_res.stream_enc, + false); - if (del_pipe->stream_res.audio) - update_audio_usage( - &new_ctx->res_ctx, - dc->res_pool, - del_pipe->stream_res.audio, - false); + if (del_pipe->stream_res.audio) + update_audio_usage( + &new_ctx->res_ctx, + dc->res_pool, + del_pipe->stream_res.audio, + false); - resource_unreference_clock_source(&new_ctx->res_ctx, - dc->res_pool, - del_pipe->clock_source); + resource_unreference_clock_source(&new_ctx->res_ctx, + dc->res_pool, + del_pipe->clock_source); - if (dc->res_pool->funcs->remove_stream_from_ctx) - dc->res_pool->funcs->remove_stream_from_ctx(dc, new_ctx, stream); + if (dc->res_pool->funcs->remove_stream_from_ctx) + dc->res_pool->funcs->remove_stream_from_ctx(dc, new_ctx, stream); - memset(del_pipe, 0, sizeof(*del_pipe)); - if (odm_pipe) - memset(odm_pipe, 0, sizeof(*odm_pipe)); + while (odm_pipe) { + struct pipe_ctx *next_odm_pipe = odm_pipe->next_odm_pipe; - break; - } - } - - if (!del_pipe) { - DC_ERROR("Pipe not found for stream %p !\n", stream); - return DC_ERROR_UNEXPECTED; + memset(odm_pipe, 0, sizeof(*odm_pipe)); + odm_pipe = next_odm_pipe; } + memset(del_pipe, 0, sizeof(*del_pipe)); for (i = 0; i < new_ctx->stream_count; i++) if (new_ctx->streams[i] == stream) @@ -1880,7 +1848,7 @@ static int acquire_resource_from_hw_enabled_state( struct dc_stream_state *stream) { struct dc_link *link = stream->link; - unsigned int inst; + unsigned int inst, tg_inst; /* Check for enabled DIG to identify enabled display */ if (!link->link_enc->funcs->is_dig_enabled(link->link_enc)) @@ -1892,28 +1860,37 @@ static int acquire_resource_from_hw_enabled_state( * current implementation always map 1-to-1, so this code makes * the same assumption and doesn't check OTG source. */ - inst = link->link_enc->funcs->get_dig_frontend(link->link_enc) - 1; + inst = link->link_enc->funcs->get_dig_frontend(link->link_enc); /* Instance should be within the range of the pool */ if (inst >= pool->pipe_count) return -1; - if (!res_ctx->pipe_ctx[inst].stream) { - struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[inst]; + if (inst >= pool->stream_enc_count) + return -1; + + tg_inst = pool->stream_enc[inst]->funcs->dig_source_otg(pool->stream_enc[inst]); - pipe_ctx->stream_res.tg = pool->timing_generators[inst]; - pipe_ctx->plane_res.mi = pool->mis[inst]; - pipe_ctx->plane_res.hubp = pool->hubps[inst]; - pipe_ctx->plane_res.ipp = pool->ipps[inst]; - pipe_ctx->plane_res.xfm = pool->transforms[inst]; - pipe_ctx->plane_res.dpp = pool->dpps[inst]; - pipe_ctx->stream_res.opp = pool->opps[inst]; - if (pool->dpps[inst]) - pipe_ctx->plane_res.mpcc_inst = pool->dpps[inst]->inst; - pipe_ctx->pipe_idx = inst; + if (tg_inst >= pool->timing_generator_count) + return false; + + if (!res_ctx->pipe_ctx[tg_inst].stream) { + struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[tg_inst]; + + pipe_ctx->stream_res.tg = pool->timing_generators[tg_inst]; + pipe_ctx->plane_res.mi = pool->mis[tg_inst]; + pipe_ctx->plane_res.hubp = pool->hubps[tg_inst]; + pipe_ctx->plane_res.ipp = pool->ipps[tg_inst]; + pipe_ctx->plane_res.xfm = pool->transforms[tg_inst]; + pipe_ctx->plane_res.dpp = pool->dpps[tg_inst]; + pipe_ctx->stream_res.opp = pool->opps[tg_inst]; + + if (pool->dpps[tg_inst]) + pipe_ctx->plane_res.mpcc_inst = pool->dpps[tg_inst]->inst; + pipe_ctx->pipe_idx = tg_inst; pipe_ctx->stream = stream; - return inst; + return tg_inst; } return -1; @@ -2475,6 +2452,12 @@ void dc_resource_state_copy_construct( if (cur_pipe->bottom_pipe) cur_pipe->bottom_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx]; + + if (cur_pipe->next_odm_pipe) + cur_pipe->next_odm_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->next_odm_pipe->pipe_idx]; + + if (cur_pipe->prev_odm_pipe) + cur_pipe->prev_odm_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->prev_odm_pipe->pipe_idx]; } for (i = 0; i < dst_ctx->stream_count; i++) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 352862370390..bf1d7bb90e0f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -566,6 +566,7 @@ bool dc_stream_get_scanoutpos(const struct dc_stream_state *stream, return ret; } + #if defined(CONFIG_DRM_AMD_DC_DCN2_0) bool dc_stream_dmdata_status_done(struct dc *dc, struct dc_stream_state *stream) { @@ -597,6 +598,14 @@ bool dc_stream_set_dynamic_metadata(struct dc *dc, struct hubp *hubp; int i; + /* Dynamic metadata is only supported on HDMI or DP */ + if (!dc_is_hdmi_signal(stream->signal) && !dc_is_dp_signal(stream->signal)) + return false; + + /* Check hardware support */ + if (!dc->hwss.program_dmdata_engine) + return false; + for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream == stream) @@ -612,23 +621,7 @@ bool dc_stream_set_dynamic_metadata(struct dc *dc, pipe_ctx->stream->dmdata_address = attr->address; - if (pipe_ctx->stream_res.stream_enc && - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata != NULL) { - if (pipe_ctx->stream->dmdata_address.quad_part != 0) { - /* if using dynamic meta, don't set up generic infopackets */ - pipe_ctx->stream_res.encoder_info_frame.hdrsmd.valid = false; - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - true, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } else - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - false, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } + dc->hwss.program_dmdata_engine(pipe_ctx); if (hubp->funcs->dmdata_set_attributes != NULL && pipe_ctx->stream->dmdata_address.quad_part != 0) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c index f40e4fd52fa2..b9d6a5bd8522 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c @@ -60,7 +60,6 @@ static void construct(struct dc_context *ctx, struct dc_plane_state *plane_state plane_state->lut3d_func = dc_create_3dlut_func(); if (plane_state->lut3d_func != NULL) { plane_state->lut3d_func->ctx = ctx; - plane_state->lut3d_func->initialized = false; } plane_state->blend_tf = dc_create_transfer_func(); if (plane_state->blend_tf != NULL) { @@ -279,7 +278,7 @@ struct dc_3dlut *dc_create_3dlut_func(void) goto alloc_fail; kref_init(&lut->refcount); - lut->initialized = false; + lut->state.raw = 0; return lut; diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index e513028faefa..a82352a87808 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -39,7 +39,7 @@ #include "inc/hw/dmcu.h" #include "dml/display_mode_lib.h" -#define DC_VER "3.2.35" +#define DC_VER "3.2.48" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -121,6 +121,7 @@ struct dc_caps { struct dc_bug_wa { bool no_connect_phy_config; bool dedcn20_305_wa; + bool skip_clock_update; }; #endif @@ -219,7 +220,7 @@ struct dc_config { bool power_down_display_on_boot; bool edp_not_connected; bool forced_clocks; - + bool multi_mon_pp_mclk_switch; }; enum visual_confirm { @@ -252,7 +253,10 @@ enum wm_report_mode { struct dc_clocks { int dispclk_khz; int max_supported_dppclk_khz; + int max_supported_dispclk_khz; int dppclk_khz; + int bw_dppclk_khz; /*a copy of dppclk_khz*/ + int bw_dispclk_khz; int dcfclk_khz; int socclk_khz; int dcfclk_deep_sleep_khz; @@ -260,6 +264,12 @@ struct dc_clocks { int phyclk_khz; int dramclk_khz; bool p_state_change_support; + + /* + * Elements below are not compared for the purposes of + * optimization required + */ + bool prev_p_state_change_support; }; struct dc_bw_validation_profile { @@ -341,6 +351,7 @@ struct dc_debug_options { bool disable_pplib_wm_range; enum wm_report_mode pplib_wm_report_mode; unsigned int min_disp_clk_khz; + unsigned int min_dpp_clk_khz; int sr_exit_time_dpm0_ns; int sr_enter_plus_exit_time_dpm0_ns; int sr_exit_time_ns; @@ -367,6 +378,7 @@ struct dc_debug_options { bool scl_reset_length10; bool hdmi20_disable; bool skip_detection_link_training; + bool remove_disconnect_edp; unsigned int force_odm_combine; //bit vector based on otg inst unsigned int force_fclk_khz; bool disable_tri_buf; @@ -374,10 +386,18 @@ struct dc_debug_options { #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT bool disable_fec; #endif +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 + bool disable_48mhz_pwrdwn; +#endif /* This forces a hard min on the DCFCLK requested to SMU/PP * watermarks are not affected. */ unsigned int force_min_dcfclk_mhz; + bool disable_timing_sync; +#if defined(CONFIG_DRM_AMD_DC_DCN2_0) + bool cm_in_bypass; +#endif + int force_clock_mode;/*every mode change.*/ }; struct dc_debug_data { @@ -406,6 +426,7 @@ struct dc_phy_addr_space_config { } gart_config; bool valid; + uint64_t page_table_default_page_addr; }; struct dc_virtual_addr_space_config { @@ -597,9 +618,12 @@ enum dc_transfer_func_predefined { TRANSFER_FUNCTION_UNITY, TRANSFER_FUNCTION_HLG, TRANSFER_FUNCTION_HLG12, - TRANSFER_FUNCTION_GAMMA22 + TRANSFER_FUNCTION_GAMMA22, + TRANSFER_FUNCTION_GAMMA24, + TRANSFER_FUNCTION_GAMMA26 }; + struct dc_transfer_func { struct kref refcount; enum dc_transfer_func_type type; @@ -615,12 +639,26 @@ struct dc_transfer_func { #if defined(CONFIG_DRM_AMD_DC_DCN2_0) +union dc_3dlut_state { + struct { + uint32_t initialized:1; /*if 3dlut is went through color module for initialization */ + uint32_t rmu_idx_valid:1; /*if mux settings are valid*/ + uint32_t rmu_mux_num:3; /*index of mux to use*/ + uint32_t mpc_rmu0_mux:4; /*select mpcc on mux, one of the following : mpcc0, mpcc1, mpcc2, mpcc3*/ + uint32_t mpc_rmu1_mux:4; + uint32_t mpc_rmu2_mux:4; + uint32_t reserved:15; + } bits; + uint32_t raw; +}; + struct dc_3dlut { struct kref refcount; struct tetrahedral_params lut_3d; uint32_t hdr_multiplier; - bool initialized; + bool initialized; /*remove after diag fix*/ + union dc_3dlut_state state; struct dc_context *ctx; }; #endif @@ -682,7 +720,7 @@ struct dc_plane_state { struct rect dst_rect; struct rect clip_rect; - union plane_size plane_size; + struct plane_size plane_size; union dc_tiling_info tiling_info; struct dc_plane_dcc_param dcc; @@ -716,6 +754,7 @@ struct dc_plane_state { bool visible; bool flip_immediate; bool horizontal_mirror; + int layer_index; union surface_update_flags update_flags; /* private to DC core */ @@ -731,7 +770,7 @@ struct dc_plane_state { }; struct dc_plane_info { - union plane_size plane_size; + struct plane_size plane_size; union dc_tiling_info tiling_info; struct dc_plane_dcc_param dcc; enum surface_pixel_format format; @@ -745,6 +784,7 @@ struct dc_plane_info { bool global_alpha; int global_alpha_value; bool input_csc_enabled; + int layer_index; }; struct dc_scaling_info { @@ -834,6 +874,9 @@ enum dc_status dc_validate_plane(struct dc *dc, const struct dc_plane_state *pla void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx *info); +bool dc_set_generic_gpio_for_stereo(bool enable, + struct gpio_service *gpio_service); + /* * fast_validate: we return after determining if we can support the new state, * but before we populate the programming info @@ -1020,6 +1063,8 @@ unsigned int dc_get_target_backlight_pwm(struct dc *dc); bool dc_is_dmcu_initialized(struct dc *dc); +enum dc_status dc_set_clock(struct dc *dc, enum dc_clock_type clock_type, uint32_t clk_khz, uint32_t stepping); +void dc_get_clock(struct dc *dc, enum dc_clock_type clock_type, struct dc_clock_config *clock_cfg); #if defined(CONFIG_DRM_AMD_DC_DSC_SUPPORT) /******************************************************************************* * DSC Interfaces diff --git a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h index 78c3b300ec45..b1dd0d60d98e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h @@ -61,9 +61,6 @@ struct dc_vbios_funcs { struct graphics_object_id connector_object_id, uint32_t device_tag_index, struct connector_device_tag_info *info); - enum bp_result (*get_firmware_info)( - struct dc_bios *bios, - struct dc_firmware_info *info); enum bp_result (*get_spread_spectrum_info)( struct dc_bios *bios, enum as_signal_type signal, @@ -152,6 +149,8 @@ struct dc_bios { struct dc_context *ctx; const struct bios_registers *regs; struct integrated_info *integrated_info; + struct dc_firmware_info fw_info; + bool fw_info_valid; }; #endif /* DC_BIOS_TYPES_H */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index dfcec4d3e9c0..ef79a686e4c2 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -90,6 +90,13 @@ enum dc_post_cursor2 { POST_CURSOR2_MAX_LEVEL = POST_CURSOR2_LEVEL3, }; +enum dc_dp_training_pattern { + DP_TRAINING_PATTERN_SEQUENCE_1 = 0, + DP_TRAINING_PATTERN_SEQUENCE_2, + DP_TRAINING_PATTERN_SEQUENCE_3, + DP_TRAINING_PATTERN_SEQUENCE_4, +}; + struct dc_link_settings { enum dc_lane_count lane_count; enum dc_link_rate link_rate; @@ -109,6 +116,23 @@ struct dc_link_training_settings { struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]; }; +struct dc_link_training_overrides { + enum dc_voltage_swing *voltage_swing; + enum dc_pre_emphasis *pre_emphasis; + enum dc_post_cursor2 *post_cursor2; + + uint16_t *cr_pattern_time; + uint16_t *eq_pattern_time; + enum dc_dp_training_pattern *pattern_for_eq; + + enum dc_link_spread *downspread; + bool *alternate_scrambler_reset; + bool *enhanced_framing; + bool *mst_enable; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + bool *fec_enable; +#endif +}; union dpcd_rev { struct { diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index 22db5682aa6c..0b8700a8a94a 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -115,42 +115,40 @@ struct rect { int height; }; -union plane_size { - /* Grph or Video will be selected - * based on format above: - * Use Video structure if - * format >= DalPixelFormat_VideoBegin - * else use Grph structure +struct plane_size { + /* Graphic surface pitch in pixels. + * In LINEAR_GENERAL mode, pitch + * is 32 pixel aligned. */ - struct { - struct rect surface_size; - /* Graphic surface pitch in pixels. - * In LINEAR_GENERAL mode, pitch - * is 32 pixel aligned. - */ - int surface_pitch; - } grph; + int surface_pitch; + int chroma_pitch; + struct rect surface_size; + struct rect chroma_size; - struct { - struct rect luma_size; - /* Graphic surface pitch in pixels. - * In LINEAR_GENERAL mode, pitch is - * 32 pixel aligned. - */ - int luma_pitch; + union { + struct { + struct rect surface_size; + int surface_pitch; + } grph; - struct rect chroma_size; - /* Graphic surface pitch in pixels. - * In LINEAR_GENERAL mode, pitch is - * 32 pixel aligned. - */ - int chroma_pitch; - } video; + struct { + struct rect luma_size; + int luma_pitch; + struct rect chroma_size; + int chroma_pitch; + } video; + }; }; struct dc_plane_dcc_param { bool enable; + int meta_pitch; + bool independent_64b_blks; + + int meta_pitch_c; + bool independent_64b_blks_c; + union { struct { int meta_pitch; @@ -482,7 +480,6 @@ struct dc_gamma { * is_logical_identity indicates the given gamma ramp regardless of type is identity. */ bool is_identity; - bool is_logical_identity; }; /* Used by both ipp amd opp functions*/ @@ -519,7 +516,8 @@ union dc_cursor_attribute_flags { uint32_t INVERT_PIXEL_DATA:1; uint32_t ZERO_EXPANSION:1; uint32_t MIN_MAX_INVERT:1; - uint32_t RESERVED:25; + uint32_t ENABLE_CURSOR_DEGAMMA:1; + uint32_t RESERVED:24; } bits; uint32_t value; }; @@ -615,6 +613,7 @@ struct scaling_taps { uint32_t h_taps; uint32_t v_taps_c; uint32_t h_taps_c; + bool integer_scaling; }; enum dc_timing_standard { @@ -758,6 +757,8 @@ struct crtc_trigger_info { struct dc_crtc_timing_adjust { uint32_t v_total_min; uint32_t v_total_max; + uint32_t v_total_mid; + uint32_t v_total_mid_frame_num; }; #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index 6f0b80111e58..9ea75db3484e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -83,6 +83,8 @@ struct dc_link { bool is_hpd_filter_disabled; bool dp_ss_off; bool link_state_valid; + bool aux_access_disabled; + bool sync_lt_in_progress; /* caps is the same as reported_link_cap. link_traing use * reported_link_cap. Will clean up. TODO @@ -92,6 +94,7 @@ struct dc_link { struct dc_link_settings cur_link_settings; struct dc_lane_settings cur_lane_setting; struct dc_link_settings preferred_link_setting; + struct dc_link_training_overrides preferred_training_settings; uint8_t ddc_hw_inst; @@ -217,11 +220,24 @@ void dc_link_dp_set_drive_settings( struct dc_link *link, struct link_training_settings *lt_settings); +bool dc_link_dp_perform_link_training_skip_aux( + struct dc_link *link, + const struct dc_link_settings *link_setting); + enum link_training_result dc_link_dp_perform_link_training( struct dc_link *link, const struct dc_link_settings *link_setting, bool skip_video_pattern); +bool dc_link_dp_sync_lt_begin(struct dc_link *link); + +enum link_training_result dc_link_dp_sync_lt_attempt( + struct dc_link *link, + struct dc_link_settings *link_setting, + struct dc_link_training_overrides *lt_settings); + +bool dc_link_dp_sync_lt_end(struct dc_link *link, bool link_down); + void dc_link_dp_enable_hpd(const struct dc_link *link); void dc_link_dp_disable_hpd(const struct dc_link *link); @@ -251,6 +267,11 @@ void dc_link_perform_link_training(struct dc *dc, void dc_link_set_preferred_link_settings(struct dc *dc, struct dc_link_settings *link_setting, struct dc_link *link); +void dc_link_set_preferred_training_settings(struct dc *dc, + struct dc_link_settings *link_setting, + struct dc_link_training_overrides *lt_overrides, + struct dc_link *link, + bool skip_immediate_retrain); void dc_link_enable_hpd(const struct dc_link *link); void dc_link_disable_hpd(const struct dc_link *link); void dc_link_set_test_pattern(struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 6eabb6491a3d..b273735b6a3e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -202,6 +202,7 @@ struct dc_panel_patch { unsigned int dppowerup_delay; unsigned int extra_t12_ms; unsigned int extra_delay_backlight_off; + unsigned int extra_t7_ms; }; struct dc_edid_caps { @@ -725,6 +726,19 @@ struct AsicStateEx { unsigned int phyClock; }; + +enum dc_clock_type { + DC_CLOCK_TYPE_DISPCLK = 0, + DC_CLOCK_TYPE_DPPCLK = 1, +}; + +struct dc_clock_config { + uint32_t max_clock_khz; + uint32_t min_clock_khz; + uint32_t bw_requirequired_clock_khz; + uint32_t current_clock_khz;/*current clock in use*/ +}; + #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT /* DSC DPCD capabilities */ union dsc_slice_caps1 { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c index 4a10a5d22c90..5a35495bc11d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c @@ -145,20 +145,20 @@ static void check_audio_bandwidth_hdmi( if (channel_count > 2) { /* Based on HDMI spec 1.3 Table 7.5 */ - if ((crtc_info->requested_pixel_clock <= 27000) && + if ((crtc_info->requested_pixel_clock_100Hz <= 270000) && (crtc_info->v_active <= 576) && !(crtc_info->interlaced) && !(crtc_info->pixel_repetition == 2 || crtc_info->pixel_repetition == 4)) { limit_freq_to_48_khz = true; - } else if ((crtc_info->requested_pixel_clock <= 27000) && + } else if ((crtc_info->requested_pixel_clock_100Hz <= 270000) && (crtc_info->v_active <= 576) && (crtc_info->interlaced) && (crtc_info->pixel_repetition == 2)) { limit_freq_to_88_2_khz = true; - } else if ((crtc_info->requested_pixel_clock <= 54000) && + } else if ((crtc_info->requested_pixel_clock_100Hz <= 540000) && (crtc_info->v_active <= 576) && !(crtc_info->interlaced)) { limit_freq_to_174_4_khz = true; @@ -613,6 +613,8 @@ void dce_aud_az_configure( AZ_REG_WRITE(AZALIA_F0_CODEC_PIN_CONTROL_SINK_INFO1, value); + DC_LOG_HW_AUDIO("\n\tAUDIO:az_configure: index: %u data, 0x%x, displayName %s: \n", + audio->inst, value, audio_info->display_name); /* *write the port ID: @@ -737,8 +739,8 @@ void dce_aud_az_configure( /* search pixel clock value for Azalia HDMI Audio */ static void get_azalia_clock_info_hdmi( - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100hz, + uint32_t actual_pixel_clock_100Hz, struct azalia_clock_info *azalia_clock_info) { /* audio_dto_phase= 24 * 10,000; @@ -749,11 +751,11 @@ static void get_azalia_clock_info_hdmi( /* audio_dto_module = PCLKFrequency * 10,000; * [khz] -> [100Hz] */ azalia_clock_info->audio_dto_module = - actual_pixel_clock_in_khz * 10; + actual_pixel_clock_100Hz; } static void get_azalia_clock_info_dp( - uint32_t requested_pixel_clock_in_khz, + uint32_t requested_pixel_clock_100Hz, const struct audio_pll_info *pll_info, struct azalia_clock_info *azalia_clock_info) { @@ -792,15 +794,15 @@ void dce_aud_wall_dto_setup( /* calculate DTO settings */ get_azalia_clock_info_hdmi( - crtc_info->requested_pixel_clock, - crtc_info->calculated_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, + crtc_info->calculated_pixel_clock_100Hz, &clock_info); - DC_LOG_HW_AUDIO("\n%s:Input::requested_pixel_clock = %d"\ - "calculated_pixel_clock =%d\n"\ + DC_LOG_HW_AUDIO("\n%s:Input::requested_pixel_clock_100Hz = %d"\ + "calculated_pixel_clock_100Hz =%d\n"\ "audio_dto_module = %d audio_dto_phase =%d \n\n", __func__,\ - crtc_info->requested_pixel_clock,\ - crtc_info->calculated_pixel_clock,\ + crtc_info->requested_pixel_clock_100Hz,\ + crtc_info->calculated_pixel_clock_100Hz,\ clock_info.audio_dto_module,\ clock_info.audio_dto_phase); @@ -833,7 +835,7 @@ void dce_aud_wall_dto_setup( calculate DTO settings */ get_azalia_clock_info_dp( - crtc_info->requested_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, pll_info, &clock_info); @@ -922,7 +924,6 @@ static const struct audio_funcs funcs = { .az_configure = dce_aud_az_configure, .destroy = dce_aud_destroy, }; - void dce_aud_destroy(struct audio **audio) { struct dce_audio *aud = DCE_AUD(*audio); @@ -936,7 +937,7 @@ struct audio *dce_audio_create( unsigned int inst, const struct dce_audio_registers *reg, const struct dce_audio_shift *shifts, - const struct dce_aduio_mask *masks + const struct dce_audio_mask *masks ) { struct dce_audio *audio = kzalloc(sizeof(*audio), GFP_KERNEL); @@ -953,7 +954,6 @@ struct audio *dce_audio_create( audio->regs = reg; audio->shifts = shifts; audio->masks = masks; - return &audio->base; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h index a0d5724aab31..1392fab0860b 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h @@ -101,7 +101,7 @@ struct dce_audio_shift { uint32_t DCCG_AUDIO_DTO1_USE_512FBR_DTO; }; -struct dce_aduio_mask { +struct dce_audio_mask { uint32_t AZALIA_ENDPOINT_REG_INDEX; uint32_t AZALIA_ENDPOINT_REG_DATA; @@ -125,7 +125,7 @@ struct dce_audio { struct audio base; const struct dce_audio_registers *regs; const struct dce_audio_shift *shifts; - const struct dce_aduio_mask *masks; + const struct dce_audio_mask *masks; }; struct audio *dce_audio_create( @@ -133,7 +133,7 @@ struct audio *dce_audio_create( unsigned int inst, const struct dce_audio_registers *reg, const struct dce_audio_shift *shifts, - const struct dce_aduio_mask *masks); + const struct dce_audio_mask *masks); void dce_aud_destroy(struct audio **audio); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index f2295e780031..c3f9f4185ce8 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -30,6 +30,7 @@ #include "core_types.h" #include "dce_aux.h" #include "dce/dce_11_0_sh_mask.h" +#include "dm_event_log.h" #define CTX \ aux110->base.ctx @@ -252,6 +253,8 @@ static void submit_channel_request( } REG_UPDATE(AUX_SW_CONTROL, AUX_SW_GO, 1); + EVENT_LOG_AUX_REQ(engine->ddc->pin_data->en, EVENT_LOG_AUX_ORIGIN_NATIVE, + request->action, request->address, request->length, request->data); } static int read_channel_reply(struct dce_aux *engine, uint32_t size, @@ -480,9 +483,13 @@ int dce_aux_transfer_raw(struct ddc_service *ddc, *operation_result = get_channel_status(aux_engine, &returned_bytes); if (*operation_result == AUX_CHANNEL_OPERATION_SUCCEEDED) { - read_channel_reply(aux_engine, payload->length, + int bytes_replied = 0; + bytes_replied = read_channel_reply(aux_engine, payload->length, payload->data, payload->reply, &status); + EVENT_LOG_AUX_REP(aux_engine->ddc->pin_data->en, + EVENT_LOG_AUX_ORIGIN_NATIVE, *payload->reply, + bytes_replied, payload->data); res = returned_bytes; } else { res = -1; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 5fae77e201d5..f787a6b94781 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -1061,7 +1061,8 @@ static bool dcn20_program_pix_clk( static const struct clock_source_funcs dcn20_clk_src_funcs = { .cs_power_down = dce110_clock_source_power_down, .program_pix_clk = dcn20_program_pix_clk, - .get_pix_clk_dividers = dce112_get_pix_clk_dividers + .get_pix_clk_dividers = dce112_get_pix_clk_dividers, + .get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz }; #endif @@ -1234,37 +1235,36 @@ static bool calc_pll_max_vco_construct( struct calc_pll_clock_source_init_data *init_data) { uint32_t i; - struct dc_firmware_info fw_info = { { 0 } }; + struct dc_firmware_info *fw_info; if (calc_pll_cs == NULL || init_data == NULL || init_data->bp == NULL) return false; - if (init_data->bp->funcs->get_firmware_info( - init_data->bp, - &fw_info) != BP_RESULT_OK) + if (!init_data->bp->fw_info_valid) return false; + fw_info = &init_data->bp->fw_info; calc_pll_cs->ctx = init_data->ctx; - calc_pll_cs->ref_freq_khz = fw_info.pll_info.crystal_frequency; + calc_pll_cs->ref_freq_khz = fw_info->pll_info.crystal_frequency; calc_pll_cs->min_vco_khz = - fw_info.pll_info.min_output_pxl_clk_pll_frequency; + fw_info->pll_info.min_output_pxl_clk_pll_frequency; calc_pll_cs->max_vco_khz = - fw_info.pll_info.max_output_pxl_clk_pll_frequency; + fw_info->pll_info.max_output_pxl_clk_pll_frequency; if (init_data->max_override_input_pxl_clk_pll_freq_khz != 0) calc_pll_cs->max_pll_input_freq_khz = init_data->max_override_input_pxl_clk_pll_freq_khz; else calc_pll_cs->max_pll_input_freq_khz = - fw_info.pll_info.max_input_pxl_clk_pll_frequency; + fw_info->pll_info.max_input_pxl_clk_pll_frequency; if (init_data->min_override_input_pxl_clk_pll_freq_khz != 0) calc_pll_cs->min_pll_input_freq_khz = init_data->min_override_input_pxl_clk_pll_freq_khz; else calc_pll_cs->min_pll_input_freq_khz = - fw_info.pll_info.min_input_pxl_clk_pll_frequency; + fw_info->pll_info.min_input_pxl_clk_pll_frequency; calc_pll_cs->min_pix_clock_pll_post_divider = init_data->min_pix_clk_pll_post_divider; @@ -1316,7 +1316,6 @@ bool dce110_clk_src_construct( const struct dce110_clk_src_shift *cs_shift, const struct dce110_clk_src_mask *cs_mask) { - struct dc_firmware_info fw_info = { { 0 } }; struct calc_pll_clock_source_init_data calc_pll_cs_init_data_hdmi; struct calc_pll_clock_source_init_data calc_pll_cs_init_data; @@ -1329,14 +1328,12 @@ bool dce110_clk_src_construct( clk_src->cs_shift = cs_shift; clk_src->cs_mask = cs_mask; - if (clk_src->bios->funcs->get_firmware_info( - clk_src->bios, &fw_info) != BP_RESULT_OK) { + if (!clk_src->bios->fw_info_valid) { ASSERT_CRITICAL(false); goto unexpected_failure; } - clk_src->ext_clk_khz = - fw_info.external_clock_source_frequency_for_dp; + clk_src->ext_clk_khz = clk_src->bios->fw_info.external_clock_source_frequency_for_dp; /* structure normally used with PLL ranges from ATOMBIOS; DS on by default */ calc_pll_cs_init_data.bp = bios; @@ -1376,7 +1373,7 @@ bool dce110_clk_src_construct( FRACT_FB_DIVIDER_DEC_POINTS_MAX_NUM; calc_pll_cs_init_data_hdmi.ctx = ctx; - clk_src->ref_freq_khz = fw_info.pll_info.crystal_frequency; + clk_src->ref_freq_khz = clk_src->bios->fw_info.pll_info.crystal_frequency; if (clk_src->base.id == CLOCK_SOURCE_ID_EXTERNAL) return true; @@ -1419,8 +1416,6 @@ bool dce112_clk_src_construct( const struct dce110_clk_src_shift *cs_shift, const struct dce110_clk_src_mask *cs_mask) { - struct dc_firmware_info fw_info = { { 0 } }; - clk_src->base.ctx = ctx; clk_src->bios = bios; clk_src->base.id = id; @@ -1430,13 +1425,12 @@ bool dce112_clk_src_construct( clk_src->cs_shift = cs_shift; clk_src->cs_mask = cs_mask; - if (clk_src->bios->funcs->get_firmware_info( - clk_src->bios, &fw_info) != BP_RESULT_OK) { + if (!clk_src->bios->fw_info_valid) { ASSERT_CRITICAL(false); return false; } - clk_src->ext_clk_khz = fw_info.external_clock_source_frequency_for_dp; + clk_src->ext_clk_khz = clk_src->bios->fw_info.external_clock_source_frequency_for_dp; return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h index adae03b1f3a7..43c1bf60b83c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h @@ -78,6 +78,23 @@ SRII(PIXEL_RATE_CNTL, OTG, 5) #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#define CS_COMMON_REG_LIST_DCN2_1(index, pllid) \ + SRI(PIXCLK_RESYNC_CNTL, PHYPLL, pllid),\ + SRII(PHASE, DP_DTO, 0),\ + SRII(PHASE, DP_DTO, 1),\ + SRII(PHASE, DP_DTO, 2),\ + SRII(PHASE, DP_DTO, 3),\ + SRII(MODULO, DP_DTO, 0),\ + SRII(MODULO, DP_DTO, 1),\ + SRII(MODULO, DP_DTO, 2),\ + SRII(MODULO, DP_DTO, 3),\ + SRII(PIXEL_RATE_CNTL, OTG, 0),\ + SRII(PIXEL_RATE_CNTL, OTG, 1),\ + SRII(PIXEL_RATE_CNTL, OTG, 2),\ + SRII(PIXEL_RATE_CNTL, OTG, 3) +#endif + #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define CS_COMMON_MASK_SH_LIST_DCN2_0(mask_sh)\ CS_SF(DP_DTO0_PHASE, DP_DTO0_PHASE, mask_sh),\ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h index cb0a037b1c4a..ac04d77058f0 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h @@ -62,6 +62,10 @@ SRII(BLND_CONTROL, BLND, 4), \ SRII(BLND_CONTROL, BLND, 5) +#define HSWEQ_DCN_PIXEL_RATE_REG_LIST(blk, inst) \ + SRII(PIXEL_RATE_CNTL, blk, inst), \ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, inst) + #define HWSEQ_PIXEL_RATE_REG_LIST(blk) \ SRII(PIXEL_RATE_CNTL, blk, 0), \ SRII(PIXEL_RATE_CNTL, blk, 1), \ @@ -151,7 +155,10 @@ SR(DCCG_GATE_DISABLE_CNTL2), \ SR(DCFCLK_CNTL),\ SR(DCFCLK_CNTL), \ - SR(DC_MEM_GLOBAL_PWR_REQ_CNTL), \ + SR(DC_MEM_GLOBAL_PWR_REQ_CNTL) + + +#define MMHUB_DCN_REG_LIST()\ /* todo: get these from GVM instead of reading registers ourselves */\ MMHUB_SR(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32),\ MMHUB_SR(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32),\ @@ -166,10 +173,14 @@ MMHUB_SR(MC_VM_SYSTEM_APERTURE_LOW_ADDR),\ MMHUB_SR(MC_VM_SYSTEM_APERTURE_HIGH_ADDR) + #define HWSEQ_DCN1_REG_LIST()\ HWSEQ_DCN_REG_LIST(), \ - HWSEQ_PIXEL_RATE_REG_LIST(OTG), \ - HWSEQ_PHYPLL_REG_LIST(OTG), \ + MMHUB_DCN_REG_LIST(), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 0), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 1), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 2), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 3), \ SR(DCHUBBUB_SDPIF_FB_BASE),\ SR(DCHUBBUB_SDPIF_FB_OFFSET),\ SR(DCHUBBUB_SDPIF_AGP_BASE),\ @@ -202,8 +213,12 @@ #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define HWSEQ_DCN2_REG_LIST()\ HWSEQ_DCN_REG_LIST(), \ - HWSEQ_PIXEL_RATE_REG_LIST(OTG), \ - HWSEQ_PHYPLL_REG_LIST(OTG), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 0), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 1), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 2), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 3), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 4), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 5), \ SR(MICROSECOND_TIME_BASE_DIV), \ SR(MILLISECOND_TIME_BASE_DIV), \ SR(DISPCLK_FREQ_CHANGE_CNTL), \ @@ -227,8 +242,8 @@ SR(DOMAIN7_PG_CONFIG), \ SR(DOMAIN8_PG_CONFIG), \ SR(DOMAIN9_PG_CONFIG), \ - SR(DOMAIN10_PG_CONFIG), \ - SR(DOMAIN11_PG_CONFIG), \ +/* SR(DOMAIN10_PG_CONFIG), Navi1x HUBP5 not powergate-able*/\ +/* SR(DOMAIN11_PG_CONFIG), Navi1x DPP5 is not powergate-able */\ SR(DOMAIN16_PG_CONFIG), \ SR(DOMAIN17_PG_CONFIG), \ SR(DOMAIN18_PG_CONFIG), \ @@ -263,6 +278,59 @@ BL_REG_LIST() #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#define HWSEQ_DCN21_REG_LIST()\ + HWSEQ_DCN_REG_LIST(), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 0), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 1), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 2), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 3), \ + MMHUB_DCN_REG_LIST(), \ + SR(MICROSECOND_TIME_BASE_DIV), \ + SR(MILLISECOND_TIME_BASE_DIV), \ + SR(DISPCLK_FREQ_CHANGE_CNTL), \ + SR(RBBMIF_TIMEOUT_DIS), \ + SR(RBBMIF_TIMEOUT_DIS_2), \ + SR(DCHUBBUB_CRC_CTRL), \ + SR(DPP_TOP0_DPP_CRC_CTRL), \ + SR(DPP_TOP0_DPP_CRC_VAL_B_A), \ + SR(DPP_TOP0_DPP_CRC_VAL_R_G), \ + SR(MPC_CRC_CTRL), \ + SR(MPC_CRC_RESULT_GB), \ + SR(MPC_CRC_RESULT_C), \ + SR(MPC_CRC_RESULT_AR), \ + SR(DOMAIN0_PG_CONFIG), \ + SR(DOMAIN1_PG_CONFIG), \ + SR(DOMAIN2_PG_CONFIG), \ + SR(DOMAIN3_PG_CONFIG), \ + SR(DOMAIN4_PG_CONFIG), \ + SR(DOMAIN5_PG_CONFIG), \ + SR(DOMAIN6_PG_CONFIG), \ + SR(DOMAIN7_PG_CONFIG), \ + SR(DOMAIN16_PG_CONFIG), \ + SR(DOMAIN17_PG_CONFIG), \ + SR(DOMAIN18_PG_CONFIG), \ + SR(DOMAIN0_PG_STATUS), \ + SR(DOMAIN1_PG_STATUS), \ + SR(DOMAIN2_PG_STATUS), \ + SR(DOMAIN3_PG_STATUS), \ + SR(DOMAIN4_PG_STATUS), \ + SR(DOMAIN5_PG_STATUS), \ + SR(DOMAIN6_PG_STATUS), \ + SR(DOMAIN7_PG_STATUS), \ + SR(DOMAIN16_PG_STATUS), \ + SR(DOMAIN17_PG_STATUS), \ + SR(DOMAIN18_PG_STATUS), \ + SR(D1VGA_CONTROL), \ + SR(D2VGA_CONTROL), \ + SR(D3VGA_CONTROL), \ + SR(D4VGA_CONTROL), \ + SR(D5VGA_CONTROL), \ + SR(D6VGA_CONTROL), \ + SR(DC_IP_REQUEST_CNTL), \ + BL_REG_LIST() +#endif + struct dce_hwseq_registers { /* Backlight registers */ @@ -401,36 +469,34 @@ struct dce_hwseq_registers { HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PHYPLL_PIXEL_RATE_SOURCE, mask_sh),\ HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PIXEL_RATE_PLL_SOURCE, mask_sh) +#define HWSEQ_LVTMA_MASK_SH_LIST(mask_sh)\ + HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ + HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh),\ + HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh),\ + HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + #define HWSEQ_DCE8_MASK_SH_LIST(mask_sh)\ .DCFE_CLOCK_ENABLE = CRTC_DCFE_CLOCK_CONTROL__CRTC_DCFE_CLOCK_ENABLE ## mask_sh, \ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_DCP_GRPH_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_SCL_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_DCP_GRPH_SURF_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, CONTROL, BLND_MODE, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh),\ - HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_) + HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #define HWSEQ_DCE10_MASK_SH_LIST(mask_sh)\ HWSEQ_DCEF_MASK_SH_LIST(mask_sh, DCFE_),\ HWSEQ_BLND_MASK_SH_LIST(mask_sh, BLND_),\ - HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #define HWSEQ_DCE11_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE10_MASK_SH_LIST(mask_sh),\ SF(DCFEV_CLOCK_CONTROL, DCFEV_CLOCK_ENABLE, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh),\ HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_) #define HWSEQ_DCE112_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE10_MASK_SH_LIST(mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh),\ HWSEQ_PHYPLL_MASK_SH_LIST(mask_sh, CRTC0_) #define HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh)\ @@ -438,18 +504,15 @@ struct dce_hwseq_registers { SF(DCHUB_FB_LOCATION, FB_BASE, mask_sh),\ SF(DCHUB_AGP_BASE, AGP_BASE, mask_sh),\ SF(DCHUB_AGP_BOT, AGP_BOT, mask_sh),\ - SF(DCHUB_AGP_TOP, AGP_TOP, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + SF(DCHUB_AGP_TOP, AGP_TOP, mask_sh) #define HWSEQ_DCE12_MASK_SH_LIST(mask_sh)\ HWSEQ_DCEF_MASK_SH_LIST(mask_sh, DCFE0_DCFE_),\ HWSEQ_BLND_MASK_SH_LIST(mask_sh, BLND0_BLND_),\ HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ HWSEQ_PHYPLL_MASK_SH_LIST(mask_sh, CRTC0_),\ - HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh),\ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #define HWSEQ_VG20_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE12_MASK_SH_LIST(mask_sh),\ @@ -512,10 +575,7 @@ struct dce_hwseq_registers { HWS_SF(, D4VGA_CONTROL, D4VGA_MODE_ENABLE, mask_sh),\ HWS_SF(, VGA_TEST_CONTROL, VGA_TEST_ENABLE, mask_sh),\ HWS_SF(, VGA_TEST_CONTROL, VGA_TEST_RENDER_START, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define HWSEQ_DCN2_MASK_SH_LIST(mask_sh)\ @@ -576,6 +636,49 @@ struct dce_hwseq_registers { HWS_SF(, DOMAIN20_PG_STATUS, DOMAIN20_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DOMAIN21_PG_STATUS, DOMAIN21_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh), \ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) +#endif + +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#define HWSEQ_DCN21_MASK_SH_LIST(mask_sh)\ + HWSEQ_DCN_MASK_SH_LIST(mask_sh), \ + HWS_SF(, DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, mask_sh), \ + HWS_SF(, MMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, PAGE_DIRECTORY_ENTRY_HI32, mask_sh),\ + HWS_SF(, MMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, PAGE_DIRECTORY_ENTRY_LO32, mask_sh),\ + HWS_SF(, DOMAIN0_PG_CONFIG, DOMAIN0_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN0_PG_CONFIG, DOMAIN0_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN1_PG_CONFIG, DOMAIN1_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN1_PG_CONFIG, DOMAIN1_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN2_PG_CONFIG, DOMAIN2_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN2_PG_CONFIG, DOMAIN2_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN3_PG_CONFIG, DOMAIN3_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN3_PG_CONFIG, DOMAIN3_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN4_PG_CONFIG, DOMAIN4_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN4_PG_CONFIG, DOMAIN4_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN5_PG_CONFIG, DOMAIN5_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN5_PG_CONFIG, DOMAIN5_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN6_PG_CONFIG, DOMAIN6_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN6_PG_CONFIG, DOMAIN6_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN7_PG_CONFIG, DOMAIN7_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN7_PG_CONFIG, DOMAIN7_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN16_PG_CONFIG, DOMAIN16_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN16_PG_CONFIG, DOMAIN16_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN17_PG_CONFIG, DOMAIN17_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN17_PG_CONFIG, DOMAIN17_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN18_PG_CONFIG, DOMAIN18_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN18_PG_CONFIG, DOMAIN18_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN0_PG_STATUS, DOMAIN0_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN1_PG_STATUS, DOMAIN1_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN2_PG_STATUS, DOMAIN2_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN3_PG_STATUS, DOMAIN3_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN4_PG_STATUS, DOMAIN4_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN5_PG_STATUS, DOMAIN5_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN6_PG_STATUS, DOMAIN6_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN7_PG_STATUS, DOMAIN7_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN16_PG_STATUS, DOMAIN16_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN17_PG_STATUS, DOMAIN17_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN18_PG_STATUS, DOMAIN18_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh), \ HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) #endif @@ -612,9 +715,9 @@ struct dce_hwseq_registers { type ENABLE_L1_TLB;\ type SYSTEM_ACCESS_MODE;\ type LVTMA_BLON;\ - type LVTMA_PWRSEQ_TARGET_STATE_R;\ type LVTMA_DIGON;\ - type LVTMA_DIGON_OVRD; + type LVTMA_DIGON_OVRD;\ + type LVTMA_PWRSEQ_TARGET_STATE_R; #define HWSEQ_DCN_REG_FIELD_LIST(type) \ type HUBP_VTG_SEL; \ @@ -696,7 +799,8 @@ struct dce_hwseq_registers { type D2VGA_MODE_ENABLE; \ type D3VGA_MODE_ENABLE; \ type D4VGA_MODE_ENABLE; \ - type AZALIA_AUDIO_DTO_MODULE; + type AZALIA_AUDIO_DTO_MODULE;\ + type HPO_HDMISTREAMCLK_GATE_DIS; struct dce_hwseq_shift { HWSEQ_REG_FIELD_LIST(uint8_t) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c index a9061aaf1562..aad7b52165be 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c @@ -25,6 +25,7 @@ #include <linux/delay.h> +#include "resource.h" #include "dce_i2c.h" #include "dce_i2c_hw.h" #include "reg_helper.h" @@ -99,17 +100,6 @@ static uint32_t get_hw_buffer_available_size( dce_i2c_hw->buffer_used_bytes; } -uint32_t get_reference_clock( - struct dc_bios *bios) -{ - struct dc_firmware_info info = { { 0 } }; - - if (bios->funcs->get_firmware_info(bios, &info) != BP_RESULT_OK) - return 0; - - return info.pll_info.crystal_frequency; -} - static uint32_t get_speed( const struct dce_i2c_hw *dce_i2c_hw) { @@ -401,7 +391,7 @@ struct dce_i2c_hw *acquire_i2c_hw_engine( if (ddc->hw_info.hw_supported) { enum gpio_ddc_line line = dal_ddc_get_line(ddc); - if (line < pool->pipe_count) + if (line < pool->res_cap->num_ddc) dce_i2c_hw = pool->hw_i2cs[line]; } @@ -632,7 +622,7 @@ void dce_i2c_hw_construct( { dce_i2c_hw->ctx = ctx; dce_i2c_hw->engine_id = engine_id; - dce_i2c_hw->reference_frequency = get_reference_clock(ctx->dc_bios) >> 1; + dce_i2c_hw->reference_frequency = (ctx->dc_bios->fw_info.pll_info.crystal_frequency) >> 1; dce_i2c_hw->regs = regs; dce_i2c_hw->shifts = shifts; dce_i2c_hw->masks = masks; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c index a24a2bda8656..31b698bf9cfc 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c @@ -148,7 +148,7 @@ static void dce_mi_program_pte_vm( pte->min_pte_before_flip_horiz_scan; REG_UPDATE(GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, - GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, 0xff); + GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, 0x7f); REG_UPDATE_3(DVMM_PTE_CONTROL, DVMM_PAGE_WIDTH, page_width, @@ -157,7 +157,7 @@ static void dce_mi_program_pte_vm( REG_UPDATE_2(DVMM_PTE_ARB_CONTROL, DVMM_PTE_REQ_PER_CHUNK, pte->pte_req_per_chunk, - DVMM_MAX_PTE_REQ_OUTSTANDING, 0xff); + DVMM_MAX_PTE_REQ_OUTSTANDING, 0x7f); } static void program_urgency_watermark( @@ -391,10 +391,10 @@ static void program_tiling( static void program_size_and_rotation( struct dce_mem_input *dce_mi, enum dc_rotation_angle rotation, - const union plane_size *plane_size) + const struct plane_size *plane_size) { - const struct rect *in_rect = &plane_size->grph.surface_size; - struct rect hw_rect = plane_size->grph.surface_size; + const struct rect *in_rect = &plane_size->surface_size; + struct rect hw_rect = plane_size->surface_size; const uint32_t rotation_angles[ROTATION_ANGLE_COUNT] = { [ROTATION_ANGLE_0] = 0, [ROTATION_ANGLE_90] = 1, @@ -423,7 +423,7 @@ static void program_size_and_rotation( GRPH_Y_END, hw_rect.height); REG_SET(GRPH_PITCH, 0, - GRPH_PITCH, plane_size->grph.surface_pitch); + GRPH_PITCH, plane_size->surface_pitch); REG_SET(HW_ROTATION, 0, GRPH_ROTATION_ANGLE, rotation_angles[rotation]); @@ -505,7 +505,7 @@ static void dce_mi_program_surface_config( struct mem_input *mi, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c index 5e2b4d47c548..6ed922a3c1cd 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c @@ -1038,6 +1038,24 @@ static void dce110_stream_encoder_set_avmute( } +static void dce110_reset_hdmi_stream_attribute( + struct stream_encoder *enc) +{ + struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc); + if (enc110->se_mask->HDMI_DATA_SCRAMBLE_EN) + REG_UPDATE_5(HDMI_CONTROL, + HDMI_PACKET_GEN_VERSION, 1, + HDMI_KEEPOUT_MODE, 1, + HDMI_DEEP_COLOR_ENABLE, 0, + HDMI_DATA_SCRAMBLE_EN, 0, + HDMI_CLOCK_CHANNEL_RATE, 0); + else + REG_UPDATE_3(HDMI_CONTROL, + HDMI_PACKET_GEN_VERSION, 1, + HDMI_KEEPOUT_MODE, 1, + HDMI_DEEP_COLOR_ENABLE, 0); +} + #define DP_SEC_AUD_N__DP_SEC_AUD_N__DEFAULT 0x8000 #define DP_SEC_TIMESTAMP__DP_SEC_TIMESTAMP_MODE__AUTO_CALC 1 @@ -1251,13 +1269,13 @@ static uint32_t calc_max_audio_packets_per_line( static void get_audio_clock_info( enum dc_color_depth color_depth, - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100Hz, + uint32_t actual_pixel_clock_100Hz, struct audio_clock_info *audio_clock_info) { const struct audio_clock_info *clock_info; uint32_t index; - uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_in_khz / 10; + uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_100Hz / 100; uint32_t audio_array_size; switch (color_depth) { @@ -1294,16 +1312,16 @@ static void get_audio_clock_info( } /* not found */ - if (actual_pixel_clock_in_khz == 0) - actual_pixel_clock_in_khz = crtc_pixel_clock_in_khz; + if (actual_pixel_clock_100Hz == 0) + actual_pixel_clock_100Hz = crtc_pixel_clock_100Hz; /* See HDMI spec the table entry under * pixel clock of "Other". */ audio_clock_info->pixel_clock_in_10khz = - actual_pixel_clock_in_khz / 10; - audio_clock_info->cts_32khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_44khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_48khz = actual_pixel_clock_in_khz; + actual_pixel_clock_100Hz / 100; + audio_clock_info->cts_32khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_44khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_48khz = actual_pixel_clock_100Hz / 10; audio_clock_info->n_32khz = 4096; audio_clock_info->n_44khz = 6272; @@ -1369,14 +1387,14 @@ static void dce110_se_setup_hdmi_audio( /* Program audio clock sample/regeneration parameters */ get_audio_clock_info(crtc_info->color_depth, - crtc_info->requested_pixel_clock, - crtc_info->calculated_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, + crtc_info->calculated_pixel_clock_100Hz, &audio_clock_info); DC_LOG_HW_AUDIO( - "\n%s:Input::requested_pixel_clock = %d" \ - "calculated_pixel_clock = %d \n", __func__, \ - crtc_info->requested_pixel_clock, \ - crtc_info->calculated_pixel_clock); + "\n%s:Input::requested_pixel_clock_100Hz = %d" \ + "calculated_pixel_clock_100Hz = %d \n", __func__, \ + crtc_info->requested_pixel_clock_100Hz, \ + crtc_info->calculated_pixel_clock_100Hz); /* HDMI_ACR_32_0__HDMI_ACR_CTS_32_MASK */ REG_UPDATE(HDMI_ACR_32_0, HDMI_ACR_CTS_32, audio_clock_info.cts_32khz); @@ -1584,6 +1602,17 @@ static void dig_connect_to_otg( REG_UPDATE(DIG_FE_CNTL, DIG_SOURCE_SELECT, tg_inst); } +static unsigned int dig_source_otg( + struct stream_encoder *enc) +{ + uint32_t tg_inst = 0; + struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc); + + REG_GET(DIG_FE_CNTL, DIG_SOURCE_SELECT, &tg_inst); + + return tg_inst; +} + static const struct stream_encoder_funcs dce110_str_enc_funcs = { .dp_set_stream_attribute = dce110_stream_encoder_dp_set_stream_attribute, @@ -1618,6 +1647,8 @@ static const struct stream_encoder_funcs dce110_str_enc_funcs = { .setup_stereo_sync = setup_stereo_sync, .set_avmute = dce110_stream_encoder_set_avmute, .dig_connect_to_otg = dig_connect_to_otg, + .hdmi_reset_stream_attribute = dce110_reset_hdmi_stream_attribute, + .dig_source_otg = dig_source_otg, }; void dce110_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c index 6248c8455314..1787b9bf800a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c @@ -304,7 +304,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -910,7 +910,6 @@ static bool construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -921,8 +920,7 @@ static bool construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce100_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1093,6 +1091,7 @@ struct resource_pool *dce100_create_resource_pool( if (construct(num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index fafb4b470140..01a924bf477a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -667,29 +667,7 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx) link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc, pipe_ctx->stream_res.stream_enc->id, true); - /* update AVI info frame (HDMI, DP)*/ - /* TODO: FPGA may change to hwss.update_info_frame */ - -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - if (pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata != NULL && - pipe_ctx->plane_res.hubp != NULL) { - if (pipe_ctx->stream->dmdata_address.quad_part != 0) { - /* if using dynamic meta, don't set up generic infopackets */ - pipe_ctx->stream_res.encoder_info_frame.hdrsmd.valid = false; - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - true, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } else - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - false, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } -#endif - dce110_update_info_frame(pipe_ctx); + link->dc->hwss.update_info_frame(pipe_ctx); /* enable early control to avoid corruption on DP monitor*/ active_total_with_borders = @@ -753,7 +731,7 @@ static enum bp_result link_transmitter_control( * @brief * eDP only. */ -void hwss_edp_wait_for_hpd_ready( +void dce110_edp_wait_for_hpd_ready( struct dc_link *link, bool power_up) { @@ -821,7 +799,7 @@ void hwss_edp_wait_for_hpd_ready( } } -void hwss_edp_power_control( +void dce110_edp_power_control( struct dc_link *link, bool power_up) { @@ -903,7 +881,7 @@ void hwss_edp_power_control( * @brief * eDP only. Control the backlight of the eDP panel */ -void hwss_edp_backlight_control( +void dce110_edp_backlight_control( struct dc_link *link, bool enable) { @@ -1003,7 +981,7 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx) } } -void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option) +void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx) { struct dc *dc; struct pp_smu_funcs *pp_smu = NULL; @@ -1026,24 +1004,13 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option) if (dc->res_pool->pp_smu) pp_smu = dc->res_pool->pp_smu; - if (option != KEEP_ACQUIRED_RESOURCE || - !dc->debug.az_endpoint_mute_only) - /*only disalbe az_endpoint if power down or free*/ - pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); - if (dc_is_dp_signal(pipe_ctx->stream->signal)) pipe_ctx->stream_res.stream_enc->funcs->dp_audio_disable( pipe_ctx->stream_res.stream_enc); else pipe_ctx->stream_res.stream_enc->funcs->hdmi_audio_disable( pipe_ctx->stream_res.stream_enc); - /*don't free audio if it is from retrain or internal disable stream*/ - if (option == FREE_ACQUIRED_RESOURCE && dc->caps.dynamic_audio == true) { - /*we have to dynamic arbitrate the audio endpoints*/ - /*we free the resource, need reset is_audio_acquired*/ - update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, pipe_ctx->stream_res.audio, false); - pipe_ctx->stream_res.audio = NULL; - } + if (clk_mgr->funcs->enable_pme_wa) /*this is the first audio. apply the PME w/a in order to wake AZ from D3*/ clk_mgr->funcs->enable_pme_wa(clk_mgr); @@ -1056,21 +1023,24 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option) } } -void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option) +void dce110_disable_stream(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; struct dc *dc = pipe_ctx->stream->ctx->dc; - if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) + if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) { pipe_ctx->stream_res.stream_enc->funcs->stop_hdmi_info_packets( pipe_ctx->stream_res.stream_enc); + pipe_ctx->stream_res.stream_enc->funcs->hdmi_reset_stream_attribute( + pipe_ctx->stream_res.stream_enc); + } if (dc_is_dp_signal(pipe_ctx->stream->signal)) pipe_ctx->stream_res.stream_enc->funcs->stop_dp_info_packets( pipe_ctx->stream_res.stream_enc); - dc->hwss.disable_audio_stream(pipe_ctx, option); + dc->hwss.disable_audio_stream(pipe_ctx); link->link_enc->funcs->connect_dig_be_to_fe( link->link_enc, @@ -1174,27 +1144,27 @@ static void build_audio_output( stream->timing.flags.INTERLACE; audio_output->crtc_info.refresh_rate = - (stream->timing.pix_clk_100hz*10000)/ + (stream->timing.pix_clk_100hz*100)/ (stream->timing.h_total*stream->timing.v_total); audio_output->crtc_info.color_depth = stream->timing.display_color_depth; - audio_output->crtc_info.requested_pixel_clock = - pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10; + audio_output->crtc_info.requested_pixel_clock_100Hz = + pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz; - audio_output->crtc_info.calculated_pixel_clock = - pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10; + audio_output->crtc_info.calculated_pixel_clock_100Hz = + pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz; /*for HDMI, audio ACR is with deep color ratio factor*/ if (dc_is_hdmi_signal(pipe_ctx->stream->signal) && - audio_output->crtc_info.requested_pixel_clock == - (stream->timing.pix_clk_100hz / 10)) { + audio_output->crtc_info.requested_pixel_clock_100Hz == + (stream->timing.pix_clk_100hz)) { if (pipe_ctx->stream_res.pix_clk_params.pixel_encoding == PIXEL_ENCODING_YCBCR420) { - audio_output->crtc_info.requested_pixel_clock = - audio_output->crtc_info.requested_pixel_clock/2; - audio_output->crtc_info.calculated_pixel_clock = - pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz/20; + audio_output->crtc_info.requested_pixel_clock_100Hz = + audio_output->crtc_info.requested_pixel_clock_100Hz/2; + audio_output->crtc_info.calculated_pixel_clock_100Hz = + pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz/2; } } @@ -1360,7 +1330,7 @@ static enum dc_status apply_single_controller_ctx_to_hw( struct drr_params params = {0}; unsigned int event_triggers = 0; #if defined(CONFIG_DRM_AMD_DC_DCN2_0) - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; #endif if (dc->hwss.disable_stream_gating) { @@ -1428,7 +1398,7 @@ static enum dc_status apply_single_controller_ctx_to_hw( &stream->bit_depth_params, &stream->clamping); #if defined(CONFIG_DRM_AMD_DC_DCN2_0) - if (odm_pipe) { + while (odm_pipe) { odm_pipe->stream_res.opp->funcs->opp_set_dyn_expansion( odm_pipe->stream_res.opp, COLOR_SPACE_YCBCR601, @@ -1439,6 +1409,7 @@ static enum dc_status apply_single_controller_ctx_to_hw( odm_pipe->stream_res.opp, &stream->bit_depth_params, &stream->clamping); + odm_pipe = odm_pipe->next_odm_pipe; } #endif @@ -1748,7 +1719,8 @@ void dce110_set_safe_displaymarks( ******************************************************************************/ static void set_drr(struct pipe_ctx **pipe_ctx, - int num_pipes, int vmin, int vmax) + int num_pipes, unsigned int vmin, unsigned int vmax, + unsigned int vmid, unsigned int vmid_frame_number) { int i = 0; struct drr_params params = {0}; @@ -1932,8 +1904,25 @@ static void dce110_reset_hw_ctx_wrap( /* Disable if new stream is null. O/w, if stream is * disabled already, no need to disable again. */ - if (!pipe_ctx->stream || !pipe_ctx->stream->dpms_off) - core_link_disable_stream(pipe_ctx_old, FREE_ACQUIRED_RESOURCE); + if (!pipe_ctx->stream || !pipe_ctx->stream->dpms_off) { + core_link_disable_stream(pipe_ctx_old); + + /* free acquired resources*/ + if (pipe_ctx_old->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx_old->stream_res.audio->funcs-> + az_disable(pipe_ctx_old->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx_old->stream_res.audio, false); + pipe_ctx_old->stream_res.audio = NULL; + } + } + } pipe_ctx_old->stream_res.tg->funcs->set_blank(pipe_ctx_old->stream_res.tg, true); if (!hwss_wait_for_blank_complete(pipe_ctx_old->stream_res.tg)) { @@ -2098,7 +2087,7 @@ enum dc_status dce110_apply_ctx_to_hw( if (pipe_ctx_old->stream && !pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) continue; - if (pipe_ctx->top_pipe) + if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe) continue; status = apply_single_controller_ctx_to_hw( @@ -2777,9 +2766,9 @@ static const struct hw_sequencer_funcs dce110_funcs = { .setup_stereo = NULL, .set_avmute = dce110_set_avmute, .wait_for_mpcc_disconnect = dce110_wait_for_mpcc_disconnect, - .edp_backlight_control = hwss_edp_backlight_control, - .edp_power_control = hwss_edp_power_control, - .edp_wait_for_hpd_ready = hwss_edp_wait_for_hpd_ready, + .edp_backlight_control = dce110_edp_backlight_control, + .edp_power_control = dce110_edp_power_control, + .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, .set_cursor_position = dce110_set_cursor_position, .set_cursor_attribute = dce110_set_cursor_attribute }; diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h index cd3e36d52a52..2f9b7dbdf415 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h @@ -42,7 +42,7 @@ enum dc_status dce110_apply_ctx_to_hw( void dce110_enable_stream(struct pipe_ctx *pipe_ctx); -void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option); +void dce110_disable_stream(struct pipe_ctx *pipe_ctx); void dce110_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); @@ -50,7 +50,7 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx, void dce110_blank_stream(struct pipe_ctx *pipe_ctx); void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx); -void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option); +void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx); void dce110_update_info_frame(struct pipe_ctx *pipe_ctx); @@ -73,15 +73,15 @@ void dce110_optimize_bandwidth( void dp_receiver_power_ctrl(struct dc_link *link, bool on); -void hwss_edp_power_control( +void dce110_edp_power_control( struct dc_link *link, bool power_up); -void hwss_edp_backlight_control( +void dce110_edp_backlight_control( struct dc_link *link, bool enable); -void hwss_edp_wait_for_hpd_ready( +void dce110_edp_wait_for_hpd_ready( struct dc_link *link, bool power_up); diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c index 9b9fc3d96c07..d54172d88f5f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c @@ -229,26 +229,26 @@ static void program_tiling( static void program_size_and_rotation( struct dce_mem_input *mem_input110, enum dc_rotation_angle rotation, - const union plane_size *plane_size) + const struct plane_size *plane_size) { uint32_t value = 0; - union plane_size local_size = *plane_size; + struct plane_size local_size = *plane_size; if (rotation == ROTATION_ANGLE_90 || rotation == ROTATION_ANGLE_270) { - swap(local_size.video.luma_size.x, - local_size.video.luma_size.y); - swap(local_size.video.luma_size.width, - local_size.video.luma_size.height); - swap(local_size.video.chroma_size.x, - local_size.video.chroma_size.y); - swap(local_size.video.chroma_size.width, - local_size.video.chroma_size.height); + swap(local_size.surface_size.x, + local_size.surface_size.y); + swap(local_size.surface_size.width, + local_size.surface_size.height); + swap(local_size.chroma_size.x, + local_size.chroma_size.y); + swap(local_size.chroma_size.width, + local_size.chroma_size.height); } value = 0; - set_reg_field_value(value, local_size.video.luma_pitch, + set_reg_field_value(value, local_size.surface_pitch, UNP_GRPH_PITCH_L, GRPH_PITCH_L); dm_write_reg( @@ -257,7 +257,7 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.chroma_pitch, + set_reg_field_value(value, local_size.chroma_pitch, UNP_GRPH_PITCH_C, GRPH_PITCH_C); dm_write_reg( mem_input110->base.ctx, @@ -297,8 +297,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.luma_size.x + - local_size.video.luma_size.width, + set_reg_field_value(value, local_size.surface_size.x + + local_size.surface_size.width, UNP_GRPH_X_END_L, GRPH_X_END_L); dm_write_reg( mem_input110->base.ctx, @@ -306,8 +306,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.chroma_size.x + - local_size.video.chroma_size.width, + set_reg_field_value(value, local_size.chroma_size.x + + local_size.chroma_size.width, UNP_GRPH_X_END_C, GRPH_X_END_C); dm_write_reg( mem_input110->base.ctx, @@ -315,8 +315,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.luma_size.y + - local_size.video.luma_size.height, + set_reg_field_value(value, local_size.surface_size.y + + local_size.surface_size.height, UNP_GRPH_Y_END_L, GRPH_Y_END_L); dm_write_reg( mem_input110->base.ctx, @@ -324,8 +324,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.chroma_size.y + - local_size.video.chroma_size.height, + set_reg_field_value(value, local_size.chroma_size.y + + local_size.chroma_size.height, UNP_GRPH_Y_END_C, GRPH_Y_END_C); dm_write_reg( mem_input110->base.ctx, @@ -637,7 +637,7 @@ void dce_mem_input_v_program_surface_config( struct mem_input *mem_input, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizotal_mirror) diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c index 764329264c3b..318e9c2e2ca8 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c @@ -331,7 +331,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -1274,7 +1274,6 @@ static bool construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -1300,8 +1299,7 @@ static bool construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce110_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1464,6 +1462,7 @@ struct resource_pool *dce110_create_resource_pool( if (construct(num_virtual_links, dc, pool, asic_id)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c index c6136e0ed1a4..83e1878161c9 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c @@ -337,7 +337,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -987,6 +987,10 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) struct dm_pp_clock_levels_with_latency mem_clks = {0}; struct dm_pp_wm_sets_with_clock_ranges clk_ranges = {0}; struct dm_pp_clock_levels clks = {0}; + int memory_type_multiplier = MEMORY_TYPE_MULTIPLIER_CZ; + + if (dc->bw_vbios && dc->bw_vbios->memory_type == bw_def_hbm) + memory_type_multiplier = MEMORY_TYPE_HBM; /*do system clock TODO PPLIB: after PPLIB implement, * then remove old way @@ -1026,12 +1030,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) &clks); dc->bw_vbios->low_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[0] * MEMORY_TYPE_MULTIPLIER_CZ, 1000); + clks.clocks_in_khz[0] * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[clks.num_levels>>1] * MEMORY_TYPE_MULTIPLIER_CZ, + clks.clocks_in_khz[clks.num_levels>>1] * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[clks.num_levels-1] * MEMORY_TYPE_MULTIPLIER_CZ, + clks.clocks_in_khz[clks.num_levels-1] * memory_type_multiplier, 1000); return; @@ -1067,12 +1071,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) * YCLK = UMACLK*m_memoryTypeMultiplier */ dc->bw_vbios->low_yclk = bw_frc_to_fixed( - mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, 1000); + mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, 1000); /* Now notify PPLib/SMU about which Watermarks sets they should select @@ -1338,6 +1342,7 @@ struct resource_pool *dce112_create_resource_pool( if (construct(num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c index 4a6ba3173a5a..8b85e5274bba 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c @@ -352,7 +352,7 @@ static const struct dce_audio_shift audio_shift = { DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -847,6 +847,8 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) int i; unsigned int clk; unsigned int latency; + /*original logic in dal3*/ + int memory_type_multiplier = MEMORY_TYPE_MULTIPLIER_CZ; /*do system clock*/ if (!dm_pp_get_clock_levels_by_type_with_latency( @@ -905,13 +907,16 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) * ALSO always convert UMA clock (from PPLIB) to YCLK (HW formula): * YCLK = UMACLK*m_memoryTypeMultiplier */ + if (dc->bw_vbios->memory_type == bw_def_hbm) + memory_type_multiplier = MEMORY_TYPE_HBM; + dc->bw_vbios->low_yclk = bw_frc_to_fixed( - mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, 1000); + mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, 1000); /* Now notify PPLib/SMU about which Watermarks sets they should select @@ -1203,6 +1208,7 @@ struct resource_pool *dce120_create_resource_pool( if (construct(num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c index 860a524ebcfa..4625df9f9fd2 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c @@ -322,7 +322,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -876,7 +876,6 @@ static bool dce80_construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -902,8 +901,7 @@ static bool dce80_construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce80_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1075,7 +1073,6 @@ static bool dce81_construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -1101,8 +1098,7 @@ static bool dce81_construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce80_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1274,7 +1270,6 @@ static bool dce83_construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -1300,8 +1295,7 @@ static bool dce83_construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce80_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index 7469333a2c8a..01c7e30b9ce1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -343,8 +343,8 @@ bool cm_helper_translate_curve_to_hw_format( region_start = -MAX_LOW_POINT; region_end = NUMBER_REGIONS - MAX_LOW_POINT; } else { - /* 10 segments - * segment is from 2^-10 to 2^0 + /* 11 segments + * segment is from 2^-10 to 2^1 * There are less than 256 points, for optimization */ seg_distr[0] = 3; @@ -357,9 +357,10 @@ bool cm_helper_translate_curve_to_hw_format( seg_distr[7] = 4; seg_distr[8] = 4; seg_distr[9] = 4; + seg_distr[10] = 1; region_start = -10; - region_end = 0; + region_end = 1; } for (i = region_end - region_start; i < MAX_REGIONS_NUMBER ; i++) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c index b95ec73fcae3..d8b2da18db39 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c @@ -426,8 +426,9 @@ void dpp1_cnv_setup ( void dpp1_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format) + struct dc_cursor_attributes *cursor_attributes) { + enum dc_cursor_color_format color_format = cursor_attributes->color_format; struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base); REG_UPDATE_2(CURSOR0_CONTROL, @@ -456,6 +457,19 @@ void dpp1_set_cursor_position( int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y; uint32_t cur_en = pos->enable ? 1 : 0; + // Cursor width/height and hotspots need to be rotated for offset calculation + if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) { + swap(width, height); + if (param->rotation == ROTATION_ANGLE_90) { + src_x_offset = pos->x - pos->y_hotspot - param->viewport.x; + src_y_offset = pos->y - pos->x_hotspot - param->viewport.y; + } + } else if (param->rotation == ROTATION_ANGLE_180) { + src_x_offset = pos->x - param->viewport.x; + src_y_offset = pos->y - param->viewport.y; + } + + if (src_x_offset >= (int)param->viewport.width) cur_en = 0; /* not visible beyond right edge*/ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h index 8a5517eebb7c..e2c613611ac9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h @@ -1368,7 +1368,7 @@ enum dcn10_input_csc_select { void dpp1_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format); + struct dc_cursor_attributes *cursor_attributes); void dpp1_set_cursor_position( struct dpp *dpp_base, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index a780057e2dbc..a02c10e23e0d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -104,7 +104,7 @@ void hubbub1_allow_self_refresh_control(struct hubbub *hubbub, bool allow) DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE, !allow); } -bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubbub) +bool hubbub1_is_allow_self_refresh_enabled(struct hubbub *hubbub) { struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); uint32_t enable = 0; @@ -945,6 +945,8 @@ static const struct hubbub_funcs hubbub1_funcs = { .get_dcc_compression_cap = hubbub1_get_dcc_compression_cap, .wm_read_state = hubbub1_wm_read_state, .program_watermarks = hubbub1_program_watermarks, + .is_allow_self_refresh_enabled = hubbub1_is_allow_self_refresh_enabled, + .allow_self_refresh_control = hubbub1_allow_self_refresh_control, }; void hubbub1_construct(struct hubbub *hubbub, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h index 7c2559c9ae23..69d903d68661 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h @@ -119,6 +119,28 @@ struct dcn_hubbub_registers { uint32_t DCN_VM_AGP_BOT; uint32_t DCN_VM_AGP_TOP; uint32_t DCN_VM_AGP_BASE; + uint32_t DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB; + uint32_t DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB; +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_A; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_B; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_C; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_D; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D; + uint32_t DCHUBBUB_ARB_HOSTVM_CNTL; + uint32_t DCHVM_CTRL0; + uint32_t DCHVM_MEM_CTRL; + uint32_t DCHVM_CLK_CTRL; + uint32_t DCHVM_RIOMMU_CTRL0; + uint32_t DCHVM_RIOMMU_STAT0; +#endif }; /* set field name */ @@ -196,7 +218,9 @@ struct dcn_hubbub_registers { type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A;\ type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B;\ type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C;\ - type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D + type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D;\ + type DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB;\ + type DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB #define HUBBUB_STUTTER_REG_FIELD_LIST(type) \ type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A;\ @@ -208,15 +232,68 @@ struct dcn_hubbub_registers { type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C;\ type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#define HUBBUB_HVM_REG_FIELD_LIST(type) \ + type DCHUBBUB_ARB_MIN_REQ_OUTSTAND_COMMIT_THRESHOLD;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_D;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_D;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_D;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_A;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_B;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_C;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_D;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D;\ + type DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD;\ + type HOSTVM_INIT_REQ; \ + type HVM_GPUVMRET_PWR_REQ_DIS; \ + type HVM_GPUVMRET_FORCE_REQ; \ + type HVM_GPUVMRET_POWER_STATUS; \ + type HVM_DISPCLK_R_GATE_DIS; \ + type HVM_DISPCLK_G_GATE_DIS; \ + type HVM_DCFCLK_R_GATE_DIS; \ + type HVM_DCFCLK_G_GATE_DIS; \ + type TR_REQ_REQCLKREQ_MODE; \ + type TW_RSP_COMPCLKREQ_MODE; \ + type HOSTVM_PREFETCH_REQ; \ + type HOSTVM_POWERSTATUS; \ + type RIOMMU_ACTIVE; \ + type HOSTVM_PREFETCH_DONE +#endif struct dcn_hubbub_shift { DCN_HUBBUB_REG_FIELD_LIST(uint8_t); HUBBUB_STUTTER_REG_FIELD_LIST(uint8_t); +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + HUBBUB_HVM_REG_FIELD_LIST(uint8_t); +#endif }; struct dcn_hubbub_mask { DCN_HUBBUB_REG_FIELD_LIST(uint32_t); HUBBUB_STUTTER_REG_FIELD_LIST(uint32_t); +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + HUBBUB_HVM_REG_FIELD_LIST(uint32_t); +#endif }; struct dc; @@ -247,7 +324,7 @@ void hubbub1_program_watermarks( void hubbub1_allow_self_refresh_control(struct hubbub *hubbub, bool allow); -bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubub); +bool hubbub1_is_allow_self_refresh_enabled(struct hubbub *hubub); void hubbub1_toggle_watermark_change_req( struct hubbub *hubbub); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index 934bacc0c6ad..001db49e4bb2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -163,7 +163,7 @@ void hubp1_program_tiling( void hubp1_program_size( struct hubp *hubp, enum surface_pixel_format format, - const union plane_size *plane_size, + const struct plane_size *plane_size, struct dc_plane_dcc_param *dcc) { struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); @@ -173,16 +173,16 @@ void hubp1_program_size( * 444 or 420 luma */ if (format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN && format < SURFACE_PIXEL_FORMAT_SUBSAMPLE_END) { - ASSERT(plane_size->video.chroma_pitch != 0); + ASSERT(plane_size->chroma_pitch != 0); /* Chroma pitch zero can cause system hang! */ - pitch = plane_size->video.luma_pitch - 1; - meta_pitch = dcc->video.meta_pitch_l - 1; - pitch_c = plane_size->video.chroma_pitch - 1; - meta_pitch_c = dcc->video.meta_pitch_c - 1; + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; + pitch_c = plane_size->chroma_pitch - 1; + meta_pitch_c = dcc->meta_pitch_c - 1; } else { - pitch = plane_size->grph.surface_pitch - 1; - meta_pitch = dcc->grph.meta_pitch - 1; + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; pitch_c = 0; meta_pitch_c = 0; } @@ -509,7 +509,7 @@ bool hubp1_program_surface_flip_and_addr( } void hubp1_dcc_control(struct hubp *hubp, bool enable, - bool independent_64b_blks) + enum hubp_ind_block_size independent_64b_blks) { uint32_t dcc_en = enable ? 1 : 0; uint32_t dcc_ind_64b_blk = independent_64b_blks ? 1 : 0; @@ -526,13 +526,13 @@ void hubp1_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, unsigned int compat_level) { - hubp1_dcc_control(hubp, dcc->enable, dcc->grph.independent_64b_blks); + hubp1_dcc_control(hubp, dcc->enable, dcc->independent_64b_blks); hubp1_program_tiling(hubp, tiling_info, format); hubp1_program_size(hubp, format, plane_size, dcc); hubp1_program_rotation(hubp, rotation, horizontal_mirror); @@ -843,7 +843,7 @@ void min_set_viewport( PRI_VIEWPORT_Y_START_C, viewport_c->y); } -void hubp1_read_state(struct hubp *hubp) +void hubp1_read_state_common(struct hubp *hubp) { struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); struct dcn_hubp_state *s = &hubp1->state; @@ -859,24 +859,6 @@ void hubp1_read_state(struct hubp *hubp) PRQ_EXPANSION_MODE, &rq_regs->prq_expansion_mode, MRQ_EXPANSION_MODE, &rq_regs->mrq_expansion_mode, CRQ_EXPANSION_MODE, &rq_regs->crq_expansion_mode); - REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, - CHUNK_SIZE, &rq_regs->rq_regs_l.chunk_size, - MIN_CHUNK_SIZE, &rq_regs->rq_regs_l.min_chunk_size, - META_CHUNK_SIZE, &rq_regs->rq_regs_l.meta_chunk_size, - MIN_META_CHUNK_SIZE, &rq_regs->rq_regs_l.min_meta_chunk_size, - DPTE_GROUP_SIZE, &rq_regs->rq_regs_l.dpte_group_size, - MPTE_GROUP_SIZE, &rq_regs->rq_regs_l.mpte_group_size, - SWATH_HEIGHT, &rq_regs->rq_regs_l.swath_height, - PTE_ROW_HEIGHT_LINEAR, &rq_regs->rq_regs_l.pte_row_height_linear); - REG_GET_8(DCHUBP_REQ_SIZE_CONFIG_C, - CHUNK_SIZE_C, &rq_regs->rq_regs_c.chunk_size, - MIN_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_chunk_size, - META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.meta_chunk_size, - MIN_META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_meta_chunk_size, - DPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.dpte_group_size, - MPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.mpte_group_size, - SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height, - PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear); /* DLG - Per hubp */ REG_GET_2(BLANK_OFFSET_0, @@ -1030,8 +1012,38 @@ void hubp1_read_state(struct hubp *hubp) REG_GET_2(DCN_TTU_QOS_WM, QoS_LEVEL_LOW_WM, &s->qos_level_low_wm, QoS_LEVEL_HIGH_WM, &s->qos_level_high_wm); + } +void hubp1_read_state(struct hubp *hubp) +{ + struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); + struct dcn_hubp_state *s = &hubp1->state; + struct _vcs_dpi_display_rq_regs_st *rq_regs = &s->rq_regs; + + hubp1_read_state_common(hubp); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, + CHUNK_SIZE, &rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, &rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, &rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, &rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, &rq_regs->rq_regs_l.dpte_group_size, + MPTE_GROUP_SIZE, &rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, &rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, &rq_regs->rq_regs_l.pte_row_height_linear); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG_C, + CHUNK_SIZE_C, &rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.dpte_group_size, + MPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.mpte_group_size, + SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear); + +} enum cursor_pitch hubp1_get_cursor_pitch(unsigned int pitch) { enum cursor_pitch hw_pitch; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h index 31c8fdd3206c..cb20d10288c0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h @@ -125,8 +125,6 @@ SRI(DCN_VM_SYSTEM_APERTURE_LOW_ADDR_LSB, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_MSB, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB, HUBPREQ, id),\ - SR(DCHUBBUB_SDPIF_FB_BASE),\ - SR(DCHUBBUB_SDPIF_FB_OFFSET),\ SRI(CURSOR_SETTINS, HUBPREQ, id), \ SRI(CURSOR_SURFACE_ADDRESS_HIGH, CURSOR, id), \ SRI(CURSOR_SURFACE_ADDRESS, CURSOR, id), \ @@ -226,14 +224,6 @@ uint32_t DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB; \ uint32_t DCN_VM_SYSTEM_APERTURE_LOW_ADDR; \ uint32_t DCN_VM_SYSTEM_APERTURE_HIGH_ADDR; \ - uint32_t DCHUBBUB_SDPIF_FB_BASE; \ - uint32_t DCHUBBUB_SDPIF_FB_OFFSET; \ - uint32_t DCN_VM_FB_LOCATION_TOP; \ - uint32_t DCN_VM_FB_LOCATION_BASE; \ - uint32_t DCN_VM_FB_OFFSET; \ - uint32_t DCN_VM_AGP_BASE; \ - uint32_t DCN_VM_AGP_BOT; \ - uint32_t DCN_VM_AGP_TOP; \ uint32_t CURSOR_SETTINS; \ uint32_t CURSOR_SETTINGS; \ uint32_t CURSOR_SURFACE_ADDRESS_HIGH; \ @@ -249,7 +239,8 @@ .field_name = reg_name ## __ ## field_name ## post_fix /* Mask/shift struct generation macro for all ASICs (including those with reduced functionality) */ -#define HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh)\ +/*1.x, 2.x, and 3.x*/ +#define HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh)\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_BLANK_EN, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_TTU_DISABLE, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_UNDERFLOW_STATUS, mask_sh),\ @@ -265,7 +256,6 @@ HUBP_SF(HUBP0_DCSURF_ADDR_CONFIG, MAX_COMPRESSED_FRAGS, mask_sh),\ HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, SW_MODE, mask_sh),\ HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, META_LINEAR, mask_sh),\ - HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh),\ HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, PIPE_ALIGNED, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SURFACE_PITCH, META_PITCH, mask_sh),\ @@ -372,12 +362,17 @@ HUBP_SF(HUBPREQ0_DCN_SURF0_TTU_CNTL0, QoS_RAMP_DISABLE, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_SURF0_TTU_CNTL1, REFCYC_PER_REQ_DELIVERY_PRE, mask_sh),\ HUBP_SF(HUBP0_HUBP_CLK_CNTL, HUBP_CLOCK_ENABLE, mask_sh) - -#define HUBP_MASK_SH_LIST_DCN(mask_sh)\ - HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh),\ +/*2.x and 1.x only*/ +#define HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, MPTE_GROUP_SIZE, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, MPTE_GROUP_SIZE_C, mask_sh) +/*2.x and 1.x only*/ +#define HUBP_MASK_SH_LIST_DCN(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh) + /* Mask/shift struct generation macro for ASICs with VM */ #define HUBP_MASK_SH_LIST_DCN_VM(mask_sh)\ HUBP_SF(HUBPREQ0_NOM_PARAMETERS_0, DST_Y_PER_PTE_ROW_NOM_L, mask_sh),\ @@ -412,8 +407,6 @@ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_LOW_ADDR_LSB, MC_VM_SYSTEM_APERTURE_LOW_ADDR_LSB, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_MSB, MC_VM_SYSTEM_APERTURE_HIGH_ADDR_MSB, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB, MC_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB, mask_sh),\ - HUBP_SF(DCHUBBUB_SDPIF_FB_BASE, SDPIF_FB_BASE, mask_sh),\ - HUBP_SF(DCHUBBUB_SDPIF_FB_OFFSET, SDPIF_FB_OFFSET, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, MC_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, mask_sh),\ @@ -434,7 +427,7 @@ HUBP_SF(CURSOR0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_Y, mask_sh), \ HUBP_SF(CURSOR0_CURSOR_DST_OFFSET, CURSOR_DST_X_OFFSET, mask_sh) -#define DCN_HUBP_REG_FIELD_LIST(type) \ +#define DCN_HUBP_REG_FIELD_BASE_LIST(type) \ type HUBP_BLANK_EN;\ type HUBP_DISABLE;\ type HUBP_TTU_DISABLE;\ @@ -459,7 +452,6 @@ type ROTATION_ANGLE;\ type H_MIRROR_EN;\ type SURFACE_PIXEL_FORMAT;\ - type ALPHA_PLANE_EN;\ type SURFACE_FLIP_TYPE;\ type SURFACE_FLIP_MODE_FOR_STEREOSYNC;\ type SURFACE_FLIP_IN_STEREOSYNC;\ @@ -589,18 +581,6 @@ type MC_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB;\ type MC_VM_SYSTEM_APERTURE_LOW_ADDR;\ type MC_VM_SYSTEM_APERTURE_HIGH_ADDR;\ - type SDPIF_FB_TOP;\ - type SDPIF_FB_BASE;\ - type SDPIF_FB_OFFSET;\ - type SDPIF_AGP_BASE;\ - type SDPIF_AGP_BOT;\ - type SDPIF_AGP_TOP;\ - type FB_TOP;\ - type FB_BASE;\ - type FB_OFFSET;\ - type AGP_BASE;\ - type AGP_BOT;\ - type AGP_TOP;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB;\ @@ -632,6 +612,10 @@ type CURSOR_DST_X_OFFSET; \ type OUTPUT_FP +#define DCN_HUBP_REG_FIELD_LIST(type) \ + DCN_HUBP_REG_FIELD_BASE_LIST(type);\ + type ALPHA_PLANE_EN + struct dcn_mi_registers { HUBP_COMMON_REG_VARIABLE_LIST; }; @@ -677,7 +661,7 @@ void hubp1_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, @@ -699,7 +683,7 @@ void hubp1_program_pixel_format( void hubp1_program_size( struct hubp *hubp, enum surface_pixel_format format, - const union plane_size *plane_size, + const struct plane_size *plane_size, struct dc_plane_dcc_param *dcc); void hubp1_program_rotation( @@ -714,7 +698,7 @@ void hubp1_program_tiling( void hubp1_dcc_control(struct hubp *hubp, bool enable, - bool independent_64b_blks); + enum hubp_ind_block_size independent_64b_blks); #ifdef CONFIG_DRM_AMD_DC_DCN2_0 bool hubp1_program_surface_flip_and_addr( @@ -760,5 +744,6 @@ void hubp1_vready_workaround(struct hubp *hubp, struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest); void hubp1_init(struct hubp *hubp); +void hubp1_read_state_common(struct hubp *hubp); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 2118ea21d7e9..60123db7ba02 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -438,7 +438,7 @@ bool dcn10_did_underflow_occur(struct dc *dc, struct pipe_ctx *pipe_ctx) return false; } -static void enable_power_gating_plane( +static void dcn10_enable_power_gating_plane( struct dce_hwseq *hws, bool enable) { @@ -460,7 +460,7 @@ static void enable_power_gating_plane( REG_UPDATE(DOMAIN7_PG_CONFIG, DOMAIN7_POWER_FORCEON, force_on); } -static void disable_vga( +static void dcn10_disable_vga( struct dce_hwseq *hws) { unsigned int in_vga1_mode = 0; @@ -493,7 +493,7 @@ static void disable_vga( REG_UPDATE(VGA_TEST_CONTROL, VGA_TEST_RENDER_START, 1); } -static void dpp_pg_control( +static void dcn10_dpp_pg_control( struct dce_hwseq *hws, unsigned int dpp_inst, bool power_on) @@ -545,7 +545,7 @@ static void dpp_pg_control( } } -static void hubp_pg_control( +static void dcn10_hubp_pg_control( struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on) @@ -605,8 +605,8 @@ static void power_on_plane( if (REG(DC_IP_REQUEST_CNTL)) { REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - dpp_pg_control(hws, plane_id, true); - hubp_pg_control(hws, plane_id, true); + hws->ctx->dc->hwss.dpp_pg_control(hws, plane_id, true); + hws->ctx->dc->hwss.hubp_pg_control(hws, plane_id, true); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); DC_LOG_DEBUG( @@ -627,7 +627,7 @@ static void undo_DEGVIDCN10_253_wa(struct dc *dc) REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - hubp_pg_control(hws, 0, false); + dc->hwss.hubp_pg_control(hws, 0, false); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); @@ -656,7 +656,7 @@ static void apply_DEGVIDCN10_253_wa(struct dc *dc) REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - hubp_pg_control(hws, 0, true); + dc->hwss.hubp_pg_control(hws, 0, true); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); @@ -664,10 +664,23 @@ static void apply_DEGVIDCN10_253_wa(struct dc *dc) hws->wa_state.DEGVIDCN10_253_applied = true; } -static void bios_golden_init(struct dc *dc) +static void dcn10_bios_golden_init(struct dc *dc) { struct dc_bios *bp = dc->ctx->dc_bios; int i; + bool allow_self_fresh_force_enable = true; + + if (dc->res_pool->hubbub->funcs->is_allow_self_refresh_enabled) + allow_self_fresh_force_enable = + dc->res_pool->hubbub->funcs->is_allow_self_refresh_enabled(dc->res_pool->hubbub); + + + /* WA for making DF sleep when idle after resume from S0i3. + * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE is set to 1 by + * command table, if DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0 + * before calling command table and it changed to 1 after, + * it should be set back to 0. + */ /* initialize dcn global */ bp->funcs->enable_disp_power_gating(bp, @@ -678,6 +691,12 @@ static void bios_golden_init(struct dc *dc) bp->funcs->enable_disp_power_gating(bp, CONTROLLER_ID_D0 + i, ASIC_PIPE_DISABLE); } + + if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) + if (allow_self_fresh_force_enable == false && + dc->res_pool->hubbub->funcs->is_allow_self_refresh_enabled(dc->res_pool->hubbub)) + dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, true); + } static void false_optc_underflow_wa( @@ -702,7 +721,8 @@ static void false_optc_underflow_wa( dc->hwss.wait_for_mpcc_disconnect(dc, dc->res_pool, old_pipe_ctx); } - tg->funcs->set_blank_data_double_buffer(tg, true); + if (tg->funcs->set_blank_data_double_buffer) + tg->funcs->set_blank_data_double_buffer(tg, true); if (tg->funcs->is_optc_underflow_occurred(tg) && !underflow) tg->funcs->clear_optc_underflow(tg); @@ -808,11 +828,23 @@ static void dcn10_reset_back_end_for_pipe( if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { /* DPMS may already disable */ if (!pipe_ctx->stream->dpms_off) - core_link_disable_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); - else if (pipe_ctx->stream_res.audio) { - dc->hwss.disable_audio_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); + core_link_disable_stream(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; + } } - } /* by upper caller loop, parent pipe: pipe0, will be reset last. @@ -823,6 +855,9 @@ static void dcn10_reset_back_end_for_pipe( pipe_ctx->stream_res.tg->funcs->disable_crtc(pipe_ctx->stream_res.tg); pipe_ctx->stream_res.tg->funcs->enable_optc_clock(pipe_ctx->stream_res.tg, false); + if (pipe_ctx->stream_res.tg->funcs->set_drr) + pipe_ctx->stream_res.tg->funcs->set_drr( + pipe_ctx->stream_res.tg, NULL); } for (i = 0; i < dc->res_pool->pipe_count; i++) @@ -968,7 +1003,7 @@ void hwss1_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) dcn10_verify_allow_pstate_change_high(dc); } -static void plane_atomic_power_down(struct dc *dc, +static void dcn10_plane_atomic_power_down(struct dc *dc, struct dpp *dpp, struct hubp *hubp) { @@ -978,8 +1013,8 @@ static void plane_atomic_power_down(struct dc *dc, if (REG(DC_IP_REQUEST_CNTL)) { REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - dpp_pg_control(hws, dpp->inst, false); - hubp_pg_control(hws, hubp->inst, false); + dc->hwss.dpp_pg_control(hws, dpp->inst, false); + dc->hwss.hubp_pg_control(hws, hubp->inst, false); dpp->funcs->dpp_reset(dpp); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); @@ -991,7 +1026,7 @@ static void plane_atomic_power_down(struct dc *dc, /* disable HW used by plane. * note: cannot disable until disconnect is complete */ -static void plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) +static void dcn10_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) { struct hubp *hubp = pipe_ctx->plane_res.hubp; struct dpp *dpp = pipe_ctx->plane_res.dpp; @@ -1011,7 +1046,7 @@ static void plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) hubp->power_gated = true; dc->optimized_required = false; /* We're powering off, no need to optimize */ - plane_atomic_power_down(dc, + dc->hwss.plane_atomic_power_down(dc, pipe_ctx->plane_res.dpp, pipe_ctx->plane_res.hubp); @@ -1030,7 +1065,7 @@ static void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) if (!pipe_ctx->plane_res.hubp || pipe_ctx->plane_res.hubp->power_gated) return; - plane_atomic_disable(dc, pipe_ctx); + dc->hwss.plane_atomic_disable(dc, pipe_ctx); apply_DEGVIDCN10_253_wa(dc); @@ -1065,15 +1100,27 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context) * command table. */ if (tg->funcs->is_tg_enabled(tg)) { - tg->funcs->lock(tg); - tg->funcs->set_blank(tg, true); - hwss_wait_for_blank_complete(tg); + if (dc->hwss.init_blank != NULL) { + dc->hwss.init_blank(dc, tg); + tg->funcs->lock(tg); + } else { + tg->funcs->lock(tg); + tg->funcs->set_blank(tg, true); + hwss_wait_for_blank_complete(tg); + } } } - /* Cannot reset the MPC mux if seamless boot */ - if (!can_apply_seamless_boot) - dc->res_pool->mpc->funcs->mpc_init(dc->res_pool->mpc); + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + /* Cannot reset the MPC mux if seamless boot */ + if (pipe_ctx->stream != NULL && can_apply_seamless_boot) + continue; + + dc->res_pool->mpc->funcs->mpc_init_single_inst( + dc->res_pool->mpc, i); + } for (i = 0; i < dc->res_pool->pipe_count; i++) { struct timing_generator *tg = dc->res_pool->timing_generators[i]; @@ -1111,12 +1158,12 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context) dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; - hwss1_plane_atomic_disconnect(dc, pipe_ctx); + dc->hwss.plane_atomic_disconnect(dc, pipe_ctx); if (tg->funcs->is_tg_enabled(tg)) tg->funcs->unlock(tg); - dcn10_disable_plane(dc, pipe_ctx); + dc->hwss.disable_plane(dc, pipe_ctx); pipe_ctx->stream_res.tg = NULL; pipe_ctx->plane_res.hubp = NULL; @@ -1132,8 +1179,17 @@ static void dcn10_init_hw(struct dc *dc) struct dmcu *dmcu = dc->res_pool->dmcu; struct dce_hwseq *hws = dc->hwseq; struct dc_bios *dcb = dc->ctx->dc_bios; + struct resource_pool *res_pool = dc->res_pool; + + if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) + dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); + + // Initialize the dccg + if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->dccg_init) + dc->res_pool->dccg->funcs->dccg_init(res_pool->dccg); if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + REG_WRITE(REFCLK_CNTL, 0); REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); REG_WRITE(DIO_MEM_PWR_CTRL, 0); @@ -1147,31 +1203,40 @@ static void dcn10_init_hw(struct dc *dc) REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); } - enable_power_gating_plane(dc->hwseq, true); + //Enable ability to power gate / don't force power on permanently + dc->hwss.enable_power_gating_plane(hws, true); - /* end of FPGA. Below if real ASIC */ return; } - if (!dcb->funcs->is_accelerated_mode(dcb)) { - bool allow_self_fresh_force_enable = - hububu1_is_allow_self_refresh_enabled( - dc->res_pool->hubbub); + if (!dcb->funcs->is_accelerated_mode(dcb)) + dc->hwss.disable_vga(dc->hwseq); - bios_golden_init(dc); + dc->hwss.bios_golden_init(dc); + if (dc->ctx->dc_bios->fw_info_valid) { + res_pool->ref_clocks.xtalin_clock_inKhz = + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; - /* WA for making DF sleep when idle after resume from S0i3. - * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE is set to 1 by - * command table, if DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0 - * before calling command table and it changed to 1 after, - * it should be set back to 0. - */ - if (allow_self_fresh_force_enable == false && - hububu1_is_allow_self_refresh_enabled(dc->res_pool->hubbub)) - hubbub1_allow_self_refresh_control(dc->res_pool->hubbub, true); + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + if (res_pool->dccg && res_pool->hubbub) { - disable_vga(dc->hwseq); - } + (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, + &res_pool->ref_clocks.dccg_ref_clock_inKhz); + + (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, + res_pool->ref_clocks.dccg_ref_clock_inKhz, + &res_pool->ref_clocks.dchub_ref_clock_inKhz); + } else { + // Not all ASICs have DCCG sw component + res_pool->ref_clocks.dccg_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + res_pool->ref_clocks.dchub_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + } + } + } else + ASSERT_CRITICAL(false); for (i = 0; i < dc->link_count; i++) { /* Power up AND update implementation according to the @@ -1188,6 +1253,13 @@ static void dcn10_init_hw(struct dc *dc) link->link_status.link_active = true; } + /* Power gate DSCs */ +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + for (i = 0; i < res_pool->res_cap->num_dsc; i++) + if (dc->hwss.dsc_pg_control != NULL) + dc->hwss.dsc_pg_control(hws, res_pool->dscs[i]->inst, false); +#endif + /* If taking control over from VBIOS, we may want to optimize our first * mode set, so we need to skip powering down pipes until we know which * pipes we want to use. @@ -1198,8 +1270,8 @@ static void dcn10_init_hw(struct dc *dc) dc->hwss.init_pipes(dc, dc->current_state); } - for (i = 0; i < dc->res_pool->audio_count; i++) { - struct audio *audio = dc->res_pool->audios[i]; + for (i = 0; i < res_pool->audio_count; i++) { + struct audio *audio = res_pool->audios[i]; audio->funcs->hw_init(audio); } @@ -1227,9 +1299,7 @@ static void dcn10_init_hw(struct dc *dc) REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); } - enable_power_gating_plane(dc->hwseq, true); - - memset(&dc->clk_mgr->clks, 0, sizeof(dc->clk_mgr->clks)); + dc->hwss.enable_power_gating_plane(dc->hwseq, true); } static void dcn10_reset_hw_ctx_wrap( @@ -1366,6 +1436,34 @@ static bool dcn10_set_input_transfer_func(struct pipe_ctx *pipe_ctx, return result; } +#define MAX_NUM_HW_POINTS 0x200 + +static void log_tf(struct dc_context *ctx, + struct dc_transfer_func *tf, uint32_t hw_points_num) +{ + // DC_LOG_GAMMA is default logging of all hw points + // DC_LOG_ALL_GAMMA logs all points, not only hw points + // DC_LOG_ALL_TF_POINTS logs all channels of the tf + int i = 0; + + DC_LOGGER_INIT(ctx->logger); + DC_LOG_GAMMA("Gamma Correction TF"); + DC_LOG_ALL_GAMMA("Logging all tf points..."); + DC_LOG_ALL_TF_CHANNELS("Logging all channels..."); + + for (i = 0; i < hw_points_num; i++) { + DC_LOG_GAMMA("R\t%d\t%llu\n", i, tf->tf_pts.red[i].value); + DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu\n", i, tf->tf_pts.green[i].value); + DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu\n", i, tf->tf_pts.blue[i].value); + } + + for (i = hw_points_num; i < MAX_NUM_HW_POINTS; i++) { + DC_LOG_ALL_GAMMA("R\t%d\t%llu\n", i, tf->tf_pts.red[i].value); + DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu\n", i, tf->tf_pts.green[i].value); + DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu\n", i, tf->tf_pts.blue[i].value); + } +} + static bool dcn10_set_output_transfer_func(struct pipe_ctx *pipe_ctx, const struct dc_stream_state *stream) @@ -1394,6 +1492,13 @@ dcn10_set_output_transfer_func(struct pipe_ctx *pipe_ctx, } else dpp->funcs->dpp_program_regamma_pwl(dpp, NULL, OPP_REGAMMA_BYPASS); + if (stream != NULL && stream->ctx != NULL && + stream->out_transfer_func != NULL) { + log_tf(stream->ctx, + stream->out_transfer_func, + dpp->regamma_params.hw_points_num); + } + return true; } @@ -1786,7 +1891,7 @@ static void dcn10_enable_plane( } } -static void program_gamut_remap(struct pipe_ctx *pipe_ctx) +static void dcn10_program_gamut_remap(struct pipe_ctx *pipe_ctx) { int i = 0; struct dpp_grph_csc_adjustment adjust; @@ -1804,6 +1909,36 @@ static void program_gamut_remap(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.dpp->funcs->dpp_set_gamut_remap(pipe_ctx->plane_res.dpp, &adjust); } + +static bool dcn10_is_rear_mpo_fix_required(struct pipe_ctx *pipe_ctx, enum dc_color_space colorspace) +{ + if (pipe_ctx->plane_state && pipe_ctx->plane_state->layer_index > 0 && is_rgb_cspace(colorspace)) { + if (pipe_ctx->top_pipe) { + struct pipe_ctx *top = pipe_ctx->top_pipe; + + while (top->top_pipe) + top = top->top_pipe; // Traverse to top pipe_ctx + if (top->plane_state && top->plane_state->layer_index == 0) + return true; // Front MPO plane not hidden + } + } + return false; +} + +static void dcn10_set_csc_adjustment_rgb_mpo_fix(struct pipe_ctx *pipe_ctx, uint16_t *matrix) +{ + // Override rear plane RGB bias to fix MPO brightness + uint16_t rgb_bias = matrix[3]; + + matrix[3] = 0; + matrix[7] = 0; + matrix[11] = 0; + pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix); + matrix[3] = rgb_bias; + matrix[7] = rgb_bias; + matrix[11] = rgb_bias; +} + static void dcn10_program_output_csc(struct dc *dc, struct pipe_ctx *pipe_ctx, enum dc_color_space colorspace, @@ -1811,8 +1946,25 @@ static void dcn10_program_output_csc(struct dc *dc, int opp_id) { if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) { - if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment != NULL) - pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix); + if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment != NULL) { + + /* MPO is broken with RGB colorspaces when OCSC matrix + * brightness offset >= 0 on DCN1 due to OCSC before MPC + * Blending adds offsets from front + rear to rear plane + * + * Fix is to set RGB bias to 0 on rear plane, top plane + * black value pixels add offset instead of rear + front + */ + + int16_t rgb_bias = matrix[3]; + // matrix[3/7/11] are all the same offset value + + if (rgb_bias > 0 && dcn10_is_rear_mpo_fix_required(pipe_ctx, colorspace)) { + dcn10_set_csc_adjustment_rgb_mpo_fix(pipe_ctx, matrix); + } else { + pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix); + } + } } else { if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default != NULL) pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default(pipe_ctx->plane_res.dpp, colorspace); @@ -2132,7 +2284,7 @@ void update_dchubp_dpp( struct hubp *hubp = pipe_ctx->plane_res.hubp; struct dpp *dpp = pipe_ctx->plane_res.dpp; struct dc_plane_state *plane_state = pipe_ctx->plane_state; - union plane_size size = plane_state->plane_size; + struct plane_size size = plane_state->plane_size; unsigned int compat_level = 0; /* depends on DML calculation, DPP clock value may change dynamically */ @@ -2152,7 +2304,8 @@ void update_dchubp_dpp( dc->res_pool->dccg->funcs->update_dpp_dto( dc->res_pool->dccg, dpp->inst, - pipe_ctx->plane_res.bw.dppclk_khz); + pipe_ctx->plane_res.bw.dppclk_khz, + false); else dc->clk_mgr->clks.dppclk_khz = should_divided_by_2 ? dc->clk_mgr->clks.dispclk_khz / 2 : @@ -2178,7 +2331,7 @@ void update_dchubp_dpp( &pipe_ctx->ttu_regs); } - size.grph.surface_size = pipe_ctx->plane_res.scl_data.viewport; + size.surface_size = pipe_ctx->plane_res.scl_data.viewport; if (plane_state->update_flags.bits.full_update || plane_state->update_flags.bits.bpp_change) @@ -2216,7 +2369,7 @@ void update_dchubp_dpp( if (plane_state->update_flags.bits.full_update) { /*gamut remap*/ - program_gamut_remap(pipe_ctx); + dc->hwss.program_gamut_remap(pipe_ctx); dc->hwss.program_output_csc(dc, pipe_ctx, @@ -2388,7 +2541,7 @@ struct pipe_ctx *find_top_pipe_for_stream( if (pipe_ctx->stream != stream) continue; - if (!pipe_ctx->top_pipe) + if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe) return pipe_ctx; } return NULL; @@ -2453,7 +2606,7 @@ static void dcn10_apply_ctx_for_surface( if (old_pipe_ctx->stream_res.tg == tg && old_pipe_ctx->plane_res.hubp && old_pipe_ctx->plane_res.hubp->opp_id != OPP_ID_INVALID) - dcn10_disable_plane(dc, old_pipe_ctx); + dc->hwss.disable_plane(dc, old_pipe_ctx); } if ((!pipe_ctx->plane_state || @@ -2501,7 +2654,7 @@ static void dcn10_apply_ctx_for_surface( for (i = 0; i < dc->res_pool->pipe_count; i++) if (removed_pipe[i]) - dcn10_disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + dc->hwss.disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); for (i = 0; i < dc->res_pool->pipe_count; i++) if (removed_pipe[i]) { @@ -2593,8 +2746,9 @@ static void dcn10_optimize_bandwidth( dcn10_verify_allow_pstate_change_high(dc); } -static void set_drr(struct pipe_ctx **pipe_ctx, - int num_pipes, int vmin, int vmax) +static void dcn10_set_drr(struct pipe_ctx **pipe_ctx, + int num_pipes, unsigned int vmin, unsigned int vmax, + unsigned int vmid, unsigned int vmid_frame_number) { int i = 0; struct drr_params params = {0}; @@ -2603,6 +2757,8 @@ static void set_drr(struct pipe_ctx **pipe_ctx, params.vertical_total_max = vmax; params.vertical_total_min = vmin; + params.vertical_total_mid = vmid; + params.vertical_total_mid_frame_num = vmid_frame_number; /* TODO: If multiple pipes are to be supported, you need * some GSL stuff. Static screen triggers may be programmed differently @@ -2618,7 +2774,7 @@ static void set_drr(struct pipe_ctx **pipe_ctx, } } -static void get_position(struct pipe_ctx **pipe_ctx, +static void dcn10_get_position(struct pipe_ctx **pipe_ctx, int num_pipes, struct crtc_position *position) { @@ -2630,7 +2786,7 @@ static void get_position(struct pipe_ctx **pipe_ctx, pipe_ctx[i]->stream_res.tg->funcs->get_position(pipe_ctx[i]->stream_res.tg, position); } -static void set_static_screen_control(struct pipe_ctx **pipe_ctx, +static void dcn10_set_static_screen_control(struct pipe_ctx **pipe_ctx, int num_pipes, const struct dc_static_screen_events *events) { unsigned int i; @@ -2692,6 +2848,13 @@ static void dcn10_setup_stereo(struct pipe_ctx *pipe_ctx, struct dc *dc) dcn10_config_stereo_parameters(stream, &flags); + if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_SIDEBAND_FA) { + if (!dc_set_generic_gpio_for_stereo(true, dc->ctx->gpio_service)) + dc_set_generic_gpio_for_stereo(false, dc->ctx->gpio_service); + } else { + dc_set_generic_gpio_for_stereo(false, dc->ctx->gpio_service); + } + pipe_ctx->stream_res.opp->funcs->opp_program_stereo( pipe_ctx->stream_res.opp, flags.PROGRAM_STEREO == 1 ? true:false, @@ -2782,14 +2945,10 @@ static void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx) static void dcn10_update_dchub(struct dce_hwseq *hws, struct dchub_init_data *dh_data) { - if (hws->ctx->dc->res_pool->hubbub != NULL) { - struct hubp *hubp = hws->ctx->dc->res_pool->hubps[0]; + struct hubbub *hubbub = hws->ctx->dc->res_pool->hubbub; - if (hubp->funcs->hubp_update_dchub) - hubp->funcs->hubp_update_dchub(hubp, dh_data); - else - hubbub1_update_dchub(hws->ctx->dc->res_pool->hubbub, dh_data); - } + /* In DCN, this programming sequence is owned by the hubbub */ + hubbub->funcs->update_dchub(hubbub, dh_data); } static void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) @@ -2820,6 +2979,40 @@ static void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) pos_cpy.enable = false; + // Swap axis and mirror horizontally + if (param.rotation == ROTATION_ANGLE_90) { + uint32_t temp_x = pos_cpy.x; + pos_cpy.x = pipe_ctx->plane_res.scl_data.viewport.width - + (pos_cpy.y - pipe_ctx->plane_res.scl_data.viewport.x) + pipe_ctx->plane_res.scl_data.viewport.x; + pos_cpy.y = temp_x; + } + // Swap axis and mirror vertically + else if (param.rotation == ROTATION_ANGLE_270) { + uint32_t temp_y = pos_cpy.y; + if (pos_cpy.x > pipe_ctx->plane_res.scl_data.viewport.height) { + pos_cpy.x = pos_cpy.x - pipe_ctx->plane_res.scl_data.viewport.height; + pos_cpy.y = pipe_ctx->plane_res.scl_data.viewport.height - pos_cpy.x; + } else { + pos_cpy.y = 2 * pipe_ctx->plane_res.scl_data.viewport.height - pos_cpy.x; + } + pos_cpy.x = temp_y; + } + // Mirror horizontally and vertically + else if (param.rotation == ROTATION_ANGLE_180) { + if (pos_cpy.x >= pipe_ctx->plane_res.scl_data.viewport.width + pipe_ctx->plane_res.scl_data.viewport.x) { + pos_cpy.x = 2 * pipe_ctx->plane_res.scl_data.viewport.width + - pos_cpy.x + 2 * pipe_ctx->plane_res.scl_data.viewport.x; + } else { + uint32_t temp_x = pos_cpy.x; + pos_cpy.x = 2 * pipe_ctx->plane_res.scl_data.viewport.x - pos_cpy.x; + if (temp_x >= pipe_ctx->plane_res.scl_data.viewport.x + (int)hubp->curs_attr.width + || pos_cpy.x <= (int)hubp->curs_attr.width + pipe_ctx->plane_state->src_rect.x) { + pos_cpy.x = temp_x + pipe_ctx->plane_res.scl_data.viewport.width; + } + } + pos_cpy.y = pipe_ctx->plane_res.scl_data.viewport.height - pos_cpy.y; + } + hubp->funcs->set_cursor_position(hubp, &pos_cpy, ¶m); dpp->funcs->set_cursor_position(dpp, &pos_cpy, ¶m, hubp->curs_attr.width, hubp->curs_attr.height); } @@ -2831,7 +3024,7 @@ static void dcn10_set_cursor_attribute(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.hubp->funcs->set_cursor_attributes( pipe_ctx->plane_res.hubp, attributes); pipe_ctx->plane_res.dpp->funcs->set_cursor_attributes( - pipe_ctx->plane_res.dpp, attributes->color_format); + pipe_ctx->plane_res.dpp, attributes); } static void dcn10_set_cursor_sdr_white_level(struct pipe_ctx *pipe_ctx) @@ -3062,9 +3255,59 @@ static void dcn10_send_immediate_sdp_message(struct pipe_ctx *pipe_ctx, sdp_message_size); } } +static enum dc_status dcn10_set_clock(struct dc *dc, + enum dc_clock_type clock_type, + uint32_t clk_khz, + uint32_t stepping) +{ + struct dc_state *context = dc->current_state; + struct dc_clock_config clock_cfg = {0}; + struct dc_clocks *current_clocks = &context->bw_ctx.bw.dcn.clk; + + if (dc->clk_mgr && dc->clk_mgr->funcs->get_clock) + dc->clk_mgr->funcs->get_clock(dc->clk_mgr, + context, clock_type, &clock_cfg); + + if (!dc->clk_mgr->funcs->get_clock) + return DC_FAIL_UNSUPPORTED_1; + + if (clk_khz > clock_cfg.max_clock_khz) + return DC_FAIL_CLK_EXCEED_MAX; + + if (clk_khz < clock_cfg.min_clock_khz) + return DC_FAIL_CLK_BELOW_MIN; + + if (clk_khz < clock_cfg.bw_requirequired_clock_khz) + return DC_FAIL_CLK_BELOW_CFG_REQUIRED; + + /*update internal request clock for update clock use*/ + if (clock_type == DC_CLOCK_TYPE_DISPCLK) + current_clocks->dispclk_khz = clk_khz; + else if (clock_type == DC_CLOCK_TYPE_DPPCLK) + current_clocks->dppclk_khz = clk_khz; + else + return DC_ERROR_UNEXPECTED; + + if (dc->clk_mgr && dc->clk_mgr->funcs->update_clocks) + dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, + context, true); + return DC_OK; + +} + +static void dcn10_get_clock(struct dc *dc, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg) +{ + struct dc_state *context = dc->current_state; + + if (dc->clk_mgr && dc->clk_mgr->funcs->get_clock) + dc->clk_mgr->funcs->get_clock(dc->clk_mgr, context, clock_type, clock_cfg); + +} static const struct hw_sequencer_funcs dcn10_funcs = { - .program_gamut_remap = program_gamut_remap, + .program_gamut_remap = dcn10_program_gamut_remap, .init_hw = dcn10_init_hw, .init_pipes = dcn10_init_pipes, .apply_ctx_to_hw = dce110_apply_ctx_to_hw, @@ -3097,18 +3340,18 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .optimize_bandwidth = dcn10_optimize_bandwidth, .reset_hw_ctx_wrap = dcn10_reset_hw_ctx_wrap, .enable_stream_timing = dcn10_enable_stream_timing, - .set_drr = set_drr, - .get_position = get_position, - .set_static_screen_control = set_static_screen_control, + .set_drr = dcn10_set_drr, + .get_position = dcn10_get_position, + .set_static_screen_control = dcn10_set_static_screen_control, .setup_stereo = dcn10_setup_stereo, .set_avmute = dce110_set_avmute, .log_hw_state = dcn10_log_hw_state, .get_hw_state = dcn10_get_hw_state, .clear_status_bits = dcn10_clear_status_bits, .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, - .edp_backlight_control = hwss_edp_backlight_control, - .edp_power_control = hwss_edp_power_control, - .edp_wait_for_hpd_ready = hwss_edp_wait_for_hpd_ready, + .edp_backlight_control = dce110_edp_backlight_control, + .edp_power_control = dce110_edp_power_control, + .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, .set_cursor_position = dcn10_set_cursor_position, .set_cursor_attribute = dcn10_set_cursor_attribute, .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, @@ -3116,7 +3359,18 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .enable_stream_gating = NULL, .setup_periodic_interrupt = dcn10_setup_periodic_interrupt, .setup_vupdate_interrupt = dcn10_setup_vupdate_interrupt, - .did_underflow_occur = dcn10_did_underflow_occur + .set_clock = dcn10_set_clock, + .get_clock = dcn10_get_clock, + .did_underflow_occur = dcn10_did_underflow_occur, + .init_blank = NULL, + .disable_vga = dcn10_disable_vga, + .bios_golden_init = dcn10_bios_golden_init, + .plane_atomic_disable = dcn10_plane_atomic_disable, + .plane_atomic_power_down = dcn10_plane_atomic_power_down, + .enable_power_gating_plane = dcn10_enable_power_gating_plane, + .dpp_pg_control = dcn10_dpp_pg_control, + .hubp_pg_control = dcn10_hubp_pg_control, + .dsc_pg_control = NULL, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c index 6e47444109d7..7f4766e45dff 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c @@ -45,7 +45,7 @@ #include "dcn10_cm_common.h" #include "clk_mgr.h" -static unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...) +unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...) { unsigned int ret_vsnprintf; unsigned int chars_printed; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c index 549d423a01f6..1a37c90e9d43 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -89,6 +89,7 @@ static const struct link_encoder_funcs dcn10_lnk_enc_funcs = { .disable_hpd = dcn10_link_encoder_disable_hpd, .is_dig_enabled = dcn10_is_dig_enabled, .get_dig_frontend = dcn10_get_dig_frontend, + .get_dig_mode = dcn10_get_dig_mode, .destroy = dcn10_link_encoder_destroy }; @@ -446,6 +447,46 @@ static uint8_t get_frontend_source( } } +unsigned int dcn10_get_dig_frontend(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + int32_t value; + enum engine_id result; + + REG_GET(DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, &value); + + switch (value) { + case DCN10_DIG_FE_SOURCE_SELECT_DIGA: + result = ENGINE_ID_DIGA; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGB: + result = ENGINE_ID_DIGB; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGC: + result = ENGINE_ID_DIGC; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGD: + result = ENGINE_ID_DIGD; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGE: + result = ENGINE_ID_DIGE; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGF: + result = ENGINE_ID_DIGF; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGG: + result = ENGINE_ID_DIGG; + break; + default: + // invalid source select DIG + ASSERT(false); + result = ENGINE_ID_UNKNOWN; + } + + return result; + +} + void enc1_configure_encoder( struct dcn10_link_encoder *enc10, const struct dc_link_settings *link_settings) @@ -501,15 +542,6 @@ bool dcn10_is_dig_enabled(struct link_encoder *enc) return value; } -unsigned int dcn10_get_dig_frontend(struct link_encoder *enc) -{ - struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); - uint32_t value; - - REG_GET(DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, &value); - return value; -} - static void link_encoder_disable(struct dcn10_link_encoder *enc10) { /* reset training pattern */ @@ -1366,3 +1398,25 @@ void dcn10_aux_initialize(struct dcn10_link_encoder *enc10) AUX_REG_UPDATE(AUX_DPHY_RX_CONTROL0, AUX_RX_RECEIVE_WINDOW, 0); } + +enum signal_type dcn10_get_dig_mode( + struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t value; + REG_GET(DIG_BE_CNTL, DIG_MODE, &value); + switch (value) { + case 1: + return SIGNAL_TYPE_DISPLAY_PORT; + case 2: + return SIGNAL_TYPE_DVI_SINGLE_LINK; + case 3: + return SIGNAL_TYPE_HDMI_TYPE_A; + case 5: + return SIGNAL_TYPE_DISPLAY_PORT_MST; + default: + return SIGNAL_TYPE_NONE; + } + return SIGNAL_TYPE_NONE; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h index 33b2af1a181c..8bf5f0f2301d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h @@ -337,6 +337,7 @@ struct dcn10_link_enc_registers { type RDPCS_TX_FIFO_ERROR_MASK;\ type RDPCS_DPALT_DISABLE_TOGGLE_MASK;\ type RDPCS_DPALT_4LANE_TOGGLE_MASK;\ + type RDPCS_PHY_DPALT_DISABLE;\ type RDPCS_PHY_DPALT_DISABLE_ACK;\ type RDPCS_PHY_DP_MPLLB_V2I;\ type RDPCS_PHY_DP_MPLLB_FREQ_VCO;\ @@ -514,4 +515,6 @@ unsigned int dcn10_get_dig_frontend(struct link_encoder *enc); void dcn10_aux_initialize(struct dcn10_link_encoder *enc10); +enum signal_type dcn10_get_dig_mode( + struct link_encoder *enc); #endif /* __DC_LINK_ENCODER__DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c index 0bca011ed7c9..8b2f29f6dabd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c @@ -211,7 +211,7 @@ struct mpcc *mpc1_insert_plane( } else { new_mpcc->mpcc_bot = NULL; REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); - REG_UPDATE(MPCC_CONTROL[mpcc_id], MPCC_MODE, MPCC_BLEND_MODE_TOP_LAYER_PASSTHROUGH); + REG_UPDATE(MPCC_CONTROL[mpcc_id], MPCC_MODE, MPCC_BLEND_MODE_TOP_LAYER_ONLY); } REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, dpp_id); REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, tree->opp_id); @@ -364,6 +364,24 @@ void mpc1_mpc_init(struct mpc *mpc) } } +void mpc1_mpc_init_single_inst(struct mpc *mpc, unsigned int mpcc_id) +{ + struct dcn10_mpc *mpc10 = TO_DCN10_MPC(mpc); + int opp_id; + + REG_GET(MPCC_OPP_ID[mpcc_id], MPCC_OPP_ID, &opp_id); + + REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf); + REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); + REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf); + + mpc1_init_mpcc(&(mpc->mpcc_array[mpcc_id]), mpcc_id); + + if (opp_id < MAX_OPP && REG(MUX[opp_id])) + REG_UPDATE(MUX[opp_id], MPC_OUT_MUX, 0xf); +} + + void mpc1_init_mpcc_list_from_hw( struct mpc *mpc, struct mpc_tree *tree) @@ -433,6 +451,7 @@ static const struct mpc_funcs dcn10_mpc_funcs = { .insert_plane = mpc1_insert_plane, .remove_mpcc = mpc1_remove_mpcc, .mpc_init = mpc1_mpc_init, + .mpc_init_single_inst = mpc1_mpc_init_single_inst, .get_mpcc_for_dpp = mpc1_get_mpcc_for_dpp, .wait_for_idle = mpc1_assert_idle_mpcc, .assert_mpcc_idle_before_connect = mpc1_assert_mpcc_idle_before_connect, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h index d3d16c4cbea3..962a68e322ee 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h @@ -149,6 +149,10 @@ void mpc1_remove_mpcc( void mpc1_mpc_init( struct mpc *mpc); +void mpc1_mpc_init_single_inst( + struct mpc *mpc, + unsigned int mpcc_id); + void mpc1_assert_idle_mpcc( struct mpc *mpc, int id); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index a546c2bc9129..e74a07d03fde 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -154,7 +154,7 @@ void optc1_program_timing( uint32_t h_sync_polarity, v_sync_polarity; uint32_t start_point = 0; uint32_t field_num = 0; - uint32_t h_div_2; + enum h_timing_div_mode h_div = H_TIMING_NO_DIV; struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -285,10 +285,11 @@ void optc1_program_timing( * of stereo handled in explicit call */ - h_div_2 = optc1_is_two_pixels_per_containter(&patched_crtc_timing); - REG_UPDATE(OTG_H_TIMING_CNTL, - OTG_H_TIMING_DIV_BY2, h_div_2 || optc1->comb_opp_id != 0xf); + if (optc1_is_two_pixels_per_containter(&patched_crtc_timing) || optc1->opp_count == 2) + h_div = H_TIMING_DIV_BY2; + REG_UPDATE(OTG_H_TIMING_CNTL, + OTG_H_TIMING_DIV_BY2, h_div); } void optc1_set_vtg_params(struct timing_generator *optc, @@ -824,6 +825,9 @@ void optc1_program_manual_trigger(struct timing_generator *optc) REG_SET(OTG_MANUAL_FLOW_CONTROL, 0, MANUAL_FLOW_CONTROL, 1); + + REG_SET(OTG_MANUAL_FLOW_CONTROL, 0, + MANUAL_FLOW_CONTROL, 0); } @@ -846,6 +850,18 @@ void optc1_set_drr( params->vertical_total_max > 0 && params->vertical_total_min > 0) { + if (params->vertical_total_mid != 0) { + + REG_SET(OTG_V_TOTAL_MID, 0, + OTG_V_TOTAL_MID, params->vertical_total_mid - 1); + + REG_UPDATE_2(OTG_V_TOTAL_CONTROL, + OTG_VTOTAL_MID_REPLACING_MAX_EN, 1, + OTG_VTOTAL_MID_FRAME_NUM, + (uint8_t)params->vertical_total_mid_frame_num); + + } + REG_SET(OTG_V_TOTAL_MAX, 0, OTG_V_TOTAL_MAX, params->vertical_total_max - 1); @@ -1513,7 +1529,6 @@ void dcn10_timing_generator_init(struct optc *optc1) optc1->min_v_blank_interlace = 5; optc1->min_h_sync_width = 8; optc1->min_v_sync_width = 1; - optc1->comb_opp_id = 0xf; } #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h index 02599eb92ca6..83575599672e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h @@ -54,6 +54,7 @@ SRI(OTG_3D_STRUCTURE_CONTROL, OTG, inst),\ SRI(OTG_STEREO_STATUS, OTG, inst),\ SRI(OTG_V_TOTAL_MAX, OTG, inst),\ + SRI(OTG_V_TOTAL_MID, OTG, inst),\ SRI(OTG_V_TOTAL_MIN, OTG, inst),\ SRI(OTG_V_TOTAL_CONTROL, OTG, inst),\ SRI(OTG_TRIGA_CNTL, OTG, inst),\ @@ -125,6 +126,7 @@ struct dcn_optc_registers { uint32_t OTG_3D_STRUCTURE_CONTROL; uint32_t OTG_STEREO_STATUS; uint32_t OTG_V_TOTAL_MAX; + uint32_t OTG_V_TOTAL_MID; uint32_t OTG_V_TOTAL_MIN; uint32_t OTG_V_TOTAL_CONTROL; uint32_t OTG_TRIGA_CNTL; @@ -214,12 +216,15 @@ struct dcn_optc_registers { SF(OTG0_OTG_3D_STRUCTURE_CONTROL, OTG_3D_STRUCTURE_V_UPDATE_MODE, mask_sh),\ SF(OTG0_OTG_3D_STRUCTURE_CONTROL, OTG_3D_STRUCTURE_STEREO_SEL_OVR, mask_sh),\ SF(OTG0_OTG_V_TOTAL_MAX, OTG_V_TOTAL_MAX, mask_sh),\ + SF(OTG0_OTG_V_TOTAL_MID, OTG_V_TOTAL_MID, mask_sh),\ SF(OTG0_OTG_V_TOTAL_MIN, OTG_V_TOTAL_MIN, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_V_TOTAL_MIN_SEL, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_V_TOTAL_MAX_SEL, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_FORCE_LOCK_ON_EVENT, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_SET_V_TOTAL_MIN_MASK_EN, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_SET_V_TOTAL_MIN_MASK, mask_sh),\ + SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_VTOTAL_MID_REPLACING_MAX_EN, mask_sh),\ + SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_VTOTAL_MID_FRAME_NUM, mask_sh),\ SF(OTG0_OTG_FORCE_COUNT_NOW_CNTL, OTG_FORCE_COUNT_NOW_CLEAR, mask_sh),\ SF(OTG0_OTG_FORCE_COUNT_NOW_CNTL, OTG_FORCE_COUNT_NOW_MODE, mask_sh),\ SF(OTG0_OTG_FORCE_COUNT_NOW_CNTL, OTG_FORCE_COUNT_NOW_OCCURRED, mask_sh),\ @@ -348,9 +353,12 @@ struct dcn_optc_registers { type OTG_3D_STRUCTURE_V_UPDATE_MODE;\ type OTG_3D_STRUCTURE_STEREO_SEL_OVR;\ type OTG_V_TOTAL_MAX;\ + type OTG_V_TOTAL_MID;\ type OTG_V_TOTAL_MIN;\ type OTG_V_TOTAL_MIN_SEL;\ type OTG_V_TOTAL_MAX_SEL;\ + type OTG_VTOTAL_MID_REPLACING_MAX_EN;\ + type OTG_VTOTAL_MID_FRAME_NUM;\ type OTG_FORCE_LOCK_ON_EVENT;\ type OTG_SET_V_TOTAL_MIN_MASK_EN;\ type OTG_SET_V_TOTAL_MIN_MASK;\ @@ -494,7 +502,7 @@ struct optc { const struct dcn_optc_shift *tg_shift; const struct dcn_optc_mask *tg_mask; - int comb_opp_id; + int opp_count; uint32_t max_h_total; uint32_t max_v_total; @@ -539,6 +547,10 @@ struct dcn_otg_state { void optc1_read_otg_state(struct optc *optc1, struct dcn_otg_state *s); +bool optc1_is_matching_timing( + struct timing_generator *tg, + const struct dc_crtc_timing *otg_timing); + bool optc1_validate_timing( struct timing_generator *optc, const struct dc_crtc_timing *timing); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index a12530a3ab9c..59305e411a66 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -270,7 +270,7 @@ static const struct dce_audio_shift audio_shift = { DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -1416,6 +1416,14 @@ static bool construct( pool->base.pp_smu = dcn10_pp_smu_create(ctx); + /* + * Right now SMU/PPLIB and DAL all have the AZ D3 force PME notification * + * implemented. So AZ D3 should work.For issue 197007. * + */ + if (pool->base.pp_smu != NULL + && pool->base.pp_smu->rv_funcs.set_pme_wa_enable != NULL) + dc->debug.az_endpoint_mute_only = false; + if (!dc->debug.disable_pplib_clock_request) dcn_bw_update_from_pplib(dc); dcn_bw_sync_calcs_and_dml(dc); @@ -1562,6 +1570,7 @@ struct resource_pool *dcn10_create_resource_pool( if (construct(init_data->num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c index b9ffbf6b58ff..9aa258f3550b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c @@ -512,11 +512,12 @@ void enc1_stream_encoder_hdmi_set_stream_attribute( enc1_stream_encoder_set_stream_attribute_helper(enc1, crtc_timing); /* setup HDMI engine */ - REG_UPDATE_5(HDMI_CONTROL, + REG_UPDATE_6(HDMI_CONTROL, HDMI_PACKET_GEN_VERSION, 1, HDMI_KEEPOUT_MODE, 1, HDMI_DEEP_COLOR_ENABLE, 0, HDMI_DATA_SCRAMBLE_EN, 0, + HDMI_NO_EXTRA_NULL_PACKET_FILLED, 1, HDMI_CLOCK_CHANNEL_RATE, 0); @@ -1003,6 +1004,19 @@ void enc1_stream_encoder_set_avmute( REG_UPDATE(HDMI_GC, HDMI_GC_AVMUTE, value); } +void enc1_reset_hdmi_stream_attribute( + struct stream_encoder *enc) +{ + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); + + REG_UPDATE_5(HDMI_CONTROL, + HDMI_PACKET_GEN_VERSION, 1, + HDMI_KEEPOUT_MODE, 1, + HDMI_DEEP_COLOR_ENABLE, 0, + HDMI_DATA_SCRAMBLE_EN, 0, + HDMI_CLOCK_CHANNEL_RATE, 0); +} + #define DP_SEC_AUD_N__DP_SEC_AUD_N__DEFAULT 0x8000 #define DP_SEC_TIMESTAMP__DP_SEC_TIMESTAMP_MODE__AUTO_CALC 1 @@ -1196,13 +1210,13 @@ static union audio_cea_channels speakers_to_channels( void get_audio_clock_info( enum dc_color_depth color_depth, - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100Hz, + uint32_t actual_pixel_clock_100Hz, struct audio_clock_info *audio_clock_info) { const struct audio_clock_info *clock_info; uint32_t index; - uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_in_khz / 10; + uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_100Hz / 100; uint32_t audio_array_size; switch (color_depth) { @@ -1239,16 +1253,16 @@ void get_audio_clock_info( } /* not found */ - if (actual_pixel_clock_in_khz == 0) - actual_pixel_clock_in_khz = crtc_pixel_clock_in_khz; + if (actual_pixel_clock_100Hz == 0) + actual_pixel_clock_100Hz = crtc_pixel_clock_100Hz; /* See HDMI spec the table entry under * pixel clock of "Other". */ audio_clock_info->pixel_clock_in_10khz = - actual_pixel_clock_in_khz / 10; - audio_clock_info->cts_32khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_44khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_48khz = actual_pixel_clock_in_khz; + actual_pixel_clock_100Hz / 100; + audio_clock_info->cts_32khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_44khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_48khz = actual_pixel_clock_100Hz / 10; audio_clock_info->n_32khz = 4096; audio_clock_info->n_44khz = 6272; @@ -1308,14 +1322,14 @@ static void enc1_se_setup_hdmi_audio( /* Program audio clock sample/regeneration parameters */ get_audio_clock_info(crtc_info->color_depth, - crtc_info->requested_pixel_clock, - crtc_info->calculated_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, + crtc_info->calculated_pixel_clock_100Hz, &audio_clock_info); DC_LOG_HW_AUDIO( - "\n%s:Input::requested_pixel_clock = %d" \ - "calculated_pixel_clock = %d \n", __func__, \ - crtc_info->requested_pixel_clock, \ - crtc_info->calculated_pixel_clock); + "\n%s:Input::requested_pixel_clock_100Hz = %d" \ + "calculated_pixel_clock_100Hz = %d \n", __func__, \ + crtc_info->requested_pixel_clock_100Hz, \ + crtc_info->calculated_pixel_clock_100Hz); /* HDMI_ACR_32_0__HDMI_ACR_CTS_32_MASK */ REG_UPDATE(HDMI_ACR_32_0, HDMI_ACR_CTS_32, audio_clock_info.cts_32khz); @@ -1528,6 +1542,17 @@ void enc1_dig_connect_to_otg( REG_UPDATE(DIG_FE_CNTL, DIG_SOURCE_SELECT, tg_inst); } +unsigned int enc1_dig_source_otg( + struct stream_encoder *enc) +{ + uint32_t tg_inst = 0; + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); + + REG_GET(DIG_FE_CNTL, DIG_SOURCE_SELECT, &tg_inst); + + return tg_inst; +} + static const struct stream_encoder_funcs dcn10_str_enc_funcs = { .dp_set_stream_attribute = enc1_stream_encoder_dp_set_stream_attribute, @@ -1562,6 +1587,8 @@ static const struct stream_encoder_funcs dcn10_str_enc_funcs = { .setup_stereo_sync = enc1_setup_stereo_sync, .set_avmute = enc1_stream_encoder_set_avmute, .dig_connect_to_otg = enc1_dig_connect_to_otg, + .hdmi_reset_stream_attribute = enc1_reset_hdmi_stream_attribute, + .dig_source_otg = enc1_dig_source_otg, }; void dcn10_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h index bc2b4af9543b..a512cbea00d1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h @@ -89,7 +89,8 @@ SRI(DP_VID_STREAM_CNTL, DP, id), \ SRI(DP_VID_TIMING, DP, id), \ SRI(DP_SEC_AUD_N, DP, id), \ - SRI(DP_SEC_TIMESTAMP, DP, id) + SRI(DP_SEC_TIMESTAMP, DP, id), \ + SRI(DIG_CLOCK_PATTERN, DIG, id) #define SE_DCN_REG_LIST(id)\ SE_COMMON_DCN_REG_LIST(id) @@ -170,6 +171,7 @@ struct dcn10_stream_enc_registers { uint32_t HDMI_METADATA_PACKET_CONTROL; uint32_t DP_SEC_FRAMING4; #endif + uint32_t DIG_CLOCK_PATTERN; }; @@ -189,6 +191,7 @@ struct dcn10_stream_enc_registers { SE_SF(DIG0_HDMI_CONTROL, HDMI_DEEP_COLOR_ENABLE, mask_sh),\ SE_SF(DIG0_HDMI_CONTROL, HDMI_DEEP_COLOR_DEPTH, mask_sh),\ SE_SF(DIG0_HDMI_CONTROL, HDMI_DATA_SCRAMBLE_EN, mask_sh),\ + SE_SF(DIG0_HDMI_CONTROL, HDMI_NO_EXTRA_NULL_PACKET_FILLED, mask_sh),\ SE_SF(DIG0_HDMI_VBI_PACKET_CONTROL, HDMI_GC_CONT, mask_sh),\ SE_SF(DIG0_HDMI_VBI_PACKET_CONTROL, HDMI_GC_SEND, mask_sh),\ SE_SF(DIG0_HDMI_VBI_PACKET_CONTROL, HDMI_NULL_SEND, mask_sh),\ @@ -297,7 +300,8 @@ struct dcn10_stream_enc_registers { SE_SF(DP0_DP_MSA_TIMING_PARAM4, DP_MSA_VHEIGHT, mask_sh),\ SE_SF(DIG0_HDMI_DB_CONTROL, HDMI_DB_DISABLE, mask_sh),\ SE_SF(DP0_DP_VID_TIMING, DP_VID_N_MUL, mask_sh),\ - SE_SF(DIG0_DIG_FE_CNTL, DIG_SOURCE_SELECT, mask_sh) + SE_SF(DIG0_DIG_FE_CNTL, DIG_SOURCE_SELECT, mask_sh),\ + SE_SF(DIG0_DIG_CLOCK_PATTERN, DIG_CLOCK_PATTERN, mask_sh) #define SE_COMMON_MASK_SH_LIST_SOC(mask_sh)\ SE_COMMON_MASK_SH_LIST_SOC_BASE(mask_sh) @@ -374,6 +378,7 @@ struct dcn10_stream_enc_registers { type HDMI_GC_SEND;\ type HDMI_NULL_SEND;\ type HDMI_DATA_SCRAMBLE_EN;\ + type HDMI_NO_EXTRA_NULL_PACKET_FILLED;\ type HDMI_AUDIO_INFO_SEND;\ type AFMT_AUDIO_INFO_UPDATE;\ type HDMI_AUDIO_INFO_LINE;\ @@ -458,7 +463,8 @@ struct dcn10_stream_enc_registers { type HDMI_DB_DISABLE;\ type DP_VID_N_MUL;\ type DP_VID_M_DOUBLE_VALUE_EN;\ - type DIG_SOURCE_SELECT + type DIG_SOURCE_SELECT;\ + type DIG_CLOCK_PATTERN #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define SE_REG_FIELD_LIST_DCN2_0(type) \ @@ -592,6 +598,9 @@ void enc1_dig_connect_to_otg( struct stream_encoder *enc, int tg_inst); +unsigned int enc1_dig_source_otg( + struct stream_encoder *enc); + void enc1_stream_encoder_set_stream_attribute_helper( struct dcn10_stream_encoder *enc1, struct dc_crtc_timing *crtc_timing); @@ -605,8 +614,11 @@ void enc1_se_enable_dp_audio( void get_audio_clock_info( enum dc_color_depth color_depth, - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100Hz, + uint32_t actual_pixel_clock_100Hz, struct audio_clock_info *audio_clock_info); +void enc1_reset_hdmi_stream_attribute( + struct stream_encoder *enc); + #endif /* __DC_STREAM_ENCODER_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile index e9721a906592..ddb8d5649e79 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile @@ -16,7 +16,11 @@ else ifneq ($(call cc-option, -mstack-alignment=16),) cc_stack_align := -mstack-alignment=16 endif -CFLAGS_dcn20_resource.o := -mhard-float -msse $(cc_stack_align) +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse $(cc_stack_align) + +ifdef CONFIG_CC_IS_CLANG +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2 +endif AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c index 31aa6ee5cd5b..16476ed25536 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c @@ -44,12 +44,16 @@ #define DC_LOGGER \ dccg->ctx->logger -void dccg2_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) +void dccg2_update_dpp_dto(struct dccg *dccg, + int dpp_inst, + int req_dppclk, + bool reduce_divider_only) { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); if (dccg->ref_dppclk && req_dppclk) { int ref_dppclk = dccg->ref_dppclk; + int current_phase, current_modulo; ASSERT(req_dppclk <= ref_dppclk); /* need to clamp to 8 bits */ @@ -61,9 +65,28 @@ void dccg2_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) if (req_dppclk > ref_dppclk) req_dppclk = ref_dppclk; } - REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0, - DPPCLK0_DTO_PHASE, req_dppclk, - DPPCLK0_DTO_MODULO, ref_dppclk); + + REG_GET_2(DPPCLK_DTO_PARAM[dpp_inst], + DPPCLK0_DTO_PHASE, ¤t_phase, + DPPCLK0_DTO_MODULO, ¤t_modulo); + + if (reduce_divider_only) { + // requested phase/modulo greater than current + if (req_dppclk * current_modulo >= current_phase * ref_dppclk) { + REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0, + DPPCLK0_DTO_PHASE, req_dppclk, + DPPCLK0_DTO_MODULO, ref_dppclk); + } else { + REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0, + DPPCLK0_DTO_PHASE, current_phase, + DPPCLK0_DTO_MODULO, current_modulo); + } + } else { + REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0, + DPPCLK0_DTO_PHASE, req_dppclk, + DPPCLK0_DTO_MODULO, ref_dppclk); + } + REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 1); } else { diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h index 2205cb0204e7..74a074a873cd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h @@ -97,7 +97,7 @@ struct dcn_dccg { const struct dccg_mask *dccg_mask; }; -void dccg2_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk); +void dccg2_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk, bool raise_divider_only); void dccg2_get_dccg_ref_freq(struct dccg *dccg, unsigned int xtalin_freq_inKhz, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c index 9bc5dd23d297..2f5aade1e882 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c @@ -72,6 +72,21 @@ void dpp20_read_state(struct dpp *dpp_base, } } +void dpp2_power_on_obuf( + struct dpp *dpp_base, + bool power_on) +{ + struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); + + REG_UPDATE(CM_MEM_PWR_CTRL, SHARED_MEM_PWR_DIS, power_on == true ? 1:0); + + REG_UPDATE(OBUF_MEM_PWR_CTRL, + OBUF_MEM_PWR_FORCE, power_on == true ? 0:1); + + REG_UPDATE(DSCL_MEM_PWR_CTRL, + LUT_MEM_PWR_FORCE, power_on == true ? 0:1); +} + void dpp2_dummy_program_input_lut( struct dpp *dpp_base, const struct dc_gamma *gamma) @@ -227,6 +242,7 @@ static void dpp2_cnv_setup ( CUR0_ENABLE, 0); } + dpp2_power_on_obuf(dpp_base, true); } @@ -326,14 +342,18 @@ void dpp2_cnv_set_alpha_keyer( void dpp2_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format) + struct dc_cursor_attributes *cursor_attributes) { + enum dc_cursor_color_format color_format = cursor_attributes->color_format; struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); int cur_rom_en = 0; if (color_format == CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA || - color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) - cur_rom_en = 1; + color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) { + if (cursor_attributes->attribute_flags.bits.ENABLE_CURSOR_DEGAMMA) { + cur_rom_en = 1; + } + } REG_UPDATE_3(CURSOR0_CONTROL, CUR0_MODE, color_format, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h index 59b67ed57c19..290b2854bd2c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h @@ -162,7 +162,9 @@ SRI(COLOR_KEYER_GREEN, CNVC_CFG, id), \ SRI(COLOR_KEYER_BLUE, CNVC_CFG, id), \ SRI(CM_SHAPER_LUT_DATA, CM, id), \ - SRI(CURSOR_CONTROL, CURSOR0_, id) + SRI(CURSOR_CONTROL, CURSOR0_, id),\ + SRI(OBUF_MEM_PWR_CTRL, DSCL, id),\ + SRI(DSCL_MEM_PWR_CTRL, DSCL, id) #define TF_REG_LIST_SH_MASK_DCN20(mask_sh)\ TF_REG_LIST_SH_MASK_DCN(mask_sh), \ @@ -554,7 +556,9 @@ TF_SF(CNVC_CFG0_COLOR_KEYER_BLUE, COLOR_KEYER_BLUE_HIGH, mask_sh), \ TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_PIX_INV_MODE, mask_sh), \ TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_PIXEL_ALPHA_MOD_EN, mask_sh), \ - TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_ROM_EN, mask_sh) + TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_ROM_EN, mask_sh),\ + TF_SF(DSCL0_OBUF_MEM_PWR_CTRL, OBUF_MEM_PWR_FORCE, mask_sh),\ + TF_SF(DSCL0_DSCL_MEM_PWR_CTRL, LUT_MEM_PWR_FORCE, mask_sh) #define TF_REG_FIELD_LIST_DCN2_0(type) \ TF_REG_FIELD_LIST(type) \ @@ -585,7 +589,9 @@ type COLOR_KEYER_BLUE_HIGH; \ type CUR0_PIX_INV_MODE; \ type CUR0_PIXEL_ALPHA_MOD_EN; \ - type CUR0_ROM_EN + type CUR0_ROM_EN;\ + type OBUF_MEM_PWR_FORCE;\ + type LUT_MEM_PWR_FORCE struct dcn2_dpp_shift { TF_REG_FIELD_LIST_DCN2_0(uint8_t); @@ -609,7 +615,9 @@ struct dcn2_dpp_mask { uint32_t COLOR_KEYER_ALPHA; \ uint32_t COLOR_KEYER_RED; \ uint32_t COLOR_KEYER_GREEN; \ - uint32_t COLOR_KEYER_BLUE + uint32_t COLOR_KEYER_BLUE; \ + uint32_t OBUF_MEM_PWR_CTRL;\ + uint32_t DSCL_MEM_PWR_CTRL struct dcn2_dpp_registers { DPP_DCN2_REG_VARIABLE_LIST; @@ -668,7 +676,7 @@ void dscl2_calc_lb_num_partitions( void dpp2_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format); + struct dc_cursor_attributes *cursor_attributes); void dpp2_dummy_program_input_lut( struct dpp *dpp_base, @@ -695,4 +703,7 @@ bool dpp2_construct(struct dcn20_dpp *dpp2, const struct dcn2_dpp_shift *tf_shift, const struct dcn2_dpp_mask *tf_mask); +void dpp2_power_on_obuf( + struct dpp *dpp_base, + bool power_on); #endif /* __DC_HWSS_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c index e28b8e7bedf5..2d112c316424 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c @@ -52,7 +52,12 @@ static void dpp2_enable_cm_block( { struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); - REG_UPDATE(CM_CONTROL, CM_BYPASS, 0); + unsigned int cm_bypass_mode = 0; + //Temp, put CM in bypass mode + if (dpp_base->ctx->dc->debug.cm_in_bypass) + cm_bypass_mode = 1; + + REG_UPDATE(CM_CONTROL, CM_BYPASS, cm_bypass_mode); } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c index ffd0014ec3b5..1b419407af94 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c @@ -29,7 +29,7 @@ #include "dsc/dscc_types.h" static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps); -static bool dsc_prepare_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, +static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals, struct dsc_optc_config *dsc_optc_cfg); static void dsc_init_reg_values(struct dsc_reg_values *reg_vals); static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params); @@ -42,7 +42,8 @@ static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock static void dsc2_read_state(struct display_stream_compressor *dsc, struct dcn_dsc_state *s); static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg); static void dsc2_set_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, - struct dsc_optc_config *dsc_optc_cfg, uint8_t *dsc_packed_pps); + struct dsc_optc_config *dsc_optc_cfg); +static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps); static void dsc2_enable(struct display_stream_compressor *dsc, int opp_pipe); static void dsc2_disable(struct display_stream_compressor *dsc); @@ -51,6 +52,7 @@ const struct dsc_funcs dcn20_dsc_funcs = { .dsc_read_state = dsc2_read_state, .dsc_validate_stream = dsc2_validate_stream, .dsc_set_config = dsc2_set_config, + .dsc_get_packed_pps = dsc2_get_packed_pps, .dsc_enable = dsc2_enable, .dsc_disable = dsc2_disable, }; @@ -116,8 +118,8 @@ static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock dsc_enc_caps->color_formats.bits.RGB = 1; dsc_enc_caps->color_formats.bits.YCBCR_444 = 1; - dsc_enc_caps->color_formats.bits.YCBCR_SIMPLE_422 = 1; - dsc_enc_caps->color_formats.bits.YCBCR_NATIVE_422 = 1; + dsc_enc_caps->color_formats.bits.YCBCR_SIMPLE_422 = 0; + dsc_enc_caps->color_formats.bits.YCBCR_NATIVE_422 = 0; dsc_enc_caps->color_formats.bits.YCBCR_NATIVE_420 = 1; dsc_enc_caps->color_depth.bits.COLOR_DEPTH_8_BPC = 1; @@ -162,40 +164,61 @@ static void dsc2_read_state(struct display_stream_compressor *dsc, struct dcn_ds static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg) { struct dsc_optc_config dsc_optc_cfg; + struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); - if (dsc_cfg->pic_width > TO_DCN20_DSC(dsc)->max_image_width) + if (dsc_cfg->pic_width > dsc20->max_image_width) return false; - return dsc_prepare_config(dsc, dsc_cfg, &dsc_optc_cfg); + return dsc_prepare_config(dsc_cfg, &dsc20->reg_vals, &dsc_optc_cfg); } -static void dsc_config_log(struct display_stream_compressor *dsc, - const struct dsc_config *config) +static void dsc_config_log(struct display_stream_compressor *dsc, const struct dsc_config *config) { - DC_LOG_DSC("Setting DSC Config at DSC inst %d", dsc->inst); - DC_LOG_DSC("\n\tnum_slices_h %d\n\tnum_slices_v %d\n\tbits_per_pixel %d\n\tcolor_depth %d", - config->dc_dsc_cfg.num_slices_h, - config->dc_dsc_cfg.num_slices_v, + DC_LOG_DSC("\tnum_slices_h %d", config->dc_dsc_cfg.num_slices_h); + DC_LOG_DSC("\tnum_slices_v %d", config->dc_dsc_cfg.num_slices_v); + DC_LOG_DSC("\tbits_per_pixel %d (%d.%04d)", config->dc_dsc_cfg.bits_per_pixel, - config->color_depth); + config->dc_dsc_cfg.bits_per_pixel / 16, + ((config->dc_dsc_cfg.bits_per_pixel % 16) * 10000) / 16); + DC_LOG_DSC("\tcolor_depth %d", config->color_depth); } static void dsc2_set_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, - struct dsc_optc_config *dsc_optc_cfg, uint8_t *dsc_packed_pps) + struct dsc_optc_config *dsc_optc_cfg) { bool is_config_ok; struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); + DC_LOG_DSC(" "); + DC_LOG_DSC("Setting DSC Config at DSC inst %d", dsc->inst); dsc_config_log(dsc, dsc_cfg); - is_config_ok = dsc_prepare_config(dsc, dsc_cfg, dsc_optc_cfg); + is_config_ok = dsc_prepare_config(dsc_cfg, &dsc20->reg_vals, dsc_optc_cfg); ASSERT(is_config_ok); - drm_dsc_pps_payload_pack((struct drm_dsc_picture_parameter_set *)dsc_packed_pps, &dsc20->reg_vals.pps); + DC_LOG_DSC("programming DSC Picture Parameter Set (PPS):"); dsc_log_pps(dsc, &dsc20->reg_vals.pps); dsc_write_to_registers(dsc, &dsc20->reg_vals); } +static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps) +{ + bool is_config_ok; + struct dsc_reg_values dsc_reg_vals; + struct dsc_optc_config dsc_optc_cfg; + + DC_LOG_DSC("Getting packed DSC PPS for DSC Config:"); + dsc_config_log(dsc, dsc_cfg); + DC_LOG_DSC("DSC Picture Parameter Set (PPS):"); + is_config_ok = dsc_prepare_config(dsc_cfg, &dsc_reg_vals, &dsc_optc_cfg); + ASSERT(is_config_ok); + drm_dsc_pps_payload_pack((struct drm_dsc_picture_parameter_set *)dsc_packed_pps, &dsc_reg_vals.pps); + dsc_log_pps(dsc, &dsc_reg_vals.pps); + + return is_config_ok; +} + + static void dsc2_enable(struct display_stream_compressor *dsc, int opp_pipe) { struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); @@ -232,7 +255,6 @@ static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_co int i; int bits_per_pixel = pps->bits_per_pixel; - DC_LOG_DSC("programming DSC Picture Parameter Set (PPS):"); DC_LOG_DSC("\tdsc_version_major %d", pps->dsc_version_major); DC_LOG_DSC("\tdsc_version_minor %d", pps->dsc_version_minor); DC_LOG_DSC("\tbits_per_component %d", pps->bits_per_component); @@ -282,13 +304,11 @@ static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_co } } -static bool dsc_prepare_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, +static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals, struct dsc_optc_config *dsc_optc_cfg) { struct dsc_parameters dsc_params; - struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); - /* Validate input parameters */ ASSERT(dsc_cfg->dc_dsc_cfg.num_slices_h); ASSERT(dsc_cfg->dc_dsc_cfg.num_slices_v); @@ -302,7 +322,7 @@ static bool dsc_prepare_config(struct display_stream_compressor *dsc, const stru dsc_cfg->dc_dsc_cfg.linebuf_depth == 0))); ASSERT(96 <= dsc_cfg->dc_dsc_cfg.bits_per_pixel && dsc_cfg->dc_dsc_cfg.bits_per_pixel <= 0x3ff); // 6.0 <= bits_per_pixel <= 63.9375 - if (!dsc_cfg->dc_dsc_cfg.num_slices_v || !dsc_cfg->dc_dsc_cfg.num_slices_v || + if (!dsc_cfg->dc_dsc_cfg.num_slices_v || !dsc_cfg->dc_dsc_cfg.num_slices_h || !(dsc_cfg->dc_dsc_cfg.version_minor == 1 || dsc_cfg->dc_dsc_cfg.version_minor == 2) || !dsc_cfg->pic_width || !dsc_cfg->pic_height || !((dsc_cfg->dc_dsc_cfg.version_minor == 1 && // v1.1 line buffer depth range: @@ -315,54 +335,54 @@ static bool dsc_prepare_config(struct display_stream_compressor *dsc, const stru return false; } - dsc_init_reg_values(&dsc20->reg_vals); + dsc_init_reg_values(dsc_reg_vals); /* Copy input config */ - dsc20->reg_vals.pixel_format = dsc_dc_pixel_encoding_to_dsc_pixel_format(dsc_cfg->pixel_encoding, dsc_cfg->dc_dsc_cfg.ycbcr422_simple); - dsc20->reg_vals.num_slices_h = dsc_cfg->dc_dsc_cfg.num_slices_h; - dsc20->reg_vals.num_slices_v = dsc_cfg->dc_dsc_cfg.num_slices_v; - dsc20->reg_vals.pps.dsc_version_minor = dsc_cfg->dc_dsc_cfg.version_minor; - dsc20->reg_vals.pps.pic_width = dsc_cfg->pic_width; - dsc20->reg_vals.pps.pic_height = dsc_cfg->pic_height; - dsc20->reg_vals.pps.bits_per_component = dsc_dc_color_depth_to_dsc_bits_per_comp(dsc_cfg->color_depth); - dsc20->reg_vals.pps.block_pred_enable = dsc_cfg->dc_dsc_cfg.block_pred_enable; - dsc20->reg_vals.pps.line_buf_depth = dsc_cfg->dc_dsc_cfg.linebuf_depth; - dsc20->reg_vals.alternate_ich_encoding_en = dsc20->reg_vals.pps.dsc_version_minor == 1 ? 0 : 1; + dsc_reg_vals->pixel_format = dsc_dc_pixel_encoding_to_dsc_pixel_format(dsc_cfg->pixel_encoding, dsc_cfg->dc_dsc_cfg.ycbcr422_simple); + dsc_reg_vals->num_slices_h = dsc_cfg->dc_dsc_cfg.num_slices_h; + dsc_reg_vals->num_slices_v = dsc_cfg->dc_dsc_cfg.num_slices_v; + dsc_reg_vals->pps.dsc_version_minor = dsc_cfg->dc_dsc_cfg.version_minor; + dsc_reg_vals->pps.pic_width = dsc_cfg->pic_width; + dsc_reg_vals->pps.pic_height = dsc_cfg->pic_height; + dsc_reg_vals->pps.bits_per_component = dsc_dc_color_depth_to_dsc_bits_per_comp(dsc_cfg->color_depth); + dsc_reg_vals->pps.block_pred_enable = dsc_cfg->dc_dsc_cfg.block_pred_enable; + dsc_reg_vals->pps.line_buf_depth = dsc_cfg->dc_dsc_cfg.linebuf_depth; + dsc_reg_vals->alternate_ich_encoding_en = dsc_reg_vals->pps.dsc_version_minor == 1 ? 0 : 1; // TODO: in addition to validating slice height (pic height must be divisible by slice height), // see what happens when the same condition doesn't apply for slice_width/pic_width. - dsc20->reg_vals.pps.slice_width = dsc_cfg->pic_width / dsc_cfg->dc_dsc_cfg.num_slices_h; - dsc20->reg_vals.pps.slice_height = dsc_cfg->pic_height / dsc_cfg->dc_dsc_cfg.num_slices_v; + dsc_reg_vals->pps.slice_width = dsc_cfg->pic_width / dsc_cfg->dc_dsc_cfg.num_slices_h; + dsc_reg_vals->pps.slice_height = dsc_cfg->pic_height / dsc_cfg->dc_dsc_cfg.num_slices_v; - ASSERT(dsc20->reg_vals.pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height); - if (!(dsc20->reg_vals.pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height)) { + ASSERT(dsc_reg_vals->pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height); + if (!(dsc_reg_vals->pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height)) { dm_output_to_console("%s: pix height %d not divisible by num_slices_v %d\n\n", __func__, dsc_cfg->pic_height, dsc_cfg->dc_dsc_cfg.num_slices_v); return false; } - dsc20->reg_vals.bpp_x32 = dsc_cfg->dc_dsc_cfg.bits_per_pixel << 1; - if (dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR420 || dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR422) - dsc20->reg_vals.pps.bits_per_pixel = dsc20->reg_vals.bpp_x32; + dsc_reg_vals->bpp_x32 = dsc_cfg->dc_dsc_cfg.bits_per_pixel << 1; + if (dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR420 || dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR422) + dsc_reg_vals->pps.bits_per_pixel = dsc_reg_vals->bpp_x32; else - dsc20->reg_vals.pps.bits_per_pixel = dsc20->reg_vals.bpp_x32 >> 1; + dsc_reg_vals->pps.bits_per_pixel = dsc_reg_vals->bpp_x32 >> 1; - dsc20->reg_vals.pps.convert_rgb = dsc20->reg_vals.pixel_format == DSC_PIXFMT_RGB ? 1 : 0; - dsc20->reg_vals.pps.native_422 = (dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR422); - dsc20->reg_vals.pps.native_420 = (dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR420); - dsc20->reg_vals.pps.simple_422 = (dsc20->reg_vals.pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422); + dsc_reg_vals->pps.convert_rgb = dsc_reg_vals->pixel_format == DSC_PIXFMT_RGB ? 1 : 0; + dsc_reg_vals->pps.native_422 = (dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR422); + dsc_reg_vals->pps.native_420 = (dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR420); + dsc_reg_vals->pps.simple_422 = (dsc_reg_vals->pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422); - if (dscc_compute_dsc_parameters(&dsc20->reg_vals.pps, &dsc_params)) { + if (dscc_compute_dsc_parameters(&dsc_reg_vals->pps, &dsc_params)) { dm_output_to_console("%s: DSC config failed\n", __func__); return false; } - dsc_update_from_dsc_parameters(&dsc20->reg_vals, &dsc_params); + dsc_update_from_dsc_parameters(dsc_reg_vals, &dsc_params); dsc_optc_cfg->bytes_per_pixel = dsc_params.bytes_per_pixel; - dsc_optc_cfg->slice_width = dsc20->reg_vals.pps.slice_width; - dsc_optc_cfg->is_pixel_format_444 = dsc20->reg_vals.pixel_format == DSC_PIXFMT_RGB || - dsc20->reg_vals.pixel_format == DSC_PIXFMT_YCBCR444 || - dsc20->reg_vals.pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422; + dsc_optc_cfg->slice_width = dsc_reg_vals->pps.slice_width; + dsc_optc_cfg->is_pixel_format_444 = dsc_reg_vals->pixel_format == DSC_PIXFMT_RGB || + dsc_reg_vals->pixel_format == DSC_PIXFMT_YCBCR444 || + dsc_reg_vals->pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422; return true; } @@ -427,6 +447,8 @@ static void dsc_init_reg_values(struct dsc_reg_values *reg_vals) { int i; + memset(reg_vals, 0, sizeof(struct dsc_reg_values)); + /* Non-PPS values */ reg_vals->dsc_clock_enable = 1; reg_vals->dsc_clock_gating_disable = 0; @@ -436,7 +458,7 @@ static void dsc_init_reg_values(struct dsc_reg_values *reg_vals) reg_vals->ich_reset_at_eol = 0; reg_vals->alternate_ich_encoding_en = 0; reg_vals->rc_buffer_model_size = 0; - reg_vals->disable_ich = 0; + /*reg_vals->disable_ich = 0;*/ reg_vals->dsc_dbg_en = 0; for (i = 0; i < 4; i++) @@ -518,9 +540,11 @@ static void dsc_write_to_registers(struct display_stream_compressor *dsc, const ALTERNATE_ICH_ENCODING_EN, reg_vals->alternate_ich_encoding_en, NUMBER_OF_SLICES_IN_VERTICAL_DIRECTION, reg_vals->num_slices_v - 1); - REG_SET_2(DSCC_CONFIG1, 0, + REG_SET(DSCC_CONFIG1, 0, + DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE, reg_vals->rc_buffer_model_size); + /*REG_SET_2(DSCC_CONFIG1, 0, DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE, reg_vals->rc_buffer_model_size, - DSCC_DISABLE_ICH, reg_vals->disable_ich); + DSCC_DISABLE_ICH, reg_vals->disable_ich);*/ REG_SET_4(DSCC_INTERRUPT_CONTROL_STATUS, 0, DSCC_RATE_CONTROL_BUFFER_MODEL0_OVERFLOW_OCCURRED_INT_EN, reg_vals->rc_buffer_model_overflow_int_en[0], diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h index 168865a16288..4e2fb38390a4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h @@ -103,7 +103,7 @@ DSC_SF(DSCC0_DSCC_CONFIG0, ALTERNATE_ICH_ENCODING_EN, mask_sh), \ DSC_SF(DSCC0_DSCC_CONFIG0, NUMBER_OF_SLICES_IN_VERTICAL_DIRECTION, mask_sh), \ DSC_SF(DSCC0_DSCC_CONFIG1, DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE, mask_sh), \ - DSC_SF(DSCC0_DSCC_CONFIG1, DSCC_DISABLE_ICH, mask_sh), \ + /*DSC_SF(DSCC0_DSCC_CONFIG1, DSCC_DISABLE_ICH, mask_sh),*/ \ DSC_SF(DSCC0_DSCC_STATUS, DSCC_DOUBLE_BUFFER_REG_UPDATE_PENDING, mask_sh), \ DSC_SF(DSCC0_DSCC_INTERRUPT_CONTROL_STATUS, DSCC_RATE_BUFFER0_OVERFLOW_OCCURRED, mask_sh), \ DSC_SF(DSCC0_DSCC_INTERRUPT_CONTROL_STATUS, DSCC_RATE_BUFFER1_OVERFLOW_OCCURRED, mask_sh), \ @@ -278,7 +278,7 @@ type ALTERNATE_ICH_ENCODING_EN; \ type NUMBER_OF_SLICES_IN_VERTICAL_DIRECTION; \ type DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE; \ - type DSCC_DISABLE_ICH; \ + /*type DSCC_DISABLE_ICH;*/ \ type DSCC_DOUBLE_BUFFER_REG_UPDATE_PENDING; \ type DSCC_RATE_BUFFER0_OVERFLOW_OCCURRED; \ type DSCC_RATE_BUFFER1_OVERFLOW_OCCURRED; \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c index 6e2dbd03f9bf..b83c022e2c6f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c @@ -26,6 +26,7 @@ #include "dcn20_hubbub.h" #include "reg_helper.h" +#include "clk_mgr.h" #define REG(reg)\ hubbub1->regs->reg @@ -379,6 +380,11 @@ int hubbub2_init_dchub_sys_ctx(struct hubbub *hubbub, REG_SET(DCN_VM_AGP_BASE, 0, AGP_BASE, pa_config->system_aperture.agp_base >> 24); + REG_SET(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, 0, + DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, (pa_config->page_table_default_page_addr >> 44) & 0xF); + REG_SET(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, 0, + DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, (pa_config->page_table_default_page_addr >> 12) & 0xFFFFFFFF); + if (pa_config->gart_config.page_table_start_addr != pa_config->gart_config.page_table_end_addr) { phys_config.page_table_start_addr = pa_config->gart_config.page_table_start_addr >> 12; phys_config.page_table_end_addr = pa_config->gart_config.page_table_end_addr >> 12; @@ -397,54 +403,67 @@ void hubbub2_update_dchub(struct hubbub *hubbub, { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); - if (REG(DCHUBBUB_SDPIF_FB_TOP) == 0) { - ASSERT(false); - /*should not come here*/ + if (REG(DCN_VM_FB_LOCATION_TOP) == 0) return; - } - /* TODO: port code from dal2 */ + switch (dh_data->fb_mode) { case FRAME_BUFFER_MODE_ZFB_ONLY: /*For ZFB case need to put DCHUB FB BASE and TOP upside down to indicate ZFB mode*/ - REG_UPDATE(DCHUBBUB_SDPIF_FB_TOP, - SDPIF_FB_TOP, 0); - - REG_UPDATE(DCHUBBUB_SDPIF_FB_BASE, - SDPIF_FB_BASE, 0x0FFFF); - - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BASE, - SDPIF_AGP_BASE, dh_data->zfb_phys_addr_base >> 22); - - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BOT, - SDPIF_AGP_BOT, dh_data->zfb_mc_base_addr >> 22); - - REG_UPDATE(DCHUBBUB_SDPIF_AGP_TOP, - SDPIF_AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 22); + REG_UPDATE(DCN_VM_FB_LOCATION_TOP, + FB_TOP, 0); + + REG_UPDATE(DCN_VM_FB_LOCATION_BASE, + FB_BASE, 0xFFFFFF); + + /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ + REG_UPDATE(DCN_VM_AGP_BASE, + AGP_BASE, dh_data->zfb_phys_addr_base >> 24); + + /*This field defines the bottom range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_BOT, + AGP_BOT, dh_data->zfb_mc_base_addr >> 24); + + /*This field defines the top range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_TOP, + AGP_TOP, (dh_data->zfb_mc_base_addr + + dh_data->zfb_size_in_byte - 1) >> 24); break; case FRAME_BUFFER_MODE_MIXED_ZFB_AND_LOCAL: /*Should not touch FB LOCATION (done by VBIOS on AsicInit table)*/ - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BASE, - SDPIF_AGP_BASE, dh_data->zfb_phys_addr_base >> 22); + /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ + REG_UPDATE(DCN_VM_AGP_BASE, + AGP_BASE, dh_data->zfb_phys_addr_base >> 24); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BOT, - SDPIF_AGP_BOT, dh_data->zfb_mc_base_addr >> 22); + /*This field defines the bottom range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_BOT, + AGP_BOT, dh_data->zfb_mc_base_addr >> 24); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_TOP, - SDPIF_AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 22); + /*This field defines the top range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_TOP, + AGP_TOP, (dh_data->zfb_mc_base_addr + + dh_data->zfb_size_in_byte - 1) >> 24); break; case FRAME_BUFFER_MODE_LOCAL_ONLY: - /*Should not touch FB LOCATION (done by VBIOS on AsicInit table)*/ - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BASE, - SDPIF_AGP_BASE, 0); + /*Should not touch FB LOCATION (should be done by VBIOS)*/ - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BOT, - SDPIF_AGP_BOT, 0X03FFFF); + /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ + REG_UPDATE(DCN_VM_AGP_BASE, + AGP_BASE, 0); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_TOP, - SDPIF_AGP_TOP, 0); + /*This field defines the bottom range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_BOT, + AGP_BOT, 0xFFFFFF); + + /*This field defines the top range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_TOP, + AGP_TOP, 0); break; default: break; @@ -553,6 +572,16 @@ static void hubbub2_program_watermarks( */ hubbub1_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); hubbub1_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + + /* + * There's a special case when going from p-state support to p-state unsupported + * here we are going to LOWER watermarks to go to dummy p-state only, but this has + * to be done prepare_bandwidth, not optimize + */ + if (hubbub1->base.ctx->dc->clk_mgr->clks.prev_p_state_change_support == true && + hubbub1->base.ctx->dc->clk_mgr->clks.p_state_change_support == false) + safe_to_lower = true; + hubbub1_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); REG_SET(DCHUBBUB_ARB_SAT_LEVEL, 0, @@ -571,7 +600,7 @@ static const struct hubbub_funcs hubbub2_funcs = { .get_dcc_compression_cap = hubbub2_get_dcc_compression_cap, .wm_read_state = hubbub2_wm_read_state, .get_dchub_ref_freq = hubbub2_get_dchub_ref_freq, - .program_watermarks = hubbub2_program_watermarks, + .program_watermarks = hubbub2_program_watermarks }; void hubbub2_construct(struct dcn20_hubbub *hubbub, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h index a7b6ca26a9ad..626117d3b4e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h @@ -29,13 +29,21 @@ #include "dcn10/dcn10_hubbub.h" #include "dcn20_vmid.h" +#define HUBBUB_REG_LIST_DCN20_COMMON()\ + HUBBUB_REG_LIST_DCN_COMMON(), \ + SR(DCHUBBUB_CRC_CTRL), \ + SR(DCN_VM_FB_LOCATION_BASE),\ + SR(DCN_VM_FB_LOCATION_TOP),\ + SR(DCN_VM_FB_OFFSET),\ + SR(DCN_VM_AGP_BOT),\ + SR(DCN_VM_AGP_TOP),\ + SR(DCN_VM_AGP_BASE) + #define TO_DCN20_HUBBUB(hubbub)\ container_of(hubbub, struct dcn20_hubbub, base) -#define HUBBUB_REG_LIST_DCN20(id)\ +#define HUBBUB_REG_LIST_DCN20_COMMON()\ HUBBUB_REG_LIST_DCN_COMMON(), \ - HUBBUB_VM_REG_LIST(), \ - HUBBUB_SR_WATERMARK_REG_LIST(), \ SR(DCHUBBUB_CRC_CTRL), \ SR(DCN_VM_FB_LOCATION_BASE),\ SR(DCN_VM_FB_LOCATION_TOP),\ @@ -44,6 +52,14 @@ SR(DCN_VM_AGP_TOP),\ SR(DCN_VM_AGP_BASE) +#define HUBBUB_REG_LIST_DCN20(id)\ + HUBBUB_REG_LIST_DCN20_COMMON(), \ + HUBBUB_SR_WATERMARK_REG_LIST(), \ + HUBBUB_VM_REG_LIST(),\ + SR(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB),\ + SR(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB) + + #define HUBBUB_MASK_SH_LIST_DCN20(mask_sh)\ HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh), \ HUBBUB_MASK_SH_LIST_STUTTER(mask_sh), \ @@ -53,7 +69,9 @@ HUBBUB_SF(DCN_VM_FB_OFFSET, FB_OFFSET, mask_sh), \ HUBBUB_SF(DCN_VM_AGP_BOT, AGP_BOT, mask_sh), \ HUBBUB_SF(DCN_VM_AGP_TOP, AGP_TOP, mask_sh), \ - HUBBUB_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh) + HUBBUB_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh), \ + HUBBUB_SF(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, mask_sh), \ + HUBBUB_SF(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, mask_sh) struct dcn20_hubbub { struct hubbub base; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c index d3f7dd374d50..69e2aae42394 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c @@ -40,81 +40,6 @@ #define FN(reg_name, field_name) \ hubp2->hubp_shift->field_name, hubp2->hubp_mask->field_name -void hubp2_update_dchub( - struct hubp *hubp, - struct dchub_init_data *dh_data) -{ - struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); - if (REG(DCN_VM_FB_LOCATION_TOP) == 0) - return; - - switch (dh_data->fb_mode) { - case FRAME_BUFFER_MODE_ZFB_ONLY: - /*For ZFB case need to put DCHUB FB BASE and TOP upside down to indicate ZFB mode*/ - REG_UPDATE(DCN_VM_FB_LOCATION_TOP, - FB_TOP, 0); - - REG_UPDATE(DCN_VM_FB_LOCATION_BASE, - FB_BASE, 0xFFFFFF); - - /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ - REG_UPDATE(DCN_VM_AGP_BASE, - AGP_BASE, dh_data->zfb_phys_addr_base >> 24); - - /*This field defines the bottom range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_BOT, - AGP_BOT, dh_data->zfb_mc_base_addr >> 24); - - /*This field defines the top range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_TOP, - AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 24); - break; - case FRAME_BUFFER_MODE_MIXED_ZFB_AND_LOCAL: - /*Should not touch FB LOCATION (done by VBIOS on AsicInit table)*/ - - /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ - REG_UPDATE(DCN_VM_AGP_BASE, - AGP_BASE, dh_data->zfb_phys_addr_base >> 24); - - /*This field defines the bottom range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_BOT, - AGP_BOT, dh_data->zfb_mc_base_addr >> 24); - - /*This field defines the top range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_TOP, - AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 24); - break; - case FRAME_BUFFER_MODE_LOCAL_ONLY: - /*Should not touch FB LOCATION (should be done by VBIOS)*/ - - /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ - REG_UPDATE(DCN_VM_AGP_BASE, - AGP_BASE, 0); - - /*This field defines the bottom range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_BOT, - AGP_BOT, 0xFFFFFF); - - /*This field defines the top range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_TOP, - AGP_TOP, 0); - break; - default: - break; - } - - dh_data->dchub_initialzied = true; - dh_data->dchub_info_valid = false; -} - void hubp2_set_vm_system_aperture_settings(struct hubp *hubp, struct vm_system_aperture_param *apt) { @@ -156,7 +81,85 @@ void hubp2_program_deadline( { struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); - hubp1_program_deadline(hubp, dlg_attr, ttu_attr); + /* DLG - Per hubp */ + REG_SET_2(BLANK_OFFSET_0, 0, + REFCYC_H_BLANK_END, dlg_attr->refcyc_h_blank_end, + DLG_V_BLANK_END, dlg_attr->dlg_vblank_end); + + REG_SET(BLANK_OFFSET_1, 0, + MIN_DST_Y_NEXT_START, dlg_attr->min_dst_y_next_start); + + REG_SET(DST_DIMENSIONS, 0, + REFCYC_PER_HTOTAL, dlg_attr->refcyc_per_htotal); + + REG_SET_2(DST_AFTER_SCALER, 0, + REFCYC_X_AFTER_SCALER, dlg_attr->refcyc_x_after_scaler, + DST_Y_AFTER_SCALER, dlg_attr->dst_y_after_scaler); + + REG_SET(REF_FREQ_TO_PIX_FREQ, 0, + REF_FREQ_TO_PIX_FREQ, dlg_attr->ref_freq_to_pix_freq); + + /* DLG - Per luma/chroma */ + REG_SET(VBLANK_PARAMETERS_1, 0, + REFCYC_PER_PTE_GROUP_VBLANK_L, dlg_attr->refcyc_per_pte_group_vblank_l); + + if (REG(NOM_PARAMETERS_0)) + REG_SET(NOM_PARAMETERS_0, 0, + DST_Y_PER_PTE_ROW_NOM_L, dlg_attr->dst_y_per_pte_row_nom_l); + + if (REG(NOM_PARAMETERS_1)) + REG_SET(NOM_PARAMETERS_1, 0, + REFCYC_PER_PTE_GROUP_NOM_L, dlg_attr->refcyc_per_pte_group_nom_l); + + REG_SET(NOM_PARAMETERS_4, 0, + DST_Y_PER_META_ROW_NOM_L, dlg_attr->dst_y_per_meta_row_nom_l); + + REG_SET(NOM_PARAMETERS_5, 0, + REFCYC_PER_META_CHUNK_NOM_L, dlg_attr->refcyc_per_meta_chunk_nom_l); + + REG_SET_2(PER_LINE_DELIVERY, 0, + REFCYC_PER_LINE_DELIVERY_L, dlg_attr->refcyc_per_line_delivery_l, + REFCYC_PER_LINE_DELIVERY_C, dlg_attr->refcyc_per_line_delivery_c); + + REG_SET(VBLANK_PARAMETERS_2, 0, + REFCYC_PER_PTE_GROUP_VBLANK_C, dlg_attr->refcyc_per_pte_group_vblank_c); + + if (REG(NOM_PARAMETERS_2)) + REG_SET(NOM_PARAMETERS_2, 0, + DST_Y_PER_PTE_ROW_NOM_C, dlg_attr->dst_y_per_pte_row_nom_c); + + if (REG(NOM_PARAMETERS_3)) + REG_SET(NOM_PARAMETERS_3, 0, + REFCYC_PER_PTE_GROUP_NOM_C, dlg_attr->refcyc_per_pte_group_nom_c); + + REG_SET(NOM_PARAMETERS_6, 0, + DST_Y_PER_META_ROW_NOM_C, dlg_attr->dst_y_per_meta_row_nom_c); + + REG_SET(NOM_PARAMETERS_7, 0, + REFCYC_PER_META_CHUNK_NOM_C, dlg_attr->refcyc_per_meta_chunk_nom_c); + + /* TTU - per hubp */ + REG_SET_2(DCN_TTU_QOS_WM, 0, + QoS_LEVEL_LOW_WM, ttu_attr->qos_level_low_wm, + QoS_LEVEL_HIGH_WM, ttu_attr->qos_level_high_wm); + + /* TTU - per luma/chroma */ + /* Assumed surf0 is luma and 1 is chroma */ + + REG_SET_3(DCN_SURF0_TTU_CNTL0, 0, + REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_l, + QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_l, + QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_l); + + REG_SET_3(DCN_SURF1_TTU_CNTL0, 0, + REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_c, + QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_c, + QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_c); + + REG_SET_3(DCN_CUR0_TTU_CNTL0, 0, + REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_cur0, + QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_cur0, + QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_cur0); REG_SET(FLIP_PARAMETERS_1, 0, REFCYC_PER_PTE_GROUP_FLIP_L, dlg_attr->refcyc_per_pte_group_flip_l); @@ -184,6 +187,39 @@ void hubp2_vready_at_or_After_vsync(struct hubp *hubp, REG_UPDATE(DCHUBP_CNTL, HUBP_VREADY_AT_OR_AFTER_VSYNC, value); } +void hubp2_program_requestor( + struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, rq_regs->plane1_base_address); + REG_SET_4(DCN_EXPANSION_MODE, 0, + DRQ_EXPANSION_MODE, rq_regs->drq_expansion_mode, + PRQ_EXPANSION_MODE, rq_regs->prq_expansion_mode, + MRQ_EXPANSION_MODE, rq_regs->mrq_expansion_mode, + CRQ_EXPANSION_MODE, rq_regs->crq_expansion_mode); + REG_SET_8(DCHUBP_REQ_SIZE_CONFIG, 0, + CHUNK_SIZE, rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, rq_regs->rq_regs_l.dpte_group_size, + MPTE_GROUP_SIZE, rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, rq_regs->rq_regs_l.pte_row_height_linear); + REG_SET_8(DCHUBP_REQ_SIZE_CONFIG_C, 0, + CHUNK_SIZE_C, rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, rq_regs->rq_regs_c.dpte_group_size, + MPTE_GROUP_SIZE_C, rq_regs->rq_regs_c.mpte_group_size, + SWATH_HEIGHT_C, rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, rq_regs->rq_regs_c.pte_row_height_linear); +} + static void hubp2_setup( struct hubp *hubp, struct _vcs_dpi_display_dlg_regs_st *dlg_attr, @@ -196,7 +232,7 @@ static void hubp2_setup( */ hubp2_vready_at_or_After_vsync(hubp, pipe_dest); - hubp1_program_requestor(hubp, rq_regs); + hubp2_program_requestor(hubp, rq_regs); hubp2_program_deadline(hubp, dlg_attr, ttu_attr); } @@ -283,11 +319,205 @@ static void hubp2_program_tiling( PIPE_ALIGNED, 0); } +void hubp2_program_size( + struct hubp *hubp, + enum surface_pixel_format format, + const struct plane_size *plane_size, + struct dc_plane_dcc_param *dcc) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t pitch, meta_pitch, pitch_c, meta_pitch_c; + bool use_pitch_c = false; + + /* Program data and meta surface pitch (calculation from addrlib) + * 444 or 420 luma + */ + use_pitch_c = format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN + && format < SURFACE_PIXEL_FORMAT_SUBSAMPLE_END; + if (use_pitch_c) { + ASSERT(plane_size->chroma_pitch != 0); + /* Chroma pitch zero can cause system hang! */ + + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; + pitch_c = plane_size->chroma_pitch - 1; + meta_pitch_c = dcc->meta_pitch_c - 1; + } else { + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; + pitch_c = 0; + meta_pitch_c = 0; + } + + if (!dcc->enable) { + meta_pitch = 0; + meta_pitch_c = 0; + } + + REG_UPDATE_2(DCSURF_SURFACE_PITCH, + PITCH, pitch, META_PITCH, meta_pitch); + + use_pitch_c = format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN; + if (use_pitch_c) + REG_UPDATE_2(DCSURF_SURFACE_PITCH_C, + PITCH_C, pitch_c, META_PITCH_C, meta_pitch_c); +} + +void hubp2_program_rotation( + struct hubp *hubp, + enum dc_rotation_angle rotation, + bool horizontal_mirror) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t mirror; + + + if (horizontal_mirror) + mirror = 1; + else + mirror = 0; + + /* Program rotation angle and horz mirror - no mirror */ + if (rotation == ROTATION_ANGLE_0) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 0, + H_MIRROR_EN, mirror); + else if (rotation == ROTATION_ANGLE_90) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 1, + H_MIRROR_EN, mirror); + else if (rotation == ROTATION_ANGLE_180) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 2, + H_MIRROR_EN, mirror); + else if (rotation == ROTATION_ANGLE_270) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 3, + H_MIRROR_EN, mirror); +} + +void hubp2_dcc_control(struct hubp *hubp, bool enable, + enum hubp_ind_block_size independent_64b_blks) +{ + uint32_t dcc_en = enable ? 1 : 0; + uint32_t dcc_ind_64b_blk = independent_64b_blks ? 1 : 0; + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE_4(DCSURF_SURFACE_CONTROL, + PRIMARY_SURFACE_DCC_EN, dcc_en, + PRIMARY_SURFACE_DCC_IND_64B_BLK, dcc_ind_64b_blk, + SECONDARY_SURFACE_DCC_EN, dcc_en, + SECONDARY_SURFACE_DCC_IND_64B_BLK, dcc_ind_64b_blk); +} + +void hubp2_program_pixel_format( + struct hubp *hubp, + enum surface_pixel_format format) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t red_bar = 3; + uint32_t blue_bar = 2; + + /* swap for ABGR format */ + if (format == SURFACE_PIXEL_FORMAT_GRPH_ABGR8888 + || format == SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010 + || format == SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010_XR_BIAS + || format == SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F) { + red_bar = 2; + blue_bar = 3; + } + + REG_UPDATE_2(HUBPRET_CONTROL, + CROSSBAR_SRC_CB_B, blue_bar, + CROSSBAR_SRC_CR_R, red_bar); + + /* Mapping is same as ipp programming (cnvc) */ + + switch (format) { + case SURFACE_PIXEL_FORMAT_GRPH_ARGB1555: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 1); + break; + case SURFACE_PIXEL_FORMAT_GRPH_RGB565: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 3); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB8888: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR8888: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 8); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB2101010: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010_XR_BIAS: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 10); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 22); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F:/*we use crossbar already*/ + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 24); + break; + + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 65); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 64); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 67); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 66); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_AYCrCb8888: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 12); + break; +#if defined(CONFIG_DRM_AMD_DC_DCN2_0) + case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FIX: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 112); + break; + case SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FIX: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 113); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_ACrYCb2101010: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 114); + break; + case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FLOAT: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 118); + break; + case SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FLOAT: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 119); + break; +#endif + default: + BREAK_TO_DEBUGGER(); + break; + } + + /* don't see the need of program the xbar in DCN 1.0 */ +} + void hubp2_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, @@ -295,11 +525,11 @@ void hubp2_program_surface_config( { struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); - hubp1_dcc_control(hubp, dcc->enable, dcc->grph.independent_64b_blks); + hubp2_dcc_control(hubp, dcc->enable, dcc->independent_64b_blks); hubp2_program_tiling(hubp2, tiling_info, format); - hubp1_program_size(hubp, format, plane_size, dcc); - hubp1_program_rotation(hubp, rotation, horizontal_mirror); - hubp1_program_pixel_format(hubp, format); + hubp2_program_size(hubp, format, plane_size, dcc); + hubp2_program_rotation(hubp, rotation, horizontal_mirror); + hubp2_program_pixel_format(hubp, format); } enum cursor_lines_per_chunk hubp2_get_lines_per_chunk( @@ -652,28 +882,388 @@ void hubp2_set_flip_control_surface_gsl(struct hubp *hubp, bool enable) REG_UPDATE(DCSURF_FLIP_CONTROL2, SURFACE_GSL_ENABLE, enable ? 1 : 0); } +bool hubp2_is_flip_pending(struct hubp *hubp) +{ + uint32_t flip_pending = 0; + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + struct dc_plane_address earliest_inuse_address; + + REG_GET(DCSURF_FLIP_CONTROL, + SURFACE_FLIP_PENDING, &flip_pending); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE, + SURFACE_EARLIEST_INUSE_ADDRESS, &earliest_inuse_address.grph.addr.low_part); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE_HIGH, + SURFACE_EARLIEST_INUSE_ADDRESS_HIGH, &earliest_inuse_address.grph.addr.high_part); + + if (flip_pending) + return true; + + if (earliest_inuse_address.grph.addr.quad_part != hubp->request_address.grph.addr.quad_part) + return true; + + return false; +} + +void hubp2_set_blank(struct hubp *hubp, bool blank) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t blank_en = blank ? 1 : 0; + + REG_UPDATE_2(DCHUBP_CNTL, + HUBP_BLANK_EN, blank_en, + HUBP_TTU_DISABLE, blank_en); + + if (blank) { + uint32_t reg_val = REG_READ(DCHUBP_CNTL); + + if (reg_val) { + /* init sequence workaround: in case HUBP is + * power gated, this wait would timeout. + * + * we just wrote reg_val to non-0, if it stay 0 + * it means HUBP is gated + */ + REG_WAIT(DCHUBP_CNTL, + HUBP_NO_OUTSTANDING_REQ, 1, + 1, 200); + } + + hubp->mpcc_id = 0xf; + hubp->opp_id = OPP_ID_INVALID; + } +} + +void hubp2_cursor_set_position( + struct hubp *hubp, + const struct dc_cursor_position *pos, + const struct dc_cursor_mi_param *param) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + int src_x_offset = pos->x - pos->x_hotspot - param->viewport.x; + int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y; + int x_hotspot = pos->x_hotspot; + int y_hotspot = pos->y_hotspot; + int cursor_height = (int)hubp->curs_attr.height; + int cursor_width = (int)hubp->curs_attr.width; + uint32_t dst_x_offset; + uint32_t cur_en = pos->enable ? 1 : 0; + + /* + * Guard aganst cursor_set_position() from being called with invalid + * attributes + * + * TODO: Look at combining cursor_set_position() and + * cursor_set_attributes() into cursor_update() + */ + if (hubp->curs_attr.address.quad_part == 0) + return; + + // Rotated cursor width/height and hotspots tweaks for offset calculation + if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) { + swap(cursor_height, cursor_width); + if (param->rotation == ROTATION_ANGLE_90) { + src_x_offset = pos->x - pos->y_hotspot - param->viewport.x; + src_y_offset = pos->y - pos->x_hotspot - param->viewport.y; + } + } else if (param->rotation == ROTATION_ANGLE_180) { + src_x_offset = pos->x - param->viewport.x; + src_y_offset = pos->y - param->viewport.y; + } + + if (param->mirror) { + x_hotspot = param->viewport.width - x_hotspot; + src_x_offset = param->viewport.x + param->viewport.width - src_x_offset; + } + + dst_x_offset = (src_x_offset >= 0) ? src_x_offset : 0; + dst_x_offset *= param->ref_clk_khz; + dst_x_offset /= param->pixel_clk_khz; + + ASSERT(param->h_scale_ratio.value); + + if (param->h_scale_ratio.value) + dst_x_offset = dc_fixpt_floor(dc_fixpt_div( + dc_fixpt_from_int(dst_x_offset), + param->h_scale_ratio)); + + if (src_x_offset >= (int)param->viewport.width) + cur_en = 0; /* not visible beyond right edge*/ + + if (src_x_offset + cursor_width <= 0) + cur_en = 0; /* not visible beyond left edge*/ + + if (src_y_offset >= (int)param->viewport.height) + cur_en = 0; /* not visible beyond bottom edge*/ + + if (src_y_offset + cursor_height <= 0) + cur_en = 0; /* not visible beyond top edge*/ + + if (cur_en && REG_READ(CURSOR_SURFACE_ADDRESS) == 0) + hubp->funcs->set_cursor_attributes(hubp, &hubp->curs_attr); + + REG_UPDATE(CURSOR_CONTROL, + CURSOR_ENABLE, cur_en); + + REG_SET_2(CURSOR_POSITION, 0, + CURSOR_X_POSITION, pos->x, + CURSOR_Y_POSITION, pos->y); + + REG_SET_2(CURSOR_HOT_SPOT, 0, + CURSOR_HOT_SPOT_X, x_hotspot, + CURSOR_HOT_SPOT_Y, y_hotspot); + + REG_SET(CURSOR_DST_OFFSET, 0, + CURSOR_DST_X_OFFSET, dst_x_offset); + /* TODO Handle surface pixel formats other than 4:4:4 */ +} + +void hubp2_clk_cntl(struct hubp *hubp, bool enable) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t clk_enable = enable ? 1 : 0; + + REG_UPDATE(HUBP_CLK_CNTL, HUBP_CLOCK_ENABLE, clk_enable); +} + +void hubp2_vtg_sel(struct hubp *hubp, uint32_t otg_inst) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE(DCHUBP_CNTL, HUBP_VTG_SEL, otg_inst); +} + +void hubp2_clear_underflow(struct hubp *hubp) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE(DCHUBP_CNTL, HUBP_UNDERFLOW_CLEAR, 1); +} + +void hubp2_read_state_common(struct hubp *hubp) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + struct dcn_hubp_state *s = &hubp2->state; + struct _vcs_dpi_display_dlg_regs_st *dlg_attr = &s->dlg_attr; + struct _vcs_dpi_display_ttu_regs_st *ttu_attr = &s->ttu_attr; + struct _vcs_dpi_display_rq_regs_st *rq_regs = &s->rq_regs; + + /* Requester */ + REG_GET(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, &rq_regs->plane1_base_address); + REG_GET_4(DCN_EXPANSION_MODE, + DRQ_EXPANSION_MODE, &rq_regs->drq_expansion_mode, + PRQ_EXPANSION_MODE, &rq_regs->prq_expansion_mode, + MRQ_EXPANSION_MODE, &rq_regs->mrq_expansion_mode, + CRQ_EXPANSION_MODE, &rq_regs->crq_expansion_mode); + + /* DLG - Per hubp */ + REG_GET_2(BLANK_OFFSET_0, + REFCYC_H_BLANK_END, &dlg_attr->refcyc_h_blank_end, + DLG_V_BLANK_END, &dlg_attr->dlg_vblank_end); + + REG_GET(BLANK_OFFSET_1, + MIN_DST_Y_NEXT_START, &dlg_attr->min_dst_y_next_start); + + REG_GET(DST_DIMENSIONS, + REFCYC_PER_HTOTAL, &dlg_attr->refcyc_per_htotal); + + REG_GET_2(DST_AFTER_SCALER, + REFCYC_X_AFTER_SCALER, &dlg_attr->refcyc_x_after_scaler, + DST_Y_AFTER_SCALER, &dlg_attr->dst_y_after_scaler); + + if (REG(PREFETCH_SETTINS)) + REG_GET_2(PREFETCH_SETTINS, + DST_Y_PREFETCH, &dlg_attr->dst_y_prefetch, + VRATIO_PREFETCH, &dlg_attr->vratio_prefetch); + else + REG_GET_2(PREFETCH_SETTINGS, + DST_Y_PREFETCH, &dlg_attr->dst_y_prefetch, + VRATIO_PREFETCH, &dlg_attr->vratio_prefetch); + + REG_GET_2(VBLANK_PARAMETERS_0, + DST_Y_PER_VM_VBLANK, &dlg_attr->dst_y_per_vm_vblank, + DST_Y_PER_ROW_VBLANK, &dlg_attr->dst_y_per_row_vblank); + + REG_GET(REF_FREQ_TO_PIX_FREQ, + REF_FREQ_TO_PIX_FREQ, &dlg_attr->ref_freq_to_pix_freq); + + /* DLG - Per luma/chroma */ + REG_GET(VBLANK_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_VBLANK_L, &dlg_attr->refcyc_per_pte_group_vblank_l); + + REG_GET(VBLANK_PARAMETERS_3, + REFCYC_PER_META_CHUNK_VBLANK_L, &dlg_attr->refcyc_per_meta_chunk_vblank_l); + + if (REG(NOM_PARAMETERS_0)) + REG_GET(NOM_PARAMETERS_0, + DST_Y_PER_PTE_ROW_NOM_L, &dlg_attr->dst_y_per_pte_row_nom_l); + + if (REG(NOM_PARAMETERS_1)) + REG_GET(NOM_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_NOM_L, &dlg_attr->refcyc_per_pte_group_nom_l); + + REG_GET(NOM_PARAMETERS_4, + DST_Y_PER_META_ROW_NOM_L, &dlg_attr->dst_y_per_meta_row_nom_l); + + REG_GET(NOM_PARAMETERS_5, + REFCYC_PER_META_CHUNK_NOM_L, &dlg_attr->refcyc_per_meta_chunk_nom_l); + + REG_GET_2(PER_LINE_DELIVERY_PRE, + REFCYC_PER_LINE_DELIVERY_PRE_L, &dlg_attr->refcyc_per_line_delivery_pre_l, + REFCYC_PER_LINE_DELIVERY_PRE_C, &dlg_attr->refcyc_per_line_delivery_pre_c); + + REG_GET_2(PER_LINE_DELIVERY, + REFCYC_PER_LINE_DELIVERY_L, &dlg_attr->refcyc_per_line_delivery_l, + REFCYC_PER_LINE_DELIVERY_C, &dlg_attr->refcyc_per_line_delivery_c); + + if (REG(PREFETCH_SETTINS_C)) + REG_GET(PREFETCH_SETTINS_C, + VRATIO_PREFETCH_C, &dlg_attr->vratio_prefetch_c); + else + REG_GET(PREFETCH_SETTINGS_C, + VRATIO_PREFETCH_C, &dlg_attr->vratio_prefetch_c); + + REG_GET(VBLANK_PARAMETERS_2, + REFCYC_PER_PTE_GROUP_VBLANK_C, &dlg_attr->refcyc_per_pte_group_vblank_c); + + REG_GET(VBLANK_PARAMETERS_4, + REFCYC_PER_META_CHUNK_VBLANK_C, &dlg_attr->refcyc_per_meta_chunk_vblank_c); + + if (REG(NOM_PARAMETERS_2)) + REG_GET(NOM_PARAMETERS_2, + DST_Y_PER_PTE_ROW_NOM_C, &dlg_attr->dst_y_per_pte_row_nom_c); + + if (REG(NOM_PARAMETERS_3)) + REG_GET(NOM_PARAMETERS_3, + REFCYC_PER_PTE_GROUP_NOM_C, &dlg_attr->refcyc_per_pte_group_nom_c); + + REG_GET(NOM_PARAMETERS_6, + DST_Y_PER_META_ROW_NOM_C, &dlg_attr->dst_y_per_meta_row_nom_c); + + REG_GET(NOM_PARAMETERS_7, + REFCYC_PER_META_CHUNK_NOM_C, &dlg_attr->refcyc_per_meta_chunk_nom_c); + + /* TTU - per hubp */ + REG_GET_2(DCN_TTU_QOS_WM, + QoS_LEVEL_LOW_WM, &ttu_attr->qos_level_low_wm, + QoS_LEVEL_HIGH_WM, &ttu_attr->qos_level_high_wm); + + REG_GET_2(DCN_GLOBAL_TTU_CNTL, + MIN_TTU_VBLANK, &ttu_attr->min_ttu_vblank, + QoS_LEVEL_FLIP, &ttu_attr->qos_level_flip); + + /* TTU - per luma/chroma */ + /* Assumed surf0 is luma and 1 is chroma */ + + REG_GET_3(DCN_SURF0_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr->refcyc_per_req_delivery_l, + QoS_LEVEL_FIXED, &ttu_attr->qos_level_fixed_l, + QoS_RAMP_DISABLE, &ttu_attr->qos_ramp_disable_l); + + REG_GET(DCN_SURF0_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, + &ttu_attr->refcyc_per_req_delivery_pre_l); + + REG_GET_3(DCN_SURF1_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr->refcyc_per_req_delivery_c, + QoS_LEVEL_FIXED, &ttu_attr->qos_level_fixed_c, + QoS_RAMP_DISABLE, &ttu_attr->qos_ramp_disable_c); + + REG_GET(DCN_SURF1_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, + &ttu_attr->refcyc_per_req_delivery_pre_c); + + /* Rest of hubp */ + REG_GET(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, &s->pixel_format); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE_HIGH, + SURFACE_EARLIEST_INUSE_ADDRESS_HIGH, &s->inuse_addr_hi); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE, + SURFACE_EARLIEST_INUSE_ADDRESS, &s->inuse_addr_lo); + + REG_GET_2(DCSURF_PRI_VIEWPORT_DIMENSION, + PRI_VIEWPORT_WIDTH, &s->viewport_width, + PRI_VIEWPORT_HEIGHT, &s->viewport_height); + + REG_GET_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, &s->rotation_angle, + H_MIRROR_EN, &s->h_mirror_en); + + REG_GET(DCSURF_TILING_CONFIG, + SW_MODE, &s->sw_mode); + + REG_GET(DCSURF_SURFACE_CONTROL, + PRIMARY_SURFACE_DCC_EN, &s->dcc_en); + + REG_GET_3(DCHUBP_CNTL, + HUBP_BLANK_EN, &s->blank_en, + HUBP_TTU_DISABLE, &s->ttu_disable, + HUBP_UNDERFLOW_STATUS, &s->underflow_status); + + REG_GET(DCN_GLOBAL_TTU_CNTL, + MIN_TTU_VBLANK, &s->min_ttu_vblank); + + REG_GET_2(DCN_TTU_QOS_WM, + QoS_LEVEL_LOW_WM, &s->qos_level_low_wm, + QoS_LEVEL_HIGH_WM, &s->qos_level_high_wm); + +} + +void hubp2_read_state(struct hubp *hubp) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + struct dcn_hubp_state *s = &hubp2->state; + struct _vcs_dpi_display_rq_regs_st *rq_regs = &s->rq_regs; + + hubp2_read_state_common(hubp); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, + CHUNK_SIZE, &rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, &rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, &rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, &rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, &rq_regs->rq_regs_l.dpte_group_size, + MPTE_GROUP_SIZE, &rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, &rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, &rq_regs->rq_regs_l.pte_row_height_linear); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG_C, + CHUNK_SIZE_C, &rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.dpte_group_size, + MPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.mpte_group_size, + SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear); + +} + static struct hubp_funcs dcn20_hubp_funcs = { .hubp_enable_tripleBuffer = hubp2_enable_triplebuffer, .hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled, .hubp_program_surface_flip_and_addr = hubp2_program_surface_flip_and_addr, .hubp_program_surface_config = hubp2_program_surface_config, - .hubp_is_flip_pending = hubp1_is_flip_pending, + .hubp_is_flip_pending = hubp2_is_flip_pending, .hubp_setup = hubp2_setup, .hubp_setup_interdependent = hubp2_setup_interdependent, .hubp_set_vm_system_aperture_settings = hubp2_set_vm_system_aperture_settings, - .set_blank = hubp1_set_blank, - .dcc_control = hubp1_dcc_control, - .hubp_update_dchub = hubp2_update_dchub, + .set_blank = hubp2_set_blank, + .dcc_control = hubp2_dcc_control, .mem_program_viewport = min_set_viewport, .set_cursor_attributes = hubp2_cursor_set_attributes, - .set_cursor_position = hubp1_cursor_set_position, - .hubp_clk_cntl = hubp1_clk_cntl, - .hubp_vtg_sel = hubp1_vtg_sel, + .set_cursor_position = hubp2_cursor_set_position, + .hubp_clk_cntl = hubp2_clk_cntl, + .hubp_vtg_sel = hubp2_vtg_sel, .dmdata_set_attributes = hubp2_dmdata_set_attributes, .dmdata_load = hubp2_dmdata_load, .dmdata_status_done = hubp2_dmdata_status_done, - .hubp_read_state = hubp1_read_state, - .hubp_clear_underflow = hubp1_clear_underflow, + .hubp_read_state = hubp2_read_state, + .hubp_clear_underflow = hubp2_clear_underflow, .hubp_set_flip_control_surface_gsl = hubp2_set_flip_control_surface_gsl, .hubp_init = hubp1_init, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h index d5acc348be22..d5c8615af45e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h @@ -38,12 +38,6 @@ SRI(PREFETCH_SETTINGS_C, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_LOW_ADDR, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, HUBPREQ, id),\ - SR(DCN_VM_FB_LOCATION_TOP),\ - SR(DCN_VM_FB_LOCATION_BASE),\ - SR(DCN_VM_FB_OFFSET),\ - SR(DCN_VM_AGP_BASE),\ - SR(DCN_VM_AGP_BOT),\ - SR(DCN_VM_AGP_TOP),\ SRI(CURSOR_SETTINGS, HUBPREQ, id), \ SRI(CURSOR_SURFACE_ADDRESS_HIGH, CURSOR0_, id), \ SRI(CURSOR_SURFACE_ADDRESS, CURSOR0_, id), \ @@ -72,8 +66,8 @@ SR(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB),\ SR(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB) -#define HUBP_MASK_SH_LIST_DCN2_COMMON(mask_sh)\ - HUBP_MASK_SH_LIST_DCN(mask_sh),\ +#define HUBP_MASK_SH_LIST_DCN2_SHARE_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh),\ HUBP_MASK_SH_LIST_DCN_VM(mask_sh),\ HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, ROTATION_ANGLE, mask_sh),\ HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, H_MIRROR_EN, mask_sh),\ @@ -82,12 +76,6 @@ HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS_C, VRATIO_PREFETCH_C, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_LOW_ADDR, MC_VM_SYSTEM_APERTURE_LOW_ADDR, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, MC_VM_SYSTEM_APERTURE_HIGH_ADDR, mask_sh),\ - HUBP_SF(DCN_VM_FB_LOCATION_TOP, FB_TOP, mask_sh),\ - HUBP_SF(DCN_VM_FB_LOCATION_BASE, FB_BASE, mask_sh),\ - HUBP_SF(DCN_VM_FB_OFFSET, FB_OFFSET, mask_sh),\ - HUBP_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh),\ - HUBP_SF(DCN_VM_AGP_BOT, AGP_BOT, mask_sh),\ - HUBP_SF(DCN_VM_AGP_TOP, AGP_TOP, mask_sh),\ HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_DST_Y_OFFSET, mask_sh), \ HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_CHUNK_HDL_ADJUST, mask_sh), \ HUBP_SF(CURSOR0_0_CURSOR_SURFACE_ADDRESS_HIGH, CURSOR_SURFACE_ADDRESS_HIGH, mask_sh), \ @@ -127,13 +115,21 @@ HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL2, SURFACE_TRIPLE_BUFFER_ENABLE, mask_sh),\ HUBP_SF(HUBPREQ0_VMID_SETTINGS_0, VMID, mask_sh) +/*DCN2.x and DCN1.x*/ +#define HUBP_MASK_SH_LIST_DCN2_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN2_SHARE_COMMON(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, MPTE_GROUP_SIZE, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, MPTE_GROUP_SIZE_C, mask_sh) + +/*DCN2.0 specific*/ #define HUBP_MASK_SH_LIST_DCN20(mask_sh)\ HUBP_MASK_SH_LIST_DCN2_COMMON(mask_sh),\ HUBP_SF(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, DCN_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM, mask_sh),\ HUBP_SF(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, mask_sh),\ HUBP_SF(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, mask_sh) - +/*DCN2.x */ #define DCN2_HUBP_REG_COMMON_VARIABLE_LIST \ HUBP_COMMON_REG_VARIABLE_LIST; \ uint32_t DMDATA_ADDRESS_HIGH; \ @@ -149,14 +145,22 @@ uint32_t FLIP_PARAMETERS_2;\ uint32_t DCN_CUR1_TTU_CNTL0;\ uint32_t DCN_CUR1_TTU_CNTL1;\ - uint32_t VMID_SETTINGS_0;\ + uint32_t VMID_SETTINGS_0 + + +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#define DCN21_HUBP_REG_COMMON_VARIABLE_LIST \ + DCN2_HUBP_REG_COMMON_VARIABLE_LIST; \ uint32_t FLIP_PARAMETERS_3;\ uint32_t FLIP_PARAMETERS_4;\ + uint32_t FLIP_PARAMETERS_5;\ + uint32_t FLIP_PARAMETERS_6;\ uint32_t VBLANK_PARAMETERS_5;\ uint32_t VBLANK_PARAMETERS_6 +#endif #define DCN2_HUBP_REG_FIELD_VARIABLE_LIST(type) \ - DCN_HUBP_REG_FIELD_LIST(type); \ + DCN_HUBP_REG_FIELD_BASE_LIST(type); \ type DMDATA_ADDRESS_HIGH;\ type DMDATA_MODE;\ type DMDATA_UPDATED;\ @@ -180,17 +184,41 @@ type SURFACE_TRIPLE_BUFFER_ENABLE;\ type VMID +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 +#define DCN21_HUBP_REG_FIELD_VARIABLE_LIST(type) \ + DCN2_HUBP_REG_FIELD_VARIABLE_LIST(type);\ + type REFCYC_PER_VM_GROUP_FLIP;\ + type REFCYC_PER_VM_REQ_FLIP;\ + type REFCYC_PER_VM_GROUP_VBLANK;\ + type REFCYC_PER_VM_REQ_VBLANK;\ + type REFCYC_PER_PTE_GROUP_FLIP_C; \ + type REFCYC_PER_META_CHUNK_FLIP_C; \ + type VM_GROUP_SIZE +#endif + struct dcn_hubp2_registers { +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + DCN21_HUBP_REG_COMMON_VARIABLE_LIST; +#else DCN2_HUBP_REG_COMMON_VARIABLE_LIST; +#endif }; struct dcn_hubp2_shift { +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + DCN21_HUBP_REG_FIELD_VARIABLE_LIST(uint8_t); +#else DCN2_HUBP_REG_FIELD_VARIABLE_LIST(uint8_t); +#endif }; struct dcn_hubp2_mask { +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + DCN21_HUBP_REG_FIELD_VARIABLE_LIST(uint32_t); +#else DCN2_HUBP_REG_FIELD_VARIABLE_LIST(uint32_t); +#endif }; struct dcn20_hubp { @@ -217,10 +245,6 @@ void hubp2_setup_interdependent( void hubp2_vready_at_or_After_vsync(struct hubp *hubp, struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest); -void hubp2_update_dchub( - struct hubp *hubp, - struct dchub_init_data *dh_data); - void hubp2_cursor_set_attributes( struct hubp *hubp, const struct dc_cursor_attributes *attr); @@ -262,16 +286,53 @@ bool hubp2_program_surface_flip_and_addr( const struct dc_plane_address *address, bool flip_immediate); +void hubp2_dcc_control(struct hubp *hubp, bool enable, + enum hubp_ind_block_size independent_64b_blks); + +void hubp2_program_size( + struct hubp *hubp, + enum surface_pixel_format format, + const struct plane_size *plane_size, + struct dc_plane_dcc_param *dcc); + +void hubp2_program_rotation( + struct hubp *hubp, + enum dc_rotation_angle rotation, + bool horizontal_mirror); + +void hubp2_program_pixel_format( + struct hubp *hubp, + enum surface_pixel_format format); + void hubp2_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, unsigned int compat_level); +bool hubp2_is_flip_pending(struct hubp *hubp); + +void hubp2_set_blank(struct hubp *hubp, bool blank); + +void hubp2_cursor_set_position( + struct hubp *hubp, + const struct dc_cursor_position *pos, + const struct dc_cursor_mi_param *param); + +void hubp2_clk_cntl(struct hubp *hubp, bool enable); + +void hubp2_vtg_sel(struct hubp *hubp, uint32_t otg_inst); + +void hubp2_clear_underflow(struct hubp *hubp); + +void hubp2_read_state_common(struct hubp *hubp); + +void hubp2_read_state(struct hubp *hubp); + #endif /* __DC_MEM_INPUT_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index d810c8940129..1212da12c414 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -64,23 +64,7 @@ #define FN(reg_name, field_name) \ hws->shifts->field_name, hws->masks->field_name -static void bios_golden_init(struct dc *dc) -{ - struct dc_bios *bp = dc->ctx->dc_bios; - int i; - - /* initialize dcn global */ - bp->funcs->enable_disp_power_gating(bp, - CONTROLLER_ID_D0, ASIC_PIPE_INIT); - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - /* initialize dcn per pipe */ - bp->funcs->enable_disp_power_gating(bp, - CONTROLLER_ID_D0 + i, ASIC_PIPE_DISABLE); - } -} - -static void enable_power_gating_plane( +static void dcn20_enable_power_gating_plane( struct dce_hwseq *hws, bool enable) { @@ -94,28 +78,34 @@ static void enable_power_gating_plane( REG_UPDATE(DOMAIN2_PG_CONFIG, DOMAIN2_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN4_PG_CONFIG, DOMAIN4_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN6_PG_CONFIG, DOMAIN6_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN8_PG_CONFIG, DOMAIN8_POWER_FORCEON, force_on); - /*Do not power gate DCHUB5, should be left at HW default, power on permanently*/ - /*REG_UPDATE(DOMAIN10_PG_CONFIG, DOMAIN10_POWER_FORCEON, force_on);*/ + if (REG(DOMAIN8_PG_CONFIG)) + REG_UPDATE(DOMAIN8_PG_CONFIG, DOMAIN8_POWER_FORCEON, force_on); + if (REG(DOMAIN10_PG_CONFIG)) + REG_UPDATE(DOMAIN10_PG_CONFIG, DOMAIN8_POWER_FORCEON, force_on); /* DPP0/1/2/3/4/5 */ REG_UPDATE(DOMAIN1_PG_CONFIG, DOMAIN1_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN3_PG_CONFIG, DOMAIN3_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN5_PG_CONFIG, DOMAIN5_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN7_PG_CONFIG, DOMAIN7_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN9_PG_CONFIG, DOMAIN9_POWER_FORCEON, force_on); - /*Do not power gate DPP5, should be left at HW default, power on permanently*/ - /*REG_UPDATE(DOMAIN11_PG_CONFIG, DOMAIN11_POWER_FORCEON, force_on);*/ + if (REG(DOMAIN9_PG_CONFIG)) + REG_UPDATE(DOMAIN9_PG_CONFIG, DOMAIN9_POWER_FORCEON, force_on); + if (REG(DOMAIN11_PG_CONFIG)) + REG_UPDATE(DOMAIN11_PG_CONFIG, DOMAIN9_POWER_FORCEON, force_on); + /* DCS0/1/2/3/4/5 */ REG_UPDATE(DOMAIN16_PG_CONFIG, DOMAIN16_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN17_PG_CONFIG, DOMAIN17_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN18_PG_CONFIG, DOMAIN18_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN19_PG_CONFIG, DOMAIN19_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN20_PG_CONFIG, DOMAIN20_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN21_PG_CONFIG, DOMAIN21_POWER_FORCEON, force_on); + if (REG(DOMAIN19_PG_CONFIG)) + REG_UPDATE(DOMAIN19_PG_CONFIG, DOMAIN19_POWER_FORCEON, force_on); + if (REG(DOMAIN20_PG_CONFIG)) + REG_UPDATE(DOMAIN20_PG_CONFIG, DOMAIN20_POWER_FORCEON, force_on); + if (REG(DOMAIN21_PG_CONFIG)) + REG_UPDATE(DOMAIN21_PG_CONFIG, DOMAIN21_POWER_FORCEON, force_on); } -static void dcn20_dccg_init(struct dce_hwseq *hws) +void dcn20_dccg_init(struct dce_hwseq *hws) { /* * set MICROSECOND_TIME_BASE_DIV @@ -138,8 +128,46 @@ static void dcn20_dccg_init(struct dce_hwseq *hws) /* This value is dependent on the hardware pipeline delay so set once per SOC */ REG_WRITE(DISPCLK_FREQ_CHANGE_CNTL, 0x801003c); } +void dcn20_display_init(struct dc *dc) +{ + struct dce_hwseq *hws = dc->hwseq; + + /* RBBMIF + * disable RBBMIF timeout detection for all clients + * Ensure RBBMIF does not drop register accesses due to the per-client timeout + */ + REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF); + REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF); -static void disable_vga( + /* DCCG */ + dcn20_dccg_init(hws); + + REG_UPDATE(DC_MEM_GLOBAL_PWR_REQ_CNTL, DC_MEM_GLOBAL_PWR_REQ_DIS, 0); + + /* DCHUB/MMHUBBUB + * set global timer refclk divider + * 100Mhz refclk -> 2 + * 27Mhz refclk -> 1 + * 48Mhz refclk -> 1 + */ + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); + REG_WRITE(REFCLK_CNTL, 0); + + /* OPTC + * OTG_CONTROL.OTG_DISABLE_POINT_CNTL = 0x3; will be set during optc2_enable_crtc + */ + + /* AZ + * default value is 0x64 for 100Mhz ref clock, if the ref clock is 100Mhz, no need to program this regiser, + * if not, it should be programmed according to the ref clock + */ + REG_UPDATE(AZALIA_AUDIO_DTO, AZALIA_AUDIO_DTO_MODULE, 0x64); + /* Enable controller clock gating */ + REG_WRITE(AZALIA_CONTROLLER_CLOCK_GATING, 0x1); +} + +void dcn20_disable_vga( struct dce_hwseq *hws) { REG_WRITE(D1VGA_CONTROL, 0); @@ -163,7 +191,7 @@ void dcn20_program_tripleBuffer( } /* Blank pixel data during initialization */ -static void dcn20_init_blank( +void dcn20_init_blank( struct dc *dc, struct timing_generator *tg) { @@ -442,29 +470,6 @@ static void dcn20_hubp_pg_control( } - -static void dcn20_plane_atomic_power_down(struct dc *dc, struct pipe_ctx *pipe_ctx) -{ - struct dce_hwseq *hws = dc->hwseq; - struct dpp *dpp = pipe_ctx->plane_res.dpp; - - DC_LOGGER_INIT(dc->ctx->logger); - - if (REG(DC_IP_REQUEST_CNTL)) { - REG_SET(DC_IP_REQUEST_CNTL, 0, - IP_REQUEST_EN, 1); - dcn20_dpp_pg_control(hws, dpp->inst, false); - dcn20_hubp_pg_control(hws, pipe_ctx->plane_res.hubp->inst, false); - dpp->funcs->dpp_reset(dpp); - REG_SET(DC_IP_REQUEST_CNTL, 0, - IP_REQUEST_EN, 0); - DC_LOG_DEBUG( - "Power gated front end %d\n", pipe_ctx->pipe_idx); - } -} - - - /* disable HW used by plane. * note: cannot disable until disconnect is complete */ @@ -490,7 +495,9 @@ static void dcn20_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) hubp->power_gated = true; dc->optimized_required = false; /* We're powering off, no need to optimize */ - dcn20_plane_atomic_power_down(dc, pipe_ctx); + dc->hwss.plane_atomic_power_down(dc, + pipe_ctx->plane_res.dpp, + pipe_ctx->plane_res.hubp); pipe_ctx->stream = NULL; memset(&pipe_ctx->stream_res, 0, sizeof(pipe_ctx->stream_res)); @@ -514,199 +521,6 @@ void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) pipe_ctx->pipe_idx); } -static void dcn20_init_hw(struct dc *dc) -{ - int i, j; - struct abm *abm = dc->res_pool->abm; - struct dmcu *dmcu = dc->res_pool->dmcu; - struct dce_hwseq *hws = dc->hwseq; - struct dc_bios *dcb = dc->ctx->dc_bios; - struct resource_pool *res_pool = dc->res_pool; - struct dc_state *context = dc->current_state; - struct dc_firmware_info fw_info = { { 0 } }; - - if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) - dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); - - // Initialize the dccg - if (res_pool->dccg->funcs->dccg_init) - res_pool->dccg->funcs->dccg_init(res_pool->dccg); - - //Enable ability to power gate / don't force power on permanently - enable_power_gating_plane(dc->hwseq, true); - - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF); - REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF); - - dcn20_dccg_init(hws); - - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); - REG_WRITE(REFCLK_CNTL, 0); - } else { - if (!dcb->funcs->is_accelerated_mode(dcb)) { - bios_golden_init(dc); - if (dc->ctx->dc_bios->funcs->get_firmware_info( - dc->ctx->dc_bios, &fw_info) == BP_RESULT_OK) { - res_pool->ref_clocks.xtalin_clock_inKhz = fw_info.pll_info.crystal_frequency; - - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (res_pool->dccg && res_pool->hubbub) { - - (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, - fw_info.pll_info.crystal_frequency, - &res_pool->ref_clocks.dccg_ref_clock_inKhz); - - (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, - res_pool->ref_clocks.dccg_ref_clock_inKhz, - &res_pool->ref_clocks.dchub_ref_clock_inKhz); - } else { - // Not all ASICs have DCCG sw component - res_pool->ref_clocks.dccg_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - res_pool->ref_clocks.dchub_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - } - } - } else - ASSERT_CRITICAL(false); - disable_vga(dc->hwseq); - } - - for (i = 0; i < dc->link_count; i++) { - /* Power up AND update implementation according to the - * required signal (which may be different from the - * default signal on connector). - */ - struct dc_link *link = dc->links[i]; - - link->link_enc->funcs->hw_init(link->link_enc); - } - } - - /* Blank pixel data with OPP DPG */ - for (i = 0; i < dc->res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - if (tg->funcs->is_tg_enabled(tg)) { - dcn20_init_blank(dc, tg); - } - } - - for (i = 0; i < res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - if (tg->funcs->is_tg_enabled(tg)) - tg->funcs->lock(tg); - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct dpp *dpp = res_pool->dpps[i]; - - dpp->funcs->dpp_reset(dpp); - } - - /* Reset all MPCC muxes */ - res_pool->mpc->funcs->mpc_init(res_pool->mpc); - - /* initialize OPP mpc_tree parameter */ - for (i = 0; i < dc->res_pool->res_cap->num_opp; i++) { - res_pool->opps[i]->mpc_tree_params.opp_id = res_pool->opps[i]->inst; - res_pool->opps[i]->mpc_tree_params.opp_list = NULL; - for (j = 0; j < MAX_PIPES; j++) - res_pool->opps[i]->mpcc_disconnect_pending[j] = false; - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - struct hubp *hubp = dc->res_pool->hubps[i]; - struct dpp *dpp = dc->res_pool->dpps[i]; - - pipe_ctx->stream_res.tg = tg; - pipe_ctx->pipe_idx = i; - - pipe_ctx->plane_res.hubp = hubp; - pipe_ctx->plane_res.dpp = dpp; - pipe_ctx->plane_res.mpcc_inst = dpp->inst; - hubp->mpcc_id = dpp->inst; - hubp->opp_id = OPP_ID_INVALID; - hubp->power_gated = false; - pipe_ctx->stream_res.opp = NULL; - - hubp->funcs->hubp_init(hubp); - - //dc->res_pool->opps[i]->mpc_tree_params.opp_id = dc->res_pool->opps[i]->inst; - //dc->res_pool->opps[i]->mpc_tree_params.opp_list = NULL; - dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; - pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; - /*to do*/ - hwss1_plane_atomic_disconnect(dc, pipe_ctx); - } - - /* initialize DWB pointer to MCIF_WB */ - for (i = 0; i < res_pool->res_cap->num_dwb; i++) - res_pool->dwbc[i]->mcif = res_pool->mcif_wb[i]; - - for (i = 0; i < dc->res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - if (tg->funcs->is_tg_enabled(tg)) - tg->funcs->unlock(tg); - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - - dc->hwss.disable_plane(dc, pipe_ctx); - - pipe_ctx->stream_res.tg = NULL; - pipe_ctx->plane_res.hubp = NULL; - } - - for (i = 0; i < dc->res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - tg->funcs->tg_init(tg); - } - - /* end of FPGA. Below if real ASIC */ - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - return; - - - for (i = 0; i < res_pool->audio_count; i++) { - struct audio *audio = res_pool->audios[i]; - - audio->funcs->hw_init(audio); - } - - if (abm != NULL) { - abm->funcs->init_backlight(abm); - abm->funcs->abm_init(abm); - } - - if (dmcu != NULL) - dmcu->funcs->dmcu_init(dmcu); - - if (abm != NULL && dmcu != NULL) - abm->dmcu_is_running = dmcu->funcs->is_dmcu_initialized(dmcu); - - /* power AFMT HDMI memory TODO: may move to dis/en output save power*/ - REG_WRITE(DIO_MEM_PWR_CTRL, 0); - - if (!dc->debug.disable_clock_gate) { - /* enable all DCN clock gating */ - REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0); - - REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0); - - REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); - } - -} - enum dc_status dcn20_enable_stream_timing( struct pipe_ctx *pipe_ctx, struct dc_state *context, @@ -715,11 +529,9 @@ enum dc_status dcn20_enable_stream_timing( struct dc_stream_state *stream = pipe_ctx->stream; struct drr_params params = {0}; unsigned int event_triggers = 0; - - -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); -#endif + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + int opp_inst[MAX_PIPES] = { pipe_ctx->stream_res.opp->inst }; /* by upper caller loop, pipe0 is parent pipe and be called first. * back end is set up by for pipe0. Other children pipe share back end @@ -730,12 +542,17 @@ enum dc_status dcn20_enable_stream_timing( /* TODO check if timing_changed, disable stream if timing changed */ - if (odm_pipe) + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + opp_inst[opp_cnt] = odm_pipe->stream_res.opp->inst; + opp_cnt++; + } + + if (opp_cnt > 1) pipe_ctx->stream_res.tg->funcs->set_odm_combine( pipe_ctx->stream_res.tg, - odm_pipe->stream_res.opp->inst, - pipe_ctx->stream->timing.h_addressable/2, - pipe_ctx->stream->timing.pixel_encoding); + opp_inst, opp_cnt, + &pipe_ctx->stream->timing); + /* HW program guide assume display already disable * by unplug sequence. OTG assume stop. */ @@ -759,11 +576,7 @@ enum dc_status dcn20_enable_stream_timing( pipe_ctx->stream->signal, true); - if (pipe_ctx->stream_res.tg->funcs->setup_global_lock) - pipe_ctx->stream_res.tg->funcs->setup_global_lock( - pipe_ctx->stream_res.tg); - - if (odm_pipe) + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) odm_pipe->stream_res.opp->funcs->opp_pipe_clock_control( odm_pipe->stream_res.opp, true); @@ -784,6 +597,8 @@ enum dc_status dcn20_enable_stream_timing( params.vertical_total_min = stream->adjust.v_total_min; params.vertical_total_max = stream->adjust.v_total_max; + params.vertical_total_mid = stream->adjust.v_total_mid; + params.vertical_total_mid_frame_num = stream->adjust.v_total_mid_frame_num; if (pipe_ctx->stream_res.tg->funcs->set_drr) pipe_ctx->stream_res.tg->funcs->set_drr( pipe_ctx->stream_res.tg, ¶ms); @@ -814,6 +629,10 @@ void dcn20_program_output_csc(struct dc *dc, { struct mpc *mpc = dc->res_pool->mpc; enum mpc_output_csc_mode ocsc_mode = MPC_OUTPUT_CSC_COEF_A; + int mpcc_id = pipe_ctx->plane_res.hubp->inst; + + if (mpc->funcs->power_on_mpc_mem_pwr) + mpc->funcs->power_on_mpc_mem_pwr(mpc, mpcc_id, true); if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) { if (mpc->funcs->set_output_csc != NULL) @@ -843,7 +662,9 @@ bool dcn20_set_output_transfer_func(struct pipe_ctx *pipe_ctx, * if programming for all pipes is required then remove condition * pipe_ctx->top_pipe == NULL ,but then fix the diagnostic. */ - if ((pipe_ctx->top_pipe == NULL || dc_res_is_odm_head_pipe(pipe_ctx)) + if (mpc->funcs->power_on_mpc_mem_pwr) + mpc->funcs->power_on_mpc_mem_pwr(mpc, mpcc_id, true); + if (pipe_ctx->top_pipe == NULL && mpc->funcs->set_output_gamma && stream->out_transfer_func) { if (stream->out_transfer_func->type == TF_TYPE_HWPWL) params = &stream->out_transfer_func->pwl; @@ -909,14 +730,14 @@ static bool dcn20_set_shaper_3dlut( result = dpp_base->funcs->dpp_program_shaper_lut(dpp_base, shaper_lut); if (plane_state->lut3d_func && - plane_state->lut3d_func->initialized == true) + plane_state->lut3d_func->state.bits.initialized == 1) result = dpp_base->funcs->dpp_program_3dlut(dpp_base, &plane_state->lut3d_func->lut_3d); else result = dpp_base->funcs->dpp_program_3dlut(dpp_base, NULL); if (plane_state->lut3d_func && - plane_state->lut3d_func->initialized == true && + plane_state->lut3d_func->state.bits.initialized == 1 && plane_state->lut3d_func->hdr_multiplier != 0) dpp_base->funcs->dpp_set_hdr_multiplier(dpp_base, plane_state->lut3d_func->hdr_multiplier); @@ -1005,14 +826,20 @@ bool dcn20_set_input_transfer_func(struct pipe_ctx *pipe_ctx, static void dcn20_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx) { - struct pipe_ctx *combine_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + int opp_inst[MAX_PIPES] = { pipe_ctx->stream_res.opp->inst }; - if (combine_pipe) + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + opp_inst[opp_cnt] = odm_pipe->stream_res.opp->inst; + opp_cnt++; + } + + if (opp_cnt > 1) pipe_ctx->stream_res.tg->funcs->set_odm_combine( pipe_ctx->stream_res.tg, - combine_pipe->stream_res.opp->inst, - pipe_ctx->plane_res.scl_data.h_active, - pipe_ctx->stream->timing.pixel_encoding); + opp_inst, opp_cnt, + &pipe_ctx->stream->timing); else pipe_ctx->stream_res.tg->funcs->set_odm_bypass( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); @@ -1028,7 +855,8 @@ void dcn20_blank_pixel_data( struct dc_stream_state *stream = pipe_ctx->stream; enum dc_color_space color_space = stream->output_color_space; enum controller_dp_test_pattern test_pattern = CONTROLLER_DP_TEST_PATTERN_SOLID_COLOR; - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int odm_cnt = 1; int width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; int height = stream->timing.v_addressable + stream->timing.v_border_bottom + stream->timing.v_border_top; @@ -1036,8 +864,10 @@ void dcn20_blank_pixel_data( /* get opp dpg blank color */ color_space_to_black_color(dc, color_space, &black_color); - if (bot_odm_pipe) - width = width / 2; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + odm_cnt++; + + width = width / odm_cnt; if (blank) { if (stream_res->abm) @@ -1057,10 +887,10 @@ void dcn20_blank_pixel_data( width, height); - if (bot_odm_pipe) { - bot_odm_pipe->stream_res.opp->funcs->opp_set_disp_pattern_generator( - bot_odm_pipe->stream_res.opp, - dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE ? + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + odm_pipe->stream_res.opp->funcs->opp_set_disp_pattern_generator( + odm_pipe->stream_res.opp, + dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE && blank ? CONTROLLER_DP_TEST_PATTERN_COLORRAMP : test_pattern, stream->timing.display_color_depth, &black_color, @@ -1106,6 +936,9 @@ void dcn20_enable_plane( /* enable DCFCLK current DCHUB */ pipe_ctx->plane_res.hubp->funcs->hubp_clk_cntl(pipe_ctx->plane_res.hubp, true); + /* initialize HUBP on power up */ + pipe_ctx->plane_res.hubp->funcs->hubp_init(pipe_ctx->plane_res.hubp); + /* make sure OPP_PIPE_CLOCK_EN = 1 */ pipe_ctx->stream_res.opp->funcs->opp_pipe_clock_control( pipe_ctx->stream_res.opp, @@ -1201,7 +1034,7 @@ static void dcn20_program_all_pipe_in_tree( struct pipe_ctx *pipe_ctx, struct dc_state *context) { - if (pipe_ctx->top_pipe == NULL) { + if (pipe_ctx->top_pipe == NULL && !pipe_ctx->prev_odm_pipe) { bool blank = !is_pipe_tree_visible(pipe_ctx); pipe_ctx->stream_res.tg->funcs->program_global_sync( @@ -1223,8 +1056,13 @@ static void dcn20_program_all_pipe_in_tree( if (pipe_ctx->plane_state != NULL) dcn20_program_pipe(dc, pipe_ctx, context); - if (pipe_ctx->bottom_pipe != NULL && pipe_ctx->bottom_pipe != pipe_ctx) + if (pipe_ctx->bottom_pipe != NULL) { + ASSERT(pipe_ctx->bottom_pipe != pipe_ctx); dcn20_program_all_pipe_in_tree(dc, pipe_ctx->bottom_pipe, context); + } else if (pipe_ctx->next_odm_pipe != NULL) { + ASSERT(pipe_ctx->next_odm_pipe != pipe_ctx); + dcn20_program_all_pipe_in_tree(dc, pipe_ctx->next_odm_pipe, context); + } } void dcn20_pipe_control_lock_global( @@ -1265,17 +1103,6 @@ void dcn20_pipe_control_lock( if (pipe->plane_state != NULL) flip_immediate = pipe->plane_state->flip_immediate; - if (flip_immediate && lock) { - while (pipe->plane_res.hubp->funcs->hubp_is_flip_pending(pipe->plane_res.hubp)) { - udelay(1); - } - - if (pipe->bottom_pipe != NULL) - while (pipe->bottom_pipe->plane_res.hubp->funcs->hubp_is_flip_pending(pipe->bottom_pipe->plane_res.hubp)) { - udelay(1); - } - } - /* In flip immediate and pipe splitting case, we need to use GSL * for synchronization. Only do setup on locking and on flip type change. */ @@ -1303,18 +1130,32 @@ static void dcn20_apply_ctx_for_surface( int num_planes, struct dc_state *context) { - + const unsigned int TIMEOUT_FOR_PIPE_ENABLE_MS = 100; int i; struct timing_generator *tg; bool removed_pipe[6] = { false }; bool interdependent_update = false; struct pipe_ctx *top_pipe_to_program = find_top_pipe_for_stream(dc, context, stream); + struct pipe_ctx *prev_top_pipe_to_program = + find_top_pipe_for_stream(dc, dc->current_state, stream); DC_LOGGER_INIT(dc->ctx->logger); if (!top_pipe_to_program) return; + /* Carry over GSL groups in case the context is changing. */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + struct pipe_ctx *old_pipe_ctx = + &dc->current_state->res_ctx.pipe_ctx[i]; + + if (pipe_ctx->stream == stream && + pipe_ctx->stream == old_pipe_ctx->stream) + pipe_ctx->stream_res.gsl_group = + old_pipe_ctx->stream_res.gsl_group; + } + tg = top_pipe_to_program->stream_res.tg; interdependent_update = top_pipe_to_program->plane_state && @@ -1345,7 +1186,7 @@ static void dcn20_apply_ctx_for_surface( if (old_pipe_ctx->stream_res.tg == tg && old_pipe_ctx->plane_res.hubp && old_pipe_ctx->plane_res.hubp->opp_id != OPP_ID_INVALID) - dcn20_disable_plane(dc, old_pipe_ctx); + dc->hwss.disable_plane(dc, old_pipe_ctx); } if ((!pipe_ctx->plane_state || @@ -1391,6 +1232,22 @@ static void dcn20_apply_ctx_for_surface( for (i = 0; i < dc->res_pool->pipe_count; i++) if (removed_pipe[i]) dcn20_disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + + /* + * If we are enabling a pipe, we need to wait for pending clear as this is a critical + * part of the enable operation otherwise, DM may request an immediate flip which + * will cause HW to perform an "immediate enable" (as opposed to "vsync enable") which + * is unsupported on DCN. + */ + i = 0; + if (num_planes > 0 && top_pipe_to_program && + (prev_top_pipe_to_program == NULL || prev_top_pipe_to_program->plane_state == NULL)) { + while (i < TIMEOUT_FOR_PIPE_ENABLE_MS && + top_pipe_to_program->plane_res.hubp->funcs->hubp_is_flip_pending(top_pipe_to_program->plane_res.hubp)) { + i += 1; + msleep(1); + } + } } @@ -1400,16 +1257,16 @@ void dcn20_prepare_bandwidth( { struct hubbub *hubbub = dc->res_pool->hubbub; + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, + false); + /* program dchubbub watermarks */ hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, false); - - dc->clk_mgr->funcs->update_clocks( - dc->clk_mgr, - context, - false); } void dcn20_optimize_bandwidth( @@ -1462,8 +1319,8 @@ bool dcn20_update_bandwidth( pipe_ctx->stream_res.tg->funcs->set_vtg_params( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); - - dc->hwss.blank_pixel_data(dc, pipe_ctx, blank); + if (pipe_ctx->prev_odm_pipe == NULL) + dc->hwss.blank_pixel_data(dc, pipe_ctx, blank); } pipe_ctx->plane_res.hubp->funcs->hubp_setup( @@ -1553,12 +1410,15 @@ static void dcn20_disable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx { #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dce_hwseq *hws = dc->hwseq; - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); if (pipe_ctx->stream_res.dsc) { + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; + dcn20_dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, true); - if (bot_odm_pipe) - dcn20_dsc_pg_control(hws, bot_odm_pipe->stream_res.dsc->inst, true); + while (odm_pipe) { + dcn20_dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, true); + odm_pipe = odm_pipe->next_odm_pipe; + } } #endif } @@ -1567,12 +1427,15 @@ static void dcn20_enable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx) { #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dce_hwseq *hws = dc->hwseq; - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); if (pipe_ctx->stream_res.dsc) { + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; + dcn20_dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, false); - if (bot_odm_pipe) - dcn20_dsc_pg_control(hws, bot_odm_pipe->stream_res.dsc->inst, false); + while (odm_pipe) { + dcn20_dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, false); + odm_pipe = odm_pipe->next_odm_pipe; + } } #endif } @@ -1597,9 +1460,9 @@ void dcn20_set_dmdata_attributes(struct pipe_ctx *pipe_ctx) hubp->funcs->dmdata_set_attributes(hubp, &attr); } -void dcn20_disable_stream(struct pipe_ctx *pipe_ctx, int option) +void dcn20_disable_stream(struct pipe_ctx *pipe_ctx) { - dce110_disable_stream(pipe_ctx, option); + dce110_disable_stream(pipe_ctx); } static void dcn20_init_vm_ctx( @@ -1637,6 +1500,7 @@ static int dcn20_init_sys_ctx(struct dce_hwseq *hws, struct dc *dc, struct dc_ph config.gart_config.page_table_start_addr = pa_config->gart_config.page_table_start_addr; config.gart_config.page_table_end_addr = pa_config->gart_config.page_table_end_addr; config.gart_config.page_table_base_addr = pa_config->gart_config.page_table_base_addr; + config.page_table_default_page_addr = pa_config->page_table_default_page_addr; return dc->res_pool->hubbub->funcs->init_dchub_sys_ctx(dc->res_pool->hubbub, &config); } @@ -1702,18 +1566,22 @@ void dcn20_unblank_stream(struct pipe_ctx *pipe_ctx, struct encoder_unblank_param params = { { 0 } }; struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; - params.odm = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + params.opp_cnt = 1; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + params.opp_cnt++; + } /* only 3 items below are used by unblank */ params.timing = pipe_ctx->stream->timing; params.link_settings.link_rate = link_settings->link_rate; if (dc_is_dp_signal(pipe_ctx->stream->signal)) { - if (optc1_is_two_pixels_per_containter(&stream->timing) || params.odm) + if (optc1_is_two_pixels_per_containter(&stream->timing) || params.opp_cnt > 1) params.timing.pix_clk_100hz /= 2; pipe_ctx->stream_res.stream_enc->funcs->dp_set_odm_combine( - pipe_ctx->stream_res.stream_enc, params.odm); + pipe_ctx->stream_res.stream_enc, params.opp_cnt > 1); pipe_ctx->stream_res.stream_enc->funcs->dp_unblank(pipe_ctx->stream_res.stream_enc, ¶ms); } @@ -1749,14 +1617,29 @@ static void dcn20_reset_back_end_for_pipe( if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { /* DPMS may already disable */ if (!pipe_ctx->stream->dpms_off) - core_link_disable_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); - else if (pipe_ctx->stream_res.audio) { - dc->hwss.disable_audio_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); + core_link_disable_stream(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + /* free acquired resources */ + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; + } } } #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - else if (pipe_ctx->stream_res.dsc) + else if (pipe_ctx->stream_res.dsc) { dp_set_dsc_enable(pipe_ctx, false); + } #endif /* by upper caller loop, parent pipe: pipe0, will be reset last. @@ -1770,6 +1653,10 @@ static void dcn20_reset_back_end_for_pipe( if (pipe_ctx->stream_res.tg->funcs->set_odm_bypass) pipe_ctx->stream_res.tg->funcs->set_odm_bypass( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); + + if (pipe_ctx->stream_res.tg->funcs->set_drr) + pipe_ctx->stream_res.tg->funcs->set_drr( + pipe_ctx->stream_res.tg, NULL); } for (i = 0; i < dc->res_pool->pipe_count; i++) @@ -1799,7 +1686,7 @@ static void dcn20_reset_hw_ctx_wrap( if (!pipe_ctx_old->stream) continue; - if (pipe_ctx_old->top_pipe) + if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe) continue; if (!pipe_ctx->stream || @@ -1819,7 +1706,7 @@ static void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) { struct hubp *hubp = pipe_ctx->plane_res.hubp; struct mpcc_blnd_cfg blnd_cfg = { {0} }; - bool per_pixel_alpha = pipe_ctx->plane_state->per_pixel_alpha && pipe_ctx->bottom_pipe; + bool per_pixel_alpha = pipe_ctx->plane_state->per_pixel_alpha; int mpcc_id; struct mpcc *new_mpcc; struct mpc *mpc = dc->res_pool->mpc; @@ -2012,14 +1899,198 @@ static void dcn20_set_flip_control_gsl( } +static void dcn20_enable_stream(struct pipe_ctx *pipe_ctx) +{ + enum dc_lane_count lane_count = + pipe_ctx->stream->link->cur_link_settings.lane_count; + + struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; + struct dc_link *link = pipe_ctx->stream->link; + + uint32_t active_total_with_borders; + uint32_t early_control = 0; + struct timing_generator *tg = pipe_ctx->stream_res.tg; + + /* For MST, there are multiply stream go to only one link. + * connect DIG back_end to front_end while enable_stream and + * disconnect them during disable_stream + * BY this, it is logic clean to separate stream and link + */ + link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc, + pipe_ctx->stream_res.stream_enc->id, true); + + if (link->dc->hwss.program_dmdata_engine) + link->dc->hwss.program_dmdata_engine(pipe_ctx); + + link->dc->hwss.update_info_frame(pipe_ctx); + + /* enable early control to avoid corruption on DP monitor*/ + active_total_with_borders = + timing->h_addressable + + timing->h_border_left + + timing->h_border_right; + + if (lane_count != 0) + early_control = active_total_with_borders % lane_count; + + if (early_control == 0) + early_control = lane_count; + + tg->funcs->set_early_control(tg, early_control); + + /* enable audio only within mode set */ + if (pipe_ctx->stream_res.audio != NULL) { + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + pipe_ctx->stream_res.stream_enc->funcs->dp_audio_enable(pipe_ctx->stream_res.stream_enc); + } +} + +static void dcn20_program_dmdata_engine(struct pipe_ctx *pipe_ctx) +{ + struct dc_stream_state *stream = pipe_ctx->stream; + struct hubp *hubp = pipe_ctx->plane_res.hubp; + bool enable = false; + struct stream_encoder *stream_enc = pipe_ctx->stream_res.stream_enc; + enum dynamic_metadata_mode mode = dc_is_dp_signal(stream->signal) + ? dmdata_dp + : dmdata_hdmi; + + /* if using dynamic meta, don't set up generic infopackets */ + if (pipe_ctx->stream->dmdata_address.quad_part != 0) { + pipe_ctx->stream_res.encoder_info_frame.hdrsmd.valid = false; + enable = true; + } + + if (!hubp) + return; + + if (!stream_enc || !stream_enc->funcs->set_dynamic_metadata) + return; + + stream_enc->funcs->set_dynamic_metadata(stream_enc, enable, + hubp->inst, mode); +} + +static void dcn20_fpga_init_hw(struct dc *dc) +{ + int i, j; + struct dce_hwseq *hws = dc->hwseq; + struct resource_pool *res_pool = dc->res_pool; + struct dc_state *context = dc->current_state; + + if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) + dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); + + // Initialize the dccg + if (res_pool->dccg->funcs->dccg_init) + res_pool->dccg->funcs->dccg_init(res_pool->dccg); + + //Enable ability to power gate / don't force power on permanently + dc->hwss.enable_power_gating_plane(hws, true); + + // Specific to FPGA dccg and registers + REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF); + REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF); + + dcn20_dccg_init(hws); + + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); + REG_WRITE(REFCLK_CNTL, 0); + // + + + /* Blank pixel data with OPP DPG */ + for (i = 0; i < dc->res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + if (tg->funcs->is_tg_enabled(tg)) + dcn20_init_blank(dc, tg); + } + + for (i = 0; i < res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + if (tg->funcs->is_tg_enabled(tg)) + tg->funcs->lock(tg); + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct dpp *dpp = res_pool->dpps[i]; + + dpp->funcs->dpp_reset(dpp); + } + + /* Reset all MPCC muxes */ + res_pool->mpc->funcs->mpc_init(res_pool->mpc); + + /* initialize OPP mpc_tree parameter */ + for (i = 0; i < dc->res_pool->res_cap->num_opp; i++) { + res_pool->opps[i]->mpc_tree_params.opp_id = res_pool->opps[i]->inst; + res_pool->opps[i]->mpc_tree_params.opp_list = NULL; + for (j = 0; j < MAX_PIPES; j++) + res_pool->opps[i]->mpcc_disconnect_pending[j] = false; + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + struct hubp *hubp = dc->res_pool->hubps[i]; + struct dpp *dpp = dc->res_pool->dpps[i]; + + pipe_ctx->stream_res.tg = tg; + pipe_ctx->pipe_idx = i; + + pipe_ctx->plane_res.hubp = hubp; + pipe_ctx->plane_res.dpp = dpp; + pipe_ctx->plane_res.mpcc_inst = dpp->inst; + hubp->mpcc_id = dpp->inst; + hubp->opp_id = OPP_ID_INVALID; + hubp->power_gated = false; + pipe_ctx->stream_res.opp = NULL; + + hubp->funcs->hubp_init(hubp); + + //dc->res_pool->opps[i]->mpc_tree_params.opp_id = dc->res_pool->opps[i]->inst; + //dc->res_pool->opps[i]->mpc_tree_params.opp_list = NULL; + dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; + pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; + /*to do*/ + hwss1_plane_atomic_disconnect(dc, pipe_ctx); + } + + /* initialize DWB pointer to MCIF_WB */ + for (i = 0; i < res_pool->res_cap->num_dwb; i++) + res_pool->dwbc[i]->mcif = res_pool->mcif_wb[i]; + + for (i = 0; i < dc->res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + if (tg->funcs->is_tg_enabled(tg)) + tg->funcs->unlock(tg); + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + dc->hwss.disable_plane(dc, pipe_ctx); + + pipe_ctx->stream_res.tg = NULL; + pipe_ctx->plane_res.hubp = NULL; + } + + for (i = 0; i < dc->res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + tg->funcs->tg_init(tg); + } +} + void dcn20_hw_sequencer_construct(struct dc *dc) { dcn10_hw_sequencer_construct(dc); - dc->hwss.init_hw = dcn20_init_hw; - dc->hwss.init_pipes = NULL; dc->hwss.unblank_stream = dcn20_unblank_stream; dc->hwss.update_plane_addr = dcn20_update_plane_addr; - dc->hwss.disable_plane = dcn20_disable_plane, dc->hwss.enable_stream_timing = dcn20_enable_stream_timing; dc->hwss.program_triplebuffer = dcn20_program_tripleBuffer; dc->hwss.set_input_transfer_func = dcn20_set_input_transfer_func; @@ -2036,6 +2107,8 @@ void dcn20_hw_sequencer_construct(struct dc *dc) dc->hwss.update_odm = dcn20_update_odm; dc->hwss.blank_pixel_data = dcn20_blank_pixel_data; dc->hwss.dmdata_status_done = dcn20_dmdata_status_done; + dc->hwss.program_dmdata_engine = dcn20_program_dmdata_engine; + dc->hwss.enable_stream = dcn20_enable_stream; dc->hwss.disable_stream = dcn20_disable_stream; dc->hwss.init_sys_ctx = dcn20_init_sys_ctx; dc->hwss.init_vm_ctx = dcn20_init_vm_ctx; @@ -2045,5 +2118,23 @@ void dcn20_hw_sequencer_construct(struct dc *dc) dc->hwss.reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap; dc->hwss.update_mpcc = dcn20_update_mpcc; dc->hwss.set_flip_control_gsl = dcn20_set_flip_control_gsl; - dc->hwss.did_underflow_occur = dcn10_did_underflow_occur; + dc->hwss.init_blank = dcn20_init_blank; + dc->hwss.disable_plane = dcn20_disable_plane; + dc->hwss.plane_atomic_disable = dcn20_plane_atomic_disable; + dc->hwss.enable_power_gating_plane = dcn20_enable_power_gating_plane; + dc->hwss.dpp_pg_control = dcn20_dpp_pg_control; + dc->hwss.hubp_pg_control = dcn20_hubp_pg_control; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + dc->hwss.dsc_pg_control = dcn20_dsc_pg_control; +#else + dc->hwss.dsc_pg_control = NULL; +#endif + dc->hwss.disable_vga = dcn20_disable_vga; + + if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + dc->hwss.init_hw = dcn20_fpga_init_hw; + dc->hwss.init_pipes = NULL; + } + + } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h index 2b0409454073..92ab3dd91814 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h @@ -75,7 +75,7 @@ bool dcn20_dmdata_status_done(struct pipe_ctx *pipe_ctx); void dcn20_set_dmdata_attributes(struct pipe_ctx *pipe_ctx); -void dcn20_disable_stream(struct pipe_ctx *pipe_ctx, int option); +void dcn20_disable_stream(struct pipe_ctx *pipe_ctx); void dcn20_program_tripleBuffer( const struct dc *dc, @@ -91,13 +91,9 @@ void dcn20_pipe_control_lock_global( void dcn20_setup_gsl_group_as_lock(const struct dc *dc, struct pipe_ctx *pipe_ctx, bool enable); -void dcn20_pipe_control_lock( - struct dc *dc, - struct pipe_ctx *pipe, - bool lock); -void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); -void dcn20_enable_plane( - struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct dc_state *context); +void dcn20_dccg_init(struct dce_hwseq *hws); +void dcn20_init_blank( + struct dc *dc, + struct timing_generator *tg); +void dcn20_display_init(struct dc *dc); #endif /* __DC_HWSS_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c index f495582e9e87..e476f27aa3a9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c @@ -341,6 +341,7 @@ static const struct link_encoder_funcs dcn20_link_enc_funcs = { .fec_set_enable = enc2_fec_set_enable, .fec_set_ready = enc2_fec_set_ready, .fec_is_active = enc2_fec_is_active, + .get_dig_mode = dcn10_get_dig_mode, .get_dig_frontend = dcn10_get_dig_frontend, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c index 240749e4cf83..5a188b2bc033 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c @@ -233,14 +233,14 @@ static void mpc2_ogam_get_reg_field( reg->masks.exp_resion_start_segment = mpc20->mpc_mask->MPCC_OGAM_RAMA_EXP_REGION_START_SEGMENT_B; } -static void mpc20_power_on_ogam_lut( +void mpc20_power_on_ogam_lut( struct mpc *mpc, int mpcc_id, bool power_on) { struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); REG_SET(MPCC_MEM_PWR_CTRL[mpcc_id], 0, - MPCC_OGAM_MEM_PWR_FORCE, power_on == true ? 0:1); + MPCC_OGAM_MEM_PWR_DIS, power_on == true ? 1:0); } @@ -368,6 +368,11 @@ void apply_DEDCN20_305_wa( { struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); + if (mpc->ctx->dc->debug.cm_in_bypass) { + REG_SET(MPCC_OGAM_MODE[mpcc_id], 0, MPCC_OGAM_MODE, 0); + return; + } + if (mpc->ctx->dc->work_arounds.dedcn20_305_wa == false) { /*hw fixed in new review*/ return; @@ -390,10 +395,16 @@ void mpc2_set_output_gamma( enum dc_lut_mode next_mode; struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); + if (mpc->ctx->dc->debug.cm_in_bypass) { + REG_SET(MPCC_OGAM_MODE[mpcc_id], 0, MPCC_OGAM_MODE, 0); + return; + } + if (params == NULL) { REG_SET(MPCC_OGAM_MODE[mpcc_id], 0, MPCC_OGAM_MODE, 0); return; } + current_mode = mpc20_get_ogam_current(mpc, mpcc_id); if (current_mode == LUT_BYPASS || current_mode == LUT_RAM_A) next_mode = LUT_RAM_B; @@ -435,23 +446,22 @@ void mpc2_assert_mpcc_idle_before_connect(struct mpc *mpc, int mpcc_id) { struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); unsigned int top_sel, mpc_busy, mpc_idle, mpc_disabled; - REG_GET(MPCC_STATUS[mpcc_id], MPCC_DISABLED, &mpc_disabled); - - if (mpc_disabled) { - ASSERT(0); - return; - } REG_GET(MPCC_TOP_SEL[mpcc_id], MPCC_TOP_SEL, &top_sel); - if (top_sel == 0xf) { - REG_GET_2(MPCC_STATUS[mpcc_id], - MPCC_BUSY, &mpc_busy, - MPCC_IDLE, &mpc_idle); + REG_GET_3(MPCC_STATUS[mpcc_id], + MPCC_BUSY, &mpc_busy, + MPCC_IDLE, &mpc_idle, + MPCC_DISABLED, &mpc_disabled); - ASSERT(mpc_busy == 0); - ASSERT(mpc_idle == 1); + if (top_sel == 0xf) { + ASSERT(!mpc_busy); + ASSERT(mpc_idle); + ASSERT(mpc_disabled); + } else { + ASSERT(!mpc_disabled); + ASSERT(!mpc_idle); } } @@ -488,6 +498,7 @@ const struct mpc_funcs dcn20_mpc_funcs = { .insert_plane = mpc1_insert_plane, .remove_mpcc = mpc1_remove_mpcc, .mpc_init = mpc1_mpc_init, + .mpc_init_single_inst = mpc1_mpc_init_single_inst, .update_blending = mpc2_update_blending, .get_mpcc_for_dpp = mpc2_get_mpcc_for_dpp, .wait_for_idle = mpc2_assert_idle_mpcc, @@ -498,6 +509,7 @@ const struct mpc_funcs dcn20_mpc_funcs = { .set_output_csc = mpc2_set_output_csc, .set_ocsc_default = mpc2_set_ocsc_default, .set_output_gamma = mpc2_set_output_gamma, + .power_on_mpc_mem_pwr = mpc20_power_on_ogam_lut, }; void dcn20_mpc_construct(struct dcn20_mpc *mpc20, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h index 9750095d2d73..9f53192da2dc 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h @@ -159,6 +159,7 @@ SF(MPCC_OGAM0_MPCC_OGAM_RAMB_START_CNTL_B, MPCC_OGAM_RAMB_EXP_REGION_START_B, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_RAMB_START_CNTL_B, MPCC_OGAM_RAMB_EXP_REGION_START_SEGMENT_B, mask_sh),\ SF(MPCC0_MPCC_MEM_PWR_CTRL, MPCC_OGAM_MEM_PWR_FORCE, mask_sh),\ + SF(MPCC0_MPCC_MEM_PWR_CTRL, MPCC_OGAM_MEM_PWR_DIS, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_LUT_INDEX, MPCC_OGAM_LUT_INDEX, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_LUT_RAM_CONTROL, MPCC_OGAM_LUT_WRITE_EN_MASK, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_LUT_RAM_CONTROL, MPCC_OGAM_LUT_RAM_SEL, mask_sh),\ @@ -173,6 +174,7 @@ SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MAX_B_CB, mask_sh),\ SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MIN_B_CB, mask_sh) + #define MPC_REG_FIELD_LIST_DCN2_0(type) \ MPC_REG_FIELD_LIST(type)\ type MPCC_BG_BPC;\ @@ -217,7 +219,8 @@ type MPC_OUT_DENORM_CLAMP_MIN_G_Y;\ type MPC_OUT_DENORM_CLAMP_MAX_B_CB;\ type MPC_OUT_DENORM_CLAMP_MIN_B_CB;\ - type MPCC_DISABLED; + type MPCC_DISABLED;\ + type MPCC_OGAM_MEM_PWR_DIS; struct dcn20_mpc_registers { MPC_REG_VARIABLE_LIST_DCN2_0 @@ -282,4 +285,5 @@ void mpc2_set_output_gamma( void mpc2_assert_idle_mpcc(struct mpc *mpc, int id); void mpc2_assert_mpcc_idle_before_connect(struct mpc *mpc, int mpcc_id); +void mpc20_power_on_ogam_lut(struct mpc *mpc, int mpcc_id, bool power_on); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c index d9e7c711a71c..40164ed015ea 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c @@ -332,7 +332,6 @@ static struct opp_funcs dcn20_opp_funcs = { .opp_set_disp_pattern_generator = opp2_set_disp_pattern_generator, .dpg_is_blanked = opp2_dpg_is_blanked, .opp_dpg_set_blank_color = opp2_dpg_set_blank_color, - .opp_convert_pti = NULL, .opp_destroy = opp1_destroy, .opp_program_left_edge_extra_pixel = opp2_program_left_edge_extra_pixel, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c index 1ae973962d53..2137e2be2140 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c @@ -191,15 +191,6 @@ void optc2_set_dsc_config(struct timing_generator *optc, uint32_t dsc_slice_width) { struct optc *optc1 = DCN10TG_FROM_TG(optc); - uint32_t data_format = 0; - /* skip if dsc mode is not changed */ - data_format = dm_read_reg(CTX, REG(OPTC_DATA_FORMAT_CONTROL)); - - data_format = data_format & 0x30; /* bit5:4 */ - data_format = data_format >> 4; - - if (data_format == dsc_mode) - return; REG_UPDATE(OPTC_DATA_FORMAT_CONTROL, OPTC_DSC_MODE, dsc_mode); @@ -224,7 +215,6 @@ void optc2_set_odm_bypass(struct timing_generator *optc, struct optc *optc1 = DCN10TG_FROM_TG(optc); uint32_t h_div_2 = 0; - optc1->comb_opp_id = 0xf; REG_SET_3(OPTC_DATA_SOURCE_SELECT, 0, OPTC_NUM_OF_INPUT_SEGMENT, 0, OPTC_SEG0_SRC_SEL, optc->inst, @@ -236,13 +226,16 @@ void optc2_set_odm_bypass(struct timing_generator *optc, OTG_H_TIMING_DIV_BY2, h_div_2); REG_SET(OPTC_MEMORY_CONFIG, 0, OPTC_MEM_SEL, 0); + optc1->opp_count = 1; } -void optc2_set_odm_combine(struct timing_generator *optc, int combine_opp_id, - int mpcc_hactive, enum dc_pixel_encoding pixel_encoding) +void optc2_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_cnt, + struct dc_crtc_timing *timing) { struct optc *optc1 = DCN10TG_FROM_TG(optc); /* 2 pieces of memory required for up to 5120 displays, 4 for up to 8192 */ + int mpcc_hactive = (timing->h_addressable + timing->h_border_left + timing->h_border_right) + / opp_cnt; int memory_mask = mpcc_hactive <= 2560 ? 0x3 : 0xf; uint32_t data_fmt = 0; @@ -257,23 +250,24 @@ void optc2_set_odm_combine(struct timing_generator *optc, int combine_opp_id, REG_SET(OPTC_MEMORY_CONFIG, 0, OPTC_MEM_SEL, memory_mask << (optc->inst * 4)); - if (pixel_encoding == PIXEL_ENCODING_YCBCR422) + if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422) data_fmt = 1; - else if (pixel_encoding == PIXEL_ENCODING_YCBCR420) + else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) data_fmt = 2; REG_UPDATE(OPTC_DATA_FORMAT_CONTROL, OPTC_DATA_FORMAT, data_fmt); + ASSERT(opp_cnt == 2); REG_SET_3(OPTC_DATA_SOURCE_SELECT, 0, OPTC_NUM_OF_INPUT_SEGMENT, 1, - OPTC_SEG0_SRC_SEL, optc->inst, - OPTC_SEG1_SRC_SEL, combine_opp_id); + OPTC_SEG0_SRC_SEL, opp_id[0], + OPTC_SEG1_SRC_SEL, opp_id[1]); REG_UPDATE(OPTC_WIDTH_CONTROL, OPTC_SEGMENT_WIDTH, mpcc_hactive); REG_SET(OTG_H_TIMING_CNTL, 0, OTG_H_TIMING_DIV_BY2, 1); - optc1->comb_opp_id = combine_opp_id; + optc1->opp_count = opp_cnt; } void optc2_get_optc_source(struct timing_generator *optc, @@ -339,65 +333,6 @@ void optc2_triplebuffer_unlock(struct timing_generator *optc) } - -void optc2_setup_global_lock(struct timing_generator *optc) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - uint32_t v_blank_start = 0; - uint32_t h_blank_start = 0, h_total = 0; - - REG_SET(OTG_GLOBAL_CONTROL1, 0, MASTER_UPDATE_LOCK_DB_EN, 1); - - REG_SET(OTG_GLOBAL_CONTROL2, 0, DIG_UPDATE_LOCATION, 20); - - REG_GET(OTG_V_BLANK_START_END, OTG_V_BLANK_START, &v_blank_start); - - REG_GET(OTG_H_BLANK_START_END, OTG_H_BLANK_START, &h_blank_start); - - REG_GET(OTG_H_TOTAL, OTG_H_TOTAL, &h_total); - REG_UPDATE_2(OTG_GLOBAL_CONTROL1, - MASTER_UPDATE_LOCK_DB_X, - h_blank_start - 200 - 1, - MASTER_UPDATE_LOCK_DB_Y, - v_blank_start - 1); -} - -void optc2_lock_global(struct timing_generator *optc) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_UPDATE(OTG_GLOBAL_CONTROL2, GLOBAL_UPDATE_LOCK_EN, 1); - - REG_SET(OTG_GLOBAL_CONTROL0, 0, - OTG_MASTER_UPDATE_LOCK_SEL, optc->inst); - REG_SET(OTG_MASTER_UPDATE_LOCK, 0, - OTG_MASTER_UPDATE_LOCK, 1); - - /* Should be fast, status does not update on maximus */ - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); -} - -void optc2_lock(struct timing_generator *optc) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_UPDATE(OTG_GLOBAL_CONTROL2, GLOBAL_UPDATE_LOCK_EN, 0); - - REG_SET(OTG_GLOBAL_CONTROL0, 0, - OTG_MASTER_UPDATE_LOCK_SEL, optc->inst); - REG_SET(OTG_MASTER_UPDATE_LOCK, 0, - OTG_MASTER_UPDATE_LOCK, 1); - - /* Should be fast, status does not update on maximus */ - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); -} - void optc2_lock_doublebuffer_enable(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -492,10 +427,8 @@ static struct timing_generator_funcs dcn20_tg_funcs = { .triplebuffer_lock = optc2_triplebuffer_lock, .triplebuffer_unlock = optc2_triplebuffer_unlock, .disable_reset_trigger = optc1_disable_reset_trigger, - .lock = optc2_lock, + .lock = optc1_lock, .unlock = optc1_unlock, - .lock_global = optc2_lock_global, - .setup_global_lock = optc2_setup_global_lock, .lock_doublebuffer_enable = optc2_lock_doublebuffer_enable, .lock_doublebuffer_disable = optc2_lock_doublebuffer_disable, .enable_optc_clock = optc1_enable_optc_clock, @@ -522,7 +455,8 @@ static struct timing_generator_funcs dcn20_tg_funcs = { .set_gsl_source_select = optc2_set_gsl_source_select, .set_vtg_params = optc1_set_vtg_params, .program_manual_trigger = optc2_program_manual_trigger, - .setup_manual_trigger = optc2_setup_manual_trigger + .setup_manual_trigger = optc2_setup_manual_trigger, + .is_matching_timing = optc1_is_matching_timing }; void dcn20_timing_generator_init(struct optc *optc1) @@ -537,6 +471,5 @@ void dcn20_timing_generator_init(struct optc *optc1) optc1->min_v_blank_interlace = 5; optc1->min_h_sync_width = 4;// Minimum HSYNC = 8 pixels asked By HW in the first place for no actual reason. Oculus Rift S will not light up with 8 as it's hsyncWidth is 6. Changing it to 4 to fix that issue. optc1->min_v_sync_width = 1; - optc1->comb_opp_id = 0xf; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h index ebf07c582da2..32a58431fd09 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h @@ -96,8 +96,8 @@ void optc2_set_dsc_config(struct timing_generator *optc, void optc2_set_odm_bypass(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); -void optc2_set_odm_combine(struct timing_generator *optc, int combine_opp_id, - int mpcc_hactive, enum dc_pixel_encoding pixel_encoding); +void optc2_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_cnt, + struct dc_crtc_timing *timing); void optc2_get_optc_source(struct timing_generator *optc, uint32_t *num_of_src_opp, @@ -106,9 +106,6 @@ void optc2_get_optc_source(struct timing_generator *optc, void optc2_triplebuffer_lock(struct timing_generator *optc); void optc2_triplebuffer_unlock(struct timing_generator *optc); -void optc2_lock(struct timing_generator *optc); -void optc2_lock_global(struct timing_generator *optc); -void optc2_setup_global_lock(struct timing_generator *optc); void optc2_lock_doublebuffer_disable(struct timing_generator *optc); void optc2_lock_doublebuffer_enable(struct timing_generator *optc); void optc2_program_manual_trigger(struct timing_generator *optc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index b949e202d6cb..b4e3ce22ed52 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -82,6 +82,7 @@ #include "amdgpu_socbb.h" +/* NV12 SOC BB is currently in FW, mark SW bounding box invalid. */ #define SOC_BOUNDING_BOX_VALID false #define DC_LOGGER_INIT(logger) @@ -156,8 +157,119 @@ struct _vcs_dpi_ip_params_st dcn2_0_ip = { .xfc_fill_constant_bytes = 0, }; -struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { 0 }; - +struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { + /* Defaults that get patched on driver load from firmware. */ + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 560.0, + .fabricclk_mhz = 560.0, + .dispclk_mhz = 513.0, + .dppclk_mhz = 513.0, + .phyclk_mhz = 540.0, + .socclk_mhz = 560.0, + .dscclk_mhz = 171.0, + .dram_speed_mts = 8960.0, + }, + { + .state = 1, + .dcfclk_mhz = 694.0, + .fabricclk_mhz = 694.0, + .dispclk_mhz = 642.0, + .dppclk_mhz = 642.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 694.0, + .dscclk_mhz = 214.0, + .dram_speed_mts = 11104.0, + }, + { + .state = 2, + .dcfclk_mhz = 875.0, + .fabricclk_mhz = 875.0, + .dispclk_mhz = 734.0, + .dppclk_mhz = 734.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 875.0, + .dscclk_mhz = 245.0, + .dram_speed_mts = 14000.0, + }, + { + .state = 3, + .dcfclk_mhz = 1000.0, + .fabricclk_mhz = 1000.0, + .dispclk_mhz = 1100.0, + .dppclk_mhz = 1100.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1000.0, + .dscclk_mhz = 367.0, + .dram_speed_mts = 16000.0, + }, + { + .state = 4, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + }, + .num_states = 5, + .sr_exit_time_us = 8.6, + .sr_enter_plus_exit_time_us = 10.9, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 40.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 40.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 2, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 131, + .urgent_out_of_order_return_per_channel_bytes = 256, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 16, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404.0, + .dummy_pstate_latency_us = 5.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3850, + .xfc_bus_transport_time_us = 20, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 0 +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { 0 }; #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL #define mmDP0_DP_DPHY_INTERNAL_CTRL 0x210f @@ -314,7 +426,7 @@ static const struct dce_audio_shift audio_shift = { DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -695,6 +807,16 @@ static const struct dc_plane_cap plane_cap = { .fp16 = 1 } }; +static const struct resource_caps res_cap_nv14 = { + .num_timing_generator = 5, + .num_opp = 5, + .num_video_plane = 5, + .num_audio = 6, + .num_stream_encoder = 5, + .num_pll = 5, + .num_dwb = 0, + .num_ddc = 5, +}; static const struct dc_debug_options debug_defaults_drv = { .disable_dmcu = true, @@ -1197,7 +1319,11 @@ static void get_pixel_clock_parameters( struct pixel_clk_params *pixel_clk_params) { const struct dc_stream_state *stream = pipe_ctx->stream; - bool odm_combine = dc_res_get_odm_bottom_pipe(pipe_ctx) != NULL; + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; pixel_clk_params->requested_pix_clk_100hz = stream->timing.pix_clk_100hz; pixel_clk_params->encoder_object_id = stream->link->link_enc->id; @@ -1215,7 +1341,9 @@ static void get_pixel_clock_parameters( if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422) pixel_clk_params->color_depth = COLOR_DEPTH_888; - if (optc1_is_two_pixels_per_containter(&stream->timing) || odm_combine) + if (opp_cnt == 4) + pixel_clk_params->requested_pix_clk_100hz /= 4; + else if (optc1_is_two_pixels_per_containter(&stream->timing) || opp_cnt == 2) pixel_clk_params->requested_pix_clk_100hz /= 2; if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING) @@ -1359,22 +1487,16 @@ static enum dc_status remove_dsc_from_stream_resource(struct dc *dc, for (i = 0; i < MAX_PIPES; i++) { if (new_ctx->res_ctx.pipe_ctx[i].stream == dc_stream && !new_ctx->res_ctx.pipe_ctx[i].top_pipe) { pipe_ctx = &new_ctx->res_ctx.pipe_ctx[i]; - break; + + if (pipe_ctx->stream_res.dsc) + release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); } } if (!pipe_ctx) return DC_ERROR_UNEXPECTED; - - if (pipe_ctx->stream_res.dsc) { - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); - - release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); - if (odm_pipe) - release_dsc(&new_ctx->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); - } - - return DC_OK; + else + return DC_OK; } #endif @@ -1473,17 +1595,92 @@ static void swizzle_to_dml_params( } } -static bool dcn20_split_stream_for_combine( +static bool dcn20_split_stream_for_odm( + struct resource_context *res_ctx, + const struct resource_pool *pool, + struct pipe_ctx *prev_odm_pipe, + struct pipe_ctx *next_odm_pipe) +{ + int pipe_idx = next_odm_pipe->pipe_idx; + + *next_odm_pipe = *prev_odm_pipe; + + next_odm_pipe->pipe_idx = pipe_idx; + next_odm_pipe->plane_res.mi = pool->mis[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.hubp = pool->hubps[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.ipp = pool->ipps[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.xfm = pool->transforms[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.dpp = pool->dpps[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.mpcc_inst = pool->dpps[next_odm_pipe->pipe_idx]->inst; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + next_odm_pipe->stream_res.dsc = NULL; +#endif + if (prev_odm_pipe->next_odm_pipe && prev_odm_pipe->next_odm_pipe != next_odm_pipe) { + ASSERT(!next_odm_pipe->next_odm_pipe); + next_odm_pipe->next_odm_pipe = prev_odm_pipe->next_odm_pipe; + next_odm_pipe->next_odm_pipe->prev_odm_pipe = next_odm_pipe; + } + prev_odm_pipe->next_odm_pipe = next_odm_pipe; + next_odm_pipe->prev_odm_pipe = prev_odm_pipe; + ASSERT(next_odm_pipe->top_pipe == NULL); + + if (prev_odm_pipe->plane_state) { + struct scaler_data *sd = &prev_odm_pipe->plane_res.scl_data; + int new_width; + + /* HACTIVE halved for odm combine */ + sd->h_active /= 2; + /* Calculate new vp and recout for left pipe */ + /* Need at least 16 pixels width per side */ + if (sd->recout.x + 16 >= sd->h_active) + return false; + new_width = sd->h_active - sd->recout.x; + sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz, sd->recout.width - new_width)); + sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz_c, sd->recout.width - new_width)); + sd->recout.width = new_width; + + /* Calculate new vp and recout for right pipe */ + sd = &next_odm_pipe->plane_res.scl_data; + /* HACTIVE halved for odm combine */ + sd->h_active /= 2; + /* Need at least 16 pixels width per side */ + if (new_width <= 16) + return false; + new_width = sd->recout.width + sd->recout.x - sd->h_active; + sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz, sd->recout.width - new_width)); + sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz_c, sd->recout.width - new_width)); + sd->recout.width = new_width; + sd->viewport.x += dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz, sd->h_active - sd->recout.x)); + sd->viewport_c.x += dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz_c, sd->h_active - sd->recout.x)); + sd->recout.x = 0; + } + next_odm_pipe->stream_res.opp = pool->opps[next_odm_pipe->pipe_idx]; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + if (next_odm_pipe->stream->timing.flags.DSC == 1) { + acquire_dsc(res_ctx, pool, &next_odm_pipe->stream_res.dsc); + ASSERT(next_odm_pipe->stream_res.dsc); + if (next_odm_pipe->stream_res.dsc == NULL) + return false; + } +#endif + + return true; +} + +static void dcn20_split_stream_for_mpc( struct resource_context *res_ctx, const struct resource_pool *pool, struct pipe_ctx *primary_pipe, - struct pipe_ctx *secondary_pipe, - bool is_odm_combine) + struct pipe_ctx *secondary_pipe) { int pipe_idx = secondary_pipe->pipe_idx; - struct scaler_data *sd = &primary_pipe->plane_res.scl_data; struct pipe_ctx *sec_bot_pipe = secondary_pipe->bottom_pipe; - int new_width; *secondary_pipe = *primary_pipe; secondary_pipe->bottom_pipe = sec_bot_pipe; @@ -1506,57 +1703,9 @@ static bool dcn20_split_stream_for_combine( primary_pipe->bottom_pipe = secondary_pipe; secondary_pipe->top_pipe = primary_pipe; - if (is_odm_combine) { - if (primary_pipe->plane_state) { - /* HACTIVE halved for odm combine */ - sd->h_active /= 2; - /* Copy scl_data to secondary pipe */ - secondary_pipe->plane_res.scl_data = *sd; - - /* Calculate new vp and recout for left pipe */ - /* Need at least 16 pixels width per side */ - if (sd->recout.x + 16 >= sd->h_active) - return false; - new_width = sd->h_active - sd->recout.x; - sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz, sd->recout.width - new_width)); - sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz_c, sd->recout.width - new_width)); - sd->recout.width = new_width; - - /* Calculate new vp and recout for right pipe */ - sd = &secondary_pipe->plane_res.scl_data; - new_width = sd->recout.width + sd->recout.x - sd->h_active; - /* Need at least 16 pixels width per side */ - if (new_width <= 16) - return false; - sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz, sd->recout.width - new_width)); - sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz_c, sd->recout.width - new_width)); - sd->recout.width = new_width; - sd->viewport.x += dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz, sd->h_active - sd->recout.x)); - sd->viewport_c.x += dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz_c, sd->h_active - sd->recout.x)); - sd->recout.x = 0; - } - secondary_pipe->stream_res.opp = pool->opps[secondary_pipe->pipe_idx]; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (secondary_pipe->stream->timing.flags.DSC == 1) { - acquire_dsc(res_ctx, pool, &secondary_pipe->stream_res.dsc); - ASSERT(secondary_pipe->stream_res.dsc); - if (secondary_pipe->stream_res.dsc == NULL) - return false; - } -#endif - } else { - ASSERT(primary_pipe->plane_state); - resource_build_scaling_params(primary_pipe); - resource_build_scaling_params(secondary_pipe); - } - - return true; + ASSERT(primary_pipe->plane_state); + resource_build_scaling_params(primary_pipe); + resource_build_scaling_params(secondary_pipe); } void dcn20_populate_dml_writeback_from_context( @@ -1669,6 +1818,19 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].dout.dp_lanes = 4; pipes[pipe_cnt].pipe.dest.vtotal_min = res_ctx->pipe_ctx[i].stream->adjust.v_total_min; pipes[pipe_cnt].pipe.dest.vtotal_max = res_ctx->pipe_ctx[i].stream->adjust.v_total_max; + pipes[pipe_cnt].pipe.dest.odm_combine = res_ctx->pipe_ctx[i].prev_odm_pipe + || res_ctx->pipe_ctx[i].next_odm_pipe; + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; + if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state + == res_ctx->pipe_ctx[i].plane_state) + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].top_pipe->pipe_idx; + else if (res_ctx->pipe_ctx[i].prev_odm_pipe) { + struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].prev_odm_pipe; + + while (first_pipe->prev_odm_pipe) + first_pipe = first_pipe->prev_odm_pipe; + pipes[pipe_cnt].pipe.src.hsplit_grp = first_pipe->pipe_idx; + } switch (res_ctx->pipe_ctx[i].stream->signal) { case SIGNAL_TYPE_DISPLAY_PORT_MST: @@ -1721,7 +1883,6 @@ int dcn20_populate_dml_pipes_from_context( break; } - switch (res_ctx->pipe_ctx[i].stream->timing.pixel_encoding) { case PIXEL_ENCODING_RGB: case PIXEL_ENCODING_YCBCR444: @@ -1743,10 +1904,6 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].dout.output_format = dm_444; pipes[pipe_cnt].dout.output_bpp = output_bpc * 3; } - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; - if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state - == res_ctx->pipe_ctx[i].plane_state) - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].top_pipe->pipe_idx; /* todo: default max for now, until there is logic reflecting this in dc*/ pipes[pipe_cnt].dout.output_bpc = 12; @@ -1795,14 +1952,6 @@ int dcn20_populate_dml_pipes_from_context( && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln) || (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln); - pipes[pipe_cnt].pipe.dest.odm_combine = (res_ctx->pipe_ctx[i].bottom_pipe - && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln - && res_ctx->pipe_ctx[i].bottom_pipe->stream_res.opp - != res_ctx->pipe_ctx[i].stream_res.opp) - || (res_ctx->pipe_ctx[i].top_pipe - && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln - && res_ctx->pipe_ctx[i].top_pipe->stream_res.opp - != res_ctx->pipe_ctx[i].stream_res.opp); pipes[pipe_cnt].pipe.src.source_scan = pln->rotation == ROTATION_ANGLE_90 || pln->rotation == ROTATION_ANGLE_270 ? dm_vert : dm_horz; pipes[pipe_cnt].pipe.src.viewport_y_y = scl->viewport.y; @@ -1812,13 +1961,13 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.src.viewport_height = scl->viewport.height; pipes[pipe_cnt].pipe.src.viewport_height_c = scl->viewport_c.height; if (pln->format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) { - pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.video.luma_pitch; - pipes[pipe_cnt].pipe.src.data_pitch_c = pln->plane_size.video.chroma_pitch; - pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.video.meta_pitch_l; - pipes[pipe_cnt].pipe.src.meta_pitch_c = pln->dcc.video.meta_pitch_c; + pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; + pipes[pipe_cnt].pipe.src.data_pitch_c = pln->plane_size.chroma_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch_c = pln->dcc.meta_pitch_c; } else { - pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.grph.surface_pitch; - pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.grph.meta_pitch; + pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; } pipes[pipe_cnt].pipe.src.dcc = pln->dcc.enable; pipes[pipe_cnt].pipe.dest.recout_width = scl->recout.width; @@ -1986,20 +2135,24 @@ static bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) struct pipe_ctx *pipe_ctx = &new_ctx->res_ctx.pipe_ctx[i]; struct dc_stream_state *stream = pipe_ctx->stream; struct dsc_config dsc_cfg; + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; /* Only need to validate top pipe */ - if (pipe_ctx->top_pipe || !stream || !stream->timing.flags.DSC) + if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe || !stream || !stream->timing.flags.DSC) continue; - dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left - + stream->timing.h_border_right; + dsc_cfg.pic_width = (stream->timing.h_addressable + stream->timing.h_border_left + + stream->timing.h_border_right) / opp_cnt; dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; - if (dc_res_get_odm_bottom_pipe(pipe_ctx)) - dsc_cfg.pic_width /= 2; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; if (!pipe_ctx->stream_res.dsc->funcs->dsc_validate_stream(pipe_ctx->stream_res.dsc, &dsc_cfg)) return false; @@ -2008,15 +2161,93 @@ static bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) } #endif -bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) +static struct pipe_ctx *dcn20_find_secondary_pipe(struct dc *dc, + struct resource_context *res_ctx, + const struct resource_pool *pool, + const struct pipe_ctx *primary_pipe) { - bool out = false; + struct pipe_ctx *secondary_pipe = NULL; + + if (dc && primary_pipe) { + int j; + int preferred_pipe_idx = 0; + + /* first check the prev dc state: + * if this primary pipe has a bottom pipe in prev. state + * and if the bottom pipe is still available (which it should be), + * pick that pipe as secondary + * Same logic applies for ODM pipes. Since mpo is not allowed with odm + * check in else case. + */ + if (dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].bottom_pipe) { + preferred_pipe_idx = dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].bottom_pipe->pipe_idx; + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + } + } else if (dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].next_odm_pipe) { + preferred_pipe_idx = dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].next_odm_pipe->pipe_idx; + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + } + } - BW_VAL_TRACE_SETUP(); + /* + * if this primary pipe does not have a bottom pipe in prev. state + * start backward and find a pipe that did not used to be a bottom pipe in + * prev. dc state. This way we make sure we keep the same assignment as + * last state and will not have to reprogram every pipe + */ + if (secondary_pipe == NULL) { + for (j = dc->res_pool->pipe_count - 1; j >= 0; j--) { + if (dc->current_state->res_ctx.pipe_ctx[j].top_pipe == NULL) { + preferred_pipe_idx = j; + + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + break; + } + } + } + } + /* + * We should never hit this assert unless assignments are shuffled around + * if this happens we will prob. hit a vsync tdr + */ + ASSERT(secondary_pipe); + /* + * search backwards for the second pipe to keep pipe + * assignment more consistent + */ + if (secondary_pipe == NULL) { + for (j = dc->res_pool->pipe_count - 1; j >= 0; j--) { + preferred_pipe_idx = j; + + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + break; + } + } + } + } + + return secondary_pipe; +} + +bool dcn20_fast_validate_bw( + struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *pipe_cnt_out, + int *pipe_split_from, + int *vlevel_out) +{ + bool out = false; int pipe_cnt, i, pipe_idx, vlevel, vlevel_unsplit; - int pipe_split_from[MAX_PIPES]; bool odm_capable = context->bw_ctx.dml.ip.odm_capable; bool force_split = false; #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT @@ -2024,15 +2255,44 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, #endif int split_threshold = dc->res_pool->pipe_count / 2; bool avoid_split = dc->debug.pipe_split_policy != MPC_SPLIT_DYNAMIC; - display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_KERNEL); - DC_LOGGER_INIT(dc->ctx->logger); - BW_VAL_TRACE_COUNT(); ASSERT(pipes); if (!pipes) return false; + /* merge previously split odm pipes since mode support needs to make the decision */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + struct pipe_ctx *odm_pipe = pipe->next_odm_pipe; + + if (pipe->prev_odm_pipe) + continue; + + pipe->next_odm_pipe = NULL; + while (odm_pipe) { + struct pipe_ctx *next_odm_pipe = odm_pipe->next_odm_pipe; + + odm_pipe->plane_state = NULL; + odm_pipe->stream = NULL; + odm_pipe->top_pipe = NULL; + odm_pipe->bottom_pipe = NULL; + odm_pipe->prev_odm_pipe = NULL; + odm_pipe->next_odm_pipe = NULL; +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + if (odm_pipe->stream_res.dsc) + release_dsc(&context->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); +#endif + /* Clear plane_res and stream_res */ + memset(&odm_pipe->plane_res, 0, sizeof(odm_pipe->plane_res)); + memset(&odm_pipe->stream_res, 0, sizeof(odm_pipe->stream_res)); + odm_pipe = next_odm_pipe; + } + if (pipe->plane_state) + resource_build_scaling_params(pipe); + } + + /* merge previously mpc split pipes since mode support needs to make the decision */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; struct pipe_ctx *hsplit_pipe = pipe->bottom_pipe; @@ -2040,7 +2300,6 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, if (!hsplit_pipe || hsplit_pipe->plane_state != pipe->plane_state) continue; - /* merge previously split pipe since mode support needs to make the decision */ pipe->bottom_pipe = hsplit_pipe->bottom_pipe; if (hsplit_pipe->bottom_pipe) hsplit_pipe->bottom_pipe->top_pipe = pipe; @@ -2048,10 +2307,7 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, hsplit_pipe->stream = NULL; hsplit_pipe->top_pipe = NULL; hsplit_pipe->bottom_pipe = NULL; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (hsplit_pipe->stream_res.dsc && hsplit_pipe->stream_res.dsc != pipe->stream_res.dsc) - release_dsc(&context->res_ctx, dc->res_pool, &hsplit_pipe->stream_res.dsc); -#endif + /* Clear plane_res and stream_res */ memset(&hsplit_pipe->plane_res, 0, sizeof(hsplit_pipe->plane_res)); memset(&hsplit_pipe->stream_res, 0, sizeof(hsplit_pipe->stream_res)); @@ -2066,8 +2322,9 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, &context->res_ctx, pipes); + *pipe_cnt_out = pipe_cnt; + if (!pipe_cnt) { - BW_VAL_TRACE_SKIP(pass); out = true; goto validate_out; } @@ -2160,17 +2417,12 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, } if (force_split && context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] == 1) context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] /= 2; - if (dc->config.forced_clocks == true) { - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] = - context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; - } if (!pipe->top_pipe && !pipe->plane_state && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) { - hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, dc->res_pool, pipe); + hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); ASSERT(hsplit_pipe); - if (!dcn20_split_stream_for_combine( + if (!dcn20_split_stream_for_odm( &context->res_ctx, dc->res_pool, - pipe, hsplit_pipe, - true)) + pipe, hsplit_pipe)) goto validate_fail; pipe_split_from[hsplit_pipe->pipe_idx] = pipe_idx; dcn20_build_mapped_resource(dc, context, pipe->stream); @@ -2206,16 +2458,20 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, if (need_split3d || need_split || force_split) { if (!hsplit_pipe || hsplit_pipe->plane_state != pipe->plane_state) { /* pipe not split previously needs split */ - hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, dc->res_pool, pipe); + hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); ASSERT(hsplit_pipe || force_split); if (!hsplit_pipe) continue; - if (!dcn20_split_stream_for_combine( + if (context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) { + if (!dcn20_split_stream_for_odm( + &context->res_ctx, dc->res_pool, + pipe, hsplit_pipe)) + goto validate_fail; + } else + dcn20_split_stream_for_mpc( &context->res_ctx, dc->res_pool, - pipe, hsplit_pipe, - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx])) - goto validate_fail; + pipe, hsplit_pipe); pipe_split_from[hsplit_pipe->pipe_idx] = pipe_idx; } } else if (hsplit_pipe && hsplit_pipe->plane_state == pipe->plane_state) { @@ -2232,13 +2488,26 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, } #endif - BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + *vlevel_out = vlevel; - if (fast_validate) { - BW_VAL_TRACE_SKIP(fast); - out = true; - goto validate_out; - } + out = true; + goto validate_out; + +validate_fail: + out = false; + +validate_out: + return out; +} + +void dcn20_calculate_wm( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *out_pipe_cnt, + int *pipe_split_from, + int vlevel) +{ + int pipe_cnt, i, pipe_idx; for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) @@ -2265,10 +2534,16 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, else pipes[pipe_cnt].pipe.dest.odm_combine = 0; } + if (dc->config.forced_clocks) { pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dispclk_mhz; pipes[pipe_cnt].clks_cfg.dppclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; } + if (dc->debug.min_disp_clk_khz > pipes[pipe_cnt].clks_cfg.dispclk_mhz * 1000) + pipes[pipe_cnt].clks_cfg.dispclk_mhz = dc->debug.min_disp_clk_khz / 1000.0; + if (dc->debug.min_dpp_clk_khz > pipes[pipe_cnt].clks_cfg.dppclk_mhz * 1000) + pipes[pipe_cnt].clks_cfg.dppclk_mhz = dc->debug.min_dpp_clk_khz / 1000.0; + pipe_cnt++; } @@ -2281,6 +2556,8 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, &context->res_ctx, pipes); } + *out_pipe_cnt = pipe_cnt; + pipes[0].clks_cfg.voltage = vlevel; pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].dcfclk_mhz; pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; @@ -2327,6 +2604,17 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; +} + +void dcn20_calculate_dlg_params( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel) +{ + int i, j, pipe_idx, pipe_idx_unsplit; + bool visited[MAX_PIPES] = { 0 }; + /* Writeback MCIF_WB arbitration parameters */ dc->res_pool->funcs->set_mcif_arb_params(dc, context, pipes, pipe_cnt); @@ -2335,32 +2623,69 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.clk.socclk_khz = context->bw_ctx.dml.vba.SOCCLK * 1000; context->bw_ctx.bw.dcn.clk.dramclk_khz = context->bw_ctx.dml.vba.DRAMSpeed * 1000 / 16; context->bw_ctx.bw.dcn.clk.dcfclk_deep_sleep_khz = context->bw_ctx.dml.vba.DCFCLKDeepSleep * 1000; - context->bw_ctx.bw.dcn.clk.fclk_khz = context->bw_ctx.dml.vba.FabricClock * 1000; + context->bw_ctx.bw.dcn.clk.fclk_khz = 0; context->bw_ctx.bw.dcn.clk.p_state_change_support = context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] != dm_dram_clock_change_unsupported; context->bw_ctx.bw.dcn.clk.dppclk_khz = 0; - BW_VAL_TRACE_END_WATERMARKS(); + /* + * An artifact of dml pipe split/odm is that pipes get merged back together for + * calculation. Therefore we need to only extract for first pipe in ascending index order + * and copy into the other split half. + */ + for (i = 0, pipe_idx = 0, pipe_idx_unsplit = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + if (!visited[pipe_idx]) { + display_pipe_source_params_st *src = &pipes[pipe_idx_unsplit].pipe.src; + display_pipe_dest_params_st *dst = &pipes[pipe_idx_unsplit].pipe.dest; + + dst->vstartup_start = context->bw_ctx.dml.vba.VStartup[pipe_idx_unsplit]; + dst->vupdate_offset = context->bw_ctx.dml.vba.VUpdateOffsetPix[pipe_idx_unsplit]; + dst->vupdate_width = context->bw_ctx.dml.vba.VUpdateWidthPix[pipe_idx_unsplit]; + dst->vready_offset = context->bw_ctx.dml.vba.VReadyOffsetPix[pipe_idx_unsplit]; + /* + * j iterates inside pipes array, unlike i which iterates inside + * pipe_ctx array + */ + if (src->is_hsplit) + for (j = pipe_idx + 1; j < pipe_cnt; j++) { + display_pipe_source_params_st *src_j = &pipes[j].pipe.src; + display_pipe_dest_params_st *dst_j = &pipes[j].pipe.dest; + + if (src_j->is_hsplit && !visited[j] + && src->hsplit_grp == src_j->hsplit_grp) { + dst_j->vstartup_start = context->bw_ctx.dml.vba.VStartup[pipe_idx_unsplit]; + dst_j->vupdate_offset = context->bw_ctx.dml.vba.VUpdateOffsetPix[pipe_idx_unsplit]; + dst_j->vupdate_width = context->bw_ctx.dml.vba.VUpdateWidthPix[pipe_idx_unsplit]; + dst_j->vready_offset = context->bw_ctx.dml.vba.VReadyOffsetPix[pipe_idx_unsplit]; + visited[j] = true; + } + } + visited[pipe_idx] = true; + pipe_idx_unsplit++; + } + pipe_idx++; + } for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) continue; - pipes[pipe_idx].pipe.dest.vstartup_start = context->bw_ctx.dml.vba.VStartup[pipe_idx]; - pipes[pipe_idx].pipe.dest.vupdate_offset = context->bw_ctx.dml.vba.VUpdateOffsetPix[pipe_idx]; - pipes[pipe_idx].pipe.dest.vupdate_width = context->bw_ctx.dml.vba.VUpdateWidthPix[pipe_idx]; - pipes[pipe_idx].pipe.dest.vready_offset = context->bw_ctx.dml.vba.VReadyOffsetPix[pipe_idx]; if (context->bw_ctx.bw.dcn.clk.dppclk_khz < pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000) context->bw_ctx.bw.dcn.clk.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - context->res_ctx.pipe_ctx[i].stream_res.dscclk_khz = - context->bw_ctx.dml.vba.DSCCLK_calculated[pipe_idx] * 1000; -#endif + ASSERT(visited[pipe_idx]); context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest; pipe_idx++; } + /*save a original dppclock copy*/ + context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz; + context->bw_ctx.bw.dcn.clk.bw_dispclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz; + context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dppclk_mhz * 1000; + context->bw_ctx.bw.dcn.clk.max_supported_dispclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dispclk_mhz * 1000; for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { bool cstate_en = context->bw_ctx.dml.vba.PrefetchMode[vlevel][context->bw_ctx.dml.vba.maxMpcComb] != 2; @@ -2383,8 +2708,43 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, pipes[pipe_idx].pipe); pipe_idx++; } +} + +static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool out = false; + + BW_VAL_TRACE_SETUP(); + + int vlevel = 0; + int pipe_split_from[MAX_PIPES]; + int pipe_cnt = 0; + display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_KERNEL); + DC_LOGGER_INIT(dc->ctx->logger); + + BW_VAL_TRACE_COUNT(); + + out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel); + + if (pipe_cnt == 0) + goto validate_out; + + if (!out) + goto validate_fail; + + BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + + if (fast_validate) { + BW_VAL_TRACE_SKIP(fast); + goto validate_out; + } + + dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel); + dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + + BW_VAL_TRACE_END_WATERMARKS(); - out = true; goto validate_out; validate_fail: @@ -2402,6 +2762,50 @@ validate_out: return out; } + +bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool voltage_supported = false; + bool full_pstate_supported = false; + bool dummy_pstate_supported = false; + double p_state_latency_us = context->bw_ctx.dml.soc.dram_clock_change_latency_us; + + if (fast_validate) + return dcn20_validate_bandwidth_internal(dc, context, true); + + + // Best case, we support full UCLK switch latency + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); + full_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; + + if (context->bw_ctx.dml.soc.dummy_pstate_latency_us == 0 || + (voltage_supported && full_pstate_supported)) { + context->bw_ctx.bw.dcn.clk.p_state_change_support = true; + goto restore_dml_state; + } + + // Fallback: Try to only support G6 temperature read latency + context->bw_ctx.dml.soc.dram_clock_change_latency_us = context->bw_ctx.dml.soc.dummy_pstate_latency_us; + + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); + dummy_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; + + if (voltage_supported && dummy_pstate_supported) { + context->bw_ctx.bw.dcn.clk.p_state_change_support = false; + goto restore_dml_state; + } + + // ERROR: fallback is supposed to always work. + ASSERT(false); + +restore_dml_state: + memcpy(&context->bw_ctx.dml, &dc->dml, sizeof(struct display_mode_lib)); + context->bw_ctx.dml.soc.dram_clock_change_latency_us = p_state_latency_us; + + return voltage_supported; +} + struct pipe_ctx *dcn20_acquire_idle_pipe_for_layer( struct dc_state *state, const struct resource_pool *pool, @@ -2576,9 +2980,6 @@ static void cap_soc_clocks( && max_clocks.uClockInKhz != 0) bb->clock_limits[i].dram_speed_mts = (max_clocks.uClockInKhz / 1000) * 16; - // HACK: Force every uclk to max for now to "disable" uclk switching. - bb->clock_limits[i].dram_speed_mts = (max_clocks.uClockInKhz / 1000) * 16; - if ((bb->clock_limits[i].fabricclk_mhz > (max_clocks.fabricClockInKhz / 1000)) && max_clocks.fabricClockInKhz != 0) bb->clock_limits[i].fabricclk_mhz = (max_clocks.fabricClockInKhz / 1000); @@ -2674,6 +3075,10 @@ static void update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_ num_calculated_states++; } + calculated_states[num_calculated_states - 1].socclk_mhz = max_clocks->socClockInKhz / 1000; + calculated_states[num_calculated_states - 1].fabricclk_mhz = max_clocks->socClockInKhz / 1000; + calculated_states[num_calculated_states - 1].dcfclk_mhz = max_clocks->dcfClockInKhz / 1000; + memcpy(bb->clock_limits, calculated_states, sizeof(bb->clock_limits)); bb->num_states = num_calculated_states; @@ -2711,6 +3116,27 @@ static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_s kernel_fpu_end(); } +static struct _vcs_dpi_soc_bounding_box_st *get_asic_rev_soc_bb( + uint32_t hw_internal_rev) +{ + if (ASICREV_IS_NAVI12_P(hw_internal_rev)) + return &dcn2_0_nv12_soc; + + return &dcn2_0_soc; +} + +static struct _vcs_dpi_ip_params_st *get_asic_rev_ip_params( + uint32_t hw_internal_rev) +{ + /* NV12 and NV10 */ + return &dcn2_0_ip; +} + +static enum dml_project get_dml_project_version(uint32_t hw_internal_rev) +{ + return DML_PROJECT_NAVI10v2; +} + #define fixed16_to_double(x) (((double) x) / ((double) (1 << 16))) #define fixed16_to_double_to_cpu(x) fixed16_to_double(le32_to_cpu(x)) @@ -2718,6 +3144,11 @@ static bool init_soc_bounding_box(struct dc *dc, struct dcn20_resource_pool *pool) { const struct gpu_info_soc_bounding_box_v1_0 *bb = dc->soc_bounding_box; + struct _vcs_dpi_soc_bounding_box_st *loaded_bb = + get_asic_rev_soc_bb(dc->ctx->asic_id.hw_internal_rev); + struct _vcs_dpi_ip_params_st *loaded_ip = + get_asic_rev_ip_params(dc->ctx->asic_id.hw_internal_rev); + DC_LOGGER_INIT(dc->ctx->logger); if (!bb && !SOC_BOUNDING_BOX_VALID) { @@ -2728,103 +3159,103 @@ static bool init_soc_bounding_box(struct dc *dc, if (bb && !SOC_BOUNDING_BOX_VALID) { int i; - dcn2_0_soc.sr_exit_time_us = + dcn2_0_nv12_soc.sr_exit_time_us = fixed16_to_double_to_cpu(bb->sr_exit_time_us); - dcn2_0_soc.sr_enter_plus_exit_time_us = + dcn2_0_nv12_soc.sr_enter_plus_exit_time_us = fixed16_to_double_to_cpu(bb->sr_enter_plus_exit_time_us); - dcn2_0_soc.urgent_latency_us = + dcn2_0_nv12_soc.urgent_latency_us = fixed16_to_double_to_cpu(bb->urgent_latency_us); - dcn2_0_soc.urgent_latency_pixel_data_only_us = + dcn2_0_nv12_soc.urgent_latency_pixel_data_only_us = fixed16_to_double_to_cpu(bb->urgent_latency_pixel_data_only_us); - dcn2_0_soc.urgent_latency_pixel_mixed_with_vm_data_us = + dcn2_0_nv12_soc.urgent_latency_pixel_mixed_with_vm_data_us = fixed16_to_double_to_cpu(bb->urgent_latency_pixel_mixed_with_vm_data_us); - dcn2_0_soc.urgent_latency_vm_data_only_us = + dcn2_0_nv12_soc.urgent_latency_vm_data_only_us = fixed16_to_double_to_cpu(bb->urgent_latency_vm_data_only_us); - dcn2_0_soc.urgent_out_of_order_return_per_channel_pixel_only_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_pixel_only_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_pixel_only_bytes); - dcn2_0_soc.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_pixel_and_vm_bytes); - dcn2_0_soc.urgent_out_of_order_return_per_channel_vm_only_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_vm_only_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_vm_only_bytes); - dcn2_0_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = + dcn2_0_nv12_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = fixed16_to_double_to_cpu(bb->pct_ideal_dram_sdp_bw_after_urgent_pixel_only); - dcn2_0_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = + dcn2_0_nv12_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = fixed16_to_double_to_cpu(bb->pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm); - dcn2_0_soc.pct_ideal_dram_sdp_bw_after_urgent_vm_only = + dcn2_0_nv12_soc.pct_ideal_dram_sdp_bw_after_urgent_vm_only = fixed16_to_double_to_cpu(bb->pct_ideal_dram_sdp_bw_after_urgent_vm_only); - dcn2_0_soc.max_avg_sdp_bw_use_normal_percent = + dcn2_0_nv12_soc.max_avg_sdp_bw_use_normal_percent = fixed16_to_double_to_cpu(bb->max_avg_sdp_bw_use_normal_percent); - dcn2_0_soc.max_avg_dram_bw_use_normal_percent = + dcn2_0_nv12_soc.max_avg_dram_bw_use_normal_percent = fixed16_to_double_to_cpu(bb->max_avg_dram_bw_use_normal_percent); - dcn2_0_soc.writeback_latency_us = + dcn2_0_nv12_soc.writeback_latency_us = fixed16_to_double_to_cpu(bb->writeback_latency_us); - dcn2_0_soc.ideal_dram_bw_after_urgent_percent = + dcn2_0_nv12_soc.ideal_dram_bw_after_urgent_percent = fixed16_to_double_to_cpu(bb->ideal_dram_bw_after_urgent_percent); - dcn2_0_soc.max_request_size_bytes = + dcn2_0_nv12_soc.max_request_size_bytes = le32_to_cpu(bb->max_request_size_bytes); - dcn2_0_soc.dram_channel_width_bytes = + dcn2_0_nv12_soc.dram_channel_width_bytes = le32_to_cpu(bb->dram_channel_width_bytes); - dcn2_0_soc.fabric_datapath_to_dcn_data_return_bytes = + dcn2_0_nv12_soc.fabric_datapath_to_dcn_data_return_bytes = le32_to_cpu(bb->fabric_datapath_to_dcn_data_return_bytes); - dcn2_0_soc.dcn_downspread_percent = + dcn2_0_nv12_soc.dcn_downspread_percent = fixed16_to_double_to_cpu(bb->dcn_downspread_percent); - dcn2_0_soc.downspread_percent = + dcn2_0_nv12_soc.downspread_percent = fixed16_to_double_to_cpu(bb->downspread_percent); - dcn2_0_soc.dram_page_open_time_ns = + dcn2_0_nv12_soc.dram_page_open_time_ns = fixed16_to_double_to_cpu(bb->dram_page_open_time_ns); - dcn2_0_soc.dram_rw_turnaround_time_ns = + dcn2_0_nv12_soc.dram_rw_turnaround_time_ns = fixed16_to_double_to_cpu(bb->dram_rw_turnaround_time_ns); - dcn2_0_soc.dram_return_buffer_per_channel_bytes = + dcn2_0_nv12_soc.dram_return_buffer_per_channel_bytes = le32_to_cpu(bb->dram_return_buffer_per_channel_bytes); - dcn2_0_soc.round_trip_ping_latency_dcfclk_cycles = + dcn2_0_nv12_soc.round_trip_ping_latency_dcfclk_cycles = le32_to_cpu(bb->round_trip_ping_latency_dcfclk_cycles); - dcn2_0_soc.urgent_out_of_order_return_per_channel_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_bytes); - dcn2_0_soc.channel_interleave_bytes = + dcn2_0_nv12_soc.channel_interleave_bytes = le32_to_cpu(bb->channel_interleave_bytes); - dcn2_0_soc.num_banks = + dcn2_0_nv12_soc.num_banks = le32_to_cpu(bb->num_banks); - dcn2_0_soc.num_chans = + dcn2_0_nv12_soc.num_chans = le32_to_cpu(bb->num_chans); - dcn2_0_soc.vmm_page_size_bytes = + dcn2_0_nv12_soc.vmm_page_size_bytes = le32_to_cpu(bb->vmm_page_size_bytes); - dcn2_0_soc.dram_clock_change_latency_us = + dcn2_0_nv12_soc.dram_clock_change_latency_us = fixed16_to_double_to_cpu(bb->dram_clock_change_latency_us); // HACK!! Lower uclock latency switch time so we don't switch - dcn2_0_soc.dram_clock_change_latency_us = 10; - dcn2_0_soc.writeback_dram_clock_change_latency_us = + dcn2_0_nv12_soc.dram_clock_change_latency_us = 10; + dcn2_0_nv12_soc.writeback_dram_clock_change_latency_us = fixed16_to_double_to_cpu(bb->writeback_dram_clock_change_latency_us); - dcn2_0_soc.return_bus_width_bytes = + dcn2_0_nv12_soc.return_bus_width_bytes = le32_to_cpu(bb->return_bus_width_bytes); - dcn2_0_soc.dispclk_dppclk_vco_speed_mhz = + dcn2_0_nv12_soc.dispclk_dppclk_vco_speed_mhz = le32_to_cpu(bb->dispclk_dppclk_vco_speed_mhz); - dcn2_0_soc.xfc_bus_transport_time_us = + dcn2_0_nv12_soc.xfc_bus_transport_time_us = le32_to_cpu(bb->xfc_bus_transport_time_us); - dcn2_0_soc.xfc_xbuf_latency_tolerance_us = + dcn2_0_nv12_soc.xfc_xbuf_latency_tolerance_us = le32_to_cpu(bb->xfc_xbuf_latency_tolerance_us); - dcn2_0_soc.use_urgent_burst_bw = + dcn2_0_nv12_soc.use_urgent_burst_bw = le32_to_cpu(bb->use_urgent_burst_bw); - dcn2_0_soc.num_states = + dcn2_0_nv12_soc.num_states = le32_to_cpu(bb->num_states); - for (i = 0; i < dcn2_0_soc.num_states; i++) { - dcn2_0_soc.clock_limits[i].state = + for (i = 0; i < dcn2_0_nv12_soc.num_states; i++) { + dcn2_0_nv12_soc.clock_limits[i].state = le32_to_cpu(bb->clock_limits[i].state); - dcn2_0_soc.clock_limits[i].dcfclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dcfclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dcfclk_mhz); - dcn2_0_soc.clock_limits[i].fabricclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].fabricclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].fabricclk_mhz); - dcn2_0_soc.clock_limits[i].dispclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dispclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dispclk_mhz); - dcn2_0_soc.clock_limits[i].dppclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dppclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dppclk_mhz); - dcn2_0_soc.clock_limits[i].phyclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].phyclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].phyclk_mhz); - dcn2_0_soc.clock_limits[i].socclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].socclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].socclk_mhz); - dcn2_0_soc.clock_limits[i].dscclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dscclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dscclk_mhz); - dcn2_0_soc.clock_limits[i].dram_speed_mts = + dcn2_0_nv12_soc.clock_limits[i].dram_speed_mts = fixed16_to_double_to_cpu(bb->clock_limits[i].dram_speed_mts); } } @@ -2833,7 +3264,6 @@ static bool init_soc_bounding_box(struct dc *dc, struct pp_smu_nv_clock_table max_clocks = {0}; unsigned int uclk_states[8] = {0}; unsigned int num_states = 0; - int i; enum pp_smu_status status; bool clock_limits_available = false; bool uclk_states_available = false; @@ -2855,19 +3285,15 @@ static bool init_soc_bounding_box(struct dc *dc, clock_limits_available = (status == PP_SMU_RESULT_OK); } - // HACK: Use the max uclk_states value for all elements. - for (i = 0; i < num_states; i++) - uclk_states[i] = uclk_states[num_states - 1]; - if (clock_limits_available && uclk_states_available && num_states) - update_bounding_box(dc, &dcn2_0_soc, &max_clocks, uclk_states, num_states); + update_bounding_box(dc, loaded_bb, &max_clocks, uclk_states, num_states); else if (clock_limits_available) - cap_soc_clocks(&dcn2_0_soc, max_clocks); + cap_soc_clocks(loaded_bb, max_clocks); } - dcn2_0_ip.max_num_otg = pool->base.res_cap->num_timing_generator; - dcn2_0_ip.max_num_dpp = pool->base.pipe_count; - patch_bounding_box(dc, &dcn2_0_soc); + loaded_ip->max_num_otg = pool->base.res_cap->num_timing_generator; + loaded_ip->max_num_dpp = pool->base.pipe_count; + patch_bounding_box(dc, loaded_bb); return true; } @@ -2880,19 +3306,30 @@ static bool construct( int i; struct dc_context *ctx = dc->ctx; struct irq_service_init_data init_data; + struct _vcs_dpi_soc_bounding_box_st *loaded_bb = + get_asic_rev_soc_bb(ctx->asic_id.hw_internal_rev); + struct _vcs_dpi_ip_params_st *loaded_ip = + get_asic_rev_ip_params(ctx->asic_id.hw_internal_rev); + enum dml_project dml_project_version = + get_dml_project_version(ctx->asic_id.hw_internal_rev); ctx->dc_bios->regs = &bios_regs; - - pool->base.res_cap = &res_cap_nv10; pool->base.funcs = &dcn20_res_pool_funcs; + if (ASICREV_IS_NAVI14_M(ctx->asic_id.hw_internal_rev)) { + pool->base.res_cap = &res_cap_nv14; + pool->base.pipe_count = 5; + pool->base.mpcc_count = 5; + } else { + pool->base.res_cap = &res_cap_nv10; + pool->base.pipe_count = 6; + pool->base.mpcc_count = 6; + } /************************************************* * Resource + asic cap harcoding * *************************************************/ pool->base.underlay_pipe_index = NO_UNDERLAY_PIPE; - pool->base.pipe_count = 6; - pool->base.mpcc_count = 6; dc->caps.max_downscale_ratio = 200; dc->caps.i2c_speed_in_khz = 100; dc->caps.max_cursor_size = 256; @@ -2998,7 +3435,7 @@ static bool construct( goto create_fail; } - dml_init_instance(&dc->dml, &dcn2_0_soc, &dcn2_0_ip, DML_PROJECT_NAVI10); + dml_init_instance(&dc->dml, loaded_bb, loaded_ip, dml_project_version); if (!dc->debug.disable_pplib_wm_range) { struct pp_smu_wm_range_sets ranges = {0}; @@ -3006,7 +3443,7 @@ static bool construct( ranges.num_reader_wm_sets = 0; - if (dcn2_0_soc.num_states == 1) { + if (loaded_bb->num_states == 1) { ranges.reader_wm_sets[0].wm_inst = i; ranges.reader_wm_sets[0].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; ranges.reader_wm_sets[0].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; @@ -3014,13 +3451,13 @@ static bool construct( ranges.reader_wm_sets[0].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; ranges.num_reader_wm_sets = 1; - } else if (dcn2_0_soc.num_states > 1) { - for (i = 0; i < 4 && i < dcn2_0_soc.num_states; i++) { + } else if (loaded_bb->num_states > 1) { + for (i = 0; i < 4 && i < loaded_bb->num_states; i++) { ranges.reader_wm_sets[i].wm_inst = i; ranges.reader_wm_sets[i].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; ranges.reader_wm_sets[i].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; - ranges.reader_wm_sets[i].min_fill_clk_mhz = (i > 0) ? (dcn2_0_soc.clock_limits[i - 1].dram_speed_mts / 16) + 1 : 0; - ranges.reader_wm_sets[i].max_fill_clk_mhz = dcn2_0_soc.clock_limits[i].dram_speed_mts / 16; + ranges.reader_wm_sets[i].min_fill_clk_mhz = (i > 0) ? (loaded_bb->clock_limits[i - 1].dram_speed_mts / 16) + 1 : 0; + ranges.reader_wm_sets[i].max_fill_clk_mhz = loaded_bb->clock_limits[i].dram_speed_mts / 16; ranges.num_reader_wm_sets = i + 1; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h index b5a75289f444..44f95aa0d61e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h @@ -116,6 +116,18 @@ void dcn20_set_mcif_arb_params( display_e2e_pipe_params_st *pipes, int pipe_cnt); bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate); +bool dcn20_fast_validate_bw( + struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *pipe_cnt_out, + int *pipe_split_from, + int *vlevel_out); +void dcn20_calculate_dlg_params( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel); enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state *context, struct dc_stream_state *stream); enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c index f5bcffc426b8..5ab9d6240498 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c @@ -207,9 +207,8 @@ static void enc2_stream_encoder_stop_hdmi_info_packets( #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - /* Update GSP7 SDP 128 byte long */ -static void enc2_send_gsp7_128_info_packet( +static void enc2_update_gsp7_128_info_packet( struct dcn10_stream_encoder *enc1, const struct dc_info_packet_128 *info_packet) { @@ -277,18 +276,9 @@ static void enc2_send_gsp7_128_info_packet( static void enc2_dp_set_dsc_config(struct stream_encoder *enc, enum optc_dsc_mode dsc_mode, uint32_t dsc_bytes_per_pixel, - uint32_t dsc_slice_width, - uint8_t *dsc_packed_pps) + uint32_t dsc_slice_width) { struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); - uint32_t dsc_value = 0; - - dsc_value = REG_READ(DP_DSC_CNTL); - - /* dsc disable skip */ - if ((dsc_value & 0x3) == 0x0) - return; - REG_UPDATE_2(DP_DSC_CNTL, DP_DSC_MODE, dsc_mode, @@ -296,8 +286,16 @@ static void enc2_dp_set_dsc_config(struct stream_encoder *enc, REG_SET(DP_DSC_BYTES_PER_PIXEL, 0, DP_DSC_BYTES_PER_PIXEL, dsc_bytes_per_pixel); +} + + +static void enc2_dp_set_dsc_pps_info_packet(struct stream_encoder *enc, + bool enable, + uint8_t *dsc_packed_pps) +{ + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); - if (dsc_mode != OPTC_DSC_DISABLED) { + if (enable) { struct dc_info_packet_128 pps_sdp; ASSERT(dsc_packed_pps); @@ -309,7 +307,7 @@ static void enc2_dp_set_dsc_config(struct stream_encoder *enc, pps_sdp.hb2 = 127; pps_sdp.hb3 = 0; memcpy(&pps_sdp.sb[0], dsc_packed_pps, sizeof(pps_sdp.sb)); - enc2_send_gsp7_128_info_packet(enc1, &pps_sdp); + enc2_update_gsp7_128_info_packet(enc1, &pps_sdp); /* Enable Generic Stream Packet 7 (GSP) transmission */ //REG_UPDATE(DP_SEC_CNTL, @@ -340,9 +338,8 @@ static void enc2_dp_set_dsc_config(struct stream_encoder *enc, REG_UPDATE(DP_SEC_CNTL2, DP_SEC_GSP7_PPS, 0); } } -#endif -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + /* this function read dsc related register fields to be logged later in dcn10_log_hw_state * into a dcn_dsc_state struct. */ @@ -373,7 +370,7 @@ static void enc2_read_state(struct stream_encoder *enc, struct enc_state *s) * * Ensure the OTG master update lock is set when changing DME configuration. */ -static void enc2_set_dynamic_metadata(struct stream_encoder *enc, +void enc2_set_dynamic_metadata(struct stream_encoder *enc, bool enable_dme, uint32_t hubp_requestor_id, enum dynamic_metadata_mode dmdata_mode) @@ -463,7 +460,7 @@ void enc2_stream_encoder_dp_unblank( uint64_t m_vid_l = n_vid; /* YCbCr 4:2:0 : Computed VID_M will be 2X the input rate */ - if (is_two_pixels_per_containter(¶m->timing) || param->odm) { + if (is_two_pixels_per_containter(¶m->timing) || param->opp_cnt > 1) { /*this logic should be the same in get_pixel_clock_parameters() */ n_multiply = 1; } @@ -580,14 +577,14 @@ static const struct stream_encoder_funcs dcn20_str_enc_funcs = { .setup_stereo_sync = enc1_setup_stereo_sync, .set_avmute = enc1_stream_encoder_set_avmute, .dig_connect_to_otg = enc1_dig_connect_to_otg, + .dig_source_otg = enc1_dig_source_otg, #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .enc_read_state = enc2_read_state, -#endif - -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .dp_set_dsc_config = enc2_dp_set_dsc_config, + .dp_set_dsc_pps_info_packet = enc2_dp_set_dsc_pps_info_packet, #endif .set_dynamic_metadata = enc2_set_dynamic_metadata, + .hdmi_reset_stream_attribute = enc1_reset_hdmi_stream_attribute, }; void dcn20_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h index 6d40e8c9b78f..3f94a9f13c4a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h @@ -104,4 +104,9 @@ void enc2_stream_encoder_dp_unblank( struct stream_encoder *enc, const struct encoder_unblank_param *param); +void enc2_set_dynamic_metadata(struct stream_encoder *enc, + bool enable_dme, + uint32_t hubp_requestor_id, + enum dynamic_metadata_mode dmdata_mode); + #endif /* __DC_STREAM_ENCODER_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile new file mode 100644 index 000000000000..8cd9de8b1a7a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile @@ -0,0 +1,10 @@ +# +# Makefile for DCN21. + +DCN21 = dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o + +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse -mpreferred-stack-boundary=4 + +AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_DCN21) diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c new file mode 100644 index 000000000000..d1266741763b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c @@ -0,0 +1,595 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "dm_services.h" +#include "dcn20/dcn20_hubbub.h" +#include "dcn21_hubbub.h" +#include "reg_helper.h" + +#define REG(reg)\ + hubbub1->regs->reg +#define DC_LOGGER \ + hubbub1->base.ctx->logger +#define CTX \ + hubbub1->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + hubbub1->shifts->field_name, hubbub1->masks->field_name + +#define REG(reg)\ + hubbub1->regs->reg + +#define CTX \ + hubbub1->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + hubbub1->shifts->field_name, hubbub1->masks->field_name + +#ifdef NUM_VMID +#undef NUM_VMID +#endif +#define NUM_VMID 1 + +static uint32_t convert_and_clamp( + uint32_t wm_ns, + uint32_t refclk_mhz, + uint32_t clamp_value) +{ + uint32_t ret_val = 0; + ret_val = wm_ns * refclk_mhz; + ret_val /= 1000; + + if (ret_val > clamp_value) + ret_val = clamp_value; + + return ret_val; +} + +void dcn21_dchvm_init(struct hubbub *hubbub) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + + //Init DCHVM block + REG_UPDATE(DCHVM_CTRL0, HOSTVM_INIT_REQ, 1); + + //Poll until RIOMMU_ACTIVE = 1 + //TODO: Figure out interval us and retry count + REG_WAIT(DCHVM_RIOMMU_STAT0, RIOMMU_ACTIVE, 1, 5, 100); + + //Reflect the power status of DCHUBBUB + REG_UPDATE(DCHVM_RIOMMU_CTRL0, HOSTVM_POWERSTATUS, 1); + + //Start rIOMMU prefetching + REG_UPDATE(DCHVM_RIOMMU_CTRL0, HOSTVM_PREFETCH_REQ, 1); + + // Enable dynamic clock gating + REG_UPDATE_4(DCHVM_CLK_CTRL, + HVM_DISPCLK_R_GATE_DIS, 0, + HVM_DISPCLK_G_GATE_DIS, 0, + HVM_DCFCLK_R_GATE_DIS, 0, + HVM_DCFCLK_G_GATE_DIS, 0); + + //Poll until HOSTVM_PREFETCH_DONE = 1 + //TODO: Figure out interval us and retry count + REG_WAIT(DCHVM_RIOMMU_STAT0, HOSTVM_PREFETCH_DONE, 1, 5, 100); +} + +static int hubbub21_init_dchub(struct hubbub *hubbub, + struct dcn_hubbub_phys_addr_config *pa_config) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + + REG_SET(DCN_VM_FB_LOCATION_BASE, 0, + FB_BASE, pa_config->system_aperture.fb_base); + REG_SET(DCN_VM_FB_LOCATION_TOP, 0, + FB_TOP, pa_config->system_aperture.fb_top); + REG_SET(DCN_VM_FB_OFFSET, 0, + FB_OFFSET, pa_config->system_aperture.fb_offset); + REG_SET(DCN_VM_AGP_BOT, 0, + AGP_BOT, pa_config->system_aperture.agp_bot); + REG_SET(DCN_VM_AGP_TOP, 0, + AGP_TOP, pa_config->system_aperture.agp_top); + REG_SET(DCN_VM_AGP_BASE, 0, + AGP_BASE, pa_config->system_aperture.agp_base); + + dcn21_dchvm_init(hubbub); + + return NUM_VMID; +} + +static void hubbub21_program_urgent_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t prog_wm_value; + + /* Repeat for water mark set A, B, C and D. */ + /* clock state A */ + if (safe_to_lower || watermarks->a.urgent_ns > hubbub1->watermarks.a.urgent_ns) { + hubbub1->watermarks.a.urgent_ns = watermarks->a.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->a.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_A, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->a.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, watermarks->a.frac_urg_bw_nom); + } + + /* clock state B */ + if (safe_to_lower || watermarks->b.urgent_ns > hubbub1->watermarks.b.urgent_ns) { + hubbub1->watermarks.b.urgent_ns = watermarks->b.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->b.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_B, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->b.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, watermarks->a.frac_urg_bw_nom); + } + + /* clock state C */ + if (safe_to_lower || watermarks->c.urgent_ns > hubbub1->watermarks.c.urgent_ns) { + hubbub1->watermarks.c.urgent_ns = watermarks->c.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->c.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_C, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->c.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, watermarks->a.frac_urg_bw_nom); + } + + /* clock state D */ + if (safe_to_lower || watermarks->d.urgent_ns > hubbub1->watermarks.d.urgent_ns) { + hubbub1->watermarks.d.urgent_ns = watermarks->d.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->d.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_D, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->d.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, watermarks->a.frac_urg_bw_nom); + } +} + +static void hubbub21_program_stutter_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t prog_wm_value; + + /* clock state A */ + if (safe_to_lower || watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->a.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns = + watermarks->a.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->a.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->a.cstate_pstate.cstate_exit_ns, prog_wm_value); + } + + /* clock state B */ + if (safe_to_lower || watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->b.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns = + watermarks->b.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->b.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->b.cstate_pstate.cstate_exit_ns, prog_wm_value); + } + + /* clock state C */ + if (safe_to_lower || watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->c.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns = + watermarks->c.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->c.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->c.cstate_pstate.cstate_exit_ns, prog_wm_value); + } + + /* clock state D */ + if (safe_to_lower || watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->d.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns = + watermarks->d.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->d.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->d.cstate_pstate.cstate_exit_ns, prog_wm_value); + } +} + +static void hubbub21_program_pstate_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t prog_wm_value; + + /* clock state A */ + if (safe_to_lower || watermarks->a.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.a.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.a.cstate_pstate.pstate_change_ns = + watermarks->a.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->a.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->a.cstate_pstate.pstate_change_ns, prog_wm_value); + } + + /* clock state B */ + if (safe_to_lower || watermarks->b.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.b.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.b.cstate_pstate.pstate_change_ns = + watermarks->b.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->b.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->b.cstate_pstate.pstate_change_ns, prog_wm_value); + } + + /* clock state C */ + if (safe_to_lower || watermarks->c.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.c.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.c.cstate_pstate.pstate_change_ns = + watermarks->c.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->c.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->c.cstate_pstate.pstate_change_ns, prog_wm_value); + } + + /* clock state D */ + if (safe_to_lower || watermarks->d.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.d.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.d.cstate_pstate.pstate_change_ns = + watermarks->d.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->d.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->d.cstate_pstate.pstate_change_ns, prog_wm_value); + } +} + +void hubbub21_program_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + + hubbub21_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + hubbub21_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + hubbub21_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + + /* + * The DCHub arbiter has a mechanism to dynamically rate limit the DCHub request stream to the fabric. + * If the memory controller is fully utilized and the DCHub requestors are + * well ahead of their amortized schedule, then it is safe to prevent the next winner + * from being committed and sent to the fabric. + * The utilization of the memory controller is approximated by ensuring that + * the number of outstanding requests is greater than a threshold specified + * by the ARB_MIN_REQ_OUTSTANDING. To determine that the DCHub requestors are well ahead of the amortized schedule, + * the slack of the next winner is compared with the ARB_SAT_LEVEL in DLG RefClk cycles. + * + * TODO: Revisit request limit after figure out right number. request limit for Renoir isn't decided yet, set maximum value (0x1FF) + * to turn off it for now. + */ + REG_SET(DCHUBBUB_ARB_SAT_LEVEL, 0, + DCHUBBUB_ARB_SAT_LEVEL, 60 * refclk_mhz); + REG_UPDATE_2(DCHUBBUB_ARB_DF_REQ_OUTSTAND, + DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 0x1FF, + DCHUBBUB_ARB_MIN_REQ_OUTSTAND_COMMIT_THRESHOLD, 0xA); + REG_UPDATE(DCHUBBUB_ARB_HOSTVM_CNTL, + DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD, 0xF); + + hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); +} + +void hubbub21_wm_read_state(struct hubbub *hubbub, + struct dcn_hubbub_wm *wm) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + struct dcn_hubbub_wm_set *s; + + memset(wm, 0, sizeof(struct dcn_hubbub_wm)); + + s = &wm->sets[0]; + s->wm_set = 0; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, &s->dram_clk_chanage); + + s = &wm->sets[1]; + s->wm_set = 1; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, &s->dram_clk_chanage); + + s = &wm->sets[2]; + s->wm_set = 2; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, &s->dram_clk_chanage); + + s = &wm->sets[3]; + s->wm_set = 3; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, &s->dram_clk_chanage); +} + + +static const struct hubbub_funcs hubbub21_funcs = { + .update_dchub = hubbub2_update_dchub, + .init_dchub_sys_ctx = hubbub21_init_dchub, + .init_vm_ctx = NULL, + .dcc_support_swizzle = hubbub2_dcc_support_swizzle, + .dcc_support_pixel_format = hubbub2_dcc_support_pixel_format, + .get_dcc_compression_cap = hubbub2_get_dcc_compression_cap, + .wm_read_state = hubbub21_wm_read_state, + .get_dchub_ref_freq = hubbub2_get_dchub_ref_freq, + .program_watermarks = hubbub21_program_watermarks, +}; + +void hubbub21_construct(struct dcn20_hubbub *hubbub, + struct dc_context *ctx, + const struct dcn_hubbub_registers *hubbub_regs, + const struct dcn_hubbub_shift *hubbub_shift, + const struct dcn_hubbub_mask *hubbub_mask) +{ + hubbub->base.ctx = ctx; + + hubbub->base.funcs = &hubbub21_funcs; + + hubbub->regs = hubbub_regs; + hubbub->shifts = hubbub_shift; + hubbub->masks = hubbub_mask; + + hubbub->debug_test_index_pstate = 0xB; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h new file mode 100644 index 000000000000..6ff3cdb89178 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h @@ -0,0 +1,132 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef DAL_DC_DCN21_DCN21_HUBBUB_H_ +#define DAL_DC_DCN21_DCN21_HUBBUB_H_ + +#include "dcn20/dcn20_hubbub.h" + +#define HUBBUB_HVM_REG_LIST() \ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_A),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_B),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_C),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_D),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D),\ + SR(DCHUBBUB_ARB_HOSTVM_CNTL), \ + SR(DCHVM_CTRL0), \ + SR(DCHVM_MEM_CTRL), \ + SR(DCHVM_CLK_CTRL), \ + SR(DCHVM_RIOMMU_CTRL0), \ + SR(DCHVM_RIOMMU_STAT0) + +#define HUBBUB_REG_LIST_DCN21()\ + HUBBUB_REG_LIST_DCN_COMMON(), \ + HUBBUB_SR_WATERMARK_REG_LIST(), \ + HUBBUB_HVM_REG_LIST(), \ + SR(DCHUBBUB_CRC_CTRL), \ + SR(DCN_VM_FB_LOCATION_BASE),\ + SR(DCN_VM_FB_LOCATION_TOP),\ + SR(DCN_VM_FB_OFFSET),\ + SR(DCN_VM_AGP_BOT),\ + SR(DCN_VM_AGP_TOP),\ + SR(DCN_VM_AGP_BASE) + +#define HUBBUB_MASK_SH_LIST_HVM(mask_sh) \ + HUBBUB_SF(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND_COMMIT_THRESHOLD, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_HOSTVM_CNTL, DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD, mask_sh), \ + HUBBUB_SF(DCHVM_CTRL0, HOSTVM_INIT_REQ, mask_sh), \ + HUBBUB_SF(DCHVM_MEM_CTRL, HVM_GPUVMRET_PWR_REQ_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_MEM_CTRL, HVM_GPUVMRET_FORCE_REQ, mask_sh), \ + HUBBUB_SF(DCHVM_MEM_CTRL, HVM_GPUVMRET_POWER_STATUS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DISPCLK_R_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DISPCLK_G_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DCFCLK_R_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DCFCLK_G_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, TR_REQ_REQCLKREQ_MODE, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, TW_RSP_COMPCLKREQ_MODE, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_CTRL0, HOSTVM_PREFETCH_REQ, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_CTRL0, HOSTVM_POWERSTATUS, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_STAT0, RIOMMU_ACTIVE, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_STAT0, HOSTVM_PREFETCH_DONE, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, mask_sh) + +#define HUBBUB_MASK_SH_LIST_DCN21(mask_sh)\ + HUBBUB_MASK_SH_LIST_HVM(mask_sh),\ + HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh), \ + HUBBUB_MASK_SH_LIST_STUTTER(mask_sh), \ + HUBBUB_SF(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, mask_sh), \ + HUBBUB_SF(DCN_VM_FB_LOCATION_BASE, FB_BASE, mask_sh), \ + HUBBUB_SF(DCN_VM_FB_LOCATION_TOP, FB_TOP, mask_sh), \ + HUBBUB_SF(DCN_VM_FB_OFFSET, FB_OFFSET, mask_sh), \ + HUBBUB_SF(DCN_VM_AGP_BOT, AGP_BOT, mask_sh), \ + HUBBUB_SF(DCN_VM_AGP_TOP, AGP_TOP, mask_sh), \ + HUBBUB_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh) + +void dcn21_dchvm_init(struct hubbub *hubbub); +void hubbub21_program_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower); + +void hubbub21_wm_read_state(struct hubbub *hubbub, + struct dcn_hubbub_wm *wm); + +void hubbub21_construct(struct dcn20_hubbub *hubbub, + struct dc_context *ctx, + const struct dcn_hubbub_registers *hubbub_regs, + const struct dcn_hubbub_shift *hubbub_shift, + const struct dcn_hubbub_mask *hubbub_mask); + +#endif /* DAL_DC_DCN21_DCN21_HUBBUB_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c new file mode 100644 index 000000000000..a00af513aa2b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c @@ -0,0 +1,244 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "dcn21_hubp.h" + +#include "dm_services.h" +#include "reg_helper.h" + +#define REG(reg)\ + hubp21->hubp_regs->reg + +#define CTX \ + hubp21->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + hubp21->hubp_shift->field_name, hubp21->hubp_mask->field_name + +/* + * In DCN2.1, the non-double buffered version of the following 4 DLG registers are used in RTL. + * As a result, if S/W updates any of these registers during a mode change, + * the current frame before the mode change will use the new value right away + * and can lead to generating incorrect request deadlines and incorrect TTU/QoS behavior. + * + * REFCYC_PER_VM_GROUP_FLIP[22:0] + * REFCYC_PER_VM_GROUP_VBLANK[22:0] + * REFCYC_PER_VM_REQ_FLIP[22:0] + * REFCYC_PER_VM_REQ_VBLANK[22:0] + * + * REFCYC_PER_VM_*_FLIP affects the deadline of the VM requests generated + * when flipping to a new surface + * + * REFCYC_PER_VM_*_VBLANK affects the deadline of the VM requests generated + * during prefetch period of a frame. The prefetch starts at a pre-determined + * number of lines before the display active per frame + * + * DCN may underflow due to incorrectly programming these registers + * during VM stage of prefetch/iflip. First lines of display active + * or a sub-region of active using a new surface will be corrupted + * until the VM data returns at flip/mode change transitions + * + * Work around: + * workaround is always opt to use the more aggressive settings. + * On any mode switch, if the new reg values are smaller than the current values, + * then update the regs with the new values. + * + * Link to the ticket: http://ontrack-internal.amd.com/browse/DEDCN21-142 + * + */ +void apply_DEDCN21_142_wa_for_hostvm_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + uint32_t cur_value; + + REG_GET(VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_group_vblank) + REG_SET(VBLANK_PARAMETERS_5, 0, + REFCYC_PER_VM_GROUP_VBLANK, dlg_attr->refcyc_per_vm_group_vblank); + + REG_GET(VBLANK_PARAMETERS_6, + REFCYC_PER_VM_REQ_VBLANK, + &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_req_vblank) + REG_SET(VBLANK_PARAMETERS_6, 0, + REFCYC_PER_VM_REQ_VBLANK, dlg_attr->refcyc_per_vm_req_vblank); + + REG_GET(FLIP_PARAMETERS_3, REFCYC_PER_VM_GROUP_FLIP, &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_group_flip) + REG_SET(FLIP_PARAMETERS_3, 0, + REFCYC_PER_VM_GROUP_FLIP, dlg_attr->refcyc_per_vm_group_flip); + + REG_GET(FLIP_PARAMETERS_4, REFCYC_PER_VM_REQ_FLIP, &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_req_flip) + REG_SET(FLIP_PARAMETERS_4, 0, + REFCYC_PER_VM_REQ_FLIP, dlg_attr->refcyc_per_vm_req_flip); + + REG_SET(FLIP_PARAMETERS_5, 0, + REFCYC_PER_PTE_GROUP_FLIP_C, dlg_attr->refcyc_per_pte_group_flip_c); + REG_SET(FLIP_PARAMETERS_6, 0, + REFCYC_PER_META_CHUNK_FLIP_C, dlg_attr->refcyc_per_meta_chunk_flip_c); +} + +void hubp21_program_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *ttu_attr) +{ + hubp2_program_deadline(hubp, dlg_attr, ttu_attr); + + apply_DEDCN21_142_wa_for_hostvm_deadline(hubp, dlg_attr); +} + +void hubp21_program_requestor( + struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + + REG_UPDATE(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, rq_regs->plane1_base_address); + REG_SET_4(DCN_EXPANSION_MODE, 0, + DRQ_EXPANSION_MODE, rq_regs->drq_expansion_mode, + PRQ_EXPANSION_MODE, rq_regs->prq_expansion_mode, + MRQ_EXPANSION_MODE, rq_regs->mrq_expansion_mode, + CRQ_EXPANSION_MODE, rq_regs->crq_expansion_mode); + REG_SET_8(DCHUBP_REQ_SIZE_CONFIG, 0, + CHUNK_SIZE, rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, rq_regs->rq_regs_l.dpte_group_size, + VM_GROUP_SIZE, rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, rq_regs->rq_regs_l.pte_row_height_linear); + REG_SET_7(DCHUBP_REQ_SIZE_CONFIG_C, 0, + CHUNK_SIZE_C, rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, rq_regs->rq_regs_c.dpte_group_size, + SWATH_HEIGHT_C, rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, rq_regs->rq_regs_c.pte_row_height_linear); +} + +static void hubp21_setup( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *ttu_attr, + struct _vcs_dpi_display_rq_regs_st *rq_regs, + struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest) +{ + /* otg is locked when this func is called. Register are double buffered. + * disable the requestors is not needed + */ + + hubp2_vready_at_or_After_vsync(hubp, pipe_dest); + hubp21_program_requestor(hubp, rq_regs); + hubp21_program_deadline(hubp, dlg_attr, ttu_attr); + +} + +void hubp21_set_vm_system_aperture_settings(struct hubp *hubp, + struct vm_system_aperture_param *apt) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + + PHYSICAL_ADDRESS_LOC mc_vm_apt_default; + PHYSICAL_ADDRESS_LOC mc_vm_apt_low; + PHYSICAL_ADDRESS_LOC mc_vm_apt_high; + + // The format of default addr is 48:12 of the 48 bit addr + mc_vm_apt_default.quad_part = apt->sys_default.quad_part >> 12; + + // The format of high/low are 48:18 of the 48 bit addr + mc_vm_apt_low.quad_part = apt->sys_low.quad_part >> 18; + mc_vm_apt_high.quad_part = apt->sys_high.quad_part >> 18; + + REG_SET(DCN_VM_SYSTEM_APERTURE_LOW_ADDR, 0, + MC_VM_SYSTEM_APERTURE_LOW_ADDR, mc_vm_apt_low.quad_part); + + REG_SET(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, 0, + MC_VM_SYSTEM_APERTURE_HIGH_ADDR, mc_vm_apt_high.quad_part); + + REG_SET_2(DCN_VM_MX_L1_TLB_CNTL, 0, + ENABLE_L1_TLB, 1, + SYSTEM_ACCESS_MODE, 0x3); +} + +void hubp21_init(struct hubp *hubp) +{ + // DEDCN21-133: Inconsistent row starting line for flip between DPTE and Meta + // This is a chicken bit to enable the ECO fix. + + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + //hubp[i].HUBPREQ_DEBUG.HUBPREQ_DEBUG[26] = 1; + REG_WRITE(HUBPREQ_DEBUG, 1 << 26); +} +static struct hubp_funcs dcn21_hubp_funcs = { + .hubp_enable_tripleBuffer = hubp2_enable_triplebuffer, + .hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled, + .hubp_program_surface_flip_and_addr = hubp2_program_surface_flip_and_addr, + .hubp_program_surface_config = hubp2_program_surface_config, + .hubp_is_flip_pending = hubp1_is_flip_pending, + .hubp_setup = hubp21_setup, + .hubp_setup_interdependent = hubp2_setup_interdependent, + .hubp_set_vm_system_aperture_settings = hubp21_set_vm_system_aperture_settings, + .set_blank = hubp1_set_blank, + .dcc_control = hubp1_dcc_control, + .mem_program_viewport = min_set_viewport, + .set_cursor_attributes = hubp2_cursor_set_attributes, + .set_cursor_position = hubp1_cursor_set_position, + .hubp_clk_cntl = hubp1_clk_cntl, + .hubp_vtg_sel = hubp1_vtg_sel, + .dmdata_set_attributes = hubp2_dmdata_set_attributes, + .dmdata_load = hubp2_dmdata_load, + .dmdata_status_done = hubp2_dmdata_status_done, + .hubp_read_state = hubp1_read_state, + .hubp_clear_underflow = hubp1_clear_underflow, + .hubp_set_flip_control_surface_gsl = hubp2_set_flip_control_surface_gsl, + .hubp_init = hubp21_init, +}; + +bool hubp21_construct( + struct dcn21_hubp *hubp21, + struct dc_context *ctx, + uint32_t inst, + const struct dcn_hubp2_registers *hubp_regs, + const struct dcn_hubp2_shift *hubp_shift, + const struct dcn_hubp2_mask *hubp_mask) +{ + hubp21->base.funcs = &dcn21_hubp_funcs; + hubp21->base.ctx = ctx; + hubp21->hubp_regs = hubp_regs; + hubp21->hubp_shift = hubp_shift; + hubp21->hubp_mask = hubp_mask; + hubp21->base.inst = inst; + hubp21->base.opp_id = OPP_ID_INVALID; + hubp21->base.mpcc_id = 0xf; + + return true; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.h new file mode 100644 index 000000000000..aeda719a2a13 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.h @@ -0,0 +1,133 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef DAL_DC_DCN21_DCN21_HUBP_H_ +#define DAL_DC_DCN21_DCN21_HUBP_H_ + +#include "../dcn20/dcn20_hubp.h" +#include "../dcn10/dcn10_hubp.h" + +#define TO_DCN21_HUBP(hubp)\ + container_of(hubp, struct dcn21_hubp, base) + +#define HUBP_REG_LIST_DCN21(id)\ + HUBP_REG_LIST_DCN2_COMMON(id),\ + SRI(FLIP_PARAMETERS_3, HUBPREQ, id),\ + SRI(FLIP_PARAMETERS_4, HUBPREQ, id),\ + SRI(FLIP_PARAMETERS_5, HUBPREQ, id),\ + SRI(FLIP_PARAMETERS_6, HUBPREQ, id),\ + SRI(VBLANK_PARAMETERS_5, HUBPREQ, id),\ + SRI(VBLANK_PARAMETERS_6, HUBPREQ, id) + +#define HUBP_MASK_SH_LIST_DCN21_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh),\ + HUBP_MASK_SH_LIST_DCN_VM(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, ROTATION_ANGLE, mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, H_MIRROR_EN, mask_sh),\ + HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS, DST_Y_PREFETCH, mask_sh),\ + HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS, VRATIO_PREFETCH, mask_sh),\ + HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS_C, VRATIO_PREFETCH_C, mask_sh),\ + HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_LOW_ADDR, MC_VM_SYSTEM_APERTURE_LOW_ADDR, mask_sh),\ + HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, MC_VM_SYSTEM_APERTURE_HIGH_ADDR, mask_sh),\ + HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_DST_Y_OFFSET, mask_sh), \ + HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_CHUNK_HDL_ADJUST, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SURFACE_ADDRESS_HIGH, CURSOR_SURFACE_ADDRESS_HIGH, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SURFACE_ADDRESS, CURSOR_SURFACE_ADDRESS, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SIZE, CURSOR_WIDTH, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SIZE, CURSOR_HEIGHT, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_MODE, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_2X_MAGNIFY, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_PITCH, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_LINES_PER_CHUNK, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_ENABLE, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_POSITION, CURSOR_X_POSITION, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_POSITION, CURSOR_Y_POSITION, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_X, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_Y, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_DST_OFFSET, CURSOR_DST_X_OFFSET, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_ADDRESS_HIGH, DMDATA_ADDRESS_HIGH, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_MODE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_UPDATED, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_REPEAT, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_SIZE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_SW_CNTL, DMDATA_SW_UPDATED, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_SW_CNTL, DMDATA_SW_REPEAT, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_SW_CNTL, DMDATA_SW_SIZE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_QOS_CNTL, DMDATA_QOS_MODE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_QOS_CNTL, DMDATA_QOS_LEVEL, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_QOS_CNTL, DMDATA_DL_DELTA, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_STATUS, DMDATA_DONE, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_0, DST_Y_PER_VM_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_0, DST_Y_PER_ROW_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_1, REFCYC_PER_PTE_GROUP_FLIP_L, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_2, REFCYC_PER_META_CHUNK_FLIP_L, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_VREADY_AT_OR_AFTER_VSYNC, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_DISABLE_STOP_DATA_DURING_VM, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL, HUBPREQ_MASTER_UPDATE_LOCK_STATUS, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL2, SURFACE_GSL_ENABLE, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL2, SURFACE_TRIPLE_BUFFER_ENABLE, mask_sh),\ + HUBP_SF(HUBPREQ0_VMID_SETTINGS_0, VMID, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_3, REFCYC_PER_VM_GROUP_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_4, REFCYC_PER_VM_REQ_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_5, REFCYC_PER_PTE_GROUP_FLIP_C, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_6, REFCYC_PER_META_CHUNK_FLIP_C, mask_sh),\ + HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, mask_sh),\ + HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_6, REFCYC_PER_VM_REQ_VBLANK, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh) + +#define HUBP_MASK_SH_LIST_DCN21(mask_sh)\ + HUBP_MASK_SH_LIST_DCN21_COMMON(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh) + + +struct dcn21_hubp { + struct hubp base; + struct dcn_hubp_state state; + const struct dcn_hubp2_registers *hubp_regs; + const struct dcn_hubp2_shift *hubp_shift; + const struct dcn_hubp2_mask *hubp_mask; +}; + +bool hubp21_construct( + struct dcn21_hubp *hubp21, + struct dc_context *ctx, + uint32_t inst, + const struct dcn_hubp2_registers *hubp_regs, + const struct dcn_hubp2_shift *hubp_shift, + const struct dcn_hubp2_mask *hubp_mask); + +void apply_DEDCN21_142_wa_for_hostvm_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr); + +void hubp21_program_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *ttu_attr); + +void hubp21_program_requestor( + struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs); +#endif /* DAL_DC_DCN21_DCN21_HUBP_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c new file mode 100644 index 000000000000..de182185fe1f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -0,0 +1,1682 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/slab.h> + +#include "dm_services.h" +#include "dc.h" + +#include "resource.h" +#include "include/irq_service_interface.h" +#include "dcn20/dcn20_resource.h" + +#include "clk_mgr.h" +#include "dcn10/dcn10_hubp.h" +#include "dcn10/dcn10_ipp.h" +#include "dcn20/dcn20_hubbub.h" +#include "dcn20/dcn20_mpc.h" +#include "dcn20/dcn20_hubp.h" +#include "dcn21_hubp.h" +#include "irq/dcn21/irq_service_dcn21.h" +#include "dcn20/dcn20_dpp.h" +#include "dcn20/dcn20_optc.h" +#include "dcn20/dcn20_hwseq.h" +#include "dce110/dce110_hw_sequencer.h" +#include "dcn20/dcn20_opp.h" +#include "dcn20/dcn20_dsc.h" +#include "dcn20/dcn20_link_encoder.h" +#include "dcn20/dcn20_stream_encoder.h" +#include "dce/dce_clock_source.h" +#include "dce/dce_audio.h" +#include "dce/dce_hwseq.h" +#include "virtual/virtual_stream_encoder.h" +#include "dce110/dce110_resource.h" +#include "dml/display_mode_vba.h" +#include "dcn20/dcn20_dccg.h" +#include "dcn21_hubbub.h" +#include "dcn10/dcn10_resource.h" + +#include "dcn20/dcn20_dwb.h" +#include "dcn20/dcn20_mmhubbub.h" + +#include "renoir_ip_offset.h" +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" + +#include "nbio/nbio_7_0_offset.h" + +#include "mmhub/mmhub_2_0_0_offset.h" +#include "mmhub/mmhub_2_0_0_sh_mask.h" + +#include "reg_helper.h" +#include "dce/dce_abm.h" +#include "dce/dce_dmcu.h" +#include "dce/dce_aux.h" +#include "dce/dce_i2c.h" +#include "dcn21_resource.h" +#include "vm_helper.h" +#include "dcn20/dcn20_vmid.h" + +#define SOC_BOUNDING_BOX_VALID false +#define DC_LOGGER_INIT(logger) + + +struct _vcs_dpi_ip_params_st dcn2_1_ip = { + .gpuvm_enable = 0, + .hostvm_enable = 0, + .gpuvm_max_page_table_levels = 1, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 2, +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + .num_dsc = 3, +#else + .num_dsc = 0, +#endif + .rob_buffer_size_kbytes = 168, + .det_buffer_size_kbytes = 164, + .dpte_buffer_size_in_pte_reqs_luma = 44, + .dpte_buffer_size_in_pte_reqs_chroma = 42,//todo + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 4, + .pte_chunk_size_kbytes = 2, + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 2, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, + .line_buffer_fixed_bpp = 0, + .dcc_supported = true, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 12, + .writeback_max_vscl_taps = 12, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 4, + .max_num_dpp = 4, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 4, + .max_vscl_ratio = 4, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.10, + .min_vblank_lines = 32, // + .dppclk_delay_subtotal = 77, // + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 8, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 87, // + .dcfclk_cstate_latency = 10, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + + .xfc_supported = false, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .ptoi_supported = 0 +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 304.0, + .fabricclk_mhz = 600.0, + .dispclk_mhz = 618.0, + .dppclk_mhz = 440.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 205.67, + .dram_speed_mts = 1600.0, + }, + { + .state = 1, + .dcfclk_mhz = 304.0, + .fabricclk_mhz = 600.0, + .dispclk_mhz = 618.0, + .dppclk_mhz = 618.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 205.67, + .dram_speed_mts = 1600.0, + }, + { + .state = 2, + .dcfclk_mhz = 608.0, + .fabricclk_mhz = 1066.0, + .dispclk_mhz = 888.0, + .dppclk_mhz = 888.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 287.67, + .dram_speed_mts = 2133.0, + }, + { + .state = 3, + .dcfclk_mhz = 676.0, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1015.0, + .dppclk_mhz = 1015.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 715.0, + .dscclk_mhz = 318.334, + .dram_speed_mts = 4266.0, + }, + { + .state = 4, + .dcfclk_mhz = 810.0, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1015.0, + .dppclk_mhz = 1015.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 318.334, + .dram_speed_mts = 4266.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 810.0, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1015.0, + .dppclk_mhz = 1015.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 318.334, + .dram_speed_mts = 4266.0, + }, + + }, + + .sr_exit_time_us = 9.0, + .sr_enter_plus_exit_time_us = 11.0, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 75.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 60.0, + .max_avg_dram_bw_use_normal_percent = 100.0, + .writeback_latency_us = 12.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 4, + .fabric_datapath_to_dcn_data_return_bytes = 32, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.5, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 128, + .urgent_out_of_order_return_per_channel_bytes = 4096, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 4, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 23.84, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3550, + .xfc_bus_transport_time_us = 4, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 1, + .num_states = 5 +}; + +#ifndef MAX +#define MAX(X, Y) ((X) > (Y) ? (X) : (Y)) +#endif +#ifndef MIN +#define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) +#endif + +/* begin ********************* + * macros to expend register list macro defined in HW object header file */ + +/* DCN */ +/* TODO awful hack. fixup dcn20_dwb.h */ +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +#define BASE(seg) BASE_INNER(seg) + +#define SR(reg_name)\ + .reg_name = BASE(mm ## reg_name ## _BASE_IDX) + \ + mm ## reg_name + +#define SRI(reg_name, block, id)\ + .reg_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define SRIR(var_name, reg_name, block, id)\ + .var_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define SRII(reg_name, block, id)\ + .reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define DCCG_SRII(reg_name, block, id)\ + .block ## _ ## reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +/* NBIO */ +#define NBIO_BASE_INNER(seg) \ + NBIF0_BASE__INST0_SEG ## seg + +#define NBIO_BASE(seg) \ + NBIO_BASE_INNER(seg) + +#define NBIO_SR(reg_name)\ + .reg_name = NBIO_BASE(mm ## reg_name ## _BASE_IDX) + \ + mm ## reg_name + +/* MMHUB */ +#define MMHUB_BASE_INNER(seg) \ + MMHUB_BASE__INST0_SEG ## seg + +#define MMHUB_BASE(seg) \ + MMHUB_BASE_INNER(seg) + +#define MMHUB_SR(reg_name)\ + .reg_name = MMHUB_BASE(mmMM ## reg_name ## _BASE_IDX) + \ + mmMM ## reg_name + +#define clk_src_regs(index, pllid)\ +[index] = {\ + CS_COMMON_REG_LIST_DCN2_1(index, pllid),\ +} + +static const struct dce110_clk_src_regs clk_src_regs[] = { + clk_src_regs(0, A), + clk_src_regs(1, B), + clk_src_regs(2, C), + clk_src_regs(3, D), + clk_src_regs(4, E), +}; + +static const struct dce110_clk_src_shift cs_shift = { + CS_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT) +}; + +static const struct dce110_clk_src_mask cs_mask = { + CS_COMMON_MASK_SH_LIST_DCN2_0(_MASK) +}; + +static const struct bios_registers bios_regs = { + NBIO_SR(BIOS_SCRATCH_3), + NBIO_SR(BIOS_SCRATCH_6) +}; + +#ifdef CONFIG_DRM_AMD_DC_DMUB +static const struct dcn21_dmcub_registers dmcub_regs = { + DMCUB_REG_LIST_DCN() +}; + +static const struct dcn21_dmcub_shift dmcub_shift = { + DMCUB_COMMON_MASK_SH_LIST_BASE(__SHIFT) +}; + +static const struct dcn21_dmcub_mask dmcub_mask = { + DMCUB_COMMON_MASK_SH_LIST_BASE(_MASK) +}; +#endif + +#define audio_regs(id)\ +[id] = {\ + AUD_COMMON_REG_LIST(id)\ +} + +static const struct dce_audio_registers audio_regs[] = { + audio_regs(0), + audio_regs(1), + audio_regs(2), + audio_regs(3), + audio_regs(4), + audio_regs(5), +}; + +#define DCE120_AUD_COMMON_MASK_SH_LIST(mask_sh)\ + SF(AZF0ENDPOINT0_AZALIA_F0_CODEC_ENDPOINT_INDEX, AZALIA_ENDPOINT_REG_INDEX, mask_sh),\ + SF(AZF0ENDPOINT0_AZALIA_F0_CODEC_ENDPOINT_DATA, AZALIA_ENDPOINT_REG_DATA, mask_sh),\ + AUD_COMMON_MASK_SH_LIST_BASE(mask_sh) + +static const struct dce_audio_shift audio_shift = { + DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_audio_mask audio_mask = { + DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) +}; + +static const struct dccg_registers dccg_regs = { + DCCG_COMMON_REG_LIST_DCN_BASE() +}; + +static const struct dccg_shift dccg_shift = { + DCCG_MASK_SH_LIST_DCN2(__SHIFT) +}; + +static const struct dccg_mask dccg_mask = { + DCCG_MASK_SH_LIST_DCN2(_MASK) +}; + +#define opp_regs(id)\ +[id] = {\ + OPP_REG_LIST_DCN20(id),\ +} + +static const struct dcn20_opp_registers opp_regs[] = { + opp_regs(0), + opp_regs(1), + opp_regs(2), + opp_regs(3), + opp_regs(4), + opp_regs(5), +}; + +static const struct dcn20_opp_shift opp_shift = { + OPP_MASK_SH_LIST_DCN20(__SHIFT) +}; + +static const struct dcn20_opp_mask opp_mask = { + OPP_MASK_SH_LIST_DCN20(_MASK) +}; + +#define tg_regs(id)\ +[id] = {TG_COMMON_REG_LIST_DCN2_0(id)} + +static const struct dcn_optc_registers tg_regs[] = { + tg_regs(0), + tg_regs(1), + tg_regs(2), + tg_regs(3) +}; + +static const struct dcn_optc_shift tg_shift = { + TG_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT) +}; + +static const struct dcn_optc_mask tg_mask = { + TG_COMMON_MASK_SH_LIST_DCN2_0(_MASK) +}; + +static const struct dcn20_mpc_registers mpc_regs = { + MPC_REG_LIST_DCN2_0(0), + MPC_REG_LIST_DCN2_0(1), + MPC_REG_LIST_DCN2_0(2), + MPC_REG_LIST_DCN2_0(3), + MPC_REG_LIST_DCN2_0(4), + MPC_REG_LIST_DCN2_0(5), + MPC_OUT_MUX_REG_LIST_DCN2_0(0), + MPC_OUT_MUX_REG_LIST_DCN2_0(1), + MPC_OUT_MUX_REG_LIST_DCN2_0(2), + MPC_OUT_MUX_REG_LIST_DCN2_0(3) +}; + +static const struct dcn20_mpc_shift mpc_shift = { + MPC_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT) +}; + +static const struct dcn20_mpc_mask mpc_mask = { + MPC_COMMON_MASK_SH_LIST_DCN2_0(_MASK) +}; + +#define hubp_regs(id)\ +[id] = {\ + HUBP_REG_LIST_DCN21(id)\ +} + +static const struct dcn_hubp2_registers hubp_regs[] = { + hubp_regs(0), + hubp_regs(1), + hubp_regs(2), + hubp_regs(3) +}; + +static const struct dcn_hubp2_shift hubp_shift = { + HUBP_MASK_SH_LIST_DCN21(__SHIFT) +}; + +static const struct dcn_hubp2_mask hubp_mask = { + HUBP_MASK_SH_LIST_DCN21(_MASK) +}; + +static const struct dcn_hubbub_registers hubbub_reg = { + HUBBUB_REG_LIST_DCN21() +}; + +static const struct dcn_hubbub_shift hubbub_shift = { + HUBBUB_MASK_SH_LIST_DCN21(__SHIFT) +}; + +static const struct dcn_hubbub_mask hubbub_mask = { + HUBBUB_MASK_SH_LIST_DCN21(_MASK) +}; + + +#define vmid_regs(id)\ +[id] = {\ + DCN20_VMID_REG_LIST(id)\ +} + +static const struct dcn_vmid_registers vmid_regs[] = { + vmid_regs(0), + vmid_regs(1), + vmid_regs(2), + vmid_regs(3), + vmid_regs(4), + vmid_regs(5), + vmid_regs(6), + vmid_regs(7), + vmid_regs(8), + vmid_regs(9), + vmid_regs(10), + vmid_regs(11), + vmid_regs(12), + vmid_regs(13), + vmid_regs(14), + vmid_regs(15) +}; + +static const struct dcn20_vmid_shift vmid_shifts = { + DCN20_VMID_MASK_SH_LIST(__SHIFT) +}; + +static const struct dcn20_vmid_mask vmid_masks = { + DCN20_VMID_MASK_SH_LIST(_MASK) +}; + +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT +#define dsc_regsDCN20(id)\ +[id] = {\ + DSC_REG_LIST_DCN20(id)\ +} + +static const struct dcn20_dsc_registers dsc_regs[] = { + dsc_regsDCN20(0), + dsc_regsDCN20(1), + dsc_regsDCN20(2), + dsc_regsDCN20(3), + dsc_regsDCN20(4), + dsc_regsDCN20(5) +}; + +static const struct dcn20_dsc_shift dsc_shift = { + DSC_REG_LIST_SH_MASK_DCN20(__SHIFT) +}; + +static const struct dcn20_dsc_mask dsc_mask = { + DSC_REG_LIST_SH_MASK_DCN20(_MASK) +}; +#endif + +#define ipp_regs(id)\ +[id] = {\ + IPP_REG_LIST_DCN20(id),\ +} + +static const struct dcn10_ipp_registers ipp_regs[] = { + ipp_regs(0), + ipp_regs(1), + ipp_regs(2), + ipp_regs(3), +}; + +static const struct dcn10_ipp_shift ipp_shift = { + IPP_MASK_SH_LIST_DCN20(__SHIFT) +}; + +static const struct dcn10_ipp_mask ipp_mask = { + IPP_MASK_SH_LIST_DCN20(_MASK), +}; + +#define opp_regs(id)\ +[id] = {\ + OPP_REG_LIST_DCN20(id),\ +} + + +#define aux_engine_regs(id)\ +[id] = {\ + AUX_COMMON_REG_LIST0(id), \ + .AUXN_IMPCAL = 0, \ + .AUXP_IMPCAL = 0, \ + .AUX_RESET_MASK = DP_AUX0_AUX_CONTROL__AUX_RESET_MASK, \ +} + +static const struct dce110_aux_registers aux_engine_regs[] = { + aux_engine_regs(0), + aux_engine_regs(1), + aux_engine_regs(2), + aux_engine_regs(3), + aux_engine_regs(4), +}; + +#define tf_regs(id)\ +[id] = {\ + TF_REG_LIST_DCN20(id),\ +} + +static const struct dcn2_dpp_registers tf_regs[] = { + tf_regs(0), + tf_regs(1), + tf_regs(2), + tf_regs(3), +}; + +static const struct dcn2_dpp_shift tf_shift = { + TF_REG_LIST_SH_MASK_DCN20(__SHIFT) +}; + +static const struct dcn2_dpp_mask tf_mask = { + TF_REG_LIST_SH_MASK_DCN20(_MASK) +}; + +#define stream_enc_regs(id)\ +[id] = {\ + SE_DCN2_REG_LIST(id)\ +} + +static const struct dcn10_stream_enc_registers stream_enc_regs[] = { + stream_enc_regs(0), + stream_enc_regs(1), + stream_enc_regs(2), + stream_enc_regs(3), + stream_enc_regs(4), +}; + +static const struct dcn10_stream_encoder_shift se_shift = { + SE_COMMON_MASK_SH_LIST_DCN20(__SHIFT) +}; + +static const struct dcn10_stream_encoder_mask se_mask = { + SE_COMMON_MASK_SH_LIST_DCN20(_MASK) +}; + +static struct input_pixel_processor *dcn21_ipp_create( + struct dc_context *ctx, uint32_t inst) +{ + struct dcn10_ipp *ipp = + kzalloc(sizeof(struct dcn10_ipp), GFP_KERNEL); + + if (!ipp) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dcn20_ipp_construct(ipp, ctx, inst, + &ipp_regs[inst], &ipp_shift, &ipp_mask); + return &ipp->base; +} + +static struct dpp *dcn21_dpp_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct dcn20_dpp *dpp = + kzalloc(sizeof(struct dcn20_dpp), GFP_KERNEL); + + if (!dpp) + return NULL; + + if (dpp2_construct(dpp, ctx, inst, + &tf_regs[inst], &tf_shift, &tf_mask)) + return &dpp->base; + + BREAK_TO_DEBUGGER(); + kfree(dpp); + return NULL; +} + +static struct dce_aux *dcn21_aux_engine_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct aux_engine_dce110 *aux_engine = + kzalloc(sizeof(struct aux_engine_dce110), GFP_KERNEL); + + if (!aux_engine) + return NULL; + + dce110_aux_engine_construct(aux_engine, ctx, inst, + SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, + &aux_engine_regs[inst]); + + return &aux_engine->base; +} + +#define i2c_inst_regs(id) { I2C_HW_ENGINE_COMMON_REG_LIST(id) } + +static const struct dce_i2c_registers i2c_hw_regs[] = { + i2c_inst_regs(1), + i2c_inst_regs(2), + i2c_inst_regs(3), + i2c_inst_regs(4), + i2c_inst_regs(5), +}; + +static const struct dce_i2c_shift i2c_shifts = { + I2C_COMMON_MASK_SH_LIST_DCN2(__SHIFT) +}; + +static const struct dce_i2c_mask i2c_masks = { + I2C_COMMON_MASK_SH_LIST_DCN2(_MASK) +}; + +struct dce_i2c_hw *dcn21_i2c_hw_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct dce_i2c_hw *dce_i2c_hw = + kzalloc(sizeof(struct dce_i2c_hw), GFP_KERNEL); + + if (!dce_i2c_hw) + return NULL; + + dcn2_i2c_hw_construct(dce_i2c_hw, ctx, inst, + &i2c_hw_regs[inst], &i2c_shifts, &i2c_masks); + + return dce_i2c_hw; +} + +static const struct resource_caps res_cap_rn = { + .num_timing_generator = 4, + .num_opp = 4, + .num_video_plane = 4, + .num_audio = 6, // 6 audio endpoints. 4 audio streams + .num_stream_encoder = 5, + .num_pll = 5, // maybe 3 because the last two used for USB-c + .num_dwb = 1, + .num_ddc = 5, +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + .num_dsc = 3, +#endif +}; + +#ifdef DIAGS_BUILD +static const struct resource_caps res_cap_rn_FPGA_4pipe = { + .num_timing_generator = 4, + .num_opp = 4, + .num_video_plane = 4, + .num_audio = 7, + .num_stream_encoder = 4, + .num_pll = 4, + .num_dwb = 1, + .num_ddc = 4, + .num_dsc = 0, +}; + +static const struct resource_caps res_cap_rn_FPGA_2pipe_dsc = { + .num_timing_generator = 2, + .num_opp = 2, + .num_video_plane = 2, + .num_audio = 7, + .num_stream_encoder = 2, + .num_pll = 4, + .num_dwb = 1, + .num_ddc = 4, +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + .num_dsc = 2, +#endif +}; +#endif + +static const struct dc_plane_cap plane_cap = { + .type = DC_PLANE_TYPE_DCN_UNIVERSAL, + .blends_with_above = true, + .blends_with_below = true, + .per_pixel_alpha = true, + + .pixel_format_support = { + .argb8888 = true, + .nv12 = true, + .fp16 = true + }, + + .max_upscale_factor = { + .argb8888 = 16000, + .nv12 = 16000, + .fp16 = 16000 + }, + + .max_downscale_factor = { + .argb8888 = 250, + .nv12 = 250, + .fp16 = 250 + } +}; + +static const struct dc_debug_options debug_defaults_drv = { + .disable_dmcu = true, + .force_abm_enable = false, + .timing_trace = false, + .clock_trace = true, + .disable_pplib_clock_request = true, + .pipe_split_policy = MPC_SPLIT_AVOID_MULT_DISP, + .force_single_disp_pipe_split = true, + .disable_dcc = DCC_ENABLE, + .vsr_support = true, + .performance_trace = false, + .max_downscale_src_width = 5120,/*upto 5K*/ + .disable_pplib_wm_range = false, + .scl_reset_length10 = true, + .sanity_checks = true, + .disable_48mhz_pwrdwn = true, +}; + +static const struct dc_debug_options debug_defaults_diags = { + .disable_dmcu = true, + .force_abm_enable = false, + .timing_trace = true, + .clock_trace = true, + .disable_dpp_power_gate = true, + .disable_hubp_power_gate = true, + .disable_clock_gate = true, + .disable_pplib_clock_request = true, + .disable_pplib_wm_range = true, + .disable_stutter = true, + .disable_48mhz_pwrdwn = true, +}; + +enum dcn20_clk_src_array_id { + DCN20_CLK_SRC_PLL0, + DCN20_CLK_SRC_PLL1, + DCN20_CLK_SRC_TOTAL_DCN21 +}; + +static void destruct(struct dcn21_resource_pool *pool) +{ + unsigned int i; + + for (i = 0; i < pool->base.stream_enc_count; i++) { + if (pool->base.stream_enc[i] != NULL) { + kfree(DCN10STRENC_FROM_STRENC(pool->base.stream_enc[i])); + pool->base.stream_enc[i] = NULL; + } + } + +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + for (i = 0; i < pool->base.res_cap->num_dsc; i++) { + if (pool->base.dscs[i] != NULL) + dcn20_dsc_destroy(&pool->base.dscs[i]); + } +#endif + + if (pool->base.mpc != NULL) { + kfree(TO_DCN20_MPC(pool->base.mpc)); + pool->base.mpc = NULL; + } + if (pool->base.hubbub != NULL) { + kfree(pool->base.hubbub); + pool->base.hubbub = NULL; + } + for (i = 0; i < pool->base.pipe_count; i++) { + if (pool->base.dpps[i] != NULL) + dcn20_dpp_destroy(&pool->base.dpps[i]); + + if (pool->base.ipps[i] != NULL) + pool->base.ipps[i]->funcs->ipp_destroy(&pool->base.ipps[i]); + + if (pool->base.hubps[i] != NULL) { + kfree(TO_DCN20_HUBP(pool->base.hubps[i])); + pool->base.hubps[i] = NULL; + } + + if (pool->base.irqs != NULL) { + dal_irq_service_destroy(&pool->base.irqs); + } + } + + for (i = 0; i < pool->base.res_cap->num_ddc; i++) { + if (pool->base.engines[i] != NULL) + dce110_engine_destroy(&pool->base.engines[i]); + if (pool->base.hw_i2cs[i] != NULL) { + kfree(pool->base.hw_i2cs[i]); + pool->base.hw_i2cs[i] = NULL; + } + if (pool->base.sw_i2cs[i] != NULL) { + kfree(pool->base.sw_i2cs[i]); + pool->base.sw_i2cs[i] = NULL; + } + } + + for (i = 0; i < pool->base.res_cap->num_opp; i++) { + if (pool->base.opps[i] != NULL) + pool->base.opps[i]->funcs->opp_destroy(&pool->base.opps[i]); + } + + for (i = 0; i < pool->base.res_cap->num_timing_generator; i++) { + if (pool->base.timing_generators[i] != NULL) { + kfree(DCN10TG_FROM_TG(pool->base.timing_generators[i])); + pool->base.timing_generators[i] = NULL; + } + } + + for (i = 0; i < pool->base.res_cap->num_dwb; i++) { + if (pool->base.dwbc[i] != NULL) { + kfree(TO_DCN20_DWBC(pool->base.dwbc[i])); + pool->base.dwbc[i] = NULL; + } + if (pool->base.mcif_wb[i] != NULL) { + kfree(TO_DCN20_MMHUBBUB(pool->base.mcif_wb[i])); + pool->base.mcif_wb[i] = NULL; + } + } + + for (i = 0; i < pool->base.audio_count; i++) { + if (pool->base.audios[i]) + dce_aud_destroy(&pool->base.audios[i]); + } + + for (i = 0; i < pool->base.clk_src_count; i++) { + if (pool->base.clock_sources[i] != NULL) { + dcn20_clock_source_destroy(&pool->base.clock_sources[i]); + pool->base.clock_sources[i] = NULL; + } + } + + if (pool->base.dp_clock_source != NULL) { + dcn20_clock_source_destroy(&pool->base.dp_clock_source); + pool->base.dp_clock_source = NULL; + } + + + if (pool->base.abm != NULL) + dce_abm_destroy(&pool->base.abm); + + if (pool->base.dmcu != NULL) + dce_dmcu_destroy(&pool->base.dmcu); + +#ifdef CONFIG_DRM_AMD_DC_DMUB + if (pool->base.dmcub != NULL) + dcn21_dmcub_destroy(&pool->base.dmcub); +#endif + + if (pool->base.dccg != NULL) + dcn_dccg_destroy(&pool->base.dccg); + + if (pool->base.pp_smu != NULL) + dcn20_pp_smu_destroy(&pool->base.pp_smu); +} + + +static void calculate_wm_set_for_vlevel( + int vlevel, + struct wm_range_table_entry *table_entry, + struct dcn_watermarks *wm_set, + struct display_mode_lib *dml, + display_e2e_pipe_params_st *pipes, + int pipe_cnt) +{ + double dram_clock_change_latency_cached = dml->soc.dram_clock_change_latency_us; + + ASSERT(vlevel < dml->soc.num_states); + /* only pipe 0 is read for voltage and dcf/soc clocks */ + pipes[0].clks_cfg.voltage = vlevel; + pipes[0].clks_cfg.dcfclk_mhz = dml->soc.clock_limits[vlevel].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = dml->soc.clock_limits[vlevel].socclk_mhz; + + dml->soc.dram_clock_change_latency_us = table_entry->pstate_latency_us; + + wm_set->urgent_ns = get_wm_urgent(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(dml, pipes, pipe_cnt) * 1000; + wm_set->pte_meta_urgent_ns = get_wm_memory_trip(dml, pipes, pipe_cnt) * 1000; +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + wm_set->frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(dml, pipes, pipe_cnt) * 1000; + wm_set->frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(dml, pipes, pipe_cnt) * 1000; +#endif + dml->soc.dram_clock_change_latency_us = dram_clock_change_latency_cached; + +} + +void dcn21_calculate_wm( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *out_pipe_cnt, + int *pipe_split_from, + int vlevel_req) +{ + int pipe_cnt, i, pipe_idx; + int vlevel, vlevel_max; + struct wm_range_table_entry *table_entry; + struct clk_bw_params *bw_params = dc->clk_mgr->bw_params; + + ASSERT(bw_params); + + for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; + pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.vba.RequiredDISPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb]; + + if (pipe_split_from[i] < 0) { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_idx] == pipe_idx) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_idx]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + pipe_idx++; + } else { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_split_from[i]]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_split_from[i]] == pipe_split_from[i]) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_split_from[i]]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + } + pipe_cnt++; + } + + if (pipe_cnt != pipe_idx) { + if (dc->res_pool->funcs->populate_dml_pipes) + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, + &context->res_ctx, pipes); + else + pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, + &context->res_ctx, pipes); + } + + *out_pipe_cnt = pipe_cnt; + + vlevel_max = bw_params->clk_table.num_entries - 1; + + + /* WM Set D */ + table_entry = &bw_params->wm_table.entries[WM_D]; + if (table_entry->wm_type == WM_TYPE_RETRAINING) + vlevel = 0; + else + vlevel = vlevel_max; + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.d, + &context->bw_ctx.dml, pipes, pipe_cnt); + /* WM Set C */ + table_entry = &bw_params->wm_table.entries[WM_C]; + vlevel = MIN(MAX(vlevel_req, 2), vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.c, + &context->bw_ctx.dml, pipes, pipe_cnt); + /* WM Set B */ + table_entry = &bw_params->wm_table.entries[WM_B]; + vlevel = MIN(MAX(vlevel_req, 1), vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.b, + &context->bw_ctx.dml, pipes, pipe_cnt); + + /* WM Set A */ + table_entry = &bw_params->wm_table.entries[WM_A]; + vlevel = MIN(vlevel_req, vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.a, + &context->bw_ctx.dml, pipes, pipe_cnt); +} + + +bool dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool out = false; + + BW_VAL_TRACE_SETUP(); + + int vlevel = 0; + int pipe_split_from[MAX_PIPES]; + int pipe_cnt = 0; + display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_KERNEL); + DC_LOGGER_INIT(dc->ctx->logger); + + BW_VAL_TRACE_COUNT(); + + out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel); + + if (pipe_cnt == 0) + goto validate_out; + + if (!out) + goto validate_fail; + + BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + + if (fast_validate) { + BW_VAL_TRACE_SKIP(fast); + goto validate_out; + } + + dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel); + dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + + BW_VAL_TRACE_END_WATERMARKS(); + + goto validate_out; + +validate_fail: + DC_LOG_WARNING("Mode Validation Warning: %s failed validation.\n", + dml_get_status_message(context->bw_ctx.dml.vba.ValidationStatus[context->bw_ctx.dml.vba.soc.num_states])); + + BW_VAL_TRACE_SKIP(fail); + out = false; + +validate_out: + kfree(pipes); + + BW_VAL_TRACE_FINISH(); + + return out; +} +static void dcn21_destroy_resource_pool(struct resource_pool **pool) +{ + struct dcn21_resource_pool *dcn21_pool = TO_DCN21_RES_POOL(*pool); + + destruct(dcn21_pool); + kfree(dcn21_pool); + *pool = NULL; +} + +static struct clock_source *dcn21_clock_source_create( + struct dc_context *ctx, + struct dc_bios *bios, + enum clock_source_id id, + const struct dce110_clk_src_regs *regs, + bool dp_clk_src) +{ + struct dce110_clk_src *clk_src = + kzalloc(sizeof(struct dce110_clk_src), GFP_KERNEL); + + if (!clk_src) + return NULL; + + if (dcn20_clk_src_construct(clk_src, ctx, bios, id, + regs, &cs_shift, &cs_mask)) { + clk_src->base.dp_clk_src = dp_clk_src; + return &clk_src->base; + } + + BREAK_TO_DEBUGGER(); + return NULL; +} + +static struct hubp *dcn21_hubp_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct dcn21_hubp *hubp21 = + kzalloc(sizeof(struct dcn21_hubp), GFP_KERNEL); + + if (!hubp21) + return NULL; + + if (hubp21_construct(hubp21, ctx, inst, + &hubp_regs[inst], &hubp_shift, &hubp_mask)) + return &hubp21->base; + + BREAK_TO_DEBUGGER(); + kfree(hubp21); + return NULL; +} + +static struct hubbub *dcn21_hubbub_create(struct dc_context *ctx) +{ + int i; + + struct dcn20_hubbub *hubbub = kzalloc(sizeof(struct dcn20_hubbub), + GFP_KERNEL); + + if (!hubbub) + return NULL; + + hubbub21_construct(hubbub, ctx, + &hubbub_reg, + &hubbub_shift, + &hubbub_mask); + + for (i = 0; i < res_cap_rn.num_vmid; i++) { + struct dcn20_vmid *vmid = &hubbub->vmid[i]; + + vmid->ctx = ctx; + + vmid->regs = &vmid_regs[i]; + vmid->shifts = &vmid_shifts; + vmid->masks = &vmid_masks; + } + + return &hubbub->base; +} + +struct output_pixel_processor *dcn21_opp_create( + struct dc_context *ctx, uint32_t inst) +{ + struct dcn20_opp *opp = + kzalloc(sizeof(struct dcn20_opp), GFP_KERNEL); + + if (!opp) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dcn20_opp_construct(opp, ctx, inst, + &opp_regs[inst], &opp_shift, &opp_mask); + return &opp->base; +} + +struct timing_generator *dcn21_timing_generator_create( + struct dc_context *ctx, + uint32_t instance) +{ + struct optc *tgn10 = + kzalloc(sizeof(struct optc), GFP_KERNEL); + + if (!tgn10) + return NULL; + + tgn10->base.inst = instance; + tgn10->base.ctx = ctx; + + tgn10->tg_regs = &tg_regs[instance]; + tgn10->tg_shift = &tg_shift; + tgn10->tg_mask = &tg_mask; + + dcn20_timing_generator_init(tgn10); + + return &tgn10->base; +} + +struct mpc *dcn21_mpc_create(struct dc_context *ctx) +{ + struct dcn20_mpc *mpc20 = kzalloc(sizeof(struct dcn20_mpc), + GFP_KERNEL); + + if (!mpc20) + return NULL; + + dcn20_mpc_construct(mpc20, ctx, + &mpc_regs, + &mpc_shift, + &mpc_mask, + 6); + + return &mpc20->base; +} + +static void read_dce_straps( + struct dc_context *ctx, + struct resource_straps *straps) +{ + generic_reg_get(ctx, mmDC_PINSTRAPS + BASE(mmDC_PINSTRAPS_BASE_IDX), + FN(DC_PINSTRAPS, DC_PINSTRAPS_AUDIO), &straps->dc_pinstraps_audio); + +} + +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + +struct display_stream_compressor *dcn21_dsc_create( + struct dc_context *ctx, uint32_t inst) +{ + struct dcn20_dsc *dsc = + kzalloc(sizeof(struct dcn20_dsc), GFP_KERNEL); + + if (!dsc) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dsc2_construct(dsc, ctx, inst, &dsc_regs[inst], &dsc_shift, &dsc_mask); + return &dsc->base; +} +#endif + +static void update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) +{ + struct dcn21_resource_pool *pool = TO_DCN21_RES_POOL(dc->res_pool); + struct clk_limit_table *clk_table = &bw_params->clk_table; + int i; + + dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator; + dcn2_1_ip.max_num_dpp = pool->base.pipe_count; + dcn2_1_soc.num_chans = bw_params->num_channels; + dcn2_1_soc.num_states = 0; + + for (i = 0; i < clk_table->num_entries; i++) { + + dcn2_1_soc.clock_limits[i].state = i; + dcn2_1_soc.clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + dcn2_1_soc.clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + dcn2_1_soc.clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz; + /* This is probably wrong, TODO: find correct calculation */ + dcn2_1_soc.clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 16 / 1000; + dcn2_1_soc.num_states++; + } +} + +/* Temporary Place holder until we can get them from fuse */ +static struct dpm_clocks dummy_clocks = { + .DcfClocks = { + {.Freq = 400, .Vol = 1}, + {.Freq = 483, .Vol = 1}, + {.Freq = 602, .Vol = 1}, + {.Freq = 738, .Vol = 1} }, + .SocClocks = { + {.Freq = 300, .Vol = 1}, + {.Freq = 400, .Vol = 1}, + {.Freq = 400, .Vol = 1}, + {.Freq = 400, .Vol = 1} }, + .FClocks = { + {.Freq = 400, .Vol = 1}, + {.Freq = 800, .Vol = 1}, + {.Freq = 1067, .Vol = 1}, + {.Freq = 1600, .Vol = 1} }, + .MemClocks = { + {.Freq = 800, .Vol = 1}, + {.Freq = 1600, .Vol = 1}, + {.Freq = 1067, .Vol = 1}, + {.Freq = 1600, .Vol = 1} }, + +}; + +enum pp_smu_status dummy_set_wm_ranges(struct pp_smu *pp, + struct pp_smu_wm_range_sets *ranges) +{ + return PP_SMU_RESULT_OK; +} + +enum pp_smu_status dummy_get_dpm_clock_table(struct pp_smu *pp, + struct dpm_clocks *clock_table) +{ + *clock_table = dummy_clocks; + return PP_SMU_RESULT_OK; +} + +struct pp_smu_funcs *dcn21_pp_smu_create(struct dc_context *ctx) +{ + struct pp_smu_funcs *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL); + + pp_smu->ctx.ver = PP_SMU_VER_RN; + + pp_smu->rn_funcs.get_dpm_clock_table = dummy_get_dpm_clock_table; + pp_smu->rn_funcs.set_wm_ranges = dummy_set_wm_ranges; + + return pp_smu; +} + +void dcn21_pp_smu_destroy(struct pp_smu_funcs **pp_smu) +{ + if (pp_smu && *pp_smu) { + kfree(*pp_smu); + *pp_smu = NULL; + } +} + +static struct audio *dcn21_create_audio( + struct dc_context *ctx, unsigned int inst) +{ + return dce_audio_create(ctx, inst, + &audio_regs[inst], &audio_shift, &audio_mask); +} + +static struct dc_cap_funcs cap_funcs = { + .get_dcc_compression_cap = dcn20_get_dcc_compression_cap +}; + +struct stream_encoder *dcn21_stream_encoder_create( + enum engine_id eng_id, + struct dc_context *ctx) +{ + struct dcn10_stream_encoder *enc1 = + kzalloc(sizeof(struct dcn10_stream_encoder), GFP_KERNEL); + + if (!enc1) + return NULL; + + dcn20_stream_encoder_construct(enc1, ctx, ctx->dc_bios, eng_id, + &stream_enc_regs[eng_id], + &se_shift, &se_mask); + + return &enc1->base; +} + +static const struct dce_hwseq_registers hwseq_reg = { + HWSEQ_DCN21_REG_LIST() +}; + +static const struct dce_hwseq_shift hwseq_shift = { + HWSEQ_DCN21_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_hwseq_mask hwseq_mask = { + HWSEQ_DCN21_MASK_SH_LIST(_MASK) +}; + +static struct dce_hwseq *dcn21_hwseq_create( + struct dc_context *ctx) +{ + struct dce_hwseq *hws = kzalloc(sizeof(struct dce_hwseq), GFP_KERNEL); + + if (hws) { + hws->ctx = ctx; + hws->regs = &hwseq_reg; + hws->shifts = &hwseq_shift; + hws->masks = &hwseq_mask; + } + return hws; +} + +static const struct resource_create_funcs res_create_funcs = { + .read_dce_straps = read_dce_straps, + .create_audio = dcn21_create_audio, + .create_stream_encoder = dcn21_stream_encoder_create, + .create_hwseq = dcn21_hwseq_create, +}; + +static const struct resource_create_funcs res_create_maximus_funcs = { + .read_dce_straps = NULL, + .create_audio = NULL, + .create_stream_encoder = NULL, + .create_hwseq = dcn21_hwseq_create, +}; + +static struct resource_funcs dcn21_res_pool_funcs = { + .destroy = dcn21_destroy_resource_pool, + .link_enc_create = dcn20_link_encoder_create, + .validate_bandwidth = dcn21_validate_bandwidth, + .add_stream_to_ctx = dcn20_add_stream_to_ctx, + .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, + .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, + .get_default_swizzle_mode = dcn20_get_default_swizzle_mode, + .set_mcif_arb_params = dcn20_set_mcif_arb_params, + .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link, + .update_bw_bounding_box = update_bw_bounding_box +}; + +static bool construct( + uint8_t num_virtual_links, + struct dc *dc, + struct dcn21_resource_pool *pool) +{ + int i; + struct dc_context *ctx = dc->ctx; + struct irq_service_init_data init_data; + + ctx->dc_bios->regs = &bios_regs; + + pool->base.res_cap = &res_cap_rn; +#ifdef DIAGS_BUILD + if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) + //pool->base.res_cap = &res_cap_nv10_FPGA_2pipe_dsc; + pool->base.res_cap = &res_cap_rn_FPGA_4pipe; +#endif + + pool->base.funcs = &dcn21_res_pool_funcs; + + /************************************************* + * Resource + asic cap harcoding * + *************************************************/ + pool->base.underlay_pipe_index = NO_UNDERLAY_PIPE; + + pool->base.pipe_count = 4; + dc->caps.max_downscale_ratio = 200; + dc->caps.i2c_speed_in_khz = 100; + dc->caps.max_cursor_size = 256; + dc->caps.dmdata_alloc_size = 2048; + dc->caps.hw_3d_lut = true; + + dc->caps.max_slave_planes = 1; + dc->caps.post_blend_color_processing = true; + dc->caps.force_dp_tps4_for_cp2520 = true; + + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) + dc->debug = debug_defaults_drv; + else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { + pool->base.pipe_count = 4; + dc->debug = debug_defaults_diags; + } else + dc->debug = debug_defaults_diags; + + // Init the vm_helper + if (dc->vm_helper) + vm_helper_init(dc->vm_helper, 16); + + /************************************************* + * Create resources * + *************************************************/ + + pool->base.clock_sources[DCN20_CLK_SRC_PLL0] = + dcn21_clock_source_create(ctx, ctx->dc_bios, + CLOCK_SOURCE_COMBO_PHY_PLL0, + &clk_src_regs[0], false); + pool->base.clock_sources[DCN20_CLK_SRC_PLL1] = + dcn21_clock_source_create(ctx, ctx->dc_bios, + CLOCK_SOURCE_COMBO_PHY_PLL1, + &clk_src_regs[1], false); + + pool->base.clk_src_count = DCN20_CLK_SRC_TOTAL_DCN21; + + /* todo: not reuse phy_pll registers */ + pool->base.dp_clock_source = + dcn21_clock_source_create(ctx, ctx->dc_bios, + CLOCK_SOURCE_ID_DP_DTO, + &clk_src_regs[0], true); + + for (i = 0; i < pool->base.clk_src_count; i++) { + if (pool->base.clock_sources[i] == NULL) { + dm_error("DC: failed to create clock sources!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + } + + pool->base.dccg = dccg2_create(ctx, &dccg_regs, &dccg_shift, &dccg_mask); + if (pool->base.dccg == NULL) { + dm_error("DC: failed to create dccg!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + +#ifdef CONFIG_DRM_AMD_DC_DMUB + pool->base.dmcub = dcn21_dmcub_create(ctx, + &dmcub_regs, + &dmcub_shift, + &dmcub_mask); + if (pool->base.dmcub == NULL) { + dm_error("DC: failed to create dmcub!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } +#endif + + pool->base.pp_smu = dcn21_pp_smu_create(ctx); + + dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); + + init_data.ctx = dc->ctx; + pool->base.irqs = dal_irq_service_dcn21_create(&init_data); + if (!pool->base.irqs) + goto create_fail; + + /* mem input -> ipp -> dpp -> opp -> TG */ + for (i = 0; i < pool->base.pipe_count; i++) { + pool->base.hubps[i] = dcn21_hubp_create(ctx, i); + if (pool->base.hubps[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create memory input!\n"); + goto create_fail; + } + + pool->base.ipps[i] = dcn21_ipp_create(ctx, i); + if (pool->base.ipps[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create input pixel processor!\n"); + goto create_fail; + } + + pool->base.dpps[i] = dcn21_dpp_create(ctx, i); + if (pool->base.dpps[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create dpps!\n"); + goto create_fail; + } + } + + for (i = 0; i < pool->base.res_cap->num_ddc; i++) { + pool->base.engines[i] = dcn21_aux_engine_create(ctx, i); + if (pool->base.engines[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC:failed to create aux engine!!\n"); + goto create_fail; + } + pool->base.hw_i2cs[i] = dcn21_i2c_hw_create(ctx, i); + if (pool->base.hw_i2cs[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC:failed to create hw i2c!!\n"); + goto create_fail; + } + pool->base.sw_i2cs[i] = NULL; + } + + for (i = 0; i < pool->base.res_cap->num_opp; i++) { + pool->base.opps[i] = dcn21_opp_create(ctx, i); + if (pool->base.opps[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create output pixel processor!\n"); + goto create_fail; + } + } + + for (i = 0; i < pool->base.res_cap->num_timing_generator; i++) { + pool->base.timing_generators[i] = dcn21_timing_generator_create( + ctx, i); + if (pool->base.timing_generators[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create tg!\n"); + goto create_fail; + } + } + + pool->base.timing_generator_count = i; + + pool->base.mpc = dcn21_mpc_create(ctx); + if (pool->base.mpc == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create mpc!\n"); + goto create_fail; + } + + pool->base.hubbub = dcn21_hubbub_create(ctx); + if (pool->base.hubbub == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create hubbub!\n"); + goto create_fail; + } + +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + for (i = 0; i < pool->base.res_cap->num_dsc; i++) { + pool->base.dscs[i] = dcn21_dsc_create(ctx, i); + if (pool->base.dscs[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create display stream compressor %d!\n", i); + goto create_fail; + } + } +#endif + + if (!dcn20_dwbc_create(ctx, &pool->base)) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create dwbc!\n"); + goto create_fail; + } + if (!dcn20_mmhubbub_create(ctx, &pool->base)) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create mcif_wb!\n"); + goto create_fail; + } + + if (!resource_construct(num_virtual_links, dc, &pool->base, + (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? + &res_create_funcs : &res_create_maximus_funcs))) + goto create_fail; + + dcn20_hw_sequencer_construct(dc); + + dc->caps.max_planes = pool->base.pipe_count; + + for (i = 0; i < dc->caps.max_planes; ++i) + dc->caps.planes[i] = plane_cap; + + dc->cap_funcs = cap_funcs; + + return true; + +create_fail: + + destruct(pool); + + return false; +} + +struct resource_pool *dcn21_create_resource_pool( + const struct dc_init_data *init_data, + struct dc *dc) +{ + struct dcn21_resource_pool *pool = + kzalloc(sizeof(struct dcn21_resource_pool), GFP_KERNEL); + + if (!pool) + return NULL; + + if (construct(init_data->num_virtual_links, dc, pool)) + return &pool->base; + + BREAK_TO_DEBUGGER(); + kfree(pool); + return NULL; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h new file mode 100644 index 000000000000..a27355171bca --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h @@ -0,0 +1,45 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DCN21_RESOURCE_H_ +#define _DCN21_RESOURCE_H_ + +#include "core_types.h" + +#define TO_DCN21_RES_POOL(pool)\ + container_of(pool, struct dcn21_resource_pool, base) + +struct dc; +struct resource_pool; +struct _vcs_dpi_display_pipe_params_st; + +struct dcn21_resource_pool { + struct resource_pool base; +}; +struct resource_pool *dcn21_create_resource_pool( + const struct dc_init_data *init_data, + struct dc *dc); + +#endif /* _DCN21_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h index 680689cab5dd..c03a441ee638 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h +++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h @@ -44,6 +44,9 @@ enum pp_smu_ver { #ifndef CONFIG_TRIM_DRM_AMD_DC_DCN2_0 PP_SMU_VER_NV, #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + PP_SMU_VER_RN, +#endif PP_SMU_VER_MAX }; @@ -246,6 +249,47 @@ struct pp_smu_funcs_nv { }; #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + +#define PP_SMU_NUM_SOCCLK_DPM_LEVELS 8 +#define PP_SMU_NUM_DCFCLK_DPM_LEVELS 4 +#define PP_SMU_NUM_FCLK_DPM_LEVELS 4 +#define PP_SMU_NUM_MEMCLK_DPM_LEVELS 4 + +struct dpm_clock { + uint32_t Freq; // In MHz + uint32_t Vol; // Millivolts with 2 fractional bits +}; + + +/* this is a copy of the structure defined in smuxx_driver_if.h*/ +struct dpm_clocks { + struct dpm_clock DcfClocks[PP_SMU_NUM_DCFCLK_DPM_LEVELS]; + struct dpm_clock SocClocks[PP_SMU_NUM_SOCCLK_DPM_LEVELS]; + struct dpm_clock FClocks[PP_SMU_NUM_FCLK_DPM_LEVELS]; + struct dpm_clock MemClocks[PP_SMU_NUM_MEMCLK_DPM_LEVELS]; +}; + + +struct pp_smu_funcs_rn { + struct pp_smu pp_smu; + + /* + * reader and writer WM's are sent together as part of one table + * + * PPSMC_MSG_SetDriverDramAddrHigh + * PPSMC_MSG_SetDriverDramAddrLow + * PPSMC_MSG_TransferTableDram2Smu + * + */ + enum pp_smu_status (*set_wm_ranges)(struct pp_smu *pp, + struct pp_smu_wm_range_sets *ranges); + + enum pp_smu_status (*get_dpm_clock_table) (struct pp_smu *pp, + struct dpm_clocks *clock_table); +}; +#endif + struct pp_smu_funcs { struct pp_smu ctx; union { @@ -253,6 +297,9 @@ struct pp_smu_funcs { #ifndef CONFIG_TRIM_DRM_AMD_DC_DCN2_0 struct pp_smu_funcs_nv nv_funcs; #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + struct pp_smu_funcs_rn rn_funcs; +#endif }; }; diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h index b426ba02b793..1a0429744630 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_services.h +++ b/drivers/gpu/drm/amd/display/dc/dm_services.h @@ -151,6 +151,7 @@ void generic_reg_wait(const struct dc_context *ctx, unsigned int delay_between_poll_us, unsigned int time_out_num_tries, const char *func_name, int line); +unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...); /* These macros need to be used with soc15 registers in order to retrieve * the actual offset. diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index 0bb7a20675c4..5b2a65b42403 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -32,26 +32,38 @@ endif dml_ccflags := -mhard-float -msse $(cc_stack_align) -CFLAGS_display_mode_lib.o := $(dml_ccflags) +ifdef CONFIG_CC_IS_CLANG +dml_ccflags += -msse2 +endif + +CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) ifdef CONFIG_DRM_AMD_DC_DCN2_0 -CFLAGS_display_mode_vba.o := $(dml_ccflags) -CFLAGS_display_mode_vba_20.o := $(dml_ccflags) -CFLAGS_display_rq_dlg_calc_20.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_ccflags) endif -ifdef CONFIG_DRM_AMD_DCN3AG -CFLAGS_display_mode_vba_3ag.o := $(dml_ccflags) +ifdef CONFIG_DRM_AMD_DC_DCN2_1 +CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_ccflags) endif -CFLAGS_dml1_display_rq_dlg_calc.o := $(dml_ccflags) -CFLAGS_display_rq_dlg_helpers.o := $(dml_ccflags) -CFLAGS_dml_common_defs.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dml_common_defs.o := $(dml_ccflags) DML = display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o \ dml_common_defs.o ifdef CONFIG_DRM_AMD_DC_DCN2_0 DML += display_mode_vba.o dcn20/display_rq_dlg_calc_20.o dcn20/display_mode_vba_20.o +DML += dcn20/display_rq_dlg_calc_20v2.o dcn20/display_mode_vba_20v2.o endif +ifdef CONFIG_DRM_AMD_DC_DCN2_1 +DML += dcn21/display_rq_dlg_calc_21.o dcn21/display_mode_vba_21.o +endif + AMD_DAL_DML = $(addprefix $(AMDDALPATH)/dc/dml/,$(DML)) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c new file mode 100644 index 000000000000..0fafd693ffb4 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c @@ -0,0 +1,5136 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../display_mode_lib.h" +#include "display_mode_vba_20v2.h" +#include "../dml_inline_defs.h" + +/* + * NOTE: + * This file is gcc-parseable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ + +#define BPP_INVALID 0 +#define BPP_BLENDED_PIPE 0xffffffff + +static double adjust_ReturnBW( + struct display_mode_lib *mode_lib, + double ReturnBW, + bool DCCEnabledAnyPlane, + double ReturnBandwidthToDCN); +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat); +static unsigned int dscComputeDelay(enum output_format_class pixelFormat); +static bool CalculateDelayAfterScaler( + struct display_mode_lib *mode_lib, + double ReturnBW, + double ReadBandwidthPlaneLuma, + double ReadBandwidthPlaneChroma, + double TotalDataReadBandwidth, + double DisplayPipeLineDeliveryTimeLuma, + double DisplayPipeLineDeliveryTimeChroma, + double DPPCLK, + double DISPCLK, + double PixelClock, + unsigned int DSCDelay, + unsigned int DPPPerPlane, + bool ScalerEnabled, + unsigned int NumberOfCursors, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int HTotal, + unsigned int SwathWidthSingleDPPY, + double BytePerPixelDETY, + double BytePerPixelDETC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler + ); +// Super monster function with some 45 argument +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double DPPCLK, + double DISPCLK, + double PixelClock, + double DCFCLKDeepSleep, + unsigned int DPPPerPlane, + unsigned int NumberOfCursors, + unsigned int VBlank, + unsigned int HTotal, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int PageTableLevels, + bool GPUVMEnable, + bool DynamicMetadataEnable, + unsigned int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatencyPixelDataOnly, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool InterlaceEnable, + bool ProgressiveToInterlaceUnitInOPP, + double DSTXAfterScaler, + double DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBW, + double *Tno_bw, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix); +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed); +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed); +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath); +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidthY, + bool GPUVMEnable, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequestsLuma, + unsigned int PDEProcessingBufIn64KBReqs, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_height, + unsigned int *meta_row_height); +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatencyPixelDataOnly, + double SREnterPlusExitTime); +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk); +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw, + double *qual_row_bw); +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double UrgentExtraLatency, + double UrgentLatencyPixelDataOnly, + unsigned int GPUVMMaxPageTableLevels, + bool GPUVMEnable, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + unsigned int ImmediateFlipBytes, + double LineTime, + double VRatio, + double Tno_bw, + double PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + double qual_row_bw, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe); +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth); + +static void dml20v2_DisplayPipeConfiguration(struct display_mode_lib *mode_lib); +static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib); + +void dml20v2_recalculate(struct display_mode_lib *mode_lib) +{ + ModeSupportAndSystemConfiguration(mode_lib); + mode_lib->vba.FabricAndDRAMBandwidth = dml_min( + mode_lib->vba.DRAMSpeed * mode_lib->vba.NumberOfChannels * mode_lib->vba.DRAMChannelWidth, + mode_lib->vba.FabricClock * mode_lib->vba.FabricDatapathToDCNDataReturn) / 1000.0; + PixelClockAdjustmentForProgressiveToInterlaceUnit(mode_lib); + dml20v2_DisplayPipeConfiguration(mode_lib); + dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation(mode_lib); +} + +static double adjust_ReturnBW( + struct display_mode_lib *mode_lib, + double ReturnBW, + bool DCCEnabledAnyPlane, + double ReturnBandwidthToDCN) +{ + double CriticalCompression; + + if (DCCEnabledAnyPlane + && ReturnBandwidthToDCN + > mode_lib->vba.DCFCLK * mode_lib->vba.ReturnBusWidth / 4.0) + ReturnBW = + dml_min( + ReturnBW, + ReturnBandwidthToDCN * 4 + * (1.0 + - mode_lib->vba.UrgentLatencyPixelDataOnly + / ((mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024 + / ReturnBandwidthToDCN + - mode_lib->vba.DCFCLK + * mode_lib->vba.ReturnBusWidth + / 4) + + mode_lib->vba.UrgentLatencyPixelDataOnly)); + + CriticalCompression = 2.0 * mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLK + * mode_lib->vba.UrgentLatencyPixelDataOnly + / (ReturnBandwidthToDCN * mode_lib->vba.UrgentLatencyPixelDataOnly + + (mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024); + + if (DCCEnabledAnyPlane && CriticalCompression > 1.0 && CriticalCompression < 4.0) + ReturnBW = + dml_min( + ReturnBW, + 4.0 * ReturnBandwidthToDCN + * (mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024 + * mode_lib->vba.ReturnBusWidth + * mode_lib->vba.DCFCLK + * mode_lib->vba.UrgentLatencyPixelDataOnly + / dml_pow( + (ReturnBandwidthToDCN + * mode_lib->vba.UrgentLatencyPixelDataOnly + + (mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024), + 2)); + + return ReturnBW; +} + +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat) +{ + // valid bpc = source bits per component in the set of {8, 10, 12} + // valid bpp = increments of 1/16 of a bit + // min = 6/7/8 in N420/N422/444, respectively + // max = such that compression is 1:1 + //valid sliceWidth = number of pixels per slice line, must be less than or equal to 5184/numSlices (or 4096/numSlices in 420 mode) + //valid numSlices = number of slices in the horiziontal direction per DSC engine in the set of {1, 2, 3, 4} + //valid pixelFormat = pixel/color format in the set of {:N444_RGB, :S422, :N422, :N420} + + // fixed value + unsigned int rcModelSize = 8192; + + // N422/N420 operate at 2 pixels per clock + unsigned int pixelsPerClock, lstall, D, initalXmitDelay, w, s, ix, wx, p, l0, a, ax, l, + Delay, pixels; + + if (pixelFormat == dm_n422 || pixelFormat == dm_420) + pixelsPerClock = 2; + // #all other modes operate at 1 pixel per clock + else + pixelsPerClock = 1; + + //initial transmit delay as per PPS + initalXmitDelay = dml_round(rcModelSize / 2.0 / bpp / pixelsPerClock); + + //compute ssm delay + if (bpc == 8) + D = 81; + else if (bpc == 10) + D = 89; + else + D = 113; + + //divide by pixel per cycle to compute slice width as seen by DSC + w = sliceWidth / pixelsPerClock; + + //422 mode has an additional cycle of delay + if (pixelFormat == dm_s422) + s = 1; + else + s = 0; + + //main calculation for the dscce + ix = initalXmitDelay + 45; + wx = (w + 2) / 3; + p = 3 * wx - w; + l0 = ix / w; + a = ix + p * l0; + ax = (a + 2) / 3 + D + 6 + 1; + l = (ax + wx - 1) / wx; + if ((ix % w) == 0 && p != 0) + lstall = 1; + else + lstall = 0; + Delay = l * wx * (numSlices - 1) + ax + s + lstall + 22; + + //dsc processes 3 pixel containers per cycle and a container can contain 1 or 2 pixels + pixels = Delay * 3 * pixelsPerClock; + return pixels; +} + +static unsigned int dscComputeDelay(enum output_format_class pixelFormat) +{ + unsigned int Delay = 0; + + if (pixelFormat == dm_420) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc gets pixels every other cycle + Delay = Delay + 2; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc gets pixels every other cycle + Delay = Delay + 13; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc gets pixels every other cycle + Delay = Delay + 3; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else if (pixelFormat == dm_n422) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 1; + // dscc - input deserializer + Delay = Delay + 5; + // dscc - input cdc fifo + Delay = Delay + 25; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 10; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc - output serializer + Delay = Delay + 1; + // dscc - cdc uncertainty + Delay = Delay + 2; + // sft + Delay = Delay + 1; + } + + return Delay; +} + +static bool CalculateDelayAfterScaler( + struct display_mode_lib *mode_lib, + double ReturnBW, + double ReadBandwidthPlaneLuma, + double ReadBandwidthPlaneChroma, + double TotalDataReadBandwidth, + double DisplayPipeLineDeliveryTimeLuma, + double DisplayPipeLineDeliveryTimeChroma, + double DPPCLK, + double DISPCLK, + double PixelClock, + unsigned int DSCDelay, + unsigned int DPPPerPlane, + bool ScalerEnabled, + unsigned int NumberOfCursors, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int HTotal, + unsigned int SwathWidthSingleDPPY, + double BytePerPixelDETY, + double BytePerPixelDETC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler + ) +{ + unsigned int DPPCycles, DISPCLKCycles; + double DataFabricLineDeliveryTimeLuma; + double DataFabricLineDeliveryTimeChroma; + double DSTTotalPixelsAfterScaler; + + DataFabricLineDeliveryTimeLuma = SwathWidthSingleDPPY * SwathHeightY * dml_ceil(BytePerPixelDETY, 1) / (mode_lib->vba.ReturnBW * ReadBandwidthPlaneLuma / TotalDataReadBandwidth); + mode_lib->vba.LastPixelOfLineExtraWatermark = dml_max(mode_lib->vba.LastPixelOfLineExtraWatermark, DataFabricLineDeliveryTimeLuma - DisplayPipeLineDeliveryTimeLuma); + + if (BytePerPixelDETC != 0) { + DataFabricLineDeliveryTimeChroma = SwathWidthSingleDPPY / 2 * SwathHeightC * dml_ceil(BytePerPixelDETC, 2) / (mode_lib->vba.ReturnBW * ReadBandwidthPlaneChroma / TotalDataReadBandwidth); + mode_lib->vba.LastPixelOfLineExtraWatermark = dml_max(mode_lib->vba.LastPixelOfLineExtraWatermark, DataFabricLineDeliveryTimeChroma - DisplayPipeLineDeliveryTimeChroma); + } + + if (ScalerEnabled) + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCL; + else + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCLLBOnly; + + DPPCycles = DPPCycles + DPPCLKDelayCNVCFormater + NumberOfCursors * DPPCLKDelayCNVCCursor; + + DISPCLKCycles = DISPCLKDelaySubtotal; + + if (DPPCLK == 0.0 || DISPCLK == 0.0) + return true; + + *DSTXAfterScaler = DPPCycles * PixelClock / DPPCLK + DISPCLKCycles * PixelClock / DISPCLK + + DSCDelay; + + if (DPPPerPlane > 1) + *DSTXAfterScaler = *DSTXAfterScaler + ScalerRecoutWidth; + + if (OutputFormat == dm_420 || (Interlace && ProgressiveToInterlaceUnitInOPP)) + *DSTYAfterScaler = 1; + else + *DSTYAfterScaler = 0; + + DSTTotalPixelsAfterScaler = ((double) (*DSTYAfterScaler * HTotal)) + *DSTXAfterScaler; + *DSTYAfterScaler = dml_floor(DSTTotalPixelsAfterScaler / HTotal, 1); + *DSTXAfterScaler = DSTTotalPixelsAfterScaler - ((double) (*DSTYAfterScaler * HTotal)); + + return true; +} + +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double DPPCLK, + double DISPCLK, + double PixelClock, + double DCFCLKDeepSleep, + unsigned int DPPPerPlane, + unsigned int NumberOfCursors, + unsigned int VBlank, + unsigned int HTotal, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int PageTableLevels, + bool GPUVMEnable, + bool DynamicMetadataEnable, + unsigned int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatencyPixelDataOnly, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool InterlaceEnable, + bool ProgressiveToInterlaceUnitInOPP, + double DSTXAfterScaler, + double DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBW, + double *Tno_bw, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix) +{ + bool MyError = false; + double TotalRepeaterDelayTime; + double Tdm, LineTime, Tsetup; + double dst_y_prefetch_equ; + double Tsw_oto; + double prefetch_bw_oto; + double Tvm_oto; + double Tr0_oto; + double Tpre_oto; + double dst_y_prefetch_oto; + double TimeForFetchingMetaPTE = 0; + double TimeForFetchingRowInVBlank = 0; + double LinesToRequestPrefetchPixelData = 0; + + *VUpdateOffsetPix = dml_ceil(HTotal / 4.0, 1); + TotalRepeaterDelayTime = MaxInterDCNTileRepeaters * (2.0 / DPPCLK + 3.0 / DISPCLK); + *VUpdateWidthPix = (14.0 / DCFCLKDeepSleep + 12.0 / DPPCLK + TotalRepeaterDelayTime) + * PixelClock; + + *VReadyOffsetPix = dml_max( + 150.0 / DPPCLK, + TotalRepeaterDelayTime + 20.0 / DCFCLKDeepSleep + 10.0 / DPPCLK) + * PixelClock; + + Tsetup = (double) (*VUpdateOffsetPix + *VUpdateWidthPix + *VReadyOffsetPix) / PixelClock; + + LineTime = (double) HTotal / PixelClock; + + if (DynamicMetadataEnable) { + double Tdmbf, Tdmec, Tdmsks; + + Tdm = dml_max(0.0, UrgentExtraLatency - TCalc); + Tdmbf = DynamicMetadataTransmittedBytes / 4.0 / DISPCLK; + Tdmec = LineTime; + if (DynamicMetadataLinesBeforeActiveRequired == 0) + Tdmsks = VBlank * LineTime / 2.0; + else + Tdmsks = DynamicMetadataLinesBeforeActiveRequired * LineTime; + if (InterlaceEnable && !ProgressiveToInterlaceUnitInOPP) + Tdmsks = Tdmsks / 2; + if (VStartup * LineTime + < Tsetup + TWait + UrgentExtraLatency + Tdmbf + Tdmec + Tdmsks) { + MyError = true; + } + } else + Tdm = 0; + + if (GPUVMEnable) { + if (PageTableLevels == 4) + *Tno_bw = UrgentExtraLatency + UrgentLatencyPixelDataOnly; + else if (PageTableLevels == 3) + *Tno_bw = UrgentExtraLatency; + else + *Tno_bw = 0; + } else if (DCCEnable) + *Tno_bw = LineTime; + else + *Tno_bw = LineTime / 4; + + dst_y_prefetch_equ = VStartup - dml_max(TCalc + TWait, XFCRemoteSurfaceFlipDelay) / LineTime + - (Tsetup + Tdm) / LineTime + - (DSTYAfterScaler + DSTXAfterScaler / HTotal); + + Tsw_oto = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime; + + prefetch_bw_oto = (MetaRowByte + PixelPTEBytesPerRow + + PrefetchSourceLinesY * SwathWidthY * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * SwathWidthY / 2 * dml_ceil(BytePerPixelDETC, 2)) + / Tsw_oto; + + if (GPUVMEnable == true) { + Tvm_oto = + dml_max( + *Tno_bw + PDEAndMetaPTEBytesFrame / prefetch_bw_oto, + dml_max( + UrgentExtraLatency + + UrgentLatencyPixelDataOnly + * (PageTableLevels + - 1), + LineTime / 4.0)); + } else + Tvm_oto = LineTime / 4.0; + + if ((GPUVMEnable == true || DCCEnable == true)) { + Tr0_oto = dml_max( + (MetaRowByte + PixelPTEBytesPerRow) / prefetch_bw_oto, + dml_max(UrgentLatencyPixelDataOnly, dml_max(LineTime - Tvm_oto, LineTime / 4))); + } else + Tr0_oto = LineTime - Tvm_oto; + + Tpre_oto = Tvm_oto + Tr0_oto + Tsw_oto; + + dst_y_prefetch_oto = Tpre_oto / LineTime; + + if (dst_y_prefetch_oto < dst_y_prefetch_equ) + *DestinationLinesForPrefetch = dst_y_prefetch_oto; + else + *DestinationLinesForPrefetch = dst_y_prefetch_equ; + + *DestinationLinesForPrefetch = dml_floor(4.0 * (*DestinationLinesForPrefetch + 0.125), 1) + / 4; + + dml_print("DML: VStartup: %d\n", VStartup); + dml_print("DML: TCalc: %f\n", TCalc); + dml_print("DML: TWait: %f\n", TWait); + dml_print("DML: XFCRemoteSurfaceFlipDelay: %f\n", XFCRemoteSurfaceFlipDelay); + dml_print("DML: LineTime: %f\n", LineTime); + dml_print("DML: Tsetup: %f\n", Tsetup); + dml_print("DML: Tdm: %f\n", Tdm); + dml_print("DML: DSTYAfterScaler: %f\n", DSTYAfterScaler); + dml_print("DML: DSTXAfterScaler: %f\n", DSTXAfterScaler); + dml_print("DML: HTotal: %d\n", HTotal); + + *PrefetchBandwidth = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBW = 0; + if (*DestinationLinesForPrefetch > 1) { + *PrefetchBandwidth = (PDEAndMetaPTEBytesFrame + 2 * MetaRowByte + + 2 * PixelPTEBytesPerRow + + PrefetchSourceLinesY * SwathWidthY * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * SwathWidthY / 2 + * dml_ceil(BytePerPixelDETC, 2)) + / (*DestinationLinesForPrefetch * LineTime - *Tno_bw); + if (GPUVMEnable) { + TimeForFetchingMetaPTE = + dml_max( + *Tno_bw + + (double) PDEAndMetaPTEBytesFrame + / *PrefetchBandwidth, + dml_max( + UrgentExtraLatency + + UrgentLatencyPixelDataOnly + * (PageTableLevels + - 1), + LineTime / 4)); + } else { + if (NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingMetaPTE = LineTime / 4; + else + TimeForFetchingMetaPTE = 0.0; + } + + if ((GPUVMEnable == true || DCCEnable == true)) { + TimeForFetchingRowInVBlank = + dml_max( + (MetaRowByte + PixelPTEBytesPerRow) + / *PrefetchBandwidth, + dml_max( + UrgentLatencyPixelDataOnly, + dml_max( + LineTime + - TimeForFetchingMetaPTE, + LineTime + / 4.0))); + } else { + if (NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingRowInVBlank = LineTime - TimeForFetchingMetaPTE; + else + TimeForFetchingRowInVBlank = 0.0; + } + + *DestinationLinesToRequestVMInVBlank = dml_floor( + 4.0 * (TimeForFetchingMetaPTE / LineTime + 0.125), + 1) / 4.0; + + *DestinationLinesToRequestRowInVBlank = dml_floor( + 4.0 * (TimeForFetchingRowInVBlank / LineTime + 0.125), + 1) / 4.0; + + LinesToRequestPrefetchPixelData = + *DestinationLinesForPrefetch + - ((NumberOfCursors > 0 || GPUVMEnable + || DCCEnable) ? + (*DestinationLinesToRequestVMInVBlank + + *DestinationLinesToRequestRowInVBlank) : + 0.0); + + if (LinesToRequestPrefetchPixelData > 0) { + + *VRatioPrefetchY = (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + if ((SwathHeightY > 4) && (VInitPreFillY > 3)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillY - 3.0) / 2.0) { + *VRatioPrefetchY = + dml_max( + (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData, + (double) MaxNumSwathY + * SwathHeightY + / (LinesToRequestPrefetchPixelData + - (VInitPreFillY + - 3.0) + / 2.0)); + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + } else { + MyError = true; + *VRatioPrefetchY = 0; + } + } + + *VRatioPrefetchC = (double) PrefetchSourceLinesC + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + + if ((SwathHeightC > 4)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillC - 3.0) / 2.0) { + *VRatioPrefetchC = + dml_max( + *VRatioPrefetchC, + (double) MaxNumSwathC + * SwathHeightC + / (LinesToRequestPrefetchPixelData + - (VInitPreFillC + - 3.0) + / 2.0)); + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + } else { + MyError = true; + *VRatioPrefetchC = 0; + } + } + + *RequiredPrefetchPixDataBW = + DPPPerPlane + * ((double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData + * dml_ceil( + BytePerPixelDETY, + 1) + + (double) PrefetchSourceLinesC + / LinesToRequestPrefetchPixelData + * dml_ceil( + BytePerPixelDETC, + 2) + / 2) + * SwathWidthY / LineTime; + } else { + MyError = true; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBW = 0; + } + + } else { + MyError = true; + } + + if (MyError) { + *PrefetchBandwidth = 0; + TimeForFetchingMetaPTE = 0; + TimeForFetchingRowInVBlank = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *DestinationLinesForPrefetch = 0; + LinesToRequestPrefetchPixelData = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBW = 0; + } + + return MyError; +} + +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_floor(VCOSpeed * 4 / Clock, 1); +} + +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_ceil(VCOSpeed * 4 / Clock, 1); +} + +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath) +{ + unsigned int MaxPartialSwath; + + if (ProgressiveToInterlaceUnitInOPP) + *VInitPreFill = dml_floor((VRatio + vtaps + 1) / 2.0, 1); + else + *VInitPreFill = dml_floor((VRatio + vtaps + 1 + Interlace * 0.5 * VRatio) / 2.0, 1); + + if (!mode_lib->vba.IgnoreViewportPositioning) { + + *MaxNumSwath = dml_ceil((*VInitPreFill - 1.0) / SwathHeight, 1) + 1.0; + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 2) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 2) + % SwathHeight; + MaxPartialSwath = dml_max(1U, MaxPartialSwath); + + } else { + + if (ViewportYStart != 0) + dml_print( + "WARNING DML: using viewport y position of 0 even though actual viewport y position is non-zero in prefetch source lines calculation\n"); + + *MaxNumSwath = dml_ceil(*VInitPreFill / SwathHeight, 1); + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 1) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 1) + % SwathHeight; + } + + return *MaxNumSwath * SwathHeight + MaxPartialSwath; +} + +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidth, + bool GPUVMEnable, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequestsLuma, + unsigned int PDEProcessingBufIn64KBReqs, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_height, + unsigned int *meta_row_height) +{ + unsigned int MetaRequestHeight; + unsigned int MetaRequestWidth; + unsigned int MetaSurfWidth; + unsigned int MetaSurfHeight; + unsigned int MPDEBytesFrame; + unsigned int MetaPTEBytesFrame; + unsigned int DCCMetaSurfaceBytes; + + unsigned int MacroTileSizeBytes; + unsigned int MacroTileHeight; + unsigned int DPDE0BytesFrame; + unsigned int ExtraDPDEBytesFrame; + unsigned int PDEAndMetaPTEBytesFrame; + + if (DCCEnable == true) { + MetaRequestHeight = 8 * BlockHeight256Bytes; + MetaRequestWidth = 8 * BlockWidth256Bytes; + if (ScanDirection == dm_horz) { + *meta_row_height = MetaRequestHeight; + MetaSurfWidth = dml_ceil((double) SwathWidth - 1, MetaRequestWidth) + + MetaRequestWidth; + *MetaRowByte = MetaSurfWidth * MetaRequestHeight * BytePerPixel / 256.0; + } else { + *meta_row_height = MetaRequestWidth; + MetaSurfHeight = dml_ceil((double) SwathWidth - 1, MetaRequestHeight) + + MetaRequestHeight; + *MetaRowByte = MetaSurfHeight * MetaRequestWidth * BytePerPixel / 256.0; + } + if (ScanDirection == dm_horz) { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil(ViewportHeight - 1, 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } else { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil( + (double) ViewportHeight - 1, + 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } + if (GPUVMEnable == true) { + MetaPTEBytesFrame = (dml_ceil( + (double) (DCCMetaSurfaceBytes - VMMPageSize) + / (8 * VMMPageSize), + 1) + 1) * 64; + MPDEBytesFrame = 128 * (mode_lib->vba.GPUVMMaxPageTableLevels - 1); + } else { + MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + } + } else { + MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + *MetaRowByte = 0; + } + + if (SurfaceTiling == dm_sw_linear || SurfaceTiling == dm_sw_gfx7_2d_thin_gl || SurfaceTiling == dm_sw_gfx7_2d_thin_lvp) { + MacroTileSizeBytes = 256; + MacroTileHeight = BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_4kb_s || SurfaceTiling == dm_sw_4kb_s_x + || SurfaceTiling == dm_sw_4kb_d || SurfaceTiling == dm_sw_4kb_d_x) { + MacroTileSizeBytes = 4096; + MacroTileHeight = 4 * BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_64kb_s || SurfaceTiling == dm_sw_64kb_s_t + || SurfaceTiling == dm_sw_64kb_s_x || SurfaceTiling == dm_sw_64kb_d + || SurfaceTiling == dm_sw_64kb_d_t || SurfaceTiling == dm_sw_64kb_d_x + || SurfaceTiling == dm_sw_64kb_r_x) { + MacroTileSizeBytes = 65536; + MacroTileHeight = 16 * BlockHeight256Bytes; + } else { + MacroTileSizeBytes = 262144; + MacroTileHeight = 32 * BlockHeight256Bytes; + } + *MacroTileWidth = MacroTileSizeBytes / BytePerPixel / MacroTileHeight; + + if (GPUVMEnable == true && mode_lib->vba.GPUVMMaxPageTableLevels > 1) { + if (ScanDirection == dm_horz) { + DPDE0BytesFrame = + 64 + * (dml_ceil( + ((Pitch + * (dml_ceil( + ViewportHeight + - 1, + MacroTileHeight) + + MacroTileHeight) + * BytePerPixel) + - MacroTileSizeBytes) + / (8 + * 2097152), + 1) + 1); + } else { + DPDE0BytesFrame = + 64 + * (dml_ceil( + ((Pitch + * (dml_ceil( + (double) SwathWidth + - 1, + MacroTileHeight) + + MacroTileHeight) + * BytePerPixel) + - MacroTileSizeBytes) + / (8 + * 2097152), + 1) + 1); + } + ExtraDPDEBytesFrame = 128 * (mode_lib->vba.GPUVMMaxPageTableLevels - 2); + } else { + DPDE0BytesFrame = 0; + ExtraDPDEBytesFrame = 0; + } + + PDEAndMetaPTEBytesFrame = MetaPTEBytesFrame + MPDEBytesFrame + DPDE0BytesFrame + + ExtraDPDEBytesFrame; + + if (GPUVMEnable == true) { + unsigned int PTERequestSize; + unsigned int PixelPTEReqHeight; + unsigned int PixelPTEReqWidth; + double FractionOfPTEReturnDrop; + unsigned int EffectivePDEProcessingBufIn64KBReqs; + + if (SurfaceTiling == dm_sw_linear) { + PixelPTEReqHeight = 1; + PixelPTEReqWidth = 8.0 * VMMPageSize / BytePerPixel; + PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } else if (MacroTileSizeBytes == 4096) { + PixelPTEReqHeight = MacroTileHeight; + PixelPTEReqWidth = 8 * *MacroTileWidth; + PTERequestSize = 64; + if (ScanDirection == dm_horz) + FractionOfPTEReturnDrop = 0; + else + FractionOfPTEReturnDrop = 7 / 8; + } else if (VMMPageSize == 4096 && MacroTileSizeBytes > 4096) { + PixelPTEReqHeight = 16 * BlockHeight256Bytes; + PixelPTEReqWidth = 16 * BlockWidth256Bytes; + PTERequestSize = 128; + FractionOfPTEReturnDrop = 0; + } else { + PixelPTEReqHeight = MacroTileHeight; + PixelPTEReqWidth = 8 * *MacroTileWidth; + PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } + + if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) + EffectivePDEProcessingBufIn64KBReqs = PDEProcessingBufIn64KBReqs / 2; + else + EffectivePDEProcessingBufIn64KBReqs = PDEProcessingBufIn64KBReqs; + + if (SurfaceTiling == dm_sw_linear) { + *dpte_row_height = + dml_min( + 128, + 1 + << (unsigned int) dml_floor( + dml_log2( + dml_min( + (double) PTEBufferSizeInRequestsLuma + * PixelPTEReqWidth, + EffectivePDEProcessingBufIn64KBReqs + * 65536.0 + / BytePerPixel) + / Pitch), + 1)); + *PixelPTEBytesPerRow = PTERequestSize + * (dml_ceil( + (double) (Pitch * *dpte_row_height - 1) + / PixelPTEReqWidth, + 1) + 1); + } else if (ScanDirection == dm_horz) { + *dpte_row_height = PixelPTEReqHeight; + *PixelPTEBytesPerRow = PTERequestSize + * (dml_ceil(((double) SwathWidth - 1) / PixelPTEReqWidth, 1) + + 1); + } else { + *dpte_row_height = dml_min(PixelPTEReqWidth, *MacroTileWidth); + *PixelPTEBytesPerRow = PTERequestSize + * (dml_ceil( + ((double) SwathWidth - 1) + / PixelPTEReqHeight, + 1) + 1); + } + if (*PixelPTEBytesPerRow * (1 - FractionOfPTEReturnDrop) + <= 64 * PTEBufferSizeInRequestsLuma) { + *PTEBufferSizeNotExceeded = true; + } else { + *PTEBufferSizeNotExceeded = false; + } + } else { + *PixelPTEBytesPerRow = 0; + *PTEBufferSizeNotExceeded = true; + } + + return PDEAndMetaPTEBytesFrame; +} + +static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib) +{ + unsigned int j, k; + + mode_lib->vba.WritebackDISPCLK = 0.0; + mode_lib->vba.DISPCLKWithRamping = 0; + mode_lib->vba.DISPCLKWithoutRamping = 0; + mode_lib->vba.GlobalDPPCLK = 0.0; + + // dml_ml->vba.DISPCLK and dml_ml->vba.DPPCLK Calculation + // + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k]) { + mode_lib->vba.WritebackDISPCLK = + dml_max( + mode_lib->vba.WritebackDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.HRatio[k] > 1) { + mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1)); + } else { + mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + + mode_lib->vba.DPPCLKUsingSingleDPPLuma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k], + 1.0)); + + if ((mode_lib->vba.htaps[k] > 6 || mode_lib->vba.vtaps[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPLuma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPLuma = 2 * mode_lib->vba.PixelClock[k]; + } + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] = 0.0; + mode_lib->vba.DPPCLKUsingSingleDPP[k] = + mode_lib->vba.DPPCLKUsingSingleDPPLuma; + } else { + if (mode_lib->vba.HRatio[k] > 1) { + mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + mode_lib->vba.DPPCLKUsingSingleDPPChroma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4 + / mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k], + 1.0)); + + if ((mode_lib->vba.HTAPsChroma[k] > 6 || mode_lib->vba.VTAPsChroma[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPChroma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPChroma = 2 + * mode_lib->vba.PixelClock[k]; + } + + mode_lib->vba.DPPCLKUsingSingleDPP[k] = dml_max( + mode_lib->vba.DPPCLKUsingSingleDPPLuma, + mode_lib->vba.DPPCLKUsingSingleDPPChroma); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] != k) + continue; + if (mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } else if (!mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } + } + + mode_lib->vba.DISPCLKWithRamping = dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.WritebackDISPCLK); + mode_lib->vba.DISPCLKWithoutRamping = dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.WritebackDISPCLK); + + ASSERT(mode_lib->vba.DISPCLKDPPCLKVCOSpeed != 0); + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDispclkRoundedToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.soc.clock_limits[mode_lib->vba.soc.num_states].dispclk_mhz, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + if (mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity; + } else if (mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = mode_lib->vba.MaxDispclkRoundedToDFSGranularity; + } else { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity; + } + DTRACE(" dispclk_mhz (calculated) = %f", mode_lib->vba.DISPCLK_calculated); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.DPPPerPlane[k] == 0) { + mode_lib->vba.DPPCLK_calculated[k] = 0; + } else { + mode_lib->vba.DPPCLK_calculated[k] = mode_lib->vba.DPPCLKUsingSingleDPP[k] + / mode_lib->vba.DPPPerPlane[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100); + } + mode_lib->vba.GlobalDPPCLK = dml_max( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DPPCLK_calculated[k]); + } + mode_lib->vba.GlobalDPPCLK = RoundToDFSGranularityUp( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.DPPCLK_calculated[k] = mode_lib->vba.GlobalDPPCLK / 255 + * dml_ceil( + mode_lib->vba.DPPCLK_calculated[k] * 255 + / mode_lib->vba.GlobalDPPCLK, + 1); + DTRACE(" dppclk_mhz[%i] (calculated) = %f", k, mode_lib->vba.DPPCLK_calculated[k]); + } + + // Urgent Watermark + mode_lib->vba.DCCEnabledAnyPlane = false; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + if (mode_lib->vba.DCCEnable[k]) + mode_lib->vba.DCCEnabledAnyPlane = true; + + mode_lib->vba.ReturnBandwidthToDCN = dml_min( + mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLK, + mode_lib->vba.FabricAndDRAMBandwidth * 1000) + * mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly / 100; + + mode_lib->vba.ReturnBW = mode_lib->vba.ReturnBandwidthToDCN; + mode_lib->vba.ReturnBW = adjust_ReturnBW( + mode_lib, + mode_lib->vba.ReturnBW, + mode_lib->vba.DCCEnabledAnyPlane, + mode_lib->vba.ReturnBandwidthToDCN); + + // Let's do this calculation again?? + mode_lib->vba.ReturnBandwidthToDCN = dml_min( + mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLK, + mode_lib->vba.FabricAndDRAMBandwidth * 1000); + mode_lib->vba.ReturnBW = adjust_ReturnBW( + mode_lib, + mode_lib->vba.ReturnBW, + mode_lib->vba.DCCEnabledAnyPlane, + mode_lib->vba.ReturnBandwidthToDCN); + + DTRACE(" dcfclk_mhz = %f", mode_lib->vba.DCFCLK); + DTRACE(" return_bw_to_dcn = %f", mode_lib->vba.ReturnBandwidthToDCN); + DTRACE(" return_bus_bw = %f", mode_lib->vba.ReturnBW); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourceScan[k] == dm_horz) + mode_lib->vba.SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportWidth[k]; + else + mode_lib->vba.SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportHeight[k]; + + if (mode_lib->vba.ODMCombineEnabled[k] == true) + MainPlaneDoesODMCombine = true; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[j] == true) + MainPlaneDoesODMCombine = true; + + if (MainPlaneDoesODMCombine == true) + mode_lib->vba.SwathWidthY[k] = dml_min( + (double) mode_lib->vba.SwathWidthSingleDPPY[k], + dml_round( + mode_lib->vba.HActive[k] / 2.0 + * mode_lib->vba.HRatio[k])); + else { + if (mode_lib->vba.DPPPerPlane[k] == 0) { + mode_lib->vba.SwathWidthY[k] = 0; + } else { + mode_lib->vba.SwathWidthY[k] = mode_lib->vba.SwathWidthSingleDPPY[k] + / mode_lib->vba.DPPPerPlane[k]; + } + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + mode_lib->vba.BytePerPixelDETY[k] = 8; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + mode_lib->vba.BytePerPixelDETY[k] = 4; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + mode_lib->vba.BytePerPixelDETY[k] = 2; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8) { + mode_lib->vba.BytePerPixelDETY[k] = 1; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + mode_lib->vba.BytePerPixelDETY[k] = 1; + mode_lib->vba.BytePerPixelDETC[k] = 2; + } else { // dm_420_10 + mode_lib->vba.BytePerPixelDETY[k] = 4.0 / 3.0; + mode_lib->vba.BytePerPixelDETC[k] = 8.0 / 3.0; + } + } + + mode_lib->vba.TotalDataReadBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.ReadBandwidthPlaneLuma[k] = mode_lib->vba.SwathWidthSingleDPPY[k] + * dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k]; + mode_lib->vba.ReadBandwidthPlaneChroma[k] = mode_lib->vba.SwathWidthSingleDPPY[k] + / 2 * dml_ceil(mode_lib->vba.BytePerPixelDETC[k], 2) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k] / 2; + DTRACE( + " read_bw[%i] = %fBps", + k, + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k]); + mode_lib->vba.TotalDataReadBandwidth += mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k]; + } + + mode_lib->vba.TotalDCCActiveDPP = 0; + mode_lib->vba.TotalActiveDPP = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalActiveDPP = mode_lib->vba.TotalActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + if (mode_lib->vba.DCCEnable[k]) + mode_lib->vba.TotalDCCActiveDPP = mode_lib->vba.TotalDCCActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + } + + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency = + (mode_lib->vba.RoundTripPingLatencyCycles + 32) / mode_lib->vba.DCFCLK + + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly + * mode_lib->vba.NumberOfChannels + / mode_lib->vba.ReturnBW; + + mode_lib->vba.LastPixelOfLineExtraWatermark = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.VRatio[k] <= 1.0) + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k] = + (double) mode_lib->vba.SwathWidthY[k] + * mode_lib->vba.DPPPerPlane[k] + / mode_lib->vba.HRatio[k] + / mode_lib->vba.PixelClock[k]; + else + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k] = + (double) mode_lib->vba.SwathWidthY[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / mode_lib->vba.DPPCLK[k]; + + if (mode_lib->vba.BytePerPixelDETC[k] == 0) + mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k] = 0.0; + else if (mode_lib->vba.VRatio[k] / 2.0 <= 1.0) + mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k] = + mode_lib->vba.SwathWidthY[k] / 2.0 + * mode_lib->vba.DPPPerPlane[k] + / (mode_lib->vba.HRatio[k] / 2.0) + / mode_lib->vba.PixelClock[k]; + else + mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k] = + mode_lib->vba.SwathWidthY[k] / 2.0 + / mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] + / mode_lib->vba.DPPCLK[k]; + } + + mode_lib->vba.UrgentExtraLatency = mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency + + (mode_lib->vba.TotalActiveDPP * mode_lib->vba.PixelChunkSizeInKByte + + mode_lib->vba.TotalDCCActiveDPP + * mode_lib->vba.MetaChunkSize) * 1024.0 + / mode_lib->vba.ReturnBW; + + if (mode_lib->vba.GPUVMEnable) + mode_lib->vba.UrgentExtraLatency += mode_lib->vba.TotalActiveDPP + * mode_lib->vba.PTEGroupSize / mode_lib->vba.ReturnBW; + + mode_lib->vba.UrgentWatermark = mode_lib->vba.UrgentLatencyPixelDataOnly + + mode_lib->vba.LastPixelOfLineExtraWatermark + + mode_lib->vba.UrgentExtraLatency; + + DTRACE(" urgent_extra_latency = %fus", mode_lib->vba.UrgentExtraLatency); + DTRACE(" wm_urgent = %fus", mode_lib->vba.UrgentWatermark); + + mode_lib->vba.UrgentLatency = mode_lib->vba.UrgentLatencyPixelDataOnly; + + mode_lib->vba.TotalActiveWriteback = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k]) + mode_lib->vba.TotalActiveWriteback = mode_lib->vba.TotalActiveWriteback + mode_lib->vba.ActiveWritebacksPerPlane[k]; + } + + if (mode_lib->vba.TotalActiveWriteback <= 1) + mode_lib->vba.WritebackUrgentWatermark = mode_lib->vba.WritebackLatency; + else + mode_lib->vba.WritebackUrgentWatermark = mode_lib->vba.WritebackLatency + + mode_lib->vba.WritebackChunkSize * 1024.0 / 32 + / mode_lib->vba.SOCCLK; + + DTRACE(" wm_wb_urgent = %fus", mode_lib->vba.WritebackUrgentWatermark); + + // NB P-State/DRAM Clock Change Watermark + mode_lib->vba.DRAMClockChangeWatermark = mode_lib->vba.DRAMClockChangeLatency + + mode_lib->vba.UrgentWatermark; + + DTRACE(" wm_pstate_change = %fus", mode_lib->vba.DRAMClockChangeWatermark); + + DTRACE(" calculating wb pstate watermark"); + DTRACE(" total wb outputs %d", mode_lib->vba.TotalActiveWriteback); + DTRACE(" socclk frequency %f Mhz", mode_lib->vba.SOCCLK); + + if (mode_lib->vba.TotalActiveWriteback <= 1) + mode_lib->vba.WritebackDRAMClockChangeWatermark = + mode_lib->vba.DRAMClockChangeLatency + + mode_lib->vba.WritebackLatency; + else + mode_lib->vba.WritebackDRAMClockChangeWatermark = + mode_lib->vba.DRAMClockChangeLatency + + mode_lib->vba.WritebackLatency + + mode_lib->vba.WritebackChunkSize * 1024.0 / 32 + / mode_lib->vba.SOCCLK; + + DTRACE(" wm_wb_pstate %fus", mode_lib->vba.WritebackDRAMClockChangeWatermark); + + // Stutter Efficiency + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.LinesInDETY[k] = mode_lib->vba.DETBufferSizeY[k] + / mode_lib->vba.BytePerPixelDETY[k] / mode_lib->vba.SwathWidthY[k]; + mode_lib->vba.LinesInDETYRoundedDownToSwath[k] = dml_floor( + mode_lib->vba.LinesInDETY[k], + mode_lib->vba.SwathHeightY[k]); + mode_lib->vba.FullDETBufferingTimeY[k] = + mode_lib->vba.LinesInDETYRoundedDownToSwath[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k]; + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + mode_lib->vba.LinesInDETC[k] = mode_lib->vba.DETBufferSizeC[k] + / mode_lib->vba.BytePerPixelDETC[k] + / (mode_lib->vba.SwathWidthY[k] / 2); + mode_lib->vba.LinesInDETCRoundedDownToSwath[k] = dml_floor( + mode_lib->vba.LinesInDETC[k], + mode_lib->vba.SwathHeightC[k]); + mode_lib->vba.FullDETBufferingTimeC[k] = + mode_lib->vba.LinesInDETCRoundedDownToSwath[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / (mode_lib->vba.VRatio[k] / 2); + } else { + mode_lib->vba.LinesInDETC[k] = 0; + mode_lib->vba.LinesInDETCRoundedDownToSwath[k] = 0; + mode_lib->vba.FullDETBufferingTimeC[k] = 999999; + } + } + + mode_lib->vba.MinFullDETBufferingTime = 999999.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.FullDETBufferingTimeY[k] + < mode_lib->vba.MinFullDETBufferingTime) { + mode_lib->vba.MinFullDETBufferingTime = + mode_lib->vba.FullDETBufferingTimeY[k]; + mode_lib->vba.FrameTimeForMinFullDETBufferingTime = + (double) mode_lib->vba.VTotal[k] * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } + if (mode_lib->vba.FullDETBufferingTimeC[k] + < mode_lib->vba.MinFullDETBufferingTime) { + mode_lib->vba.MinFullDETBufferingTime = + mode_lib->vba.FullDETBufferingTimeC[k]; + mode_lib->vba.FrameTimeForMinFullDETBufferingTime = + (double) mode_lib->vba.VTotal[k] * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } + } + + mode_lib->vba.AverageReadBandwidthGBytePerSecond = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.DCCEnable[k]) { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / mode_lib->vba.DCCRate[k] + / 1000 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / mode_lib->vba.DCCRate[k] + / 1000; + } else { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / 1000 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / 1000; + } + if (mode_lib->vba.DCCEnable[k]) { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / 1000 / 256 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / 1000 / 256; + } + if (mode_lib->vba.GPUVMEnable) { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / 1000 / 512 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / 1000 / 512; + } + } + + mode_lib->vba.PartOfBurstThatFitsInROB = + dml_min( + mode_lib->vba.MinFullDETBufferingTime + * mode_lib->vba.TotalDataReadBandwidth, + mode_lib->vba.ROBBufferSizeInKByte * 1024 + * mode_lib->vba.TotalDataReadBandwidth + / (mode_lib->vba.AverageReadBandwidthGBytePerSecond + * 1000)); + mode_lib->vba.StutterBurstTime = mode_lib->vba.PartOfBurstThatFitsInROB + * (mode_lib->vba.AverageReadBandwidthGBytePerSecond * 1000) + / mode_lib->vba.TotalDataReadBandwidth / mode_lib->vba.ReturnBW + + (mode_lib->vba.MinFullDETBufferingTime + * mode_lib->vba.TotalDataReadBandwidth + - mode_lib->vba.PartOfBurstThatFitsInROB) + / (mode_lib->vba.DCFCLK * 64); + if (mode_lib->vba.TotalActiveWriteback == 0) { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = (1 + - (mode_lib->vba.SRExitTime + mode_lib->vba.StutterBurstTime) + / mode_lib->vba.MinFullDETBufferingTime) * 100; + } else { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = 0; + } + + mode_lib->vba.SmallestVBlank = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { + mode_lib->vba.VBlankTime = (double) (mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k]) * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.VBlankTime = 0; + } + mode_lib->vba.SmallestVBlank = dml_min( + mode_lib->vba.SmallestVBlank, + mode_lib->vba.VBlankTime); + } + + mode_lib->vba.StutterEfficiency = (mode_lib->vba.StutterEfficiencyNotIncludingVBlank / 100 + * (mode_lib->vba.FrameTimeForMinFullDETBufferingTime + - mode_lib->vba.SmallestVBlank) + + mode_lib->vba.SmallestVBlank) + / mode_lib->vba.FrameTimeForMinFullDETBufferingTime * 100; + + // dml_ml->vba.DCFCLK Deep Sleep + mode_lib->vba.DCFCLKDeepSleep = 8.0; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; k++) { + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = + dml_max( + 1.1 * mode_lib->vba.SwathWidthY[k] + * dml_ceil( + mode_lib->vba.BytePerPixelDETY[k], + 1) / 32 + / mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k], + 1.1 * mode_lib->vba.SwathWidthY[k] / 2.0 + * dml_ceil( + mode_lib->vba.BytePerPixelDETC[k], + 2) / 32 + / mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k]); + } else + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = 1.1 * mode_lib->vba.SwathWidthY[k] + * dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1) / 64.0 + / mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k]; + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = dml_max( + mode_lib->vba.DCFCLKDeepSleepPerPlane[k], + mode_lib->vba.PixelClock[k] / 16.0); + mode_lib->vba.DCFCLKDeepSleep = dml_max( + mode_lib->vba.DCFCLKDeepSleep, + mode_lib->vba.DCFCLKDeepSleepPerPlane[k]); + + DTRACE( + " dcfclk_deepsleep_per_plane[%i] = %fMHz", + k, + mode_lib->vba.DCFCLKDeepSleepPerPlane[k]); + } + + DTRACE(" dcfclk_deepsleep_mhz = %fMHz", mode_lib->vba.DCFCLKDeepSleep); + + // Stutter Watermark + mode_lib->vba.StutterExitWatermark = mode_lib->vba.SRExitTime + + mode_lib->vba.LastPixelOfLineExtraWatermark + + mode_lib->vba.UrgentExtraLatency + 10 / mode_lib->vba.DCFCLKDeepSleep; + mode_lib->vba.StutterEnterPlusExitWatermark = mode_lib->vba.SREnterPlusExitTime + + mode_lib->vba.LastPixelOfLineExtraWatermark + + mode_lib->vba.UrgentExtraLatency; + + DTRACE(" wm_cstate_exit = %fus", mode_lib->vba.StutterExitWatermark); + DTRACE(" wm_cstate_enter_exit = %fus", mode_lib->vba.StutterEnterPlusExitWatermark); + + // Urgent Latency Supported + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.EffectiveDETPlusLBLinesLuma = + dml_floor( + mode_lib->vba.LinesInDETY[k] + + dml_min( + mode_lib->vba.LinesInDETY[k] + * mode_lib->vba.DPPCLK[k] + * mode_lib->vba.BytePerPixelDETY[k] + * mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]), + (double) mode_lib->vba.EffectiveLBLatencyHidingSourceLinesLuma), + mode_lib->vba.SwathHeightY[k]); + + mode_lib->vba.UrgentLatencySupportUsLuma = mode_lib->vba.EffectiveDETPlusLBLinesLuma + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k] + - mode_lib->vba.EffectiveDETPlusLBLinesLuma + * mode_lib->vba.SwathWidthY[k] + * mode_lib->vba.BytePerPixelDETY[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]); + + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + mode_lib->vba.EffectiveDETPlusLBLinesChroma = + dml_floor( + mode_lib->vba.LinesInDETC[k] + + dml_min( + mode_lib->vba.LinesInDETC[k] + * mode_lib->vba.DPPCLK[k] + * mode_lib->vba.BytePerPixelDETC[k] + * mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]), + (double) mode_lib->vba.EffectiveLBLatencyHidingSourceLinesChroma), + mode_lib->vba.SwathHeightC[k]); + mode_lib->vba.UrgentLatencySupportUsChroma = + mode_lib->vba.EffectiveDETPlusLBLinesChroma + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / (mode_lib->vba.VRatio[k] / 2) + - mode_lib->vba.EffectiveDETPlusLBLinesChroma + * (mode_lib->vba.SwathWidthY[k] + / 2) + * mode_lib->vba.BytePerPixelDETC[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]); + mode_lib->vba.UrgentLatencySupportUs[k] = dml_min( + mode_lib->vba.UrgentLatencySupportUsLuma, + mode_lib->vba.UrgentLatencySupportUsChroma); + } else { + mode_lib->vba.UrgentLatencySupportUs[k] = + mode_lib->vba.UrgentLatencySupportUsLuma; + } + } + + mode_lib->vba.MinUrgentLatencySupportUs = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.MinUrgentLatencySupportUs = dml_min( + mode_lib->vba.MinUrgentLatencySupportUs, + mode_lib->vba.UrgentLatencySupportUs[k]); + } + + // Non-Urgent Latency Tolerance + mode_lib->vba.NonUrgentLatencyTolerance = mode_lib->vba.MinUrgentLatencySupportUs + - mode_lib->vba.UrgentWatermark; + + // DSCCLK + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if ((mode_lib->vba.BlendingAndTiming[k] != k) || !mode_lib->vba.DSCEnabled[k]) { + mode_lib->vba.DSCCLK_calculated[k] = 0.0; + } else { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] == dm_n422) + mode_lib->vba.DSCFormatFactor = 2; + else + mode_lib->vba.DSCFormatFactor = 1; + if (mode_lib->vba.ODMCombineEnabled[k]) + mode_lib->vba.DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 6 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + else + mode_lib->vba.DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 3 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + } + } + + // DSC Delay + // TODO + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + double bpp = mode_lib->vba.OutputBpp[k]; + unsigned int slices = mode_lib->vba.NumberOfDSCSlices[k]; + + if (mode_lib->vba.DSCEnabled[k] && bpp != 0) { + if (!mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DSCDelay[k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + mode_lib->vba.DSCDelay[k] = + 2 + * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices / 2.0, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k])); + } + mode_lib->vba.DSCDelay[k] = mode_lib->vba.DSCDelay[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.PixelClockBackEnd[k]; + } else { + mode_lib->vba.DSCDelay[k] = 0; + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) // NumberOfPlanes + if (j != k && mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.DSCEnabled[j]) + mode_lib->vba.DSCDelay[k] = mode_lib->vba.DSCDelay[j]; + + // Prefetch + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int PDEAndMetaPTEBytesFrameY; + unsigned int PixelPTEBytesPerRowY; + unsigned int MetaRowByteY; + unsigned int MetaRowByteC; + unsigned int PDEAndMetaPTEBytesFrameC; + unsigned int PixelPTEBytesPerRowC; + + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1), + dml_ceil(mode_lib->vba.BytePerPixelDETC[k], 2), + &mode_lib->vba.BlockHeight256BytesY[k], + &mode_lib->vba.BlockHeight256BytesC[k], + &mode_lib->vba.BlockWidth256BytesY[k], + &mode_lib->vba.BlockWidth256BytesC[k]); + PDEAndMetaPTEBytesFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.BlockHeight256BytesY[k], + mode_lib->vba.BlockWidth256BytesY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + mode_lib->vba.SwathWidthY[k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &mode_lib->vba.MacroTileWidthY[k], + &MetaRowByteY, + &PixelPTEBytesPerRowY, + &mode_lib->vba.PTEBufferSizeNotExceeded[mode_lib->vba.VoltageLevel][0], + &mode_lib->vba.dpte_row_height[k], + &mode_lib->vba.meta_row_height[k]); + mode_lib->vba.PrefetchSourceLinesY[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.ViewportYStartY[k], + &mode_lib->vba.VInitPreFillY[k], + &mode_lib->vba.MaxNumSwathY[k]); + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_8)) { + PDEAndMetaPTEBytesFrameC = + CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.BlockHeight256BytesC[k], + mode_lib->vba.BlockWidth256BytesC[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil( + mode_lib->vba.BytePerPixelDETC[k], + 2), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2, + mode_lib->vba.ViewportHeight[k] / 2, + mode_lib->vba.SwathWidthY[k] / 2, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchC[k], + 0, + &mode_lib->vba.MacroTileWidthC[k], + &MetaRowByteC, + &PixelPTEBytesPerRowC, + &mode_lib->vba.PTEBufferSizeNotExceeded[mode_lib->vba.VoltageLevel][0], + &mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_height_chroma[k]); + mode_lib->vba.PrefetchSourceLinesC[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k] / 2, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightC[k], + mode_lib->vba.ViewportYStartC[k], + &mode_lib->vba.VInitPreFillC[k], + &mode_lib->vba.MaxNumSwathC[k]); + } else { + PixelPTEBytesPerRowC = 0; + PDEAndMetaPTEBytesFrameC = 0; + MetaRowByteC = 0; + mode_lib->vba.MaxNumSwathC[k] = 0; + mode_lib->vba.PrefetchSourceLinesC[k] = 0; + } + + mode_lib->vba.PixelPTEBytesPerRow[k] = PixelPTEBytesPerRowY + PixelPTEBytesPerRowC; + mode_lib->vba.PDEAndMetaPTEBytesFrame[k] = PDEAndMetaPTEBytesFrameY + + PDEAndMetaPTEBytesFrameC; + mode_lib->vba.MetaRowByte[k] = MetaRowByteY + MetaRowByteC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + MetaRowByteY, + MetaRowByteC, + mode_lib->vba.meta_row_height[k], + mode_lib->vba.meta_row_height_chroma[k], + PixelPTEBytesPerRowY, + PixelPTEBytesPerRowC, + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_bw[k], + &mode_lib->vba.dpte_row_bw[k], + &mode_lib->vba.qual_row_bw[k]); + } + + mode_lib->vba.TCalc = 24.0 / mode_lib->vba.DCFCLKDeepSleep; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) + / mode_lib->vba.DISPCLK; + } else + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = 0; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[j] == k + && mode_lib->vba.WritebackEnable[j] == true) { + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = + dml_max( + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k], + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[j], + mode_lib->vba.WritebackHRatio[j], + mode_lib->vba.WritebackVRatio[j], + mode_lib->vba.WritebackLumaHTaps[j], + mode_lib->vba.WritebackLumaVTaps[j], + mode_lib->vba.WritebackChromaHTaps[j], + mode_lib->vba.WritebackChromaVTaps[j], + mode_lib->vba.WritebackDestinationWidth[j]) + / mode_lib->vba.DISPCLK); + } + } + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j) + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][j]; + + mode_lib->vba.VStartupLines = 13; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.MaxVStartupLines[k] = + mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] + - dml_max( + 1.0, + dml_ceil( + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1)); + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + mode_lib->vba.MaximumMaxVStartupLines = dml_max( + mode_lib->vba.MaximumMaxVStartupLines, + mode_lib->vba.MaxVStartupLines[k]); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.cursor_bw[k] = 0.0; + for (j = 0; j < mode_lib->vba.NumberOfCursors[k]; ++j) + mode_lib->vba.cursor_bw[k] += mode_lib->vba.CursorWidth[k][j] + * mode_lib->vba.CursorBPP[k][j] / 8.0 + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k]; + } + + do { + double MaxTotalRDBandwidth = 0; + bool DestinationLineTimesForPrefetchLessThan2 = false; + bool VRatioPrefetchMoreThan4 = false; + bool prefetch_vm_bw_valid = true; + bool prefetch_row_bw_valid = true; + double TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.SREnterPlusExitTime); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.SwathWidthY[k], + dml_ceil( + mode_lib->vba.BytePerPixelDETY[k], + 1), + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0; + } + + CalculateDelayAfterScaler(mode_lib, mode_lib->vba.ReturnBW, mode_lib->vba.ReadBandwidthPlaneLuma[k], mode_lib->vba.ReadBandwidthPlaneChroma[k], mode_lib->vba.TotalDataReadBandwidth, + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k], mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k], + mode_lib->vba.DPPCLK[k], mode_lib->vba.DISPCLK, mode_lib->vba.PixelClock[k], mode_lib->vba.DSCDelay[k], mode_lib->vba.DPPPerPlane[k], mode_lib->vba.ScalerEnabled[k], mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.DPPCLKDelaySubtotal, mode_lib->vba.DPPCLKDelaySCL, mode_lib->vba.DPPCLKDelaySCLLBOnly, mode_lib->vba.DPPCLKDelayCNVCFormater, mode_lib->vba.DPPCLKDelayCNVCCursor, mode_lib->vba.DISPCLKDelaySubtotal, + mode_lib->vba.SwathWidthY[k] / mode_lib->vba.HRatio[k], mode_lib->vba.OutputFormat[k], mode_lib->vba.HTotal[k], + mode_lib->vba.SwathWidthSingleDPPY[k], mode_lib->vba.BytePerPixelDETY[k], mode_lib->vba.BytePerPixelDETC[k], mode_lib->vba.SwathHeightY[k], mode_lib->vba.SwathHeightC[k], mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, &mode_lib->vba.DSTXAfterScaler[k], &mode_lib->vba.DSTYAfterScaler[k]); + + mode_lib->vba.ErrorResult[k] = + CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.DPPCLK[k], + mode_lib->vba.DISPCLK, + mode_lib->vba.PixelClock[k], + mode_lib->vba.DCFCLKDeepSleep, + mode_lib->vba.DPPPerPlane[k], + mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + dml_min( + mode_lib->vba.VStartupLines, + mode_lib->vba.MaxVStartupLines[k]), + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.TCalc, + mode_lib->vba.PDEAndMetaPTEBytesFrame[k], + mode_lib->vba.MetaRowByte[k], + mode_lib->vba.PixelPTEBytesPerRow[k], + mode_lib->vba.PrefetchSourceLinesY[k], + mode_lib->vba.SwathWidthY[k], + mode_lib->vba.BytePerPixelDETY[k], + mode_lib->vba.VInitPreFillY[k], + mode_lib->vba.MaxNumSwathY[k], + mode_lib->vba.PrefetchSourceLinesC[k], + mode_lib->vba.BytePerPixelDETC[k], + mode_lib->vba.VInitPreFillC[k], + mode_lib->vba.MaxNumSwathC[k], + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.DSTXAfterScaler[k], + mode_lib->vba.DSTYAfterScaler[k], + &mode_lib->vba.DestinationLinesForPrefetch[k], + &mode_lib->vba.PrefetchBandwidth[k], + &mode_lib->vba.DestinationLinesToRequestVMInVBlank[k], + &mode_lib->vba.DestinationLinesToRequestRowInVBlank[k], + &mode_lib->vba.VRatioPrefetchY[k], + &mode_lib->vba.VRatioPrefetchC[k], + &mode_lib->vba.RequiredPrefetchPixDataBWLuma[k], + &mode_lib->vba.Tno_bw[k], + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.VStartup[k] = dml_min( + mode_lib->vba.VStartupLines, + mode_lib->vba.MaxVStartupLines[k]); + if (mode_lib->vba.VStartupRequiredWhenNotEnoughTimeForDynamicMetadata + != 0) { + mode_lib->vba.VStartup[k] = + mode_lib->vba.VStartupRequiredWhenNotEnoughTimeForDynamicMetadata; + } + } else { + mode_lib->vba.VStartup[k] = + dml_min( + mode_lib->vba.VStartupLines, + mode_lib->vba.MaxVStartupLines[mode_lib->vba.BlendingAndTiming[k]]); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + + if (mode_lib->vba.PDEAndMetaPTEBytesFrame[k] == 0) + mode_lib->vba.prefetch_vm_bw[k] = 0; + else if (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] > 0) { + mode_lib->vba.prefetch_vm_bw[k] = + (double) mode_lib->vba.PDEAndMetaPTEBytesFrame[k] + / (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } else { + mode_lib->vba.prefetch_vm_bw[k] = 0; + prefetch_vm_bw_valid = false; + } + if (mode_lib->vba.MetaRowByte[k] + mode_lib->vba.PixelPTEBytesPerRow[k] + == 0) + mode_lib->vba.prefetch_row_bw[k] = 0; + else if (mode_lib->vba.DestinationLinesToRequestRowInVBlank[k] > 0) { + mode_lib->vba.prefetch_row_bw[k] = + (double) (mode_lib->vba.MetaRowByte[k] + + mode_lib->vba.PixelPTEBytesPerRow[k]) + / (mode_lib->vba.DestinationLinesToRequestRowInVBlank[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } else { + mode_lib->vba.prefetch_row_bw[k] = 0; + prefetch_row_bw_valid = false; + } + + MaxTotalRDBandwidth = + MaxTotalRDBandwidth + mode_lib->vba.cursor_bw[k] + + dml_max( + mode_lib->vba.prefetch_vm_bw[k], + dml_max( + mode_lib->vba.prefetch_row_bw[k], + dml_max( + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k], + mode_lib->vba.RequiredPrefetchPixDataBWLuma[k]) + + mode_lib->vba.meta_row_bw[k] + + mode_lib->vba.dpte_row_bw[k])); + + if (mode_lib->vba.DestinationLinesForPrefetch[k] < 2) + DestinationLineTimesForPrefetchLessThan2 = true; + if (mode_lib->vba.VRatioPrefetchY[k] > 4 + || mode_lib->vba.VRatioPrefetchC[k] > 4) + VRatioPrefetchMoreThan4 = true; + } + + if (MaxTotalRDBandwidth <= mode_lib->vba.ReturnBW && prefetch_vm_bw_valid + && prefetch_row_bw_valid && !VRatioPrefetchMoreThan4 + && !DestinationLineTimesForPrefetchLessThan2) + mode_lib->vba.PrefetchModeSupported = true; + else { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Bandwidth violation. Results are NOT valid\n"); + } + + if (mode_lib->vba.PrefetchModeSupported == true) { + double final_flip_bw[DC__NUM_DPP__MAX]; + unsigned int ImmediateFlipBytes[DC__NUM_DPP__MAX]; + double total_dcn_read_bw_with_flip = 0; + + mode_lib->vba.BandwidthAvailableForImmediateFlip = mode_lib->vba.ReturnBW; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.BandwidthAvailableForImmediateFlip + - mode_lib->vba.cursor_bw[k] + - dml_max( + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + + mode_lib->vba.qual_row_bw[k], + mode_lib->vba.PrefetchBandwidth[k]); + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + ImmediateFlipBytes[k] = 0; + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + ImmediateFlipBytes[k] = + mode_lib->vba.PDEAndMetaPTEBytesFrame[k] + + mode_lib->vba.MetaRowByte[k] + + mode_lib->vba.PixelPTEBytesPerRow[k]; + } + } + mode_lib->vba.TotImmediateFlipBytes = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.TotImmediateFlipBytes = + mode_lib->vba.TotImmediateFlipBytes + + ImmediateFlipBytes[k]; + } + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + ImmediateFlipBytes[k], + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.Tno_bw[k], + mode_lib->vba.PDEAndMetaPTEBytesFrame[k], + mode_lib->vba.MetaRowByte[k], + mode_lib->vba.PixelPTEBytesPerRow[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.meta_row_height[k], + mode_lib->vba.qual_row_bw[k], + &mode_lib->vba.DestinationLinesToRequestVMInImmediateFlip[k], + &mode_lib->vba.DestinationLinesToRequestRowInImmediateFlip[k], + &final_flip_bw[k], + &mode_lib->vba.ImmediateFlipSupportedForPipe[k]); + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + total_dcn_read_bw_with_flip = + total_dcn_read_bw_with_flip + + mode_lib->vba.cursor_bw[k] + + dml_max( + mode_lib->vba.prefetch_vm_bw[k], + dml_max( + mode_lib->vba.prefetch_row_bw[k], + final_flip_bw[k] + + dml_max( + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k], + mode_lib->vba.RequiredPrefetchPixDataBWLuma[k]))); + } + mode_lib->vba.ImmediateFlipSupported = true; + if (total_dcn_read_bw_with_flip > mode_lib->vba.ReturnBW) { + mode_lib->vba.ImmediateFlipSupported = false; + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ImmediateFlipSupportedForPipe[k] == false) { + mode_lib->vba.ImmediateFlipSupported = false; + } + } + } else { + mode_lib->vba.ImmediateFlipSupported = false; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ErrorResult[k]) { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Prefetch schedule violation. Results are NOT valid\n"); + } + } + + mode_lib->vba.VStartupLines = mode_lib->vba.VStartupLines + 1; + } while (!((mode_lib->vba.PrefetchModeSupported + && (!mode_lib->vba.ImmediateFlipSupport + || mode_lib->vba.ImmediateFlipSupported)) + || mode_lib->vba.MaximumMaxVStartupLines < mode_lib->vba.VStartupLines)); + + //Display Pipeline Delivery Time in Prefetch + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.VRatioPrefetchY[k] <= 1) { + mode_lib->vba.DisplayPipeLineDeliveryTimeLumaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] * mode_lib->vba.DPPPerPlane[k] + / mode_lib->vba.HRatio[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.DisplayPipeLineDeliveryTimeLumaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / mode_lib->vba.DPPCLK[k]; + } + if (mode_lib->vba.BytePerPixelDETC[k] == 0) { + mode_lib->vba.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = 0; + } else { + if (mode_lib->vba.VRatioPrefetchC[k] <= 1) { + mode_lib->vba.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] + * mode_lib->vba.DPPPerPlane[k] + / mode_lib->vba.HRatio[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / mode_lib->vba.DPPCLK[k]; + } + } + } + + // Min TTUVBlank + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) { + mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k] = true; + mode_lib->vba.AllowDRAMSelfRefreshDuringVBlank[k] = true; + mode_lib->vba.MinTTUVBlank[k] = dml_max( + mode_lib->vba.DRAMClockChangeWatermark, + dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark)); + } else if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 1) { + mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k] = false; + mode_lib->vba.AllowDRAMSelfRefreshDuringVBlank[k] = true; + mode_lib->vba.MinTTUVBlank[k] = dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark); + } else { + mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k] = false; + mode_lib->vba.AllowDRAMSelfRefreshDuringVBlank[k] = false; + mode_lib->vba.MinTTUVBlank[k] = mode_lib->vba.UrgentWatermark; + } + if (!mode_lib->vba.DynamicMetadataEnable[k]) + mode_lib->vba.MinTTUVBlank[k] = mode_lib->vba.TCalc + + mode_lib->vba.MinTTUVBlank[k]; + } + + // DCC Configuration + mode_lib->vba.ActiveDPPs = 0; + // NB P-State/DRAM Clock Change Support + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.ActiveDPPs = mode_lib->vba.ActiveDPPs + mode_lib->vba.DPPPerPlane[k]; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + double EffectiveLBLatencyHidingY; + double EffectiveLBLatencyHidingC; + double DPPOutputBufferLinesY; + double DPPOutputBufferLinesC; + double DPPOPPBufferingY; + double MaxDETBufferingTimeY; + double ActiveDRAMClockChangeLatencyMarginY; + + mode_lib->vba.LBLatencyHidingSourceLinesY = + dml_min( + mode_lib->vba.MaxLineBufferLines, + (unsigned int) dml_floor( + (double) mode_lib->vba.LineBufferSize + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.SwathWidthY[k] + / dml_max( + mode_lib->vba.HRatio[k], + 1.0)), + 1)) - (mode_lib->vba.vtaps[k] - 1); + + mode_lib->vba.LBLatencyHidingSourceLinesC = + dml_min( + mode_lib->vba.MaxLineBufferLines, + (unsigned int) dml_floor( + (double) mode_lib->vba.LineBufferSize + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.SwathWidthY[k] + / 2.0 + / dml_max( + mode_lib->vba.HRatio[k] + / 2, + 1.0)), + 1)) + - (mode_lib->vba.VTAPsChroma[k] - 1); + + EffectiveLBLatencyHidingY = mode_lib->vba.LBLatencyHidingSourceLinesY + / mode_lib->vba.VRatio[k] + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]); + + EffectiveLBLatencyHidingC = mode_lib->vba.LBLatencyHidingSourceLinesC + / (mode_lib->vba.VRatio[k] / 2) + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]); + + if (mode_lib->vba.SwathWidthY[k] > 2 * mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesY = mode_lib->vba.DPPOutputBufferPixels + / mode_lib->vba.SwathWidthY[k]; + } else if (mode_lib->vba.SwathWidthY[k] > mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesY = 0.5; + } else { + DPPOutputBufferLinesY = 1; + } + + if (mode_lib->vba.SwathWidthY[k] / 2 > 2 * mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesC = mode_lib->vba.DPPOutputBufferPixels + / (mode_lib->vba.SwathWidthY[k] / 2); + } else if (mode_lib->vba.SwathWidthY[k] / 2 > mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesC = 0.5; + } else { + DPPOutputBufferLinesC = 1; + } + + DPPOPPBufferingY = (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * (DPPOutputBufferLinesY + mode_lib->vba.OPPOutputBufferLines); + MaxDETBufferingTimeY = mode_lib->vba.FullDETBufferingTimeY[k] + + (mode_lib->vba.LinesInDETY[k] + - mode_lib->vba.LinesInDETYRoundedDownToSwath[k]) + / mode_lib->vba.SwathHeightY[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + + ActiveDRAMClockChangeLatencyMarginY = DPPOPPBufferingY + EffectiveLBLatencyHidingY + + MaxDETBufferingTimeY - mode_lib->vba.DRAMClockChangeWatermark; + + if (mode_lib->vba.ActiveDPPs > 1) { + ActiveDRAMClockChangeLatencyMarginY = + ActiveDRAMClockChangeLatencyMarginY + - (1 - 1 / (mode_lib->vba.ActiveDPPs - 1)) + * mode_lib->vba.SwathHeightY[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } + + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + double DPPOPPBufferingC = (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + * (DPPOutputBufferLinesC + + mode_lib->vba.OPPOutputBufferLines); + double MaxDETBufferingTimeC = + mode_lib->vba.FullDETBufferingTimeC[k] + + (mode_lib->vba.LinesInDETC[k] + - mode_lib->vba.LinesInDETCRoundedDownToSwath[k]) + / mode_lib->vba.SwathHeightC[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + double ActiveDRAMClockChangeLatencyMarginC = DPPOPPBufferingC + + EffectiveLBLatencyHidingC + MaxDETBufferingTimeC + - mode_lib->vba.DRAMClockChangeWatermark; + + if (mode_lib->vba.ActiveDPPs > 1) { + ActiveDRAMClockChangeLatencyMarginC = + ActiveDRAMClockChangeLatencyMarginC + - (1 + - 1 + / (mode_lib->vba.ActiveDPPs + - 1)) + * mode_lib->vba.SwathHeightC[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + ActiveDRAMClockChangeLatencyMarginY, + ActiveDRAMClockChangeLatencyMarginC); + } else { + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = + ActiveDRAMClockChangeLatencyMarginY; + } + + if (mode_lib->vba.WritebackEnable[k]) { + double WritebackDRAMClockChangeLatencyMargin; + + if (mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + WritebackDRAMClockChangeLatencyMargin = + (double) (mode_lib->vba.WritebackInterfaceLumaBufferSize + + mode_lib->vba.WritebackInterfaceChromaBufferSize) + / (mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + * 4) + - mode_lib->vba.WritebackDRAMClockChangeWatermark; + } else if (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + WritebackDRAMClockChangeLatencyMargin = + dml_min( + (double) mode_lib->vba.WritebackInterfaceLumaBufferSize + * 8.0 / 10, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize + * 8 / 10) + / (mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k])) + - mode_lib->vba.WritebackDRAMClockChangeWatermark; + } else { + WritebackDRAMClockChangeLatencyMargin = + dml_min( + (double) mode_lib->vba.WritebackInterfaceLumaBufferSize, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize) + / (mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k])) + - mode_lib->vba.WritebackDRAMClockChangeWatermark; + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k], + WritebackDRAMClockChangeLatencyMargin); + } + } + + mode_lib->vba.MinActiveDRAMClockChangeMargin = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] + < mode_lib->vba.MinActiveDRAMClockChangeMargin) { + mode_lib->vba.MinActiveDRAMClockChangeMargin = + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + } + } + + mode_lib->vba.MinActiveDRAMClockChangeLatencySupported = + mode_lib->vba.MinActiveDRAMClockChangeMargin + + mode_lib->vba.DRAMClockChangeLatency; + + if (mode_lib->vba.MinActiveDRAMClockChangeMargin > 0) { + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; + } else { + if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vblank; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (!mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k]) { + mode_lib->vba.DRAMClockChangeSupport[0][0] = + dm_dram_clock_change_unsupported; + } + } + } else { + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_unsupported; + } + } + for (k = 0; k <= mode_lib->vba.soc.num_states; k++) + for (j = 0; j < 2; j++) + mode_lib->vba.DRAMClockChangeSupport[k][j] = mode_lib->vba.DRAMClockChangeSupport[0][0]; + + //XFC Parameters: + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.XFCEnabled[k] == true) { + double TWait; + + mode_lib->vba.XFCSlaveVUpdateOffset[k] = mode_lib->vba.XFCTSlvVupdateOffset; + mode_lib->vba.XFCSlaveVupdateWidth[k] = mode_lib->vba.XFCTSlvVupdateWidth; + mode_lib->vba.XFCSlaveVReadyOffset[k] = mode_lib->vba.XFCTSlvVreadyOffset; + TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.SREnterPlusExitTime); + mode_lib->vba.XFCRemoteSurfaceFlipDelay = CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.SwathWidthY[k], + dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + mode_lib->vba.XFCRemoteSurfaceFlipLatency[k] = + dml_floor( + mode_lib->vba.XFCRemoteSurfaceFlipDelay + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.XFCTransferDelay[k] = + dml_ceil( + mode_lib->vba.XFCBusTransportTime + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.XFCPrechargeDelay[k] = + dml_ceil( + (mode_lib->vba.XFCBusTransportTime + + mode_lib->vba.TInitXFill + + mode_lib->vba.TslvChk) + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.InitFillLevel = mode_lib->vba.XFCXBUFLatencyTolerance + * mode_lib->vba.SrcActiveDrainRate; + mode_lib->vba.FinalFillMargin = + (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] + + mode_lib->vba.DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k] + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.XFCFillConstant; + mode_lib->vba.FinalFillLevel = mode_lib->vba.XFCRemoteSurfaceFlipDelay + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.FinalFillMargin; + mode_lib->vba.RemainingFillLevel = dml_max( + 0.0, + mode_lib->vba.FinalFillLevel - mode_lib->vba.InitFillLevel); + mode_lib->vba.TFinalxFill = mode_lib->vba.RemainingFillLevel + / (mode_lib->vba.SrcActiveDrainRate + * mode_lib->vba.XFCFillBWOverhead / 100); + mode_lib->vba.XFCPrefetchMargin[k] = + mode_lib->vba.XFCRemoteSurfaceFlipDelay + + mode_lib->vba.TFinalxFill + + (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] + + mode_lib->vba.DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.XFCSlaveVUpdateOffset[k] = 0; + mode_lib->vba.XFCSlaveVupdateWidth[k] = 0; + mode_lib->vba.XFCSlaveVReadyOffset[k] = 0; + mode_lib->vba.XFCRemoteSurfaceFlipLatency[k] = 0; + mode_lib->vba.XFCPrechargeDelay[k] = 0; + mode_lib->vba.XFCTransferDelay[k] = 0; + mode_lib->vba.XFCPrefetchMargin[k] = 0; + } + } + { + unsigned int VStartupMargin = 0; + bool FirstMainPlane = true; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + unsigned int Margin = (mode_lib->vba.MaxVStartupLines[k] - mode_lib->vba.VStartup[k]) + * mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]; + + if (FirstMainPlane) { + VStartupMargin = Margin; + FirstMainPlane = false; + } else + VStartupMargin = dml_min(VStartupMargin, Margin); + } + + if (mode_lib->vba.UseMaximumVStartup) { + if (mode_lib->vba.VTotal_Max[k] == mode_lib->vba.VTotal[k]) { + //only use max vstart if it is not drr or lateflip. + mode_lib->vba.VStartup[k] = mode_lib->vba.MaxVStartupLines[mode_lib->vba.BlendingAndTiming[k]]; + } + } + } +} +} + +static void dml20v2_DisplayPipeConfiguration(struct display_mode_lib *mode_lib) +{ + double BytePerPixDETY; + double BytePerPixDETC; + double Read256BytesBlockHeightY; + double Read256BytesBlockHeightC; + double Read256BytesBlockWidthY; + double Read256BytesBlockWidthC; + double MaximumSwathHeightY; + double MaximumSwathHeightC; + double MinimumSwathHeightY; + double MinimumSwathHeightC; + double SwathWidth; + double SwathWidthGranularityY; + double SwathWidthGranularityC; + double RoundedUpMaxSwathSizeBytesY; + double RoundedUpMaxSwathSizeBytesC; + unsigned int j, k; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + BytePerPixDETY = 8; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + BytePerPixDETY = 4; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + BytePerPixDETY = 2; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 2; + } else { + BytePerPixDETY = 4.0 / 3.0; + BytePerPixDETC = 8.0 / 3.0; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + Read256BytesBlockHeightY = 4; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + Read256BytesBlockHeightY = 8; + } else { + Read256BytesBlockHeightY = 16; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockHeightC = 0; + Read256BytesBlockWidthC = 0; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + Read256BytesBlockHeightC = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + Read256BytesBlockHeightY = 16; + Read256BytesBlockHeightC = 8; + } else { + Read256BytesBlockHeightY = 8; + Read256BytesBlockHeightC = 8; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockWidthC = 256 / dml_ceil(BytePerPixDETC, 2) + / Read256BytesBlockHeightC; + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + MaximumSwathHeightY = Read256BytesBlockHeightY; + MaximumSwathHeightC = Read256BytesBlockHeightC; + } else { + MaximumSwathHeightY = Read256BytesBlockWidthY; + MaximumSwathHeightC = Read256BytesBlockWidthC; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8 + && mode_lib->vba.SourceScan[k] != dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + } + MinimumSwathHeightC = MaximumSwathHeightC; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightC = MaximumSwathHeightC / 2.0; + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + SwathWidth = mode_lib->vba.ViewportWidth[k]; + } else { + SwathWidth = mode_lib->vba.ViewportHeight[k]; + } + + if (mode_lib->vba.ODMCombineEnabled[k] == true) { + MainPlaneDoesODMCombine = true; + } + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[j] == true) { + MainPlaneDoesODMCombine = true; + } + } + + if (MainPlaneDoesODMCombine == true) { + SwathWidth = dml_min( + SwathWidth, + mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k]); + } else { + if (mode_lib->vba.DPPPerPlane[k] == 0) + SwathWidth = 0; + else + SwathWidth = SwathWidth / mode_lib->vba.DPPPerPlane[k]; + } + + SwathWidthGranularityY = 256 / dml_ceil(BytePerPixDETY, 1) / MaximumSwathHeightY; + RoundedUpMaxSwathSizeBytesY = (dml_ceil( + (double) (SwathWidth - 1), + SwathWidthGranularityY) + SwathWidthGranularityY) * BytePerPixDETY + * MaximumSwathHeightY; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesY = dml_ceil(RoundedUpMaxSwathSizeBytesY, 256) + + 256; + } + if (MaximumSwathHeightC > 0) { + SwathWidthGranularityC = 256.0 / dml_ceil(BytePerPixDETC, 2) + / MaximumSwathHeightC; + RoundedUpMaxSwathSizeBytesC = (dml_ceil( + (double) (SwathWidth / 2.0 - 1), + SwathWidthGranularityC) + SwathWidthGranularityC) + * BytePerPixDETC * MaximumSwathHeightC; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesC = dml_ceil( + RoundedUpMaxSwathSizeBytesC, + 256) + 256; + } + } else + RoundedUpMaxSwathSizeBytesC = 0.0; + + if (RoundedUpMaxSwathSizeBytesY + RoundedUpMaxSwathSizeBytesC + <= mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0) { + mode_lib->vba.SwathHeightY[k] = MaximumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MaximumSwathHeightC; + } else { + mode_lib->vba.SwathHeightY[k] = MinimumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MinimumSwathHeightC; + } + + if (mode_lib->vba.SwathHeightC[k] == 0) { + mode_lib->vba.DETBufferSizeY[k] = mode_lib->vba.DETBufferSizeInKByte * 1024; + mode_lib->vba.DETBufferSizeC[k] = 0; + } else if (mode_lib->vba.SwathHeightY[k] <= mode_lib->vba.SwathHeightC[k]) { + mode_lib->vba.DETBufferSizeY[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 / 2; + mode_lib->vba.DETBufferSizeC[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 / 2; + } else { + mode_lib->vba.DETBufferSizeY[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 * 2 / 3; + mode_lib->vba.DETBufferSizeC[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 / 3; + } + } +} + +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatencyPixelDataOnly, + double SREnterPlusExitTime) +{ + if (PrefetchMode == 0) { + return dml_max( + DRAMClockChangeLatency + UrgentLatencyPixelDataOnly, + dml_max(SREnterPlusExitTime, UrgentLatencyPixelDataOnly)); + } else if (PrefetchMode == 1) { + return dml_max(SREnterPlusExitTime, UrgentLatencyPixelDataOnly); + } else { + return UrgentLatencyPixelDataOnly; + } +} + +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk) +{ + double TSlvSetup, AvgfillRate, result; + + *SrcActiveDrainRate = VRatio * SwathWidth * Bpp / LineTime; + TSlvSetup = XFCTSlvVupdateOffset + XFCTSlvVupdateWidth + XFCTSlvVreadyOffset; + *TInitXFill = XFCXBUFLatencyTolerance / (1 + XFCFillBWOverhead / 100); + AvgfillRate = *SrcActiveDrainRate * (1 + XFCFillBWOverhead / 100); + *TslvChk = XFCSlvChunkSize / AvgfillRate; + dml_print( + "DML::CalculateRemoteSurfaceFlipDelay: SrcActiveDrainRate: %f\n", + *SrcActiveDrainRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TSlvSetup: %f\n", TSlvSetup); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TInitXFill: %f\n", *TInitXFill); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: AvgfillRate: %f\n", AvgfillRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TslvChk: %f\n", *TslvChk); + result = 2 * XFCBusTransportTime + TSlvSetup + TCalc + TWait + *TslvChk + *TInitXFill; // TODO: This doesn't seem to match programming guide + dml_print("DML::CalculateRemoteSurfaceFlipDelay: RemoteSurfaceFlipDelay: %f\n", result); + return result; +} + +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth) +{ + double CalculateWriteBackDelay = + dml_max( + dml_ceil(WritebackLumaHTaps / 4.0, 1) / WritebackHRatio, + WritebackLumaVTaps * dml_ceil(1.0 / WritebackVRatio, 1) + * dml_ceil( + WritebackDestinationWidth + / 4.0, + 1) + + dml_ceil(1.0 / WritebackVRatio, 1) + * (dml_ceil( + WritebackLumaVTaps + / 4.0, + 1) + 4)); + + if (WritebackPixelFormat != dm_444_32) { + CalculateWriteBackDelay = + dml_max( + CalculateWriteBackDelay, + dml_max( + dml_ceil( + WritebackChromaHTaps + / 2.0, + 1) + / (2 + * WritebackHRatio), + WritebackChromaVTaps + * dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * dml_ceil( + WritebackDestinationWidth + / 2.0 + / 2.0, + 1) + + dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * (dml_ceil( + WritebackChromaVTaps + / 4.0, + 1) + + 4))); + } + return CalculateWriteBackDelay; +} + +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw, + double *qual_row_bw) +{ + if (DCCEnable != true) { + *meta_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime) + + VRatio / 2 * MetaRowByteChroma + / (meta_row_height_chroma * LineTime); + } else { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime); + } + + if (GPUVMEnable != true) { + *dpte_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime) + + VRatio / 2 * PixelPTEBytesPerRowChroma + / (dpte_row_height_chroma * LineTime); + } else { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime); + } + + if ((SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10)) { + *qual_row_bw = *meta_row_bw + *dpte_row_bw; + } else { + *qual_row_bw = 0; + } +} + +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double UrgentExtraLatency, + double UrgentLatencyPixelDataOnly, + unsigned int GPUVMMaxPageTableLevels, + bool GPUVMEnable, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + unsigned int ImmediateFlipBytes, + double LineTime, + double VRatio, + double Tno_bw, + double PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + double qual_row_bw, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe) +{ + double min_row_time = 0.0; + + if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *DestinationLinesToRequestVMInImmediateFlip = 0.0; + *DestinationLinesToRequestRowInImmediateFlip = 0.0; + *final_flip_bw = qual_row_bw; + *ImmediateFlipSupportedForPipe = true; + } else { + double TimeForFetchingMetaPTEImmediateFlip; + double TimeForFetchingRowInVBlankImmediateFlip; + + if (GPUVMEnable == true) { + mode_lib->vba.ImmediateFlipBW[0] = BandwidthAvailableForImmediateFlip + * ImmediateFlipBytes / TotImmediateFlipBytes; + TimeForFetchingMetaPTEImmediateFlip = + dml_max( + Tno_bw + + PDEAndMetaPTEBytesFrame + / mode_lib->vba.ImmediateFlipBW[0], + dml_max( + UrgentExtraLatency + + UrgentLatencyPixelDataOnly + * (GPUVMMaxPageTableLevels + - 1), + LineTime / 4.0)); + } else { + TimeForFetchingMetaPTEImmediateFlip = 0; + } + + *DestinationLinesToRequestVMInImmediateFlip = dml_floor( + 4.0 * (TimeForFetchingMetaPTEImmediateFlip / LineTime + 0.125), + 1) / 4.0; + + if ((GPUVMEnable == true || DCCEnable == true)) { + mode_lib->vba.ImmediateFlipBW[0] = BandwidthAvailableForImmediateFlip + * ImmediateFlipBytes / TotImmediateFlipBytes; + TimeForFetchingRowInVBlankImmediateFlip = dml_max( + (MetaRowByte + PixelPTEBytesPerRow) + / mode_lib->vba.ImmediateFlipBW[0], + dml_max(UrgentLatencyPixelDataOnly, LineTime / 4.0)); + } else { + TimeForFetchingRowInVBlankImmediateFlip = 0; + } + + *DestinationLinesToRequestRowInImmediateFlip = dml_floor( + 4.0 * (TimeForFetchingRowInVBlankImmediateFlip / LineTime + 0.125), + 1) / 4.0; + + if (GPUVMEnable == true) { + *final_flip_bw = + dml_max( + PDEAndMetaPTEBytesFrame + / (*DestinationLinesToRequestVMInImmediateFlip + * LineTime), + (MetaRowByte + PixelPTEBytesPerRow) + / (TimeForFetchingRowInVBlankImmediateFlip + * LineTime)); + } else if (MetaRowByte + PixelPTEBytesPerRow > 0) { + *final_flip_bw = (MetaRowByte + PixelPTEBytesPerRow) + / (TimeForFetchingRowInVBlankImmediateFlip * LineTime); + } else { + *final_flip_bw = 0; + } + + if (GPUVMEnable && !DCCEnable) + min_row_time = dpte_row_height * LineTime / VRatio; + else if (!GPUVMEnable && DCCEnable) + min_row_time = meta_row_height * LineTime / VRatio; + else + min_row_time = dml_min(dpte_row_height, meta_row_height) * LineTime + / VRatio; + + if (*DestinationLinesToRequestVMInImmediateFlip >= 8 + || *DestinationLinesToRequestRowInImmediateFlip >= 16 + || TimeForFetchingMetaPTEImmediateFlip + + 2 * TimeForFetchingRowInVBlankImmediateFlip + > min_row_time) + *ImmediateFlipSupportedForPipe = false; + else + *ImmediateFlipSupportedForPipe = true; + } +} + +static unsigned int TruncToValidBPP( + double DecimalBPP, + bool DSCEnabled, + enum output_encoder_class Output, + enum output_format_class Format, + unsigned int DSCInputBitPerComponent) +{ + if (Output == dm_hdmi) { + if (Format == dm_420) { + if (DecimalBPP >= 18) + return 18; + else if (DecimalBPP >= 15) + return 15; + else if (DecimalBPP >= 12) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_444) { + if (DecimalBPP >= 36) + return 36; + else if (DecimalBPP >= 30) + return 30; + else if (DecimalBPP >= 24) + return 24; + else if (DecimalBPP >= 18) + return 18; + else + return BPP_INVALID; + } else { + if (DecimalBPP / 1.5 >= 24) + return 24; + else if (DecimalBPP / 1.5 >= 20) + return 20; + else if (DecimalBPP / 1.5 >= 16) + return 16; + else + return BPP_INVALID; + } + } else { + if (DSCEnabled) { + if (Format == dm_420) { + if (DecimalBPP < 6) + return BPP_INVALID; + else if (DecimalBPP >= 1.5 * DSCInputBitPerComponent - 1 / 16) + return 1.5 * DSCInputBitPerComponent - 1 / 16; + else + return dml_floor(16 * DecimalBPP, 1) / 16; + } else if (Format == dm_n422) { + if (DecimalBPP < 7) + return BPP_INVALID; + else if (DecimalBPP >= 2 * DSCInputBitPerComponent - 1 / 16) + return 2 * DSCInputBitPerComponent - 1 / 16; + else + return dml_floor(16 * DecimalBPP, 1) / 16; + } else { + if (DecimalBPP < 8) + return BPP_INVALID; + else if (DecimalBPP >= 3 * DSCInputBitPerComponent - 1 / 16) + return 3 * DSCInputBitPerComponent - 1 / 16; + else + return dml_floor(16 * DecimalBPP, 1) / 16; + } + } else if (Format == dm_420) { + if (DecimalBPP >= 18) + return 18; + else if (DecimalBPP >= 15) + return 15; + else if (DecimalBPP >= 12) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_s422 || Format == dm_n422) { + if (DecimalBPP >= 24) + return 24; + else if (DecimalBPP >= 20) + return 20; + else if (DecimalBPP >= 16) + return 16; + else + return BPP_INVALID; + } else { + if (DecimalBPP >= 36) + return 36; + else if (DecimalBPP >= 30) + return 30; + else if (DecimalBPP >= 24) + return 24; + else if (DecimalBPP >= 18) + return 18; + else + return BPP_INVALID; + } + } +} + +void dml20v2_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) +{ + struct vba_vars_st *locals = &mode_lib->vba; + + int i; + unsigned int j, k, m; + + /*MODE SUPPORT, VOLTAGE STATE AND SOC CONFIGURATION*/ + + /*Scale Ratio, taps Support Check*/ + + mode_lib->vba.ScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.ScalerEnabled[k] == false + && ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) + || mode_lib->vba.HRatio[k] != 1.0 + || mode_lib->vba.htaps[k] != 1.0 + || mode_lib->vba.VRatio[k] != 1.0 + || mode_lib->vba.vtaps[k] != 1.0)) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } else if (mode_lib->vba.vtaps[k] < 1.0 || mode_lib->vba.vtaps[k] > 8.0 + || mode_lib->vba.htaps[k] < 1.0 || mode_lib->vba.htaps[k] > 8.0 + || (mode_lib->vba.htaps[k] > 1.0 + && (mode_lib->vba.htaps[k] % 2) == 1) + || mode_lib->vba.HRatio[k] > mode_lib->vba.MaxHSCLRatio + || mode_lib->vba.VRatio[k] > mode_lib->vba.MaxVSCLRatio + || mode_lib->vba.HRatio[k] > mode_lib->vba.htaps[k] + || mode_lib->vba.VRatio[k] > mode_lib->vba.vtaps[k] + || (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8 + && (mode_lib->vba.HRatio[k] / 2.0 + > mode_lib->vba.HTAPsChroma[k] + || mode_lib->vba.VRatio[k] / 2.0 + > mode_lib->vba.VTAPsChroma[k]))) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } + } + /*Source Format, Pixel Format and Scan Support Check*/ + + mode_lib->vba.SourceFormatPixelAndScanSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + && mode_lib->vba.SourceScan[k] != dm_horz) + || ((mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_t + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d_x) + && mode_lib->vba.SourcePixelFormat[k] != dm_444_64) + || (mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_r_x + && (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)) + || (((mode_lib->vba.SurfaceTiling[k] == dm_sw_gfx7_2d_thin_gl + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_gfx7_2d_thin_lvp) + && !((mode_lib->vba.SourcePixelFormat[k] + == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] + == dm_444_32) + && mode_lib->vba.SourceScan[k] + == dm_horz + && mode_lib->vba.SupportGFX7CompatibleTilingIn32bppAnd64bpp + == true + && mode_lib->vba.DCCEnable[k] + == false)) + || (mode_lib->vba.DCCEnable[k] == true + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_linear + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)))) { + mode_lib->vba.SourceFormatPixelAndScanSupport = false; + } + } + /*Bandwidth Support Check*/ + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + locals->BytePerPixelInDETY[k] = 8.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + locals->BytePerPixelInDETY[k] = 4.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16) { + locals->BytePerPixelInDETY[k] = 2.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 2.0; + } else { + locals->BytePerPixelInDETY[k] = 4.0 / 3; + locals->BytePerPixelInDETC[k] = 8.0 / 3; + } + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportWidth[k]; + } else { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportHeight[k]; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ReadBandwidthLuma[k] = locals->SwathWidthYSingleDPP[k] * dml_ceil(locals->BytePerPixelInDETY[k], 1.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->ReadBandwidthChroma[k] = locals->SwathWidthYSingleDPP[k] / 2 * dml_ceil(locals->BytePerPixelInDETC[k], 2.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k] / 2.0; + locals->ReadBandwidth[k] = locals->ReadBandwidthLuma[k] + locals->ReadBandwidthChroma[k]; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 4.0; + } else if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 3.0; + } else if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 1.5; + } else { + locals->WriteBandwidth[k] = 0.0; + } + } + mode_lib->vba.DCCEnabledInAnyPlane = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.DCCEnable[k] == true) { + mode_lib->vba.DCCEnabledInAnyPlane = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->FabricAndDRAMBandwidthPerState[i] = dml_min( + mode_lib->vba.DRAMSpeedPerState[i] * mode_lib->vba.NumberOfChannels + * mode_lib->vba.DRAMChannelWidth, + mode_lib->vba.FabricClockPerState[i] + * mode_lib->vba.FabricDatapathToDCNDataReturn) / 1000; + locals->ReturnBWToDCNPerState = dml_min(locals->ReturnBusWidth * locals->DCFCLKPerState[i], + locals->FabricAndDRAMBandwidthPerState[i] * 1000) + * locals->PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly / 100; + + locals->ReturnBWPerState[i] = locals->ReturnBWToDCNPerState; + + if (locals->DCCEnabledInAnyPlane == true && locals->ReturnBWToDCNPerState > locals->DCFCLKPerState[i] * locals->ReturnBusWidth / 4) { + locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + locals->ReturnBWToDCNPerState * 4 * (1 - locals->UrgentLatency / + ((locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + / (locals->ReturnBWToDCNPerState - locals->DCFCLKPerState[i] + * locals->ReturnBusWidth / 4) + locals->UrgentLatency))); + } + locals->CriticalPoint = 2 * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * + locals->UrgentLatency / (locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024); + + if (locals->DCCEnabledInAnyPlane && locals->CriticalPoint > 1 && locals->CriticalPoint < 4) { + locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + 4 * locals->ReturnBWToDCNPerState * + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * locals->UrgentLatency / + dml_pow((locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024), 2)); + } + + locals->ReturnBWToDCNPerState = dml_min(locals->ReturnBusWidth * + locals->DCFCLKPerState[i], locals->FabricAndDRAMBandwidthPerState[i] * 1000); + + if (locals->DCCEnabledInAnyPlane == true && locals->ReturnBWToDCNPerState > locals->DCFCLKPerState[i] * locals->ReturnBusWidth / 4) { + locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + locals->ReturnBWToDCNPerState * 4 * (1 - locals->UrgentLatency / + ((locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + / (locals->ReturnBWToDCNPerState - locals->DCFCLKPerState[i] + * locals->ReturnBusWidth / 4) + locals->UrgentLatency))); + } + locals->CriticalPoint = 2 * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * + locals->UrgentLatency / (locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024); + + if (locals->DCCEnabledInAnyPlane && locals->CriticalPoint > 1 && locals->CriticalPoint < 4) { + locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + 4 * locals->ReturnBWToDCNPerState * + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * locals->UrgentLatency / + dml_pow((locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024), 2)); + } + } + /*Writeback Latency support check*/ + + mode_lib->vba.WritebackLatencySupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + if (locals->WriteBandwidth[k] + > (mode_lib->vba.WritebackInterfaceLumaBufferSize + + mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } else { + if (locals->WriteBandwidth[k] + > 1.5 + * dml_min( + mode_lib->vba.WritebackInterfaceLumaBufferSize, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } + } + } + /*Re-ordering Buffer Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i] = + (mode_lib->vba.RoundTripPingLatencyCycles + 32.0) / mode_lib->vba.DCFCLKPerState[i] + + locals->UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels / locals->ReturnBWPerState[i]; + if ((mode_lib->vba.ROBBufferSizeInKByte - mode_lib->vba.PixelChunkSizeInKByte) * 1024.0 / locals->ReturnBWPerState[i] + > locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i]) { + locals->ROBSupport[i] = true; + } else { + locals->ROBSupport[i] = false; + } + } + /*Writeback Mode Support Check*/ + + mode_lib->vba.TotalNumberOfActiveWriteback = 0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.ActiveWritebacksPerPlane[k] == 0) + mode_lib->vba.ActiveWritebacksPerPlane[k] = 1; + mode_lib->vba.TotalNumberOfActiveWriteback = + mode_lib->vba.TotalNumberOfActiveWriteback + + mode_lib->vba.ActiveWritebacksPerPlane[k]; + } + } + mode_lib->vba.WritebackModeSupport = true; + if (mode_lib->vba.TotalNumberOfActiveWriteback > mode_lib->vba.MaxNumWriteback) { + mode_lib->vba.WritebackModeSupport = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.Writeback10bpc420Supported != true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + mode_lib->vba.WritebackModeSupport = false; + } + } + /*Writeback Scale Ratio and Taps Support Check*/ + + mode_lib->vba.WritebackScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackLumaAndChromaScalingSupported == false + && (mode_lib->vba.WritebackHRatio[k] != 1.0 + || mode_lib->vba.WritebackVRatio[k] != 1.0)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackHRatio[k] > mode_lib->vba.WritebackMaxHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackMaxVSCLRatio + || mode_lib->vba.WritebackHRatio[k] + < mode_lib->vba.WritebackMinHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + < mode_lib->vba.WritebackMinVSCLRatio + || mode_lib->vba.WritebackLumaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackLumaHTaps[k] + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackLumaVTaps[k] + || (mode_lib->vba.WritebackLumaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackLumaHTaps[k] % 2) + == 1)) + || (mode_lib->vba.WritebackPixelFormat[k] != dm_444_32 + && (mode_lib->vba.WritebackChromaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || 2.0 + * mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackChromaHTaps[k] + || 2.0 + * mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackChromaVTaps[k] + || (mode_lib->vba.WritebackChromaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackChromaHTaps[k] % 2) == 1))))) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackVRatio[k] < 1.0) { + mode_lib->vba.WritebackLumaVExtra = + dml_max(1.0 - 2.0 / dml_ceil(1.0 / mode_lib->vba.WritebackVRatio[k], 1.0), 0.0); + } else { + mode_lib->vba.WritebackLumaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_444_32 + && mode_lib->vba.WritebackLumaVTaps[k] + > (mode_lib->vba.WritebackLineBufferLumaBufferSize + + mode_lib->vba.WritebackLineBufferChromaBufferSize) + / 3.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (2.0 * mode_lib->vba.WritebackVRatio[k] < 1) { + mode_lib->vba.WritebackChromaVExtra = 0.0; + } else { + mode_lib->vba.WritebackChromaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + } + } + /*Maximum DISPCLK/DPPCLK Support check*/ + + mode_lib->vba.WritebackRequiredDISPCLK = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.WritebackRequiredDISPCLK = + dml_max( + mode_lib->vba.WritebackRequiredDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.HRatio[k] > 1.0) { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + if (locals->BytePerPixelInDETC[k] == 0.0) { + locals->PSCL_FACTOR_CHROMA[k] = 0.0; + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max3( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } else { + if (mode_lib->vba.HRatio[k] / 2.0 > 1.0) { + locals->PSCL_FACTOR_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2.0 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max5( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2.0), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4.0 + / locals->PSCL_FACTOR_CHROMA[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0 + || mode_lib->vba.HTAPsChroma[k] > 6.0 + || mode_lib->vba.VTAPsChroma[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + dml_ceil(locals->BytePerPixelInDETC[k], 2.0), + &locals->Read256BlockHeightY[k], + &locals->Read256BlockHeightC[k], + &locals->Read256BlockWidthY[k], + &locals->Read256BlockWidthC[k]); + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MaxSwathHeightY[k] = locals->Read256BlockHeightY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockHeightC[k]; + } else { + locals->MaxSwathHeightY[k] = locals->Read256BlockWidthY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockWidthC[k]; + } + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + } + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k] + / 2.0; + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } + } + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + mode_lib->vba.MaximumSwathWidthSupport = 8192.0; + } else { + mode_lib->vba.MaximumSwathWidthSupport = 5120.0; + } + mode_lib->vba.MaximumSwathWidthInDETBuffer = + dml_min( + mode_lib->vba.MaximumSwathWidthSupport, + mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0 + / (locals->BytePerPixelInDETY[k] + * locals->MinSwathHeightY[k] + + locals->BytePerPixelInDETC[k] + / 2.0 + * locals->MinSwathHeightC[k])); + if (locals->BytePerPixelInDETC[k] == 0.0) { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + mode_lib->vba.LineBufferSize + * dml_max(mode_lib->vba.HRatio[k], 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)); + } else { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + dml_min( + mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k], + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)), + 2.0 * mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k] + / 2.0, + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.VTAPsChroma[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k] + / 2.0, + 1.0) + - 2, + 0.0))); + } + locals->MaximumSwathWidth[k] = dml_min( + mode_lib->vba.MaximumSwathWidthInDETBuffer, + mode_lib->vba.MaximumSwathWidthInLineBuffer); + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDispclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDppclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = + mode_lib->vba.PixelClock[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + * (1.0 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = mode_lib->vba.PixelClock[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * (1 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + if (mode_lib->vba.ODMCapability == false || mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine <= mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) { + locals->ODMCombineEnablePerState[i][k] = false; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine; + } else { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } + if (locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) <= mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + && locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k] + && locals->ODMCombineEnablePerState[i][k] == false) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if ((locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity) + || (mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity)) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + if (j == 1) { + while (locals->TotalNumberOfActiveDPP[i][j] < mode_lib->vba.MaxNumDPP + && locals->TotalNumberOfActiveDPP[i][j] < 2 * mode_lib->vba.NumberOfActivePlanes) { + double BWOfNonSplitPlaneOfMaximumBandwidth; + unsigned int NumberOfNonSplitPlaneOfMaximumBandwidth; + + BWOfNonSplitPlaneOfMaximumBandwidth = 0; + NumberOfNonSplitPlaneOfMaximumBandwidth = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->ReadBandwidth[k] > BWOfNonSplitPlaneOfMaximumBandwidth && locals->NoOfDPP[i][j][k] == 1) { + BWOfNonSplitPlaneOfMaximumBandwidth = locals->ReadBandwidth[k]; + NumberOfNonSplitPlaneOfMaximumBandwidth = k; + } + } + locals->NoOfDPP[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = 2; + locals->RequiredDPPCLK[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = + locals->MinDPPCLKUsingSingleDPP[NumberOfNonSplitPlaneOfMaximumBandwidth] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100) / 2; + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + 1; + } + } + if (locals->TotalNumberOfActiveDPP[i][j] > mode_lib->vba.MaxNumDPP) { + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ODMCombineEnablePerState[i][k] = false; + if (locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k]) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + if (i != mode_lib->vba.soc.num_states) { + mode_lib->vba.PlaneRequiredDISPCLK = + mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + * (1.0 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + } else { + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if (locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + || mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.WritebackRequiredDISPCLK); + if (mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity + < mode_lib->vba.WritebackRequiredDISPCLK) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + } + /*Viewport Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->ViewportSizeSupport[i] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ODMCombineEnablePerState[i][k] == true) { + if (dml_min(locals->SwathWidthYSingleDPP[k], dml_round(mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k])) + > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i] = false; + } + } else { + if (locals->SwathWidthYSingleDPP[k] / 2.0 > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i] = false; + } + } + } + } + /*Total Available Pipes Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + if (locals->TotalNumberOfActiveDPP[i][j] <= mode_lib->vba.MaxNumDPP) + locals->TotalAvailablePipesSupport[i][j] = true; + else + locals->TotalAvailablePipesSupport[i][j] = false; + } + } + /*Total Available OTG Support Check*/ + + mode_lib->vba.TotalNumberOfActiveOTG = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.TotalNumberOfActiveOTG = mode_lib->vba.TotalNumberOfActiveOTG + + 1.0; + } + } + if (mode_lib->vba.TotalNumberOfActiveOTG <= mode_lib->vba.MaxNumOTG) { + mode_lib->vba.NumberOfOTGSupport = true; + } else { + mode_lib->vba.NumberOfOTGSupport = false; + } + /*Display IO and DSC Support Check*/ + + mode_lib->vba.NonsupportedDSCInputBPC = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (!(mode_lib->vba.DSCInputBitPerComponent[k] == 12.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 10.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 8.0)) { + mode_lib->vba.NonsupportedDSCInputBPC = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->RequiresDSC[i][k] = 0; + locals->RequiresFEC[i][k] = 0; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.Output[k] == dm_hdmi) { + locals->RequiresDSC[i][k] = 0; + locals->RequiresFEC[i][k] = 0; + locals->OutputBppPerState[i][k] = TruncToValidBPP( + dml_min(600.0, mode_lib->vba.PHYCLKPerState[i]) / mode_lib->vba.PixelClockBackEnd[k] * 24, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + } else if (mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp) { + if (mode_lib->vba.Output[k] == dm_edp) { + mode_lib->vba.EffectiveFECOverhead = 0.0; + } else { + mode_lib->vba.EffectiveFECOverhead = + mode_lib->vba.FECOverhead; + } + if (mode_lib->vba.PHYCLKPerState[i] >= 270.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID && mode_lib->vba.PHYCLKPerState[i] >= 540.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID + && mode_lib->vba.PHYCLKPerState[i] + >= 810.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true || mode_lib->vba.Outbpp == BPP_INVALID) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = + mode_lib->vba.Outbpp; + } + } + } else { + locals->OutputBppPerState[i][k] = BPP_BLENDED_PIPE; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->DIOSupport[i] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->OutputBppPerState[i][k] == BPP_INVALID + || (mode_lib->vba.OutputFormat[k] == dm_420 + && mode_lib->vba.Interlace[k] == true + && mode_lib->vba.ProgressiveToInterlaceUnitInOPP == true)) { + locals->DIOSupport[i] = false; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->DSCCLKRequiredMoreThanSupported[i] = false; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if ((mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp)) { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] + == dm_n422) { + mode_lib->vba.DSCFormatFactor = 2; + } else { + mode_lib->vba.DSCFormatFactor = 1; + } + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] + == true) { + if (mode_lib->vba.PixelClockBackEnd[k] / 6.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } else { + if (mode_lib->vba.PixelClockBackEnd[k] / 3.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } + } + } + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->NotEnoughDSCUnits[i] = false; + mode_lib->vba.TotalDSCUnitsRequired = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == true) { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 2.0; + } else { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 1.0; + } + } + } + if (mode_lib->vba.TotalDSCUnitsRequired > mode_lib->vba.NumberOfDSC) { + locals->NotEnoughDSCUnits[i] = true; + } + } + /*DSC Delay per state*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] != k) { + mode_lib->vba.slices = 0; + } else if (locals->RequiresDSC[i][k] == 0 + || locals->RequiresDSC[i][k] == false) { + mode_lib->vba.slices = 0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 3200.0) { + mode_lib->vba.slices = dml_ceil( + mode_lib->vba.PixelClockBackEnd[k] / 400.0, + 4.0); + } else if (mode_lib->vba.PixelClockBackEnd[k] > 1360.0) { + mode_lib->vba.slices = 8.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 680.0) { + mode_lib->vba.slices = 4.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 340.0) { + mode_lib->vba.slices = 2.0; + } else { + mode_lib->vba.slices = 1.0; + } + if (locals->OutputBppPerState[i][k] == BPP_BLENDED_PIPE + || locals->OutputBppPerState[i][k] == BPP_INVALID) { + mode_lib->vba.bpp = 0.0; + } else { + mode_lib->vba.bpp = locals->OutputBppPerState[i][k]; + } + if (locals->RequiresDSC[i][k] == true && mode_lib->vba.bpp != 0.0) { + if (locals->ODMCombineEnablePerState[i][k] == false) { + locals->DSCDelayPerState[i][k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil( + mode_lib->vba.HActive[k] + / mode_lib->vba.slices, + 1.0), + mode_lib->vba.slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + locals->DSCDelayPerState[i][k] = + 2.0 * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil(mode_lib->vba.HActive[k] / mode_lib->vba.slices, 1.0), + mode_lib->vba.slices / 2, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay(mode_lib->vba.OutputFormat[k])); + } + locals->DSCDelayPerState[i][k] = + locals->DSCDelayPerState[i][k] * mode_lib->vba.PixelClock[k] / mode_lib->vba.PixelClockBackEnd[k]; + } else { + locals->DSCDelayPerState[i][k] = 0.0; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + for (j = 0; j <= mode_lib->vba.NumberOfActivePlanes - 1; j++) { + if (mode_lib->vba.BlendingAndTiming[k] == m && locals->RequiresDSC[i][m] == true) + locals->DSCDelayPerState[i][k] = locals->DSCDelayPerState[i][m]; + } + } + } + } + + //Prefetch Check + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ODMCombineEnablePerState[i][k] == true) + locals->SwathWidthYPerState[i][j][k] = dml_min(locals->SwathWidthYSingleDPP[k], dml_round(locals->HActive[k] / 2 * locals->HRatio[k])); + else + locals->SwathWidthYPerState[i][j][k] = locals->SwathWidthYSingleDPP[k] / locals->NoOfDPP[i][j][k]; + locals->SwathWidthGranularityY = 256 / dml_ceil(locals->BytePerPixelInDETY[k], 1) / locals->MaxSwathHeightY[k]; + locals->RoundedUpMaxSwathSizeBytesY = (dml_ceil(locals->SwathWidthYPerState[i][j][k] - 1, locals->SwathWidthGranularityY) + + locals->SwathWidthGranularityY) * locals->BytePerPixelInDETY[k] * locals->MaxSwathHeightY[k]; + if (locals->SourcePixelFormat[k] == dm_420_10) { + locals->RoundedUpMaxSwathSizeBytesY = dml_ceil(locals->RoundedUpMaxSwathSizeBytesY, 256) + 256; + } + if (locals->MaxSwathHeightC[k] > 0) { + locals->SwathWidthGranularityC = 256 / dml_ceil(locals->BytePerPixelInDETC[k], 2) / locals->MaxSwathHeightC[k]; + + locals->RoundedUpMaxSwathSizeBytesC = (dml_ceil(locals->SwathWidthYPerState[i][j][k] / 2 - 1, locals->SwathWidthGranularityC) + + locals->SwathWidthGranularityC) * locals->BytePerPixelInDETC[k] * locals->MaxSwathHeightC[k]; + } + if (locals->SourcePixelFormat[k] == dm_420_10) { + locals->RoundedUpMaxSwathSizeBytesC = dml_ceil(locals->RoundedUpMaxSwathSizeBytesC, 256) + 256; + } else { + locals->RoundedUpMaxSwathSizeBytesC = 0; + } + + if (locals->RoundedUpMaxSwathSizeBytesY + locals->RoundedUpMaxSwathSizeBytesC <= locals->DETBufferSizeInKByte * 1024 / 2) { + locals->SwathHeightYPerState[i][j][k] = locals->MaxSwathHeightY[k]; + locals->SwathHeightCPerState[i][j][k] = locals->MaxSwathHeightC[k]; + } else { + locals->SwathHeightYPerState[i][j][k] = locals->MinSwathHeightY[k]; + locals->SwathHeightCPerState[i][j][k] = locals->MinSwathHeightC[k]; + } + + if (locals->BytePerPixelInDETC[k] == 0) { + locals->LinesInDETLuma = locals->DETBufferSizeInKByte * 1024 / locals->BytePerPixelInDETY[k] / locals->SwathWidthYPerState[i][j][k]; + locals->LinesInDETChroma = 0; + } else if (locals->SwathHeightYPerState[i][j][k] <= locals->SwathHeightCPerState[i][j][k]) { + locals->LinesInDETLuma = locals->DETBufferSizeInKByte * 1024 / 2 / locals->BytePerPixelInDETY[k] / + locals->SwathWidthYPerState[i][j][k]; + locals->LinesInDETChroma = locals->DETBufferSizeInKByte * 1024 / 2 / locals->BytePerPixelInDETC[k] / (locals->SwathWidthYPerState[i][j][k] / 2); + } else { + locals->LinesInDETLuma = locals->DETBufferSizeInKByte * 1024 * 2 / 3 / locals->BytePerPixelInDETY[k] / locals->SwathWidthYPerState[i][j][k]; + locals->LinesInDETChroma = locals->DETBufferSizeInKByte * 1024 / 3 / locals->BytePerPixelInDETY[k] / (locals->SwathWidthYPerState[i][j][k] / 2); + } + + locals->EffectiveLBLatencyHidingSourceLinesLuma = dml_min(locals->MaxLineBufferLines, + dml_floor(locals->LineBufferSize / locals->LBBitPerPixel[k] / (locals->SwathWidthYPerState[i][j][k] + / dml_max(locals->HRatio[k], 1)), 1)) - (locals->vtaps[k] - 1); + + locals->EffectiveLBLatencyHidingSourceLinesChroma = dml_min(locals->MaxLineBufferLines, + dml_floor(locals->LineBufferSize / locals->LBBitPerPixel[k] + / (locals->SwathWidthYPerState[i][j][k] / 2 + / dml_max(locals->HRatio[k] / 2, 1)), 1)) - (locals->VTAPsChroma[k] - 1); + + locals->EffectiveDETLBLinesLuma = dml_floor(locals->LinesInDETLuma + dml_min( + locals->LinesInDETLuma * locals->RequiredDISPCLK[i][j] * locals->BytePerPixelInDETY[k] * + locals->PSCL_FACTOR[k] / locals->ReturnBWPerState[i], + locals->EffectiveLBLatencyHidingSourceLinesLuma), + locals->SwathHeightYPerState[i][j][k]); + + locals->EffectiveDETLBLinesChroma = dml_floor(locals->LinesInDETChroma + dml_min( + locals->LinesInDETChroma * locals->RequiredDISPCLK[i][j] * locals->BytePerPixelInDETC[k] * + locals->PSCL_FACTOR_CHROMA[k] / locals->ReturnBWPerState[i], + locals->EffectiveLBLatencyHidingSourceLinesChroma), + locals->SwathHeightCPerState[i][j][k]); + + if (locals->BytePerPixelInDETC[k] == 0) { + locals->UrgentLatencySupportUsPerState[i][j][k] = locals->EffectiveDETLBLinesLuma * (locals->HTotal[k] / locals->PixelClock[k]) + / locals->VRatio[k] - locals->EffectiveDETLBLinesLuma * locals->SwathWidthYPerState[i][j][k] * + dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i] / locals->NoOfDPP[i][j][k]); + } else { + locals->UrgentLatencySupportUsPerState[i][j][k] = dml_min( + locals->EffectiveDETLBLinesLuma * (locals->HTotal[k] / locals->PixelClock[k]) + / locals->VRatio[k] - locals->EffectiveDETLBLinesLuma * locals->SwathWidthYPerState[i][j][k] * + dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i] / locals->NoOfDPP[i][j][k]), + locals->EffectiveDETLBLinesChroma * (locals->HTotal[k] / locals->PixelClock[k]) / (locals->VRatio[k] / 2) - + locals->EffectiveDETLBLinesChroma * locals->SwathWidthYPerState[i][j][k] / 2 * + dml_ceil(locals->BytePerPixelInDETC[k], 2) / (locals->ReturnBWPerState[i] / locals->NoOfDPP[i][j][k])); + } + } + } + } + + for (i = 0; i <= locals->soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->UrgentLatencySupport[i][j] = true; + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + if (locals->UrgentLatencySupportUsPerState[i][j][k] < locals->UrgentLatency) + locals->UrgentLatencySupport[i][j] = false; + } + } + } + + + /*Prefetch Check*/ + for (i = 0; i <= locals->soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->TotalNumberOfDCCActiveDPP[i][j] = 0; + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + if (locals->DCCEnable[k] == true) { + locals->TotalNumberOfDCCActiveDPP[i][j] = + locals->TotalNumberOfDCCActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + } + } + } + + CalculateMinAndMaxPrefetchMode(locals->AllowDRAMSelfRefreshOrDRAMClockChangeInVblank, &locals->MinPrefetchMode, &locals->MaxPrefetchMode); + + locals->MaxTotalVActiveRDBandwidth = 0; + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + locals->MaxTotalVActiveRDBandwidth = locals->MaxTotalVActiveRDBandwidth + locals->ReadBandwidth[k]; + } + + for (i = 0; i <= locals->soc.num_states; i++) { + for (j = 0; j < 2; j++) { + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + locals->NoOfDPPThisState[k] = locals->NoOfDPP[i][j][k]; + locals->RequiredDPPCLKThisState[k] = locals->RequiredDPPCLK[i][j][k]; + locals->SwathHeightYThisState[k] = locals->SwathHeightYPerState[i][j][k]; + locals->SwathHeightCThisState[k] = locals->SwathHeightCPerState[i][j][k]; + locals->SwathWidthYThisState[k] = locals->SwathWidthYPerState[i][j][k]; + mode_lib->vba.ProjectedDCFCLKDeepSleep = dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.PixelClock[k] / 16.0); + if (mode_lib->vba.BytePerPixelInDETC[k] == 0.0) { + if (mode_lib->vba.VRatio[k] <= 1.0) { + mode_lib->vba.ProjectedDCFCLKDeepSleep = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep, + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 64.0 + * mode_lib->vba.HRatio[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.NoOfDPP[i][j][k]); + } else { + mode_lib->vba.ProjectedDCFCLKDeepSleep = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep, + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 64.0 + * mode_lib->vba.PSCL_FACTOR[k] + * mode_lib->vba.RequiredDPPCLK[i][j][k]); + } + } else { + if (mode_lib->vba.VRatio[k] <= 1.0) { + mode_lib->vba.ProjectedDCFCLKDeepSleep = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep, + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 32.0 + * mode_lib->vba.HRatio[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.NoOfDPP[i][j][k]); + } else { + mode_lib->vba.ProjectedDCFCLKDeepSleep = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep, + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 32.0 + * mode_lib->vba.PSCL_FACTOR[k] + * mode_lib->vba.RequiredDPPCLK[i][j][k]); + } + if (mode_lib->vba.VRatio[k] / 2.0 <= 1.0) { + mode_lib->vba.ProjectedDCFCLKDeepSleep = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep, + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETC[k], + 2.0) + / 32.0 + * mode_lib->vba.HRatio[k] + / 2.0 + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.NoOfDPP[i][j][k]); + } else { + mode_lib->vba.ProjectedDCFCLKDeepSleep = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep, + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETC[k], + 2.0) + / 32.0 + * mode_lib->vba.PSCL_FACTOR_CHROMA[k] + * mode_lib->vba.RequiredDPPCLK[i][j][k]); + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.Read256BlockHeightY[k], + mode_lib->vba.Read256BlockWidthY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelInDETY[k], 1.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + mode_lib->vba.SwathWidthYPerState[i][j][k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &mode_lib->vba.MacroTileWidthY[k], + &mode_lib->vba.MetaRowBytesY, + &mode_lib->vba.DPTEBytesPerRowY, + &mode_lib->vba.PTEBufferSizeNotExceededY[i][j][k], + &mode_lib->vba.dpte_row_height[k], + &mode_lib->vba.meta_row_height[k]); + mode_lib->vba.PrefetchLinesY[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightYPerState[i][j][k], + mode_lib->vba.ViewportYStartY[k], + &mode_lib->vba.PrefillY[k], + &mode_lib->vba.MaxNumSwY[k]); + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8)) { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.Read256BlockHeightY[k], + mode_lib->vba.Read256BlockWidthY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelInDETC[k], 2.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2.0, + mode_lib->vba.ViewportHeight[k] / 2.0, + mode_lib->vba.SwathWidthYPerState[i][j][k] / 2.0, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchC[k], + 0.0, + &mode_lib->vba.MacroTileWidthC[k], + &mode_lib->vba.MetaRowBytesC, + &mode_lib->vba.DPTEBytesPerRowC, + &mode_lib->vba.PTEBufferSizeNotExceededC[i][j][k], + &mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_height_chroma[k]); + mode_lib->vba.PrefetchLinesC[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k] / 2.0, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightCPerState[i][j][k], + mode_lib->vba.ViewportYStartC[k], + &mode_lib->vba.PrefillC[k], + &mode_lib->vba.MaxNumSwC[k]); + } else { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = 0.0; + mode_lib->vba.MetaRowBytesC = 0.0; + mode_lib->vba.DPTEBytesPerRowC = 0.0; + locals->PrefetchLinesC[k] = 0.0; + locals->PTEBufferSizeNotExceededC[i][j][k] = true; + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma + mode_lib->vba.PTEBufferSizeInRequestsChroma; + } + locals->PDEAndMetaPTEBytesPerFrame[k] = + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC; + locals->MetaRowBytes[k] = mode_lib->vba.MetaRowBytesY + mode_lib->vba.MetaRowBytesC; + locals->DPTEBytesPerRow[k] = mode_lib->vba.DPTEBytesPerRowY + mode_lib->vba.DPTEBytesPerRowC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.MetaRowBytesY, + mode_lib->vba.MetaRowBytesC, + mode_lib->vba.meta_row_height[k], + mode_lib->vba.meta_row_height_chroma[k], + mode_lib->vba.DPTEBytesPerRowY, + mode_lib->vba.DPTEBytesPerRowC, + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_bw[k], + &mode_lib->vba.dpte_row_bw[k], + &mode_lib->vba.qual_row_bw[k]); + } + mode_lib->vba.ExtraLatency = + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatencyPerState[i] + + (mode_lib->vba.TotalNumberOfActiveDPP[i][j] + * mode_lib->vba.PixelChunkSizeInKByte + + mode_lib->vba.TotalNumberOfDCCActiveDPP[i][j] + * mode_lib->vba.MetaChunkSize) + * 1024.0 + / mode_lib->vba.ReturnBWPerState[i]; + if (mode_lib->vba.GPUVMEnable == true) { + mode_lib->vba.ExtraLatency = mode_lib->vba.ExtraLatency + + mode_lib->vba.TotalNumberOfActiveDPP[i][j] + * mode_lib->vba.PTEGroupSize + / mode_lib->vba.ReturnBWPerState[i]; + } + mode_lib->vba.TimeCalc = 24.0 / mode_lib->vba.ProjectedDCFCLKDeepSleep; + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WritebackDelay[i][k] = mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) / locals->RequiredDISPCLK[i][j]; + } else { + locals->WritebackDelay[i][k] = 0.0; + } + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[m] == k + && mode_lib->vba.WritebackEnable[m] + == true) { + locals->WritebackDelay[i][k] = dml_max(locals->WritebackDelay[i][k], + mode_lib->vba.WritebackLatency + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[m], + mode_lib->vba.WritebackHRatio[m], + mode_lib->vba.WritebackVRatio[m], + mode_lib->vba.WritebackLumaHTaps[m], + mode_lib->vba.WritebackLumaVTaps[m], + mode_lib->vba.WritebackChromaHTaps[m], + mode_lib->vba.WritebackChromaVTaps[m], + mode_lib->vba.WritebackDestinationWidth[m]) / locals->RequiredDISPCLK[i][j]); + } + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[k] == m) { + locals->WritebackDelay[i][k] = locals->WritebackDelay[i][m]; + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m < locals->NumberOfCursors[k]; m++) + locals->cursor_bw[k] = locals->NumberOfCursors[k] * locals->CursorWidth[k][m] * locals->CursorBPP[k][m] + / 8 / (locals->HTotal[k] / locals->PixelClock[k]) * locals->VRatio[k]; + } + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->MaximumVStartup[k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] + - dml_max(1.0, dml_ceil(locals->WritebackDelay[i][k] / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]), 1.0)); + } + + mode_lib->vba.NextPrefetchMode = mode_lib->vba.MinPrefetchMode; + do { + mode_lib->vba.PrefetchMode[i][j] = mode_lib->vba.NextPrefetchMode; + mode_lib->vba.NextPrefetchMode = mode_lib->vba.NextPrefetchMode + 1; + + mode_lib->vba.TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[i][j], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthYPerState[i][j][k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TimeCalc, + mode_lib->vba.TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0.0; + } + + CalculateDelayAfterScaler(mode_lib, mode_lib->vba.ReturnBWPerState[i], mode_lib->vba.ReadBandwidthLuma[k], mode_lib->vba.ReadBandwidthChroma[k], mode_lib->vba.MaxTotalVActiveRDBandwidth, + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k], mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k], + mode_lib->vba.RequiredDPPCLK[i][j][k], mode_lib->vba.RequiredDISPCLK[i][j], mode_lib->vba.PixelClock[k], mode_lib->vba.DSCDelayPerState[i][k], mode_lib->vba.NoOfDPP[i][j][k], mode_lib->vba.ScalerEnabled[k], mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.DPPCLKDelaySubtotal, mode_lib->vba.DPPCLKDelaySCL, mode_lib->vba.DPPCLKDelaySCLLBOnly, mode_lib->vba.DPPCLKDelayCNVCFormater, mode_lib->vba.DPPCLKDelayCNVCCursor, mode_lib->vba.DISPCLKDelaySubtotal, + mode_lib->vba.SwathWidthYPerState[i][j][k] / mode_lib->vba.HRatio[k], mode_lib->vba.OutputFormat[k], mode_lib->vba.HTotal[k], + mode_lib->vba.SwathWidthYSingleDPP[k], mode_lib->vba.BytePerPixelInDETY[k], mode_lib->vba.BytePerPixelInDETC[k], mode_lib->vba.SwathHeightYThisState[k], mode_lib->vba.SwathHeightCThisState[k], mode_lib->vba.Interlace[k], mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + &mode_lib->vba.DSTXAfterScaler[k], &mode_lib->vba.DSTYAfterScaler[k]); + + mode_lib->vba.IsErrorResult[i][j][k] = + CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.RequiredDPPCLK[i][j][k], + mode_lib->vba.RequiredDISPCLK[i][j], + mode_lib->vba.PixelClock[k], + mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.NoOfDPP[i][j][k], + mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + mode_lib->vba.MaximumVStartup[k], + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.ExtraLatency, + mode_lib->vba.TimeCalc, + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[k], + mode_lib->vba.MetaRowBytes[k], + mode_lib->vba.DPTEBytesPerRow[k], + mode_lib->vba.PrefetchLinesY[k], + mode_lib->vba.SwathWidthYPerState[i][j][k], + mode_lib->vba.BytePerPixelInDETY[k], + mode_lib->vba.PrefillY[k], + mode_lib->vba.MaxNumSwY[k], + mode_lib->vba.PrefetchLinesC[k], + mode_lib->vba.BytePerPixelInDETC[k], + mode_lib->vba.PrefillC[k], + mode_lib->vba.MaxNumSwC[k], + mode_lib->vba.SwathHeightYPerState[i][j][k], + mode_lib->vba.SwathHeightCPerState[i][j][k], + mode_lib->vba.TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.DSTXAfterScaler[k], + mode_lib->vba.DSTYAfterScaler[k], + &mode_lib->vba.LineTimesForPrefetch[k], + &mode_lib->vba.PrefetchBW[k], + &mode_lib->vba.LinesForMetaPTE[k], + &mode_lib->vba.LinesForMetaAndDPTERow[k], + &mode_lib->vba.VRatioPreY[i][j][k], + &mode_lib->vba.VRatioPreC[i][j][k], + &mode_lib->vba.RequiredPrefetchPixelDataBWLuma[i][j][k], + &mode_lib->vba.Tno_bw[k], + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + } + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = 0.0; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = 0.0; + locals->prefetch_vm_bw_valid = true; + locals->prefetch_row_bw_valid = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->PDEAndMetaPTEBytesPerFrame[k] == 0) + locals->prefetch_vm_bw[k] = 0; + else if (locals->LinesForMetaPTE[k] > 0) + locals->prefetch_vm_bw[k] = locals->PDEAndMetaPTEBytesPerFrame[k] + / (locals->LinesForMetaPTE[k] * locals->HTotal[k] / locals->PixelClock[k]); + else { + locals->prefetch_vm_bw[k] = 0; + locals->prefetch_vm_bw_valid = false; + } + if (locals->MetaRowBytes[k] + locals->DPTEBytesPerRow[k] == 0) + locals->prefetch_row_bw[k] = 0; + else if (locals->LinesForMetaAndDPTERow[k] > 0) + locals->prefetch_row_bw[k] = (locals->MetaRowBytes[k] + locals->DPTEBytesPerRow[k]) + / (locals->LinesForMetaAndDPTERow[k] * locals->HTotal[k] / locals->PixelClock[k]); + else { + locals->prefetch_row_bw[k] = 0; + locals->prefetch_row_bw_valid = false; + } + + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = mode_lib->vba.MaximumReadBandwidthWithPrefetch + + mode_lib->vba.cursor_bw[k] + mode_lib->vba.ReadBandwidth[k] + mode_lib->vba.meta_row_bw[k] + mode_lib->vba.dpte_row_bw[k]; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = + mode_lib->vba.MaximumReadBandwidthWithPrefetch + + mode_lib->vba.cursor_bw[k] + + dml_max3( + mode_lib->vba.prefetch_vm_bw[k], + mode_lib->vba.prefetch_row_bw[k], + dml_max(mode_lib->vba.ReadBandwidth[k], + mode_lib->vba.RequiredPrefetchPixelDataBWLuma[i][j][k]) + + mode_lib->vba.meta_row_bw[k] + mode_lib->vba.dpte_row_bw[k]); + } + locals->BandwidthWithoutPrefetchSupported[i] = true; + if (mode_lib->vba.MaximumReadBandwidthWithoutPrefetch > locals->ReturnBWPerState[i]) { + locals->BandwidthWithoutPrefetchSupported[i] = false; + } + + locals->PrefetchSupported[i][j] = true; + if (mode_lib->vba.MaximumReadBandwidthWithPrefetch > locals->ReturnBWPerState[i]) { + locals->PrefetchSupported[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->LineTimesForPrefetch[k] < 2.0 + || locals->LinesForMetaPTE[k] >= 8.0 + || locals->LinesForMetaAndDPTERow[k] >= 16.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->PrefetchSupported[i][j] = false; + } + } + locals->VRatioInPrefetchSupported[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->VRatioPreY[i][j][k] > 4.0 + || locals->VRatioPreC[i][j][k] > 4.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->VRatioInPrefetchSupported[i][j] = false; + } + } + } while ((locals->PrefetchSupported[i][j] != true || locals->VRatioInPrefetchSupported[i][j] != true) + && mode_lib->vba.NextPrefetchMode < mode_lib->vba.MaxPrefetchMode); + + if (mode_lib->vba.PrefetchSupported[i][j] == true + && mode_lib->vba.VRatioInPrefetchSupported[i][j] == true) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.ReturnBWPerState[i]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.BandwidthAvailableForImmediateFlip + - mode_lib->vba.cursor_bw[k] + - dml_max( + mode_lib->vba.ReadBandwidth[k] + mode_lib->vba.qual_row_bw[k], + mode_lib->vba.PrefetchBW[k]); + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.ImmediateFlipBytes[k] = 0.0; + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.ImmediateFlipBytes[k] = + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[k] + + mode_lib->vba.MetaRowBytes[k] + + mode_lib->vba.DPTEBytesPerRow[k]; + } + } + mode_lib->vba.TotImmediateFlipBytes = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.TotImmediateFlipBytes = + mode_lib->vba.TotImmediateFlipBytes + + mode_lib->vba.ImmediateFlipBytes[k]; + } + } + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.ExtraLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.ImmediateFlipBytes[k], + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.Tno_bw[k], + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[k], + mode_lib->vba.MetaRowBytes[k], + mode_lib->vba.DPTEBytesPerRow[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.meta_row_height[k], + mode_lib->vba.qual_row_bw[k], + &mode_lib->vba.DestinationLinesToRequestVMInImmediateFlip[k], + &mode_lib->vba.DestinationLinesToRequestRowInImmediateFlip[k], + &mode_lib->vba.final_flip_bw[k], + &mode_lib->vba.ImmediateFlipSupportedForPipe[k]); + } + mode_lib->vba.total_dcn_read_bw_with_flip = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.total_dcn_read_bw_with_flip = + mode_lib->vba.total_dcn_read_bw_with_flip + + mode_lib->vba.cursor_bw[k] + + dml_max3( + mode_lib->vba.prefetch_vm_bw[k], + mode_lib->vba.prefetch_row_bw[k], + mode_lib->vba.final_flip_bw[k] + + dml_max( + mode_lib->vba.ReadBandwidth[k], + mode_lib->vba.RequiredPrefetchPixelDataBWLuma[i][j][k])); + } + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = true; + if (mode_lib->vba.total_dcn_read_bw_with_flip + > mode_lib->vba.ReturnBWPerState[i]) { + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.ImmediateFlipSupportedForPipe[k] == false) { + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = false; + } + } + } else { + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = false; + } + } + } + + /*Vertical Active BW support*/ + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i] = dml_min(mode_lib->vba.ReturnBusWidth * + mode_lib->vba.DCFCLKPerState[i], mode_lib->vba.FabricAndDRAMBandwidthPerState[i] * 1000) * + mode_lib->vba.MaxAveragePercentOfIdealDRAMBWDisplayCanUseInNormalSystemOperation / 100; + if (mode_lib->vba.MaxTotalVActiveRDBandwidth <= mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i]) + mode_lib->vba.TotalVerticalActiveBandwidthSupport[i] = true; + else + mode_lib->vba.TotalVerticalActiveBandwidthSupport[i] = false; + } + + /*PTE Buffer Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->PTEBufferSizeNotExceeded[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->PTEBufferSizeNotExceededY[i][j][k] == false + || locals->PTEBufferSizeNotExceededC[i][j][k] == false) { + locals->PTEBufferSizeNotExceeded[i][j] = false; + } + } + } + } + /*Cursor Support Check*/ + mode_lib->vba.CursorSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (j = 0; j < 2; j++) { + if (mode_lib->vba.CursorWidth[k][j] > 0.0) { + if (dml_floor( + dml_floor( + mode_lib->vba.CursorBufferSize + - mode_lib->vba.CursorChunkSize, + mode_lib->vba.CursorChunkSize) * 1024.0 + / (mode_lib->vba.CursorWidth[k][j] + * mode_lib->vba.CursorBPP[k][j] + / 8.0), + 1.0) + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k] < mode_lib->vba.UrgentLatencyPixelDataOnly + || (mode_lib->vba.CursorBPP[k][j] == 64.0 + && mode_lib->vba.Cursor64BppSupport == false)) { + mode_lib->vba.CursorSupport = false; + } + } + } + } + /*Valid Pitch Check*/ + + mode_lib->vba.PitchSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->AlignedYPitch[k] = dml_ceil( + dml_max(mode_lib->vba.PitchY[k], mode_lib->vba.ViewportWidth[k]), + locals->MacroTileWidthY[k]); + if (locals->AlignedYPitch[k] > mode_lib->vba.PitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.DCCEnable[k] == true) { + locals->AlignedDCCMetaPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.DCCMetaPitchY[k], + mode_lib->vba.ViewportWidth[k]), + 64.0 * locals->Read256BlockWidthY[k]); + } else { + locals->AlignedDCCMetaPitch[k] = mode_lib->vba.DCCMetaPitchY[k]; + } + if (locals->AlignedDCCMetaPitch[k] > mode_lib->vba.DCCMetaPitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) { + locals->AlignedCPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.PitchC[k], + mode_lib->vba.ViewportWidth[k] / 2.0), + locals->MacroTileWidthC[k]); + } else { + locals->AlignedCPitch[k] = mode_lib->vba.PitchC[k]; + } + if (locals->AlignedCPitch[k] > mode_lib->vba.PitchC[k]) { + mode_lib->vba.PitchSupport = false; + } + } + /*Mode Support, Voltage State and SOC Configuration*/ + + for (i = mode_lib->vba.soc.num_states; i >= 0; i--) { + for (j = 0; j < 2; j++) { + enum dm_validation_status status = DML_VALIDATION_OK; + + if (mode_lib->vba.ScaleRatioAndTapsSupport != true) { + status = DML_FAIL_SCALE_RATIO_TAP; + } else if (mode_lib->vba.SourceFormatPixelAndScanSupport != true) { + status = DML_FAIL_SOURCE_PIXEL_FORMAT; + } else if (locals->ViewportSizeSupport[i] != true) { + status = DML_FAIL_VIEWPORT_SIZE; + } else if (locals->DIOSupport[i] != true) { + status = DML_FAIL_DIO_SUPPORT; + } else if (locals->NotEnoughDSCUnits[i] != false) { + status = DML_FAIL_NOT_ENOUGH_DSC; + } else if (locals->DSCCLKRequiredMoreThanSupported[i] != false) { + status = DML_FAIL_DSC_CLK_REQUIRED; + } else if (locals->UrgentLatencySupport[i][j] != true) { + status = DML_FAIL_URGENT_LATENCY; + } else if (locals->ROBSupport[i] != true) { + status = DML_FAIL_REORDERING_BUFFER; + } else if (locals->DISPCLK_DPPCLK_Support[i][j] != true) { + status = DML_FAIL_DISPCLK_DPPCLK; + } else if (locals->TotalAvailablePipesSupport[i][j] != true) { + status = DML_FAIL_TOTAL_AVAILABLE_PIPES; + } else if (mode_lib->vba.NumberOfOTGSupport != true) { + status = DML_FAIL_NUM_OTG; + } else if (mode_lib->vba.WritebackModeSupport != true) { + status = DML_FAIL_WRITEBACK_MODE; + } else if (mode_lib->vba.WritebackLatencySupport != true) { + status = DML_FAIL_WRITEBACK_LATENCY; + } else if (mode_lib->vba.WritebackScaleRatioAndTapsSupport != true) { + status = DML_FAIL_WRITEBACK_SCALE_RATIO_TAP; + } else if (mode_lib->vba.CursorSupport != true) { + status = DML_FAIL_CURSOR_SUPPORT; + } else if (mode_lib->vba.PitchSupport != true) { + status = DML_FAIL_PITCH_SUPPORT; + } else if (locals->PrefetchSupported[i][j] != true) { + status = DML_FAIL_PREFETCH_SUPPORT; + } else if (locals->TotalVerticalActiveBandwidthSupport[i] != true) { + status = DML_FAIL_TOTAL_V_ACTIVE_BW; + } else if (locals->VRatioInPrefetchSupported[i][j] != true) { + status = DML_FAIL_V_RATIO_PREFETCH; + } else if (locals->PTEBufferSizeNotExceeded[i][j] != true) { + status = DML_FAIL_PTE_BUFFER_SIZE; + } else if (mode_lib->vba.NonsupportedDSCInputBPC != false) { + status = DML_FAIL_DSC_INPUT_BPC; + } + + if (status == DML_VALIDATION_OK) { + locals->ModeSupport[i][j] = true; + } else { + locals->ModeSupport[i][j] = false; + } + locals->ValidationStatus[i] = status; + } + } + { + unsigned int MaximumMPCCombine = 0; + mode_lib->vba.VoltageLevel = mode_lib->vba.soc.num_states + 1; + for (i = mode_lib->vba.VoltageOverrideLevel; i <= mode_lib->vba.soc.num_states; i++) { + if (locals->ModeSupport[i][0] == true || locals->ModeSupport[i][1] == true) { + mode_lib->vba.VoltageLevel = i; + if (locals->ModeSupport[i][1] == true && (locals->ModeSupport[i][0] == false + || mode_lib->vba.WhenToDoMPCCombine == dm_mpc_always_when_possible)) { + MaximumMPCCombine = 1; + } else { + MaximumMPCCombine = 0; + } + break; + } + } + mode_lib->vba.ImmediateFlipSupport = + locals->ImmediateFlipSupportedForState[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.DPPPerPlane[k] = locals->NoOfDPP[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + locals->DPPCLK[k] = locals->RequiredDPPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + } + mode_lib->vba.DISPCLK = locals->RequiredDISPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + mode_lib->vba.maxMpcComb = MaximumMPCCombine; + } + mode_lib->vba.DCFCLK = mode_lib->vba.DCFCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.DRAMSpeed = mode_lib->vba.DRAMSpeedPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.FabricClock = mode_lib->vba.FabricClockPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.SOCCLK = mode_lib->vba.SOCCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.ReturnBW = locals->ReturnBWPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.FabricAndDRAMBandwidth = locals->FabricAndDRAMBandwidthPerState[mode_lib->vba.VoltageLevel]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.ODMCombineEnabled[k] = + locals->ODMCombineEnablePerState[mode_lib->vba.VoltageLevel][k]; + } else { + mode_lib->vba.ODMCombineEnabled[k] = 0; + } + mode_lib->vba.DSCEnabled[k] = + locals->RequiresDSC[mode_lib->vba.VoltageLevel][k]; + mode_lib->vba.OutputBpp[k] = + locals->OutputBppPerState[mode_lib->vba.VoltageLevel][k]; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.h new file mode 100644 index 000000000000..a989d3ca1e99 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.h @@ -0,0 +1,32 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DCN20V2_DISPLAY_MODE_VBA_H_ +#define _DCN20V2_DISPLAY_MODE_VBA_H_ + +void dml20v2_recalculate(struct display_mode_lib *mode_lib); +void dml20v2_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c new file mode 100644 index 000000000000..ed8bf5f723c9 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c @@ -0,0 +1,1701 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../display_mode_lib.h" +#include "../display_mode_vba.h" +#include "display_rq_dlg_calc_20v2.h" + +// Function: dml20v2_rq_dlg_get_rq_params +// Calculate requestor related parameters that register definition agnostic +// (i.e. this layer does try to separate real values from register definition) +// Input: +// pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.) +// Output: +// rq_param - values that can be used to setup RQ (e.g. swath_height, plane1_addr, etc.) +// +static void dml20v2_rq_dlg_get_rq_params( + struct display_mode_lib *mode_lib, + display_rq_params_st * rq_param, + const display_pipe_source_params_st pipe_src_param); + +// Function: dml20v2_rq_dlg_get_dlg_params +// Calculate deadline related parameters +// +static void dml20v2_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, + const display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + display_dlg_regs_st *disp_dlg_regs, + display_ttu_regs_st *disp_ttu_regs, + const display_rq_dlg_params_st rq_dlg_param, + const display_dlg_sys_params_st dlg_sys_param, + const bool cstate_en, + const bool pstate_en); +/* + * NOTE: + * This file is gcc-parseable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ + +static void calculate_ttu_cursor(struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp); + +#include "../dml_inline_defs.h" + +static unsigned int get_bytes_per_element(enum source_format_class source_format, bool is_chroma) +{ + unsigned int ret_val = 0; + + if (source_format == dm_444_16) { + if (!is_chroma) + ret_val = 2; + } else if (source_format == dm_444_32) { + if (!is_chroma) + ret_val = 4; + } else if (source_format == dm_444_64) { + if (!is_chroma) + ret_val = 8; + } else if (source_format == dm_420_8) { + if (is_chroma) + ret_val = 2; + else + ret_val = 1; + } else if (source_format == dm_420_10) { + if (is_chroma) + ret_val = 4; + else + ret_val = 2; + } else if (source_format == dm_444_8) { + ret_val = 1; + } + return ret_val; +} + +static bool is_dual_plane(enum source_format_class source_format) +{ + bool ret_val = 0; + + if ((source_format == dm_420_8) || (source_format == dm_420_10)) + ret_val = 1; + + return ret_val; +} + +static double get_refcyc_per_delivery(struct display_mode_lib *mode_lib, + double refclk_freq_in_mhz, + double pclk_freq_in_mhz, + bool odm_combine, + unsigned int recout_width, + unsigned int hactive, + double vratio, + double hscale_pixel_rate, + unsigned int delivery_width, + unsigned int req_per_swath_ub) +{ + double refcyc_per_delivery = 0.0; + + if (vratio <= 1.0) { + if (odm_combine) + refcyc_per_delivery = (double) refclk_freq_in_mhz + * dml_min((double) recout_width, (double) hactive / 2.0) + / pclk_freq_in_mhz / (double) req_per_swath_ub; + else + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) recout_width + / pclk_freq_in_mhz / (double) req_per_swath_ub; + } else { + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) delivery_width + / (double) hscale_pixel_rate / (double) req_per_swath_ub; + } + + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: recout_width = %d\n", __func__, recout_width); + dml_print("DML_DLG: %s: vratio = %3.2f\n", __func__, vratio); + dml_print("DML_DLG: %s: req_per_swath_ub = %d\n", __func__, req_per_swath_ub); + dml_print("DML_DLG: %s: refcyc_per_delivery= %3.2f\n", __func__, refcyc_per_delivery); + + return refcyc_per_delivery; + +} + +static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_size) +{ + if (tile_size == dm_256k_tile) + return (256 * 1024); + else if (tile_size == dm_64k_tile) + return (64 * 1024); + else + return (4 * 1024); +} + +static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib, + display_data_rq_regs_st *rq_regs, + const display_data_rq_sizing_params_st rq_sizing) +{ + dml_print("DML_DLG: %s: rq_sizing param\n", __func__); + print__data_rq_sizing_params_st(mode_lib, rq_sizing); + + rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10; + + if (rq_sizing.min_chunk_bytes == 0) + rq_regs->min_chunk_size = 0; + else + rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1; + + rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10; + if (rq_sizing.min_meta_chunk_bytes == 0) + rq_regs->min_meta_chunk_size = 0; + else + rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1; + + rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6; + rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6; +} + +static void extract_rq_regs(struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_rq_params_st rq_param) +{ + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + unsigned int detile_buf_plane1_addr = 0; + + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l); + + rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_l.dpte_row_height), + 1) - 3; + + if (rq_param.yuv420) { + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c); + rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_c.dpte_row_height), + 1) - 3; + } + + rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height); + rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height); + + // FIXME: take the max between luma, chroma chunk size? + // okay for now, as we are setting chunk_bytes to 8kb anyways + if (rq_param.sizing.rq_l.chunk_bytes >= 32 * 1024) { //32kb + rq_regs->drq_expansion_mode = 0; + } else { + rq_regs->drq_expansion_mode = 2; + } + rq_regs->prq_expansion_mode = 1; + rq_regs->mrq_expansion_mode = 1; + rq_regs->crq_expansion_mode = 1; + + if (rq_param.yuv420) { + if ((double) rq_param.misc.rq_l.stored_swath_bytes + / (double) rq_param.misc.rq_c.stored_swath_bytes <= 1.5) { + detile_buf_plane1_addr = (detile_buf_size_in_bytes / 2.0 / 64.0); // half to chroma + } else { + detile_buf_plane1_addr = dml_round_to_multiple((unsigned int) ((2.0 * detile_buf_size_in_bytes) / 3.0), + 256, + 0) / 64.0; // 2/3 to chroma + } + } + rq_regs->plane1_base_address = detile_buf_plane1_addr; +} + +static void handle_det_buf_split(struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_source_params_st pipe_src_param) +{ + unsigned int total_swath_bytes = 0; + unsigned int swath_bytes_l = 0; + unsigned int swath_bytes_c = 0; + unsigned int full_swath_bytes_packed_l = 0; + unsigned int full_swath_bytes_packed_c = 0; + bool req128_l = 0; + bool req128_c = 0; + bool surf_linear = (pipe_src_param.sw_mode == dm_sw_linear); + bool surf_vert = (pipe_src_param.source_scan == dm_vert); + unsigned int log2_swath_height_l = 0; + unsigned int log2_swath_height_c = 0; + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + + full_swath_bytes_packed_l = rq_param->misc.rq_l.full_swath_bytes; + full_swath_bytes_packed_c = rq_param->misc.rq_c.full_swath_bytes; + + if (rq_param->yuv420_10bpc) { + full_swath_bytes_packed_l = dml_round_to_multiple(rq_param->misc.rq_l.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + full_swath_bytes_packed_c = dml_round_to_multiple(rq_param->misc.rq_c.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + } + + if (rq_param->yuv420) { + total_swath_bytes = 2 * full_swath_bytes_packed_l + 2 * full_swath_bytes_packed_c; + + if (total_swath_bytes <= detile_buf_size_in_bytes) { //full 256b request + req128_l = 0; + req128_c = 0; + swath_bytes_l = full_swath_bytes_packed_l; + swath_bytes_c = full_swath_bytes_packed_c; + } else { //128b request (for luma only for yuv420 8bpc) + req128_l = 1; + req128_c = 0; + swath_bytes_l = full_swath_bytes_packed_l / 2; + swath_bytes_c = full_swath_bytes_packed_c; + } + // Note: assumption, the config that pass in will fit into + // the detiled buffer. + } else { + total_swath_bytes = 2 * full_swath_bytes_packed_l; + + if (total_swath_bytes <= detile_buf_size_in_bytes) + req128_l = 0; + else + req128_l = 1; + + swath_bytes_l = total_swath_bytes; + swath_bytes_c = 0; + } + rq_param->misc.rq_l.stored_swath_bytes = swath_bytes_l; + rq_param->misc.rq_c.stored_swath_bytes = swath_bytes_c; + + if (surf_linear) { + log2_swath_height_l = 0; + log2_swath_height_c = 0; + } else if (!surf_vert) { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_height) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_height) - req128_c; + } else { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_width) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_width) - req128_c; + } + rq_param->dlg.rq_l.swath_height = 1 << log2_swath_height_l; + rq_param->dlg.rq_c.swath_height = 1 << log2_swath_height_c; + + dml_print("DML_DLG: %s: req128_l = %0d\n", __func__, req128_l); + dml_print("DML_DLG: %s: req128_c = %0d\n", __func__, req128_c); + dml_print("DML_DLG: %s: full_swath_bytes_packed_l = %0d\n", + __func__, + full_swath_bytes_packed_l); + dml_print("DML_DLG: %s: full_swath_bytes_packed_c = %0d\n", + __func__, + full_swath_bytes_packed_c); +} + +static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + display_data_rq_sizing_params_st *rq_sizing_param, + unsigned int vp_width, + unsigned int vp_height, + unsigned int data_pitch, + unsigned int meta_pitch, + unsigned int source_format, + unsigned int tiling, + unsigned int macro_tile_size, + unsigned int source_scan, + unsigned int is_chroma) +{ + bool surf_linear = (tiling == dm_sw_linear); + bool surf_vert = (source_scan == dm_vert); + + unsigned int bytes_per_element; + unsigned int bytes_per_element_y = get_bytes_per_element((enum source_format_class)(source_format), + false); + unsigned int bytes_per_element_c = get_bytes_per_element((enum source_format_class)(source_format), + true); + + unsigned int blk256_width = 0; + unsigned int blk256_height = 0; + + unsigned int blk256_width_y = 0; + unsigned int blk256_height_y = 0; + unsigned int blk256_width_c = 0; + unsigned int blk256_height_c = 0; + unsigned int log2_bytes_per_element; + unsigned int log2_blk256_width; + unsigned int log2_blk256_height; + unsigned int blk_bytes; + unsigned int log2_blk_bytes; + unsigned int log2_blk_height; + unsigned int log2_blk_width; + unsigned int log2_meta_req_bytes; + unsigned int log2_meta_req_height; + unsigned int log2_meta_req_width; + unsigned int meta_req_width; + unsigned int meta_req_height; + unsigned int log2_meta_row_height; + unsigned int meta_row_width_ub; + unsigned int log2_meta_chunk_bytes; + unsigned int log2_meta_chunk_height; + + //full sized meta chunk width in unit of data elements + unsigned int log2_meta_chunk_width; + unsigned int log2_min_meta_chunk_bytes; + unsigned int min_meta_chunk_width; + unsigned int meta_chunk_width; + unsigned int meta_chunk_per_row_int; + unsigned int meta_row_remainder; + unsigned int meta_chunk_threshold; + unsigned int meta_blk_bytes; + unsigned int meta_blk_height; + unsigned int meta_blk_width; + unsigned int meta_surface_bytes; + unsigned int vmpg_bytes; + unsigned int meta_pte_req_per_frame_ub; + unsigned int meta_pte_bytes_per_frame_ub; + const unsigned int log2_vmpg_bytes = dml_log2(mode_lib->soc.vmm_page_size_bytes); + const unsigned int dpte_buf_in_pte_reqs = mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma; + const unsigned int pde_proc_buffer_size_64k_reqs = + mode_lib->ip.pde_proc_buffer_size_64k_reqs; + + unsigned int log2_vmpg_height = 0; + unsigned int log2_vmpg_width = 0; + unsigned int log2_dpte_req_height_ptes = 0; + unsigned int log2_dpte_req_height = 0; + unsigned int log2_dpte_req_width = 0; + unsigned int log2_dpte_row_height_linear = 0; + unsigned int log2_dpte_row_height = 0; + unsigned int log2_dpte_group_width = 0; + unsigned int dpte_row_width_ub = 0; + unsigned int dpte_req_height = 0; + unsigned int dpte_req_width = 0; + unsigned int dpte_group_width = 0; + unsigned int log2_dpte_group_bytes = 0; + unsigned int log2_dpte_group_length = 0; + unsigned int pde_buf_entries; + bool yuv420 = (source_format == dm_420_8 || source_format == dm_420_10); + + Calculate256BBlockSizes((enum source_format_class)(source_format), + (enum dm_swizzle_mode)(tiling), + bytes_per_element_y, + bytes_per_element_c, + &blk256_height_y, + &blk256_height_c, + &blk256_width_y, + &blk256_width_c); + + if (!is_chroma) { + blk256_width = blk256_width_y; + blk256_height = blk256_height_y; + bytes_per_element = bytes_per_element_y; + } else { + blk256_width = blk256_width_c; + blk256_height = blk256_height_c; + bytes_per_element = bytes_per_element_c; + } + + log2_bytes_per_element = dml_log2(bytes_per_element); + + dml_print("DML_DLG: %s: surf_linear = %d\n", __func__, surf_linear); + dml_print("DML_DLG: %s: surf_vert = %d\n", __func__, surf_vert); + dml_print("DML_DLG: %s: blk256_width = %d\n", __func__, blk256_width); + dml_print("DML_DLG: %s: blk256_height = %d\n", __func__, blk256_height); + + log2_blk256_width = dml_log2((double) blk256_width); + log2_blk256_height = dml_log2((double) blk256_height); + blk_bytes = surf_linear ? + 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); + log2_blk_bytes = dml_log2((double) blk_bytes); + log2_blk_height = 0; + log2_blk_width = 0; + + // remember log rule + // "+" in log is multiply + // "-" in log is divide + // "/2" is like square root + // blk is vertical biased + if (tiling != dm_sw_linear) + log2_blk_height = log2_blk256_height + + dml_ceil((double) (log2_blk_bytes - 8) / 2.0, 1); + else + log2_blk_height = 0; // blk height of 1 + + log2_blk_width = log2_blk_bytes - log2_bytes_per_element - log2_blk_height; + + if (!surf_vert) { + rq_dlg_param->swath_width_ub = dml_round_to_multiple(vp_width - 1, blk256_width, 1) + + blk256_width; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_width; + } else { + rq_dlg_param->swath_width_ub = dml_round_to_multiple(vp_height - 1, blk256_height, 1) + + blk256_height; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_height; + } + + if (!surf_vert) + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_height + * bytes_per_element; + else + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_width + * bytes_per_element; + + rq_misc_param->blk256_height = blk256_height; + rq_misc_param->blk256_width = blk256_width; + + // ------- + // meta + // ------- + log2_meta_req_bytes = 6; // meta request is 64b and is 8x8byte meta element + + // each 64b meta request for dcn is 8x8 meta elements and + // a meta element covers one 256b block of the the data surface. + log2_meta_req_height = log2_blk256_height + 3; // meta req is 8x8 byte, each byte represent 1 blk256 + log2_meta_req_width = log2_meta_req_bytes + 8 - log2_bytes_per_element + - log2_meta_req_height; + meta_req_width = 1 << log2_meta_req_width; + meta_req_height = 1 << log2_meta_req_height; + log2_meta_row_height = 0; + meta_row_width_ub = 0; + + // the dimensions of a meta row are meta_row_width x meta_row_height in elements. + // calculate upper bound of the meta_row_width + if (!surf_vert) { + log2_meta_row_height = log2_meta_req_height; + meta_row_width_ub = dml_round_to_multiple(vp_width - 1, meta_req_width, 1) + + meta_req_width; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_width; + } else { + log2_meta_row_height = log2_meta_req_width; + meta_row_width_ub = dml_round_to_multiple(vp_height - 1, meta_req_height, 1) + + meta_req_height; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_height; + } + rq_dlg_param->meta_bytes_per_row_ub = rq_dlg_param->meta_req_per_row_ub * 64; + + rq_dlg_param->meta_row_height = 1 << log2_meta_row_height; + + log2_meta_chunk_bytes = dml_log2(rq_sizing_param->meta_chunk_bytes); + log2_meta_chunk_height = log2_meta_row_height; + + //full sized meta chunk width in unit of data elements + log2_meta_chunk_width = log2_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height; + log2_min_meta_chunk_bytes = dml_log2(rq_sizing_param->min_meta_chunk_bytes); + min_meta_chunk_width = 1 + << (log2_min_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height); + meta_chunk_width = 1 << log2_meta_chunk_width; + meta_chunk_per_row_int = (unsigned int) (meta_row_width_ub / meta_chunk_width); + meta_row_remainder = meta_row_width_ub % meta_chunk_width; + meta_chunk_threshold = 0; + meta_blk_bytes = 4096; + meta_blk_height = blk256_height * 64; + meta_blk_width = meta_blk_bytes * 256 / bytes_per_element / meta_blk_height; + meta_surface_bytes = meta_pitch + * (dml_round_to_multiple(vp_height - 1, meta_blk_height, 1) + meta_blk_height) + * bytes_per_element / 256; + vmpg_bytes = mode_lib->soc.vmm_page_size_bytes; + meta_pte_req_per_frame_ub = (dml_round_to_multiple(meta_surface_bytes - vmpg_bytes, + 8 * vmpg_bytes, + 1) + 8 * vmpg_bytes) / (8 * vmpg_bytes); + meta_pte_bytes_per_frame_ub = meta_pte_req_per_frame_ub * 64; //64B mpte request + rq_dlg_param->meta_pte_bytes_per_frame_ub = meta_pte_bytes_per_frame_ub; + + dml_print("DML_DLG: %s: meta_blk_height = %d\n", __func__, meta_blk_height); + dml_print("DML_DLG: %s: meta_blk_width = %d\n", __func__, meta_blk_width); + dml_print("DML_DLG: %s: meta_surface_bytes = %d\n", __func__, meta_surface_bytes); + dml_print("DML_DLG: %s: meta_pte_req_per_frame_ub = %d\n", + __func__, + meta_pte_req_per_frame_ub); + dml_print("DML_DLG: %s: meta_pte_bytes_per_frame_ub = %d\n", + __func__, + meta_pte_bytes_per_frame_ub); + + if (!surf_vert) + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_width; + else + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_height; + + if (meta_row_remainder <= meta_chunk_threshold) + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 1; + else + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 2; + + // ------ + // dpte + // ------ + if (surf_linear) { + log2_vmpg_height = 0; // one line high + } else { + log2_vmpg_height = (log2_vmpg_bytes - 8) / 2 + log2_blk256_height; + } + log2_vmpg_width = log2_vmpg_bytes - log2_bytes_per_element - log2_vmpg_height; + + // only 3 possible shapes for dpte request in dimensions of ptes: 8x1, 4x2, 2x4. + if (surf_linear) { //one 64B PTE request returns 8 PTEs + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_vmpg_width + 3; + log2_dpte_req_height = 0; + } else if (log2_blk_bytes == 12) { //4KB tile means 4kB page size + //one 64B req gives 8x1 PTEs for 4KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } else if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) { // tile block >= 64KB + //two 64B reqs of 2x4 PTEs give 16 PTEs to cover 64KB + log2_dpte_req_height_ptes = 4; + log2_dpte_req_width = log2_blk256_width + 4; // log2_64KB_width + log2_dpte_req_height = log2_blk256_height + 4; // log2_64KB_height + } else { //64KB page size and must 64KB tile block + //one 64B req gives 8x1 PTEs for 64KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } + + // The dpte request dimensions in data elements is dpte_req_width x dpte_req_height + // log2_vmpg_width is how much 1 pte represent, now calculating how much a 64b pte req represent + // That depends on the pte shape (i.e. 8x1, 4x2, 2x4) + //log2_dpte_req_height = log2_vmpg_height + log2_dpte_req_height_ptes; + //log2_dpte_req_width = log2_vmpg_width + log2_dpte_req_width_ptes; + dpte_req_height = 1 << log2_dpte_req_height; + dpte_req_width = 1 << log2_dpte_req_width; + + // calculate pitch dpte row buffer can hold + // round the result down to a power of two. + pde_buf_entries = yuv420 ? (pde_proc_buffer_size_64k_reqs >> 1) : pde_proc_buffer_size_64k_reqs; + if (surf_linear) { + unsigned int dpte_row_height; + + log2_dpte_row_height_linear = dml_floor(dml_log2(dml_min(64 * 1024 * pde_buf_entries + / bytes_per_element, + dpte_buf_in_pte_reqs + * dpte_req_width) + / data_pitch), + 1); + + ASSERT(log2_dpte_row_height_linear >= 3); + + if (log2_dpte_row_height_linear > 7) + log2_dpte_row_height_linear = 7; + + log2_dpte_row_height = log2_dpte_row_height_linear; + // For linear, the dpte row is pitch dependent and the pte requests wrap at the pitch boundary. + // the dpte_row_width_ub is the upper bound of data_pitch*dpte_row_height in elements with this unique buffering. + dpte_row_height = 1 << log2_dpte_row_height; + dpte_row_width_ub = dml_round_to_multiple(data_pitch * dpte_row_height - 1, + dpte_req_width, + 1) + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + // the upper bound of the dpte_row_width without dependency on viewport position follows. + // for tiled mode, row height is the same as req height and row store up to vp size upper bound + if (!surf_vert) { + log2_dpte_row_height = log2_dpte_req_height; + dpte_row_width_ub = dml_round_to_multiple(vp_width - 1, dpte_req_width, 1) + + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + log2_dpte_row_height = + (log2_blk_width < log2_dpte_req_width) ? + log2_blk_width : log2_dpte_req_width; + dpte_row_width_ub = dml_round_to_multiple(vp_height - 1, dpte_req_height, 1) + + dpte_req_height; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_height; + } + } + if (log2_blk_bytes >= 16 && log2_vmpg_bytes == 12) // tile block >= 64KB + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 128; //2*64B dpte request + else + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 64; //64B dpte request + + rq_dlg_param->dpte_row_height = 1 << log2_dpte_row_height; + + // the dpte_group_bytes is reduced for the specific case of vertical + // access of a tile surface that has dpte request of 8x1 ptes. + if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + rq_sizing_param->dpte_group_bytes = 512; + else + //full size + rq_sizing_param->dpte_group_bytes = 2048; + + //since pte request size is 64byte, the number of data pte requests per full sized group is as follows. + log2_dpte_group_bytes = dml_log2(rq_sizing_param->dpte_group_bytes); + log2_dpte_group_length = log2_dpte_group_bytes - 6; //length in 64b requests + + // full sized data pte group width in elements + if (!surf_vert) + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_width; + else + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_height; + + //But if the tile block >=64KB and the page size is 4KB, then each dPTE request is 2*64B + if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) // tile block >= 64KB + log2_dpte_group_width = log2_dpte_group_width - 1; + + dpte_group_width = 1 << log2_dpte_group_width; + + // since dpte groups are only aligned to dpte_req_width and not dpte_group_width, + // the upper bound for the dpte groups per row is as follows. + rq_dlg_param->dpte_groups_per_row_ub = dml_ceil((double) dpte_row_width_ub / dpte_group_width, + 1); +} + +static void get_surf_rq_param(struct display_mode_lib *mode_lib, + display_data_rq_sizing_params_st *rq_sizing_param, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + const display_pipe_source_params_st pipe_src_param, + bool is_chroma) +{ + bool mode_422 = 0; + unsigned int vp_width = 0; + unsigned int vp_height = 0; + unsigned int data_pitch = 0; + unsigned int meta_pitch = 0; + unsigned int ppe = mode_422 ? 2 : 1; + + // FIXME check if ppe apply for both luma and chroma in 422 case + if (is_chroma) { + vp_width = pipe_src_param.viewport_width_c / ppe; + vp_height = pipe_src_param.viewport_height_c; + data_pitch = pipe_src_param.data_pitch_c; + meta_pitch = pipe_src_param.meta_pitch_c; + } else { + vp_width = pipe_src_param.viewport_width / ppe; + vp_height = pipe_src_param.viewport_height; + data_pitch = pipe_src_param.data_pitch; + meta_pitch = pipe_src_param.meta_pitch; + } + + rq_sizing_param->chunk_bytes = 8192; + + if (rq_sizing_param->chunk_bytes == 64 * 1024) + rq_sizing_param->min_chunk_bytes = 0; + else + rq_sizing_param->min_chunk_bytes = 1024; + + rq_sizing_param->meta_chunk_bytes = 2048; + rq_sizing_param->min_meta_chunk_bytes = 256; + + rq_sizing_param->mpte_group_bytes = 2048; + + get_meta_and_pte_attr(mode_lib, + rq_dlg_param, + rq_misc_param, + rq_sizing_param, + vp_width, + vp_height, + data_pitch, + meta_pitch, + pipe_src_param.source_format, + pipe_src_param.sw_mode, + pipe_src_param.macro_tile_size, + pipe_src_param.source_scan, + is_chroma); +} + +static void dml20v2_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_source_params_st pipe_src_param) +{ + // get param for luma surface + rq_param->yuv420 = pipe_src_param.source_format == dm_420_8 + || pipe_src_param.source_format == dm_420_10; + rq_param->yuv420_10bpc = pipe_src_param.source_format == dm_420_10; + + get_surf_rq_param(mode_lib, + &(rq_param->sizing.rq_l), + &(rq_param->dlg.rq_l), + &(rq_param->misc.rq_l), + pipe_src_param, + 0); + + if (is_dual_plane((enum source_format_class)(pipe_src_param.source_format))) { + // get param for chroma surface + get_surf_rq_param(mode_lib, + &(rq_param->sizing.rq_c), + &(rq_param->dlg.rq_c), + &(rq_param->misc.rq_c), + pipe_src_param, + 1); + } + + // calculate how to split the det buffer space between luma and chroma + handle_det_buf_split(mode_lib, rq_param, pipe_src_param); + print__rq_params_st(mode_lib, *rq_param); +} + +void dml20v2_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param) +{ + display_rq_params_st rq_param = {0}; + + memset(rq_regs, 0, sizeof(*rq_regs)); + dml20v2_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param.src); + extract_rq_regs(mode_lib, rq_regs, rq_param); + + print__rq_regs_st(mode_lib, *rq_regs); +} + +// Note: currently taken in as is. +// Nice to decouple code from hw register implement and extract code that are repeated for luma and chroma. +static void dml20v2_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, + const display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + display_dlg_regs_st *disp_dlg_regs, + display_ttu_regs_st *disp_ttu_regs, + const display_rq_dlg_params_st rq_dlg_param, + const display_dlg_sys_params_st dlg_sys_param, + const bool cstate_en, + const bool pstate_en) +{ + const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src; + const display_pipe_dest_params_st *dst = &e2e_pipe_param[pipe_idx].pipe.dest; + const display_output_params_st *dout = &e2e_pipe_param[pipe_idx].dout; + const display_clocks_and_cfg_st *clks = &e2e_pipe_param[pipe_idx].clks_cfg; + const scaler_ratio_depth_st *scl = &e2e_pipe_param[pipe_idx].pipe.scale_ratio_depth; + const scaler_taps_st *taps = &e2e_pipe_param[pipe_idx].pipe.scale_taps; + + // ------------------------- + // Section 1.15.2.1: OTG dependent Params + // ------------------------- + // Timing + unsigned int htotal = dst->htotal; +// unsigned int hblank_start = dst.hblank_start; // TODO: Remove + unsigned int hblank_end = dst->hblank_end; + unsigned int vblank_start = dst->vblank_start; + unsigned int vblank_end = dst->vblank_end; + unsigned int min_vblank = mode_lib->ip.min_vblank_lines; + + double dppclk_freq_in_mhz = clks->dppclk_mhz; + double dispclk_freq_in_mhz = clks->dispclk_mhz; + double refclk_freq_in_mhz = clks->refclk_mhz; + double pclk_freq_in_mhz = dst->pixel_rate_mhz; + bool interlaced = dst->interlaced; + + double ref_freq_to_pix_freq = refclk_freq_in_mhz / pclk_freq_in_mhz; + + double min_dcfclk_mhz; + double t_calc_us; + double min_ttu_vblank; + + double min_dst_y_ttu_vblank; + unsigned int dlg_vblank_start; + bool dual_plane; + bool mode_422; + unsigned int access_dir; + unsigned int vp_height_l; + unsigned int vp_width_l; + unsigned int vp_height_c; + unsigned int vp_width_c; + + // Scaling + unsigned int htaps_l; + unsigned int htaps_c; + double hratio_l; + double hratio_c; + double vratio_l; + double vratio_c; + bool scl_enable; + + double line_time_in_us; + // double vinit_l; + // double vinit_c; + // double vinit_bot_l; + // double vinit_bot_c; + + // unsigned int swath_height_l; + unsigned int swath_width_ub_l; + // unsigned int dpte_bytes_per_row_ub_l; + unsigned int dpte_groups_per_row_ub_l; + // unsigned int meta_pte_bytes_per_frame_ub_l; + // unsigned int meta_bytes_per_row_ub_l; + + // unsigned int swath_height_c; + unsigned int swath_width_ub_c; + // unsigned int dpte_bytes_per_row_ub_c; + unsigned int dpte_groups_per_row_ub_c; + + unsigned int meta_chunks_per_row_ub_l; + unsigned int meta_chunks_per_row_ub_c; + unsigned int vupdate_offset; + unsigned int vupdate_width; + unsigned int vready_offset; + + unsigned int dppclk_delay_subtotal; + unsigned int dispclk_delay_subtotal; + unsigned int pixel_rate_delay_subtotal; + + unsigned int vstartup_start; + unsigned int dst_x_after_scaler; + unsigned int dst_y_after_scaler; + double line_wait; + double dst_y_prefetch; + double dst_y_per_vm_vblank; + double dst_y_per_row_vblank; + double dst_y_per_vm_flip; + double dst_y_per_row_flip; + double min_dst_y_per_vm_vblank; + double min_dst_y_per_row_vblank; + double lsw; + double vratio_pre_l; + double vratio_pre_c; + unsigned int req_per_swath_ub_l; + unsigned int req_per_swath_ub_c; + unsigned int meta_row_height_l; + unsigned int meta_row_height_c; + unsigned int swath_width_pixels_ub_l; + unsigned int swath_width_pixels_ub_c; + unsigned int scaler_rec_in_width_l; + unsigned int scaler_rec_in_width_c; + unsigned int dpte_row_height_l; + unsigned int dpte_row_height_c; + double hscale_pixel_rate_l; + double hscale_pixel_rate_c; + double min_hratio_fact_l; + double min_hratio_fact_c; + double refcyc_per_line_delivery_pre_l; + double refcyc_per_line_delivery_pre_c; + double refcyc_per_line_delivery_l; + double refcyc_per_line_delivery_c; + + double refcyc_per_req_delivery_pre_l; + double refcyc_per_req_delivery_pre_c; + double refcyc_per_req_delivery_l; + double refcyc_per_req_delivery_c; + + unsigned int full_recout_width; + double xfc_transfer_delay; + double xfc_precharge_delay; + double xfc_remote_surface_flip_latency; + double xfc_dst_y_delta_drq_limit; + double xfc_prefetch_margin; + double refcyc_per_req_delivery_pre_cur0; + double refcyc_per_req_delivery_cur0; + double refcyc_per_req_delivery_pre_cur1; + double refcyc_per_req_delivery_cur1; + + memset(disp_dlg_regs, 0, sizeof(*disp_dlg_regs)); + memset(disp_ttu_regs, 0, sizeof(*disp_ttu_regs)); + + dml_print("DML_DLG: %s: cstate_en = %d\n", __func__, cstate_en); + dml_print("DML_DLG: %s: pstate_en = %d\n", __func__, pstate_en); + + dml_print("DML_DLG: %s: dppclk_freq_in_mhz = %3.2f\n", __func__, dppclk_freq_in_mhz); + dml_print("DML_DLG: %s: dispclk_freq_in_mhz = %3.2f\n", __func__, dispclk_freq_in_mhz); + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: interlaced = %d\n", __func__, interlaced); + ASSERT(ref_freq_to_pix_freq < 4.0); + + disp_dlg_regs->ref_freq_to_pix_freq = + (unsigned int) (ref_freq_to_pix_freq * dml_pow(2, 19)); + disp_dlg_regs->refcyc_per_htotal = (unsigned int) (ref_freq_to_pix_freq * (double) htotal + * dml_pow(2, 8)); + disp_dlg_regs->dlg_vblank_end = interlaced ? (vblank_end / 2) : vblank_end; // 15 bits + disp_dlg_regs->refcyc_h_blank_end = (unsigned int) ((double) hblank_end + * (double) ref_freq_to_pix_freq); + ASSERT(disp_dlg_regs->refcyc_h_blank_end < (unsigned int) dml_pow(2, 13)); + + min_dcfclk_mhz = dlg_sys_param.deepsleep_dcfclk_mhz; + t_calc_us = get_tcalc(mode_lib, e2e_pipe_param, num_pipes); + min_ttu_vblank = get_min_ttu_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + min_dst_y_ttu_vblank = min_ttu_vblank * pclk_freq_in_mhz / (double) htotal; + dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start; + + disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start + + min_dst_y_ttu_vblank) * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int) dml_pow(2, 18)); + + dml_print("DML_DLG: %s: min_dcfclk_mhz = %3.2f\n", + __func__, + min_dcfclk_mhz); + dml_print("DML_DLG: %s: min_ttu_vblank = %3.2f\n", + __func__, + min_ttu_vblank); + dml_print("DML_DLG: %s: min_dst_y_ttu_vblank = %3.2f\n", + __func__, + min_dst_y_ttu_vblank); + dml_print("DML_DLG: %s: t_calc_us = %3.2f\n", + __func__, + t_calc_us); + dml_print("DML_DLG: %s: disp_dlg_regs->min_dst_y_next_start = 0x%0x\n", + __func__, + disp_dlg_regs->min_dst_y_next_start); + dml_print("DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", + __func__, + ref_freq_to_pix_freq); + + // ------------------------- + // Section 1.15.2.2: Prefetch, Active and TTU + // ------------------------- + // Prefetch Calc + // Source +// dcc_en = src.dcc; + dual_plane = is_dual_plane((enum source_format_class)(src->source_format)); + mode_422 = 0; // FIXME + access_dir = (src->source_scan == dm_vert); // vp access direction: horizontal or vertical accessed +// bytes_per_element_l = get_bytes_per_element(source_format_class(src.source_format), 0); +// bytes_per_element_c = get_bytes_per_element(source_format_class(src.source_format), 1); + vp_height_l = src->viewport_height; + vp_width_l = src->viewport_width; + vp_height_c = src->viewport_height_c; + vp_width_c = src->viewport_width_c; + + // Scaling + htaps_l = taps->htaps; + htaps_c = taps->htaps_c; + hratio_l = scl->hscl_ratio; + hratio_c = scl->hscl_ratio_c; + vratio_l = scl->vscl_ratio; + vratio_c = scl->vscl_ratio_c; + scl_enable = scl->scl_enable; + + line_time_in_us = (htotal / pclk_freq_in_mhz); +// vinit_l = scl.vinit; +// vinit_c = scl.vinit_c; +// vinit_bot_l = scl.vinit_bot; +// vinit_bot_c = scl.vinit_bot_c; + +// unsigned int swath_height_l = rq_dlg_param.rq_l.swath_height; + swath_width_ub_l = rq_dlg_param.rq_l.swath_width_ub; +// unsigned int dpte_bytes_per_row_ub_l = rq_dlg_param.rq_l.dpte_bytes_per_row_ub; + dpte_groups_per_row_ub_l = rq_dlg_param.rq_l.dpte_groups_per_row_ub; +// unsigned int meta_pte_bytes_per_frame_ub_l = rq_dlg_param.rq_l.meta_pte_bytes_per_frame_ub; +// unsigned int meta_bytes_per_row_ub_l = rq_dlg_param.rq_l.meta_bytes_per_row_ub; + +// unsigned int swath_height_c = rq_dlg_param.rq_c.swath_height; + swath_width_ub_c = rq_dlg_param.rq_c.swath_width_ub; + // dpte_bytes_per_row_ub_c = rq_dlg_param.rq_c.dpte_bytes_per_row_ub; + dpte_groups_per_row_ub_c = rq_dlg_param.rq_c.dpte_groups_per_row_ub; + + meta_chunks_per_row_ub_l = rq_dlg_param.rq_l.meta_chunks_per_row_ub; + meta_chunks_per_row_ub_c = rq_dlg_param.rq_c.meta_chunks_per_row_ub; + vupdate_offset = dst->vupdate_offset; + vupdate_width = dst->vupdate_width; + vready_offset = dst->vready_offset; + + dppclk_delay_subtotal = mode_lib->ip.dppclk_delay_subtotal; + dispclk_delay_subtotal = mode_lib->ip.dispclk_delay_subtotal; + + if (scl_enable) + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl; + else + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl_lb_only; + + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_cnvc_formatter + + src->num_cursors * mode_lib->ip.dppclk_delay_cnvc_cursor; + + if (dout->dsc_enable) { + double dsc_delay = get_dsc_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dispclk_delay_subtotal += dsc_delay; + } + + pixel_rate_delay_subtotal = dppclk_delay_subtotal * pclk_freq_in_mhz / dppclk_freq_in_mhz + + dispclk_delay_subtotal * pclk_freq_in_mhz / dispclk_freq_in_mhz; + + vstartup_start = dst->vstartup_start; + if (interlaced) { + if (vstartup_start / 2.0 + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end / 2.0) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } else { + if (vstartup_start + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } + + // TODO: Where is this coming from? + if (interlaced) + vstartup_start = vstartup_start / 2; + + // TODO: What if this min_vblank doesn't match the value in the dml_config_settings.cpp? + if (vstartup_start >= min_vblank) { + dml_print("WARNING: DML_DLG: %s: vblank_start=%d vblank_end=%d\n", + __func__, + vblank_start, + vblank_end); + dml_print("WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + min_vblank = vstartup_start + 1; + dml_print("WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + } + + dst_x_after_scaler = get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_after_scaler = get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: htotal = %d\n", __func__, htotal); + dml_print("DML_DLG: %s: pixel_rate_delay_subtotal = %d\n", + __func__, + pixel_rate_delay_subtotal); + dml_print("DML_DLG: %s: dst_x_after_scaler = %d\n", + __func__, + dst_x_after_scaler); + dml_print("DML_DLG: %s: dst_y_after_scaler = %d\n", + __func__, + dst_y_after_scaler); + + // Lwait + line_wait = mode_lib->soc.urgent_latency_us; + if (cstate_en) + line_wait = dml_max(mode_lib->soc.sr_enter_plus_exit_time_us, line_wait); + if (pstate_en) + line_wait = dml_max(mode_lib->soc.dram_clock_change_latency_us + + mode_lib->soc.urgent_latency_us, + line_wait); + line_wait = line_wait / line_time_in_us; + + dst_y_prefetch = get_dst_y_prefetch(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dml_print("DML_DLG: %s: dst_y_prefetch (after rnd) = %3.2f\n", __func__, dst_y_prefetch); + + dst_y_per_vm_vblank = get_dst_y_per_vm_vblank(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_row_vblank = get_dst_y_per_row_vblank(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_vm_flip = get_dst_y_per_vm_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_per_row_flip = get_dst_y_per_row_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + min_dst_y_per_vm_vblank = 8.0; + min_dst_y_per_row_vblank = 16.0; + + // magic! + if (htotal <= 75) { + min_vblank = 300; + min_dst_y_per_vm_vblank = 100.0; + min_dst_y_per_row_vblank = 100.0; + } + + dml_print("DML_DLG: %s: dst_y_per_vm_vblank = %3.2f\n", __func__, dst_y_per_vm_vblank); + dml_print("DML_DLG: %s: dst_y_per_row_vblank = %3.2f\n", __func__, dst_y_per_row_vblank); + + ASSERT(dst_y_per_vm_vblank < min_dst_y_per_vm_vblank); + ASSERT(dst_y_per_row_vblank < min_dst_y_per_row_vblank); + + ASSERT(dst_y_prefetch > (dst_y_per_vm_vblank + dst_y_per_row_vblank)); + lsw = dst_y_prefetch - (dst_y_per_vm_vblank + dst_y_per_row_vblank); + + dml_print("DML_DLG: %s: lsw = %3.2f\n", __func__, lsw); + + vratio_pre_l = get_vratio_prefetch_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + vratio_pre_c = get_vratio_prefetch_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: vratio_pre_l=%3.2f\n", __func__, vratio_pre_l); + dml_print("DML_DLG: %s: vratio_pre_c=%3.2f\n", __func__, vratio_pre_c); + + // Active + req_per_swath_ub_l = rq_dlg_param.rq_l.req_per_swath_ub; + req_per_swath_ub_c = rq_dlg_param.rq_c.req_per_swath_ub; + meta_row_height_l = rq_dlg_param.rq_l.meta_row_height; + meta_row_height_c = rq_dlg_param.rq_c.meta_row_height; + swath_width_pixels_ub_l = 0; + swath_width_pixels_ub_c = 0; + scaler_rec_in_width_l = 0; + scaler_rec_in_width_c = 0; + dpte_row_height_l = rq_dlg_param.rq_l.dpte_row_height; + dpte_row_height_c = rq_dlg_param.rq_c.dpte_row_height; + + if (mode_422) { + swath_width_pixels_ub_l = swath_width_ub_l * 2; // *2 for 2 pixel per element + swath_width_pixels_ub_c = swath_width_ub_c * 2; + } else { + swath_width_pixels_ub_l = swath_width_ub_l * 1; + swath_width_pixels_ub_c = swath_width_ub_c * 1; + } + + hscale_pixel_rate_l = 0.; + hscale_pixel_rate_c = 0.; + min_hratio_fact_l = 1.0; + min_hratio_fact_c = 1.0; + + if (htaps_l <= 1) + min_hratio_fact_l = 2.0; + else if (htaps_l <= 6) { + if ((hratio_l * 2.0) > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l * 2.0; + } else { + if (hratio_l > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l; + } + + hscale_pixel_rate_l = min_hratio_fact_l * dppclk_freq_in_mhz; + + if (htaps_c <= 1) + min_hratio_fact_c = 2.0; + else if (htaps_c <= 6) { + if ((hratio_c * 2.0) > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c * 2.0; + } else { + if (hratio_c > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c; + } + + hscale_pixel_rate_c = min_hratio_fact_c * dppclk_freq_in_mhz; + + refcyc_per_line_delivery_pre_l = 0.; + refcyc_per_line_delivery_pre_c = 0.; + refcyc_per_line_delivery_l = 0.; + refcyc_per_line_delivery_c = 0.; + + refcyc_per_req_delivery_pre_l = 0.; + refcyc_per_req_delivery_pre_c = 0.; + refcyc_per_req_delivery_l = 0.; + refcyc_per_req_delivery_c = 0.; + + full_recout_width = 0; + // In ODM + if (src->is_hsplit) { + // This "hack" is only allowed (and valid) for MPC combine. In ODM + // combine, you MUST specify the full_recout_width...according to Oswin + if (dst->full_recout_width == 0 && !dst->odm_combine) { + dml_print("DML_DLG: %s: Warning: full_recout_width not set in hsplit mode\n", + __func__); + full_recout_width = dst->recout_width * 2; // assume half split for dcn1 + } else + full_recout_width = dst->full_recout_width; + } else + full_recout_width = dst->recout_width; + + // As of DCN2, mpc_combine and odm_combine are mutually exclusive + refcyc_per_line_delivery_pre_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + refcyc_per_line_delivery_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + dml_print("DML_DLG: %s: full_recout_width = %d\n", + __func__, + full_recout_width); + dml_print("DML_DLG: %s: hscale_pixel_rate_l = %3.2f\n", + __func__, + hscale_pixel_rate_l); + dml_print("DML_DLG: %s: refcyc_per_line_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_l); + dml_print("DML_DLG: %s: refcyc_per_line_delivery_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_l); + + if (dual_plane) { + refcyc_per_line_delivery_pre_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + refcyc_per_line_delivery_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + dml_print("DML_DLG: %s: refcyc_per_line_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_c); + dml_print("DML_DLG: %s: refcyc_per_line_delivery_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_c); + } + + // TTU - Luma / Chroma + if (access_dir) { // vertical access + scaler_rec_in_width_l = vp_height_l; + scaler_rec_in_width_c = vp_height_c; + } else { + scaler_rec_in_width_l = vp_width_l; + scaler_rec_in_width_c = vp_width_c; + } + + refcyc_per_req_delivery_pre_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + refcyc_per_req_delivery_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + + dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_l); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_l); + + ASSERT(refcyc_per_req_delivery_pre_l < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_l < dml_pow(2, 13)); + + if (dual_plane) { + refcyc_per_req_delivery_pre_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + refcyc_per_req_delivery_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + + dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_c); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_c); + + ASSERT(refcyc_per_req_delivery_pre_c < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_c < dml_pow(2, 13)); + } + + // XFC + xfc_transfer_delay = get_xfc_transfer_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + xfc_precharge_delay = get_xfc_precharge_delay(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_remote_surface_flip_latency = get_xfc_remote_surface_flip_latency(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_dst_y_delta_drq_limit = xfc_remote_surface_flip_latency; + xfc_prefetch_margin = get_xfc_prefetch_margin(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + + // TTU - Cursor + refcyc_per_req_delivery_pre_cur0 = 0.0; + refcyc_per_req_delivery_cur0 = 0.0; + if (src->num_cursors > 0) { + calculate_ttu_cursor(mode_lib, + &refcyc_per_req_delivery_pre_cur0, + &refcyc_per_req_delivery_cur0, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur0_src_width, + (enum cursor_bpp)(src->cur0_bpp)); + } + + refcyc_per_req_delivery_pre_cur1 = 0.0; + refcyc_per_req_delivery_cur1 = 0.0; + if (src->num_cursors > 1) { + calculate_ttu_cursor(mode_lib, + &refcyc_per_req_delivery_pre_cur1, + &refcyc_per_req_delivery_cur1, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur1_src_width, + (enum cursor_bpp)(src->cur1_bpp)); + } + + // TTU - Misc + // all hard-coded + + // Assignment to register structures + disp_dlg_regs->dst_y_after_scaler = dst_y_after_scaler; // in terms of line + disp_dlg_regs->refcyc_x_after_scaler = dst_x_after_scaler * ref_freq_to_pix_freq; // in terms of refclk + ASSERT(disp_dlg_regs->refcyc_x_after_scaler < (unsigned int) dml_pow(2, 13)); + disp_dlg_regs->dst_y_prefetch = (unsigned int) (dst_y_prefetch * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_vblank = (unsigned int) (dst_y_per_vm_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_vblank = (unsigned int) (dst_y_per_row_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_flip = (unsigned int) (dst_y_per_vm_flip * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_flip = (unsigned int) (dst_y_per_row_flip * dml_pow(2, 2)); + + disp_dlg_regs->vratio_prefetch = (unsigned int) (vratio_pre_l * dml_pow(2, 19)); + disp_dlg_regs->vratio_prefetch_c = (unsigned int) (vratio_pre_c * dml_pow(2, 19)); + + disp_dlg_regs->refcyc_per_pte_group_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int) dml_pow(2, 13)); + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_vblank_c = (unsigned int) (dst_y_per_row_vblank + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c + < (unsigned int) dml_pow(2, 13)); + } + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) meta_chunks_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_meta_chunk_vblank_l < (unsigned int) dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_c = + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l; // dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + disp_dlg_regs->refcyc_per_pte_group_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_l; + disp_dlg_regs->refcyc_per_meta_chunk_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_l; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_c; + disp_dlg_regs->refcyc_per_meta_chunk_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_c; + } + + disp_dlg_regs->dst_y_per_pte_row_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_pte_row_nom_l < (unsigned int) dml_pow(2, 17)); + + if (dual_plane) { + disp_dlg_regs->dst_y_per_pte_row_nom_c = (unsigned int) ((double) dpte_row_height_c + / (double) vratio_c * dml_pow(2, 2)); + if (disp_dlg_regs->dst_y_per_pte_row_nom_c >= (unsigned int) dml_pow(2, 17)) { + dml_print("DML_DLG: %s: Warning dst_y_per_pte_row_nom_c %u larger than supported by register format U15.2 %u\n", + __func__, + disp_dlg_regs->dst_y_per_pte_row_nom_c, + (unsigned int) dml_pow(2, 17) - 1); + } + } + + disp_dlg_regs->dst_y_per_meta_row_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_meta_row_nom_l < (unsigned int) dml_pow(2, 17)); + + disp_dlg_regs->dst_y_per_meta_row_nom_c = disp_dlg_regs->dst_y_per_meta_row_nom_l; // TODO: dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + disp_dlg_regs->refcyc_per_pte_group_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_pte_group_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_l = dml_pow(2, 23) - 1; + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = dml_pow(2, 23) - 1; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_nom_c = + (unsigned int) ((double) dpte_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_pte_group_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_c = dml_pow(2, 23) - 1; + + // TODO: Is this the right calculation? Does htotal need to be halved? + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = + (unsigned int) ((double) meta_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = dml_pow(2, 23) - 1; + } + + disp_dlg_regs->refcyc_per_line_delivery_pre_l = (unsigned int) dml_floor(refcyc_per_line_delivery_pre_l, + 1); + disp_dlg_regs->refcyc_per_line_delivery_l = (unsigned int) dml_floor(refcyc_per_line_delivery_l, + 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_l < (unsigned int) dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_l < (unsigned int) dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_line_delivery_pre_c = (unsigned int) dml_floor(refcyc_per_line_delivery_pre_c, + 1); + disp_dlg_regs->refcyc_per_line_delivery_c = (unsigned int) dml_floor(refcyc_per_line_delivery_c, + 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_c < (unsigned int) dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_c < (unsigned int) dml_pow(2, 13)); + + disp_dlg_regs->chunk_hdl_adjust_cur0 = 3; + disp_dlg_regs->dst_y_offset_cur0 = 0; + disp_dlg_regs->chunk_hdl_adjust_cur1 = 3; + disp_dlg_regs->dst_y_offset_cur1 = 0; + + disp_dlg_regs->xfc_reg_transfer_delay = xfc_transfer_delay; + disp_dlg_regs->xfc_reg_precharge_delay = xfc_precharge_delay; + disp_dlg_regs->xfc_reg_remote_surface_flip_latency = xfc_remote_surface_flip_latency; + disp_dlg_regs->xfc_reg_prefetch_margin = dml_ceil(xfc_prefetch_margin * refclk_freq_in_mhz, + 1); + + // slave has to have this value also set to off + if (src->xfc_enable && !src->xfc_slave) + disp_dlg_regs->dst_y_delta_drq_limit = dml_ceil(xfc_dst_y_delta_drq_limit, 1); + else + disp_dlg_regs->dst_y_delta_drq_limit = 0x7fff; // off + + disp_ttu_regs->refcyc_per_req_delivery_pre_l = (unsigned int) (refcyc_per_req_delivery_pre_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_l = (unsigned int) (refcyc_per_req_delivery_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_c = (unsigned int) (refcyc_per_req_delivery_pre_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_c = (unsigned int) (refcyc_per_req_delivery_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur0 = + (unsigned int) (refcyc_per_req_delivery_pre_cur0 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur0 = (unsigned int) (refcyc_per_req_delivery_cur0 + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur1 = + (unsigned int) (refcyc_per_req_delivery_pre_cur1 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur1 = (unsigned int) (refcyc_per_req_delivery_cur1 + * dml_pow(2, 10)); + disp_ttu_regs->qos_level_low_wm = 0; + ASSERT(disp_ttu_regs->qos_level_low_wm < dml_pow(2, 14)); + disp_ttu_regs->qos_level_high_wm = (unsigned int) (4.0 * (double) htotal + * ref_freq_to_pix_freq); + /*ASSERT(disp_ttu_regs->qos_level_high_wm < dml_pow(2, 14));*/ + + disp_ttu_regs->qos_level_flip = 14; + disp_ttu_regs->qos_level_fixed_l = 8; + disp_ttu_regs->qos_level_fixed_c = 8; + disp_ttu_regs->qos_level_fixed_cur0 = 8; + disp_ttu_regs->qos_ramp_disable_l = 0; + disp_ttu_regs->qos_ramp_disable_c = 0; + disp_ttu_regs->qos_ramp_disable_cur0 = 0; + + disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz; + ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24)); + + print__ttu_regs_st(mode_lib, *disp_ttu_regs); + print__dlg_regs_st(mode_lib, *disp_dlg_regs); +} + +void dml20v2_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support) +{ + display_rq_params_st rq_param = {0}; + display_dlg_sys_params_st dlg_sys_param = {0}; + + // Get watermark and Tex. + dlg_sys_param.t_urg_wm_us = get_wm_urgent(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.deepsleep_dcfclk_mhz = get_clk_dcf_deepsleep(mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_extra_us = get_urgent_extra_latency(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.mem_trip_us = get_wm_memory_trip(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_mclk_wm_us = get_wm_dram_clock_change(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_sr_wm_us = get_wm_stutter_enter_exit(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.total_flip_bw = get_total_immediate_flip_bw(mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.total_flip_bytes = get_total_immediate_flip_bytes(mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency + / dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated + + print__dlg_sys_params_st(mode_lib, dlg_sys_param); + + // system parameter calculation done + + dml_print("DML_DLG: Calculation for pipe[%d] start\n\n", pipe_idx); + dml20v2_rq_dlg_get_rq_params(mode_lib, &rq_param, e2e_pipe_param[pipe_idx].pipe.src); + dml20v2_rq_dlg_get_dlg_params(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx, + dlg_regs, + ttu_regs, + rq_param.dlg, + dlg_sys_param, + cstate_en, + pstate_en); + dml_print("DML_DLG: Calculation for pipe[%d] end\n", pipe_idx); +} + +static void calculate_ttu_cursor(struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp) +{ + unsigned int cur_src_width = cur_width; + unsigned int cur_req_size = 0; + unsigned int cur_req_width = 0; + double cur_width_ub = 0.0; + double cur_req_per_width = 0.0; + double hactive_cur = 0.0; + + ASSERT(cur_src_width <= 256); + + *refcyc_per_req_delivery_pre_cur = 0.0; + *refcyc_per_req_delivery_cur = 0.0; + if (cur_src_width > 0) { + unsigned int cur_bit_per_pixel = 0; + + if (cur_bpp == dm_cur_2bit) { + cur_req_size = 64; // byte + cur_bit_per_pixel = 2; + } else { // 32bit + cur_bit_per_pixel = 32; + if (cur_src_width >= 1 && cur_src_width <= 16) + cur_req_size = 64; + else if (cur_src_width >= 17 && cur_src_width <= 31) + cur_req_size = 128; + else + cur_req_size = 256; + } + + cur_req_width = (double) cur_req_size / ((double) cur_bit_per_pixel / 8.0); + cur_width_ub = dml_ceil((double) cur_src_width / (double) cur_req_width, 1) + * (double) cur_req_width; + cur_req_per_width = cur_width_ub / (double) cur_req_width; + hactive_cur = (double) cur_src_width / hscl_ratio; // FIXME: oswin to think about what to do for cursor + + if (vratio_pre_l <= 1.0) { + *refcyc_per_req_delivery_pre_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_pre_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + ASSERT(*refcyc_per_req_delivery_pre_cur < dml_pow(2, 13)); + + if (vratio_l <= 1.0) { + *refcyc_per_req_delivery_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + dml_print("DML_DLG: %s: cur_req_width = %d\n", + __func__, + cur_req_width); + dml_print("DML_DLG: %s: cur_width_ub = %3.2f\n", + __func__, + cur_width_ub); + dml_print("DML_DLG: %s: cur_req_per_width = %3.2f\n", + __func__, + cur_req_per_width); + dml_print("DML_DLG: %s: hactive_cur = %3.2f\n", + __func__, + hactive_cur); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_pre_cur); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_cur); + + ASSERT(*refcyc_per_req_delivery_cur < dml_pow(2, 13)); + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h new file mode 100644 index 000000000000..0378406bf7e7 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h @@ -0,0 +1,74 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DML20V2_DISPLAY_RQ_DLG_CALC_H__ +#define __DML20V2_DISPLAY_RQ_DLG_CALC_H__ + +#include "../dml_common_defs.h" +#include "../display_rq_dlg_helpers.h" + +struct display_mode_lib; + + +// Function: dml_rq_dlg_get_rq_reg +// Main entry point for test to get the register values out of this DML class. +// This function calls <get_rq_param> and <extract_rq_regs> fucntions to calculate +// and then populate the rq_regs struct +// Input: +// pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.) +// Output: +// rq_regs - struct that holds all the RQ registers field value. +// See also: <display_rq_regs_st> +void dml20v2_rq_dlg_get_rq_reg( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param); + + +// Function: dml_rq_dlg_get_dlg_reg +// Calculate and return DLG and TTU register struct given the system setting +// Output: +// dlg_regs - output DLG register struct +// ttu_regs - output DLG TTU register struct +// Input: +// e2e_pipe_param - "compacted" array of e2e pipe param struct +// num_pipes - num of active "pipe" or "route" +// pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg +// cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered. +// Added for legacy or unrealistic timing tests. +void dml20v2_rq_dlg_get_dlg_reg( + struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c new file mode 100644 index 000000000000..456cd0e3289c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c @@ -0,0 +1,6123 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifdef CONFIG_DRM_AMD_DC_DCN2_0 + +#include "../display_mode_lib.h" +#include "../dml_inline_defs.h" +#include "../display_mode_vba.h" +#include "display_mode_vba_21.h" + + +/* + * NOTE: + * This file is gcc-parsable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ + +typedef unsigned int uint; + +typedef struct { + double DPPCLK; + double DISPCLK; + double PixelClock; + double DCFCLKDeepSleep; + unsigned int DPPPerPlane; + bool ScalerEnabled; + enum scan_direction_class SourceScan; + unsigned int BlockWidth256BytesY; + unsigned int BlockHeight256BytesY; + unsigned int BlockWidth256BytesC; + unsigned int BlockHeight256BytesC; + unsigned int InterlaceEnable; + unsigned int NumberOfCursors; + unsigned int VBlank; + unsigned int HTotal; +} Pipe; + +typedef struct { + bool Enable; + unsigned int MaxPageTableLevels; + unsigned int CachedPageTableLevels; +} HostVM; + +#define BPP_INVALID 0 +#define BPP_BLENDED_PIPE 0xffffffff + +static void DisplayPipeConfiguration(struct display_mode_lib *mode_lib); +static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib); +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat); +static unsigned int dscComputeDelay(enum output_format_class pixelFormat); +// Super monster function with some 45 argument +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + Pipe *myPipe, + unsigned int DSCDelay, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int MaxVStartup, + unsigned int GPUVMPageTableLevels, + bool GPUVMEnable, + HostVM *myHostVM, + bool DynamicMetadataEnable, + int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatency, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBWLuma, + double *RequiredPrefetchPixDataBWChroma, + unsigned int *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + double *Tno_bw, + double *prefetch_vmrow_bw, + unsigned int *swath_width_luma_ub, + unsigned int *swath_width_chroma_ub, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix); +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed); +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed); +static double CalculateDCCConfiguration( + bool DCCEnabled, + bool DCCProgrammingAssumesScanDirectionUnknown, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + double DETBufferSize, + unsigned int RequestHeight256Byte, + unsigned int SwathHeight, + enum dm_swizzle_mode TilingFormat, + unsigned int BytePerPixel, + enum scan_direction_class ScanOrientation, + unsigned int *MaxUncompressedBlock, + unsigned int *MaxCompressedBlock, + unsigned int *Independent64ByteBlock); +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath); +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidthY, + bool GPUVMEnable, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequests, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_width_ub, + unsigned int *dpte_row_height, + unsigned int *MetaRequestWidth, + unsigned int *MetaRequestHeight, + unsigned int *meta_row_width, + unsigned int *meta_row_height, + unsigned int *vm_group_bytes, + long *dpte_group_bytes, + unsigned int *PixelPTEReqWidth, + unsigned int *PixelPTEReqHeight, + unsigned int *PTERequestSize, + unsigned int *DPDE0BytesFrame, + unsigned int *MetaPTEBytesFrame); + +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatency, + double SREnterPlusExitTime); +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk); +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw); +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + double UrgentExtraLatency, + double UrgentLatency, + unsigned int GPUVMMaxPageTableLevels, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + bool GPUVMEnable, + double PDEAndMetaPTEBytesPerFrame, + double MetaRowBytes, + double DPTEBytesPerRow, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + double LineTime, + double VRatio, + double Tno_bw, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + unsigned int dpte_row_height_chroma, + unsigned int meta_row_height_chroma, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe); +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth); +static void CalculateWatermarksAndDRAMSpeedChangeSupport( + struct display_mode_lib *mode_lib, + unsigned int PrefetchMode, + unsigned int NumberOfActivePlanes, + unsigned int MaxLineBufferLines, + unsigned int LineBufferSize, + unsigned int DPPOutputBufferPixels, + double DETBufferSizeInKByte, + unsigned int WritebackInterfaceLumaBufferSize, + unsigned int WritebackInterfaceChromaBufferSize, + double DCFCLK, + double UrgentOutOfOrderReturn, + double ReturnBW, + bool GPUVMEnable, + long dpte_group_bytes[], + unsigned int MetaChunkSize, + double UrgentLatency, + double ExtraLatency, + double WritebackLatency, + double WritebackChunkSize, + double SOCCLK, + double DRAMClockChangeLatency, + double SRExitTime, + double SREnterPlusExitTime, + double DCFCLKDeepSleep, + int DPPPerPlane[], + bool DCCEnable[], + double DPPCLK[], + unsigned int SwathWidthSingleDPPY[], + unsigned int SwathHeightY[], + double ReadBandwidthPlaneLuma[], + unsigned int SwathHeightC[], + double ReadBandwidthPlaneChroma[], + unsigned int LBBitPerPixel[], + unsigned int SwathWidthY[], + double HRatio[], + unsigned int vtaps[], + unsigned int VTAPsChroma[], + double VRatio[], + unsigned int HTotal[], + double PixelClock[], + unsigned int BlendingAndTiming[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + bool WritebackEnable[], + enum source_format_class WritebackPixelFormat[], + double WritebackDestinationWidth[], + double WritebackDestinationHeight[], + double WritebackSourceHeight[], + enum clock_change_support *DRAMClockChangeSupport, + double *UrgentWatermark, + double *WritebackUrgentWatermark, + double *DRAMClockChangeWatermark, + double *WritebackDRAMClockChangeWatermark, + double *StutterExitWatermark, + double *StutterEnterPlusExitWatermark, + double *MinActiveDRAMClockChangeLatencySupported); +static void CalculateDCFCLKDeepSleep( + struct display_mode_lib *mode_lib, + unsigned int NumberOfActivePlanes, + double BytePerPixelDETY[], + double BytePerPixelDETC[], + double VRatio[], + unsigned int SwathWidthY[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double *DCFCLKDeepSleep); +static void CalculateDETBufferSize( + double DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double *DETBufferSizeY, + double *DETBufferSizeC); +static void CalculateUrgentBurstFactor( + unsigned int DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + unsigned int SwathWidthY, + double LineTime, + double UrgentLatency, + double CursorBufferSize, + unsigned int CursorWidth, + unsigned int CursorBPP, + double VRatio, + double VRatioPreY, + double VRatioPreC, + double BytePerPixelInDETY, + double BytePerPixelInDETC, + double *UrgentBurstFactorCursor, + double *UrgentBurstFactorCursorPre, + double *UrgentBurstFactorLuma, + double *UrgentBurstFactorLumaPre, + double *UrgentBurstFactorChroma, + double *UrgentBurstFactorChromaPre, + unsigned int *NotEnoughUrgentLatencyHiding, + unsigned int *NotEnoughUrgentLatencyHidingPre); + +static void CalculatePixelDeliveryTimes( + unsigned int NumberOfActivePlanes, + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + unsigned int swath_width_luma_ub[], + unsigned int swath_width_chroma_ub[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int BlockWidth256BytesY[], + unsigned int BlockHeight256BytesY[], + unsigned int BlockWidth256BytesC[], + unsigned int BlockHeight256BytesC[], + double DisplayPipeLineDeliveryTimeLuma[], + double DisplayPipeLineDeliveryTimeChroma[], + double DisplayPipeLineDeliveryTimeLumaPrefetch[], + double DisplayPipeLineDeliveryTimeChromaPrefetch[], + double DisplayPipeRequestDeliveryTimeLuma[], + double DisplayPipeRequestDeliveryTimeChroma[], + double DisplayPipeRequestDeliveryTimeLumaPrefetch[], + double DisplayPipeRequestDeliveryTimeChromaPrefetch[]); + +static void CalculateMetaAndPTETimes( + unsigned int NumberOfActivePlanes, + bool GPUVMEnable, + unsigned int MetaChunkSize, + unsigned int MinMetaChunkSizeBytes, + unsigned int GPUVMMaxPageTableLevels, + unsigned int HTotal[], + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + double DestinationLinesToRequestRowInVBlank[], + double DestinationLinesToRequestRowInImmediateFlip[], + double DestinationLinesToRequestVMInVBlank[], + double DestinationLinesToRequestVMInImmediateFlip[], + bool DCCEnable[], + double PixelClock[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int dpte_row_height[], + unsigned int dpte_row_height_chroma[], + unsigned int meta_row_width[], + unsigned int meta_row_height[], + unsigned int meta_req_width[], + unsigned int meta_req_height[], + long dpte_group_bytes[], + unsigned int PTERequestSizeY[], + unsigned int PTERequestSizeC[], + unsigned int PixelPTEReqWidthY[], + unsigned int PixelPTEReqHeightY[], + unsigned int PixelPTEReqWidthC[], + unsigned int PixelPTEReqHeightC[], + unsigned int dpte_row_width_luma_ub[], + unsigned int dpte_row_width_chroma_ub[], + unsigned int vm_group_bytes[], + unsigned int dpde0_bytes_per_frame_ub_l[], + unsigned int dpde0_bytes_per_frame_ub_c[], + unsigned int meta_pte_bytes_per_frame_ub_l[], + unsigned int meta_pte_bytes_per_frame_ub_c[], + double DST_Y_PER_PTE_ROW_NOM_L[], + double DST_Y_PER_PTE_ROW_NOM_C[], + double DST_Y_PER_META_ROW_NOM_L[], + double TimePerMetaChunkNominal[], + double TimePerMetaChunkVBlank[], + double TimePerMetaChunkFlip[], + double time_per_pte_group_nom_luma[], + double time_per_pte_group_vblank_luma[], + double time_per_pte_group_flip_luma[], + double time_per_pte_group_nom_chroma[], + double time_per_pte_group_vblank_chroma[], + double time_per_pte_group_flip_chroma[], + double TimePerVMGroupVBlank[], + double TimePerVMGroupFlip[], + double TimePerVMRequestVBlank[], + double TimePerVMRequestFlip[]); + +static double CalculateExtraLatency( + double UrgentRoundTripAndOutOfOrderLatency, + int TotalNumberOfActiveDPP, + int PixelChunkSizeInKByte, + int TotalNumberOfDCCActiveDPP, + int MetaChunkSize, + double ReturnBW, + bool GPUVMEnable, + bool HostVMEnable, + int NumberOfActivePlanes, + int NumberOfDPP[], + long dpte_group_bytes[], + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + int HostVMMaxPageTableLevels, + int HostVMCachedPageTableLevels); + +void dml21_recalculate(struct display_mode_lib *mode_lib) +{ + ModeSupportAndSystemConfiguration(mode_lib); + PixelClockAdjustmentForProgressiveToInterlaceUnit(mode_lib); + DisplayPipeConfiguration(mode_lib); + DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation(mode_lib); +} + +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat) +{ + // valid bpc = source bits per component in the set of {8, 10, 12} + // valid bpp = increments of 1/16 of a bit + // min = 6/7/8 in N420/N422/444, respectively + // max = such that compression is 1:1 + //valid sliceWidth = number of pixels per slice line, must be less than or equal to 5184/numSlices (or 4096/numSlices in 420 mode) + //valid numSlices = number of slices in the horiziontal direction per DSC engine in the set of {1, 2, 3, 4} + //valid pixelFormat = pixel/color format in the set of {:N444_RGB, :S422, :N422, :N420} + + // fixed value + unsigned int rcModelSize = 8192; + + // N422/N420 operate at 2 pixels per clock + unsigned int pixelsPerClock, lstall, D, initalXmitDelay, w, S, ix, wx, p, l0, a, ax, l, + Delay, pixels; + + if (pixelFormat == dm_n422 || pixelFormat == dm_420) + pixelsPerClock = 2; + // #all other modes operate at 1 pixel per clock + else + pixelsPerClock = 1; + + //initial transmit delay as per PPS + initalXmitDelay = dml_round(rcModelSize / 2.0 / bpp / pixelsPerClock); + + //compute ssm delay + if (bpc == 8) + D = 81; + else if (bpc == 10) + D = 89; + else + D = 113; + + //divide by pixel per cycle to compute slice width as seen by DSC + w = sliceWidth / pixelsPerClock; + + //422 mode has an additional cycle of delay + if (pixelFormat == dm_s422) + S = 1; + else + S = 0; + + //main calculation for the dscce + ix = initalXmitDelay + 45; + wx = (w + 2) / 3; + p = 3 * wx - w; + l0 = ix / w; + a = ix + p * l0; + ax = (a + 2) / 3 + D + 6 + 1; + l = (ax + wx - 1) / wx; + if ((ix % w) == 0 && p != 0) + lstall = 1; + else + lstall = 0; + Delay = l * wx * (numSlices - 1) + ax + S + lstall + 22; + + //dsc processes 3 pixel containers per cycle and a container can contain 1 or 2 pixels + pixels = Delay * 3 * pixelsPerClock; + return pixels; +} + +static unsigned int dscComputeDelay(enum output_format_class pixelFormat) +{ + unsigned int Delay = 0; + + if (pixelFormat == dm_420) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc gets pixels every other cycle + Delay = Delay + 2; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc gets pixels every other cycle + Delay = Delay + 13; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc gets pixels every other cycle + Delay = Delay + 3; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else if (pixelFormat == dm_n422) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 1; + // dscc - input deserializer + Delay = Delay + 5; + // dscc - input cdc fifo + Delay = Delay + 25; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 10; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc - output serializer + Delay = Delay + 1; + // dscc - cdc uncertainty + Delay = Delay + 2; + // sft + Delay = Delay + 1; + } + + return Delay; +} + +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + Pipe *myPipe, + unsigned int DSCDelay, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int MaxVStartup, + unsigned int GPUVMPageTableLevels, + bool GPUVMEnable, + HostVM *myHostVM, + bool DynamicMetadataEnable, + int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatency, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBWLuma, + double *RequiredPrefetchPixDataBWChroma, + unsigned int *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + double *Tno_bw, + double *prefetch_vmrow_bw, + unsigned int *swath_width_luma_ub, + unsigned int *swath_width_chroma_ub, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix) +{ + bool MyError = false; + unsigned int DPPCycles, DISPCLKCycles; + double DSTTotalPixelsAfterScaler, TotalRepeaterDelayTime; + double Tdm, LineTime, Tsetup; + double dst_y_prefetch_equ; + double Tsw_oto; + double prefetch_bw_oto; + double Tvm_oto; + double Tr0_oto; + double Tvm_oto_lines; + double Tr0_oto_lines; + double Tsw_oto_lines; + double dst_y_prefetch_oto; + double TimeForFetchingMetaPTE = 0; + double TimeForFetchingRowInVBlank = 0; + double LinesToRequestPrefetchPixelData = 0; + double HostVMInefficiencyFactor; + unsigned int HostVMDynamicLevels; + + if (GPUVMEnable == true && myHostVM->Enable == true) { + HostVMInefficiencyFactor = + PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData + / PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly; + HostVMDynamicLevels = myHostVM->MaxPageTableLevels + - myHostVM->CachedPageTableLevels; + } else { + HostVMInefficiencyFactor = 1; + HostVMDynamicLevels = 0; + } + + if (myPipe->ScalerEnabled) + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCL; + else + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCLLBOnly; + + DPPCycles = DPPCycles + DPPCLKDelayCNVCFormater + myPipe->NumberOfCursors * DPPCLKDelayCNVCCursor; + + DISPCLKCycles = DISPCLKDelaySubtotal; + + if (myPipe->DPPCLK == 0.0 || myPipe->DISPCLK == 0.0) + return true; + + *DSTXAfterScaler = DPPCycles * myPipe->PixelClock / myPipe->DPPCLK + + DISPCLKCycles * myPipe->PixelClock / myPipe->DISPCLK + DSCDelay; + + if (myPipe->DPPPerPlane > 1) + *DSTXAfterScaler = *DSTXAfterScaler + ScalerRecoutWidth; + + if (OutputFormat == dm_420 || (myPipe->InterlaceEnable && ProgressiveToInterlaceUnitInOPP)) + *DSTYAfterScaler = 1; + else + *DSTYAfterScaler = 0; + + DSTTotalPixelsAfterScaler = ((double) (*DSTYAfterScaler * myPipe->HTotal)) + *DSTXAfterScaler; + *DSTYAfterScaler = dml_floor(DSTTotalPixelsAfterScaler / myPipe->HTotal, 1); + *DSTXAfterScaler = DSTTotalPixelsAfterScaler - ((double) (*DSTYAfterScaler * myPipe->HTotal)); + + *VUpdateOffsetPix = dml_ceil(myPipe->HTotal / 4.0, 1); + TotalRepeaterDelayTime = MaxInterDCNTileRepeaters * (2.0 / myPipe->DPPCLK + 3.0 / myPipe->DISPCLK); + *VUpdateWidthPix = (14.0 / myPipe->DCFCLKDeepSleep + 12.0 / myPipe->DPPCLK + TotalRepeaterDelayTime) + * myPipe->PixelClock; + + *VReadyOffsetPix = dml_max( + 150.0 / myPipe->DPPCLK, + TotalRepeaterDelayTime + 20.0 / myPipe->DCFCLKDeepSleep + 10.0 / myPipe->DPPCLK) + * myPipe->PixelClock; + + Tsetup = (double) (*VUpdateOffsetPix + *VUpdateWidthPix + *VReadyOffsetPix) / myPipe->PixelClock; + + LineTime = (double) myPipe->HTotal / myPipe->PixelClock; + + if (DynamicMetadataEnable) { + double Tdmbf, Tdmec, Tdmsks; + + Tdm = dml_max(0.0, UrgentExtraLatency - TCalc); + Tdmbf = DynamicMetadataTransmittedBytes / 4.0 / myPipe->DISPCLK; + Tdmec = LineTime; + if (DynamicMetadataLinesBeforeActiveRequired == -1) + Tdmsks = myPipe->VBlank * LineTime / 2.0; + else + Tdmsks = DynamicMetadataLinesBeforeActiveRequired * LineTime; + if (myPipe->InterlaceEnable && !ProgressiveToInterlaceUnitInOPP) + Tdmsks = Tdmsks / 2; + if (VStartup * LineTime + < Tsetup + TWait + UrgentExtraLatency + Tdmbf + Tdmec + Tdmsks) { + MyError = true; + *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata = (Tsetup + TWait + + UrgentExtraLatency + Tdmbf + Tdmec + Tdmsks) / LineTime; + } else + *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata = 0.0; + } else + Tdm = 0; + + if (GPUVMEnable) { + if (GPUVMPageTableLevels >= 3) + *Tno_bw = UrgentExtraLatency + UrgentLatency * ((GPUVMPageTableLevels - 2) * (myHostVM->MaxPageTableLevels + 1) - 1); + else + *Tno_bw = 0; + } else if (!DCCEnable) + *Tno_bw = LineTime; + else + *Tno_bw = LineTime / 4; + + dst_y_prefetch_equ = VStartup - dml_max(TCalc + TWait, XFCRemoteSurfaceFlipDelay) / LineTime + - (Tsetup + Tdm) / LineTime + - (*DSTYAfterScaler + *DSTXAfterScaler / myPipe->HTotal); + + Tsw_oto = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime; + + if (myPipe->SourceScan == dm_horz) { + *swath_width_luma_ub = dml_ceil(SwathWidthY - 1, myPipe->BlockWidth256BytesY) + myPipe->BlockWidth256BytesY; + *swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockWidth256BytesC) + myPipe->BlockWidth256BytesC; + } else { + *swath_width_luma_ub = dml_ceil(SwathWidthY - 1, myPipe->BlockHeight256BytesY) + myPipe->BlockHeight256BytesY; + *swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockHeight256BytesC) + myPipe->BlockHeight256BytesC; + } + + prefetch_bw_oto = (PrefetchSourceLinesY * *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + PrefetchSourceLinesC * *swath_width_chroma_ub * dml_ceil(BytePerPixelDETC, 2)) / Tsw_oto; + + + if (GPUVMEnable == true) { + Tvm_oto = dml_max(*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / prefetch_bw_oto, + dml_max(UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1), + LineTime / 4.0)); + } else + Tvm_oto = LineTime / 4.0; + + if ((GPUVMEnable == true || DCCEnable == true)) { + Tr0_oto = dml_max( + (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / prefetch_bw_oto, + dml_max(UrgentLatency * (HostVMDynamicLevels + 1), dml_max(LineTime - Tvm_oto, LineTime / 4))); + } else + Tr0_oto = (LineTime - Tvm_oto) / 2.0; + + Tvm_oto_lines = dml_ceil(4 * Tvm_oto / LineTime, 1) / 4.0; + Tr0_oto_lines = dml_ceil(4 * Tr0_oto / LineTime, 1) / 4.0; + Tsw_oto_lines = dml_ceil(4 * Tsw_oto / LineTime, 1) / 4.0; + dst_y_prefetch_oto = Tvm_oto_lines + 2 * Tr0_oto_lines + Tsw_oto_lines + 0.75; + + dst_y_prefetch_equ = dml_floor(4.0 * (dst_y_prefetch_equ + 0.125), 1) / 4.0; + + if (dst_y_prefetch_oto < dst_y_prefetch_equ) + *DestinationLinesForPrefetch = dst_y_prefetch_oto; + else + *DestinationLinesForPrefetch = dst_y_prefetch_equ; + + dml_print("DML: VStartup: %d\n", VStartup); + dml_print("DML: TCalc: %f\n", TCalc); + dml_print("DML: TWait: %f\n", TWait); + dml_print("DML: XFCRemoteSurfaceFlipDelay: %f\n", XFCRemoteSurfaceFlipDelay); + dml_print("DML: LineTime: %f\n", LineTime); + dml_print("DML: Tsetup: %f\n", Tsetup); + dml_print("DML: Tdm: %f\n", Tdm); + dml_print("DML: DSTYAfterScaler: %f\n", *DSTYAfterScaler); + dml_print("DML: DSTXAfterScaler: %f\n", *DSTXAfterScaler); + dml_print("DML: HTotal: %d\n", myPipe->HTotal); + + *PrefetchBandwidth = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBWLuma = 0; + if (*DestinationLinesForPrefetch > 1) { + double PrefetchBandwidth1 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + 2 * MetaRowByte + + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor + + PrefetchSourceLinesY * *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * *swath_width_chroma_ub * dml_ceil(BytePerPixelDETC, 2)) + / (*DestinationLinesForPrefetch * LineTime - *Tno_bw); + + double PrefetchBandwidth2 = (PDEAndMetaPTEBytesFrame * + HostVMInefficiencyFactor + PrefetchSourceLinesY * + *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * *swath_width_chroma_ub * + dml_ceil(BytePerPixelDETC, 2)) / + (*DestinationLinesForPrefetch * LineTime - *Tno_bw - 2 * + UrgentLatency * (1 + HostVMDynamicLevels)); + + double PrefetchBandwidth3 = (2 * MetaRowByte + 2 * PixelPTEBytesPerRow + * HostVMInefficiencyFactor + PrefetchSourceLinesY * + *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * *swath_width_chroma_ub * + dml_ceil(BytePerPixelDETC, 2)) / + (*DestinationLinesForPrefetch * LineTime - + UrgentExtraLatency - UrgentLatency * (GPUVMPageTableLevels + * (HostVMDynamicLevels + 1) - 1)); + + double PrefetchBandwidth4 = (PrefetchSourceLinesY * *swath_width_luma_ub * + dml_ceil(BytePerPixelDETY, 1) + PrefetchSourceLinesC * + *swath_width_chroma_ub * dml_ceil(BytePerPixelDETC, 2)) / + (*DestinationLinesForPrefetch * LineTime - + UrgentExtraLatency - UrgentLatency * (GPUVMPageTableLevels + * (HostVMDynamicLevels + 1) - 1) - 2 * UrgentLatency * + (1 + HostVMDynamicLevels)); + + if (VStartup == MaxVStartup && (PrefetchBandwidth1 > 4 * prefetch_bw_oto) && (*DestinationLinesForPrefetch - dml_ceil(Tsw_oto_lines, 1) / 4.0 - 0.75) * LineTime - *Tno_bw > 0) { + PrefetchBandwidth1 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + 2 * MetaRowByte + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor) / ((*DestinationLinesForPrefetch - dml_ceil(Tsw_oto_lines, 1) / 4.0 - 0.75) * LineTime - *Tno_bw); + } + if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth1 >= UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1) && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / PrefetchBandwidth1 >= UrgentLatency * (1 + HostVMDynamicLevels)) { + *PrefetchBandwidth = PrefetchBandwidth1; + } else if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth2 >= UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1) && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / PrefetchBandwidth2 < UrgentLatency * (1 + HostVMDynamicLevels)) { + *PrefetchBandwidth = PrefetchBandwidth2; + } else if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth3 < UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1) && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / PrefetchBandwidth3 >= UrgentLatency * (1 + HostVMDynamicLevels)) { + *PrefetchBandwidth = PrefetchBandwidth3; + } else { + *PrefetchBandwidth = PrefetchBandwidth4; + } + + if (GPUVMEnable) { + TimeForFetchingMetaPTE = dml_max(*Tno_bw + (double) PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / *PrefetchBandwidth, + dml_max(UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1), LineTime / 4)); + } else { +// 5/30/2018 - This was an optimization requested from Sy but now NumberOfCursors is no longer a factor +// so if this needs to be reinstated, then it should be officially done in the VBA code as well. +// if (mode_lib->NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingMetaPTE = LineTime / 4; +// else +// TimeForFetchingMetaPTE = 0.0; + } + + if ((GPUVMEnable == true || DCCEnable == true)) { + TimeForFetchingRowInVBlank = + dml_max( + (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) + / *PrefetchBandwidth, + dml_max( + UrgentLatency * (1 + HostVMDynamicLevels), + dml_max( + (LineTime + - TimeForFetchingMetaPTE) / 2.0, + LineTime + / 4.0))); + } else { +// See note above dated 5/30/2018 +// if (NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingRowInVBlank = (LineTime - TimeForFetchingMetaPTE) / 2.0; +// else // TODO: Did someone else add this?? +// TimeForFetchingRowInVBlank = 0.0; + } + + *DestinationLinesToRequestVMInVBlank = dml_ceil(4.0 * TimeForFetchingMetaPTE / LineTime, 1.0) / 4.0; + + *DestinationLinesToRequestRowInVBlank = dml_ceil(4.0 * TimeForFetchingRowInVBlank / LineTime, 1.0) / 4.0; + + LinesToRequestPrefetchPixelData = *DestinationLinesForPrefetch +// See note above dated 5/30/2018 +// - ((NumberOfCursors > 0 || GPUVMEnable || DCCEnable) ? + - ((GPUVMEnable || DCCEnable) ? + (*DestinationLinesToRequestVMInVBlank + 2 * *DestinationLinesToRequestRowInVBlank) : + 0.0); // TODO: Did someone else add this?? + + if (LinesToRequestPrefetchPixelData > 0) { + + *VRatioPrefetchY = (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + if ((SwathHeightY > 4) && (VInitPreFillY > 3)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillY - 3.0) / 2.0) { + *VRatioPrefetchY = + dml_max( + (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData, + (double) MaxNumSwathY + * SwathHeightY + / (LinesToRequestPrefetchPixelData + - (VInitPreFillY + - 3.0) + / 2.0)); + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + } else { + MyError = true; + *VRatioPrefetchY = 0; + } + } + + *VRatioPrefetchC = (double) PrefetchSourceLinesC + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + + if ((SwathHeightC > 4)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillC - 3.0) / 2.0) { + *VRatioPrefetchC = + dml_max( + *VRatioPrefetchC, + (double) MaxNumSwathC + * SwathHeightC + / (LinesToRequestPrefetchPixelData + - (VInitPreFillC + - 3.0) + / 2.0)); + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + } else { + MyError = true; + *VRatioPrefetchC = 0; + } + } + + *RequiredPrefetchPixDataBWLuma = myPipe->DPPPerPlane + * (double) PrefetchSourceLinesY / LinesToRequestPrefetchPixelData + * dml_ceil(BytePerPixelDETY, 1) + * *swath_width_luma_ub / LineTime; + *RequiredPrefetchPixDataBWChroma = myPipe->DPPPerPlane + * (double) PrefetchSourceLinesC / LinesToRequestPrefetchPixelData + * dml_ceil(BytePerPixelDETC, 2) + * *swath_width_chroma_ub / LineTime; + } else { + MyError = true; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBWLuma = 0; + *RequiredPrefetchPixDataBWChroma = 0; + } + + dml_print("DML: Tvm: %fus\n", TimeForFetchingMetaPTE); + dml_print("DML: Tr0: %fus\n", TimeForFetchingRowInVBlank); + dml_print("DML: Tsw: %fus\n", (double)(*DestinationLinesForPrefetch) * LineTime - TimeForFetchingMetaPTE - TimeForFetchingRowInVBlank); + dml_print("DML: Tpre: %fus\n", (double)(*DestinationLinesForPrefetch) * LineTime); + dml_print("DML: row_bytes = dpte_row_bytes (per_pipe) = PixelPTEBytesPerRow = : %d\n", PixelPTEBytesPerRow); + + } else { + MyError = true; + } + + { + double prefetch_vm_bw; + double prefetch_row_bw; + + if (PDEAndMetaPTEBytesFrame == 0) { + prefetch_vm_bw = 0; + } else if (*DestinationLinesToRequestVMInVBlank > 0) { + prefetch_vm_bw = PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / (*DestinationLinesToRequestVMInVBlank * LineTime); + } else { + prefetch_vm_bw = 0; + MyError = true; + } + if (MetaRowByte + PixelPTEBytesPerRow == 0) { + prefetch_row_bw = 0; + } else if (*DestinationLinesToRequestRowInVBlank > 0) { + prefetch_row_bw = (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / (*DestinationLinesToRequestRowInVBlank * LineTime); + } else { + prefetch_row_bw = 0; + MyError = true; + } + + *prefetch_vmrow_bw = dml_max(prefetch_vm_bw, prefetch_row_bw); + } + + if (MyError) { + *PrefetchBandwidth = 0; + TimeForFetchingMetaPTE = 0; + TimeForFetchingRowInVBlank = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *DestinationLinesForPrefetch = 0; + LinesToRequestPrefetchPixelData = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBWLuma = 0; + *RequiredPrefetchPixDataBWChroma = 0; + } + + return MyError; +} + +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_floor(VCOSpeed * 4 / Clock, 1); +} + +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_ceil(VCOSpeed * 4 / Clock, 1); +} + +static double CalculateDCCConfiguration( + bool DCCEnabled, + bool DCCProgrammingAssumesScanDirectionUnknown, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + double DETBufferSize, + unsigned int RequestHeight256Byte, + unsigned int SwathHeight, + enum dm_swizzle_mode TilingFormat, + unsigned int BytePerPixel, + enum scan_direction_class ScanOrientation, + unsigned int *MaxUncompressedBlock, + unsigned int *MaxCompressedBlock, + unsigned int *Independent64ByteBlock) +{ + double MaximumDCCCompressionSurface = 0.0; + enum { + REQ_256Bytes, + REQ_128BytesNonContiguous, + REQ_128BytesContiguous, + REQ_NA + } Request = REQ_NA; + + if (DCCEnabled == true) { + if (DCCProgrammingAssumesScanDirectionUnknown == true) { + if (DETBufferSize >= RequestHeight256Byte * ViewportWidth * BytePerPixel + && DETBufferSize + >= 256 / RequestHeight256Byte + * ViewportHeight) { + Request = REQ_256Bytes; + } else if ((DETBufferSize + < RequestHeight256Byte * ViewportWidth * BytePerPixel + && (BytePerPixel == 2 || BytePerPixel == 4)) + || (DETBufferSize + < 256 / RequestHeight256Byte + * ViewportHeight + && BytePerPixel == 8 + && (TilingFormat == dm_sw_4kb_d + || TilingFormat + == dm_sw_4kb_d_x + || TilingFormat + == dm_sw_var_d + || TilingFormat + == dm_sw_var_d_x + || TilingFormat + == dm_sw_64kb_d + || TilingFormat + == dm_sw_64kb_d_x + || TilingFormat + == dm_sw_64kb_d_t + || TilingFormat + == dm_sw_64kb_r_x))) { + Request = REQ_128BytesNonContiguous; + } else { + Request = REQ_128BytesContiguous; + } + } else { + if (BytePerPixel == 1) { + if (ScanOrientation == dm_vert || SwathHeight == 16) { + Request = REQ_256Bytes; + } else { + Request = REQ_128BytesContiguous; + } + } else if (BytePerPixel == 2) { + if ((ScanOrientation == dm_vert && SwathHeight == 16) || (ScanOrientation != dm_vert && SwathHeight == 8)) { + Request = REQ_256Bytes; + } else if (ScanOrientation == dm_vert) { + Request = REQ_128BytesContiguous; + } else { + Request = REQ_128BytesNonContiguous; + } + } else if (BytePerPixel == 4) { + if (SwathHeight == 8) { + Request = REQ_256Bytes; + } else if (ScanOrientation == dm_vert) { + Request = REQ_128BytesContiguous; + } else { + Request = REQ_128BytesNonContiguous; + } + } else if (BytePerPixel == 8) { + if (TilingFormat == dm_sw_4kb_d || TilingFormat == dm_sw_4kb_d_x + || TilingFormat == dm_sw_var_d + || TilingFormat == dm_sw_var_d_x + || TilingFormat == dm_sw_64kb_d + || TilingFormat == dm_sw_64kb_d_x + || TilingFormat == dm_sw_64kb_d_t + || TilingFormat == dm_sw_64kb_r_x) { + if ((ScanOrientation == dm_vert && SwathHeight == 8) + || (ScanOrientation != dm_vert + && SwathHeight == 4)) { + Request = REQ_256Bytes; + } else if (ScanOrientation != dm_vert) { + Request = REQ_128BytesContiguous; + } else { + Request = REQ_128BytesNonContiguous; + } + } else { + if (ScanOrientation != dm_vert || SwathHeight == 8) { + Request = REQ_256Bytes; + } else { + Request = REQ_128BytesContiguous; + } + } + } + } + } else { + Request = REQ_NA; + } + + if (Request == REQ_256Bytes) { + *MaxUncompressedBlock = 256; + *MaxCompressedBlock = 256; + *Independent64ByteBlock = false; + MaximumDCCCompressionSurface = 4.0; + } else if (Request == REQ_128BytesContiguous) { + *MaxUncompressedBlock = 128; + *MaxCompressedBlock = 128; + *Independent64ByteBlock = false; + MaximumDCCCompressionSurface = 2.0; + } else if (Request == REQ_128BytesNonContiguous) { + *MaxUncompressedBlock = 256; + *MaxCompressedBlock = 64; + *Independent64ByteBlock = true; + MaximumDCCCompressionSurface = 4.0; + } else { + *MaxUncompressedBlock = 0; + *MaxCompressedBlock = 0; + *Independent64ByteBlock = 0; + MaximumDCCCompressionSurface = 0.0; + } + + return MaximumDCCCompressionSurface; +} + +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath) +{ + unsigned int MaxPartialSwath; + + if (ProgressiveToInterlaceUnitInOPP) + *VInitPreFill = dml_floor((VRatio + vtaps + 1) / 2.0, 1); + else + *VInitPreFill = dml_floor((VRatio + vtaps + 1 + Interlace * 0.5 * VRatio) / 2.0, 1); + + if (!mode_lib->vba.IgnoreViewportPositioning) { + + *MaxNumSwath = dml_ceil((*VInitPreFill - 1.0) / SwathHeight, 1) + 1.0; + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 2) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 2) + % SwathHeight; + MaxPartialSwath = dml_max(1U, MaxPartialSwath); + + } else { + + if (ViewportYStart != 0) + dml_print( + "WARNING DML: using viewport y position of 0 even though actual viewport y position is non-zero in prefetch source lines calculation\n"); + + *MaxNumSwath = dml_ceil(*VInitPreFill / SwathHeight, 1); + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 1) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 1) + % SwathHeight; + } + + return *MaxNumSwath * SwathHeight + MaxPartialSwath; +} + +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidth, + bool GPUVMEnable, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequests, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_width_ub, + unsigned int *dpte_row_height, + unsigned int *MetaRequestWidth, + unsigned int *MetaRequestHeight, + unsigned int *meta_row_width, + unsigned int *meta_row_height, + unsigned int *vm_group_bytes, + long *dpte_group_bytes, + unsigned int *PixelPTEReqWidth, + unsigned int *PixelPTEReqHeight, + unsigned int *PTERequestSize, + unsigned int *DPDE0BytesFrame, + unsigned int *MetaPTEBytesFrame) +{ + unsigned int MPDEBytesFrame; + unsigned int DCCMetaSurfaceBytes; + unsigned int MacroTileSizeBytes; + unsigned int MacroTileHeight; + unsigned int ExtraDPDEBytesFrame; + unsigned int PDEAndMetaPTEBytesFrame; + unsigned int PixelPTEReqHeightPTEs; + + if (DCCEnable == true) { + *MetaRequestHeight = 8 * BlockHeight256Bytes; + *MetaRequestWidth = 8 * BlockWidth256Bytes; + if (ScanDirection == dm_horz) { + *meta_row_height = *MetaRequestHeight; + *meta_row_width = dml_ceil((double) SwathWidth - 1, *MetaRequestWidth) + + *MetaRequestWidth; + *MetaRowByte = *meta_row_width * *MetaRequestHeight * BytePerPixel / 256.0; + } else { + *meta_row_height = *MetaRequestWidth; + *meta_row_width = dml_ceil((double) SwathWidth - 1, *MetaRequestHeight) + + *MetaRequestHeight; + *MetaRowByte = *meta_row_width * *MetaRequestWidth * BytePerPixel / 256.0; + } + if (ScanDirection == dm_horz) { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil(ViewportHeight - 1, 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } else { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil( + (double) ViewportHeight - 1, + 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } + if (GPUVMEnable == true) { + *MetaPTEBytesFrame = (dml_ceil( + (double) (DCCMetaSurfaceBytes - VMMPageSize) + / (8 * VMMPageSize), + 1) + 1) * 64; + MPDEBytesFrame = 128 * ((mode_lib->vba.GPUVMMaxPageTableLevels + 1) * (mode_lib->vba.HostVMMaxPageTableLevels + 1) - 2); + } else { + *MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + } + } else { + *MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + *MetaRowByte = 0; + } + + if (SurfaceTiling == dm_sw_linear || SurfaceTiling == dm_sw_gfx7_2d_thin_gl || SurfaceTiling == dm_sw_gfx7_2d_thin_lvp) { + MacroTileSizeBytes = 256; + MacroTileHeight = BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_4kb_s || SurfaceTiling == dm_sw_4kb_s_x + || SurfaceTiling == dm_sw_4kb_d || SurfaceTiling == dm_sw_4kb_d_x) { + MacroTileSizeBytes = 4096; + MacroTileHeight = 4 * BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_64kb_s || SurfaceTiling == dm_sw_64kb_s_t + || SurfaceTiling == dm_sw_64kb_s_x || SurfaceTiling == dm_sw_64kb_d + || SurfaceTiling == dm_sw_64kb_d_t || SurfaceTiling == dm_sw_64kb_d_x + || SurfaceTiling == dm_sw_64kb_r_x) { + MacroTileSizeBytes = 65536; + MacroTileHeight = 16 * BlockHeight256Bytes; + } else { + MacroTileSizeBytes = 262144; + MacroTileHeight = 32 * BlockHeight256Bytes; + } + *MacroTileWidth = MacroTileSizeBytes / BytePerPixel / MacroTileHeight; + + if (GPUVMEnable == true && (mode_lib->vba.GPUVMMaxPageTableLevels + 1) * (mode_lib->vba.HostVMMaxPageTableLevels + 1) > 2) { + if (ScanDirection == dm_horz) { + *DPDE0BytesFrame = 64 * (dml_ceil(((Pitch * (dml_ceil(ViewportHeight - 1, MacroTileHeight) + MacroTileHeight) * BytePerPixel) - MacroTileSizeBytes) / (8 * 2097152), 1) + 1); + } else { + *DPDE0BytesFrame = 64 * (dml_ceil(((Pitch * (dml_ceil((double) SwathWidth - 1, MacroTileHeight) + MacroTileHeight) * BytePerPixel) - MacroTileSizeBytes) / (8 * 2097152), 1) + 1); + } + ExtraDPDEBytesFrame = 128 * ((mode_lib->vba.GPUVMMaxPageTableLevels + 1) * (mode_lib->vba.HostVMMaxPageTableLevels + 1) - 3); + } else { + *DPDE0BytesFrame = 0; + ExtraDPDEBytesFrame = 0; + } + + PDEAndMetaPTEBytesFrame = *MetaPTEBytesFrame + MPDEBytesFrame + *DPDE0BytesFrame + + ExtraDPDEBytesFrame; + + if (HostVMEnable == true) { + PDEAndMetaPTEBytesFrame = PDEAndMetaPTEBytesFrame * (1 + 8 * (HostVMMaxPageTableLevels - HostVMCachedPageTableLevels)); + } + + if (GPUVMEnable == true) { + double FractionOfPTEReturnDrop; + + if (SurfaceTiling == dm_sw_linear) { + PixelPTEReqHeightPTEs = 1; + *PixelPTEReqHeight = 1; + *PixelPTEReqWidth = 8.0 * VMMPageSize / BytePerPixel; + *PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } else if (MacroTileSizeBytes == 4096) { + PixelPTEReqHeightPTEs = 1; + *PixelPTEReqHeight = MacroTileHeight; + *PixelPTEReqWidth = 8 * *MacroTileWidth; + *PTERequestSize = 64; + if (ScanDirection == dm_horz) + FractionOfPTEReturnDrop = 0; + else + FractionOfPTEReturnDrop = 7 / 8; + } else if (VMMPageSize == 4096 && MacroTileSizeBytes > 4096) { + PixelPTEReqHeightPTEs = 16; + *PixelPTEReqHeight = 16 * BlockHeight256Bytes; + *PixelPTEReqWidth = 16 * BlockWidth256Bytes; + *PTERequestSize = 128; + FractionOfPTEReturnDrop = 0; + } else { + PixelPTEReqHeightPTEs = 1; + *PixelPTEReqHeight = MacroTileHeight; + *PixelPTEReqWidth = 8 * *MacroTileWidth; + *PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } + + if (SurfaceTiling == dm_sw_linear) { + *dpte_row_height = dml_min(128, + 1 << (unsigned int) dml_floor( + dml_log2( + (double) PTEBufferSizeInRequests * *PixelPTEReqWidth / Pitch), + 1)); + *dpte_row_width_ub = (dml_ceil((double) (Pitch * *dpte_row_height - 1) / *PixelPTEReqWidth, 1) + 1) * *PixelPTEReqWidth; + *PixelPTEBytesPerRow = *dpte_row_width_ub / *PixelPTEReqWidth * *PTERequestSize; + } else if (ScanDirection == dm_horz) { + *dpte_row_height = *PixelPTEReqHeight; + *dpte_row_width_ub = (dml_ceil((double) (SwathWidth - 1) / *PixelPTEReqWidth, 1) + 1) * *PixelPTEReqWidth; + *PixelPTEBytesPerRow = *dpte_row_width_ub / *PixelPTEReqWidth * *PTERequestSize; + } else { + *dpte_row_height = dml_min(*PixelPTEReqWidth, *MacroTileWidth); + *dpte_row_width_ub = (dml_ceil((double) (SwathWidth - 1) / *PixelPTEReqHeight, 1) + 1) * *PixelPTEReqHeight; + *PixelPTEBytesPerRow = *dpte_row_width_ub / *PixelPTEReqHeight * *PTERequestSize; + } + if (*PixelPTEBytesPerRow * (1 - FractionOfPTEReturnDrop) + <= 64 * PTEBufferSizeInRequests) { + *PTEBufferSizeNotExceeded = true; + } else { + *PTEBufferSizeNotExceeded = false; + } + } else { + *PixelPTEBytesPerRow = 0; + *PTEBufferSizeNotExceeded = true; + } + dml_print("DML: vm_bytes = meta_pte_bytes_per_frame (per_pipe) = MetaPTEBytesFrame = : %d\n", *MetaPTEBytesFrame); + + if (HostVMEnable == true) { + *PixelPTEBytesPerRow = *PixelPTEBytesPerRow * (1 + 8 * (HostVMMaxPageTableLevels - HostVMCachedPageTableLevels)); + } + + if (HostVMEnable == true) { + *vm_group_bytes = 512; + *dpte_group_bytes = 512; + } else if (GPUVMEnable == true) { + *vm_group_bytes = 2048; + if (SurfaceTiling != dm_sw_linear && PixelPTEReqHeightPTEs == 1 && ScanDirection != dm_horz) { + *dpte_group_bytes = 512; + } else { + *dpte_group_bytes = 2048; + } + } else { + *vm_group_bytes = 0; + *dpte_group_bytes = 0; + } + + return PDEAndMetaPTEBytesFrame; +} + +static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib) +{ + struct vba_vars_st *locals = &mode_lib->vba; + unsigned int j, k; + + mode_lib->vba.WritebackDISPCLK = 0.0; + mode_lib->vba.DISPCLKWithRamping = 0; + mode_lib->vba.DISPCLKWithoutRamping = 0; + mode_lib->vba.GlobalDPPCLK = 0.0; + + // DISPCLK and DPPCLK Calculation + // + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k]) { + mode_lib->vba.WritebackDISPCLK = + dml_max( + mode_lib->vba.WritebackDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.HRatio[k] > 1) { + locals->PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1)); + } else { + locals->PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + + mode_lib->vba.DPPCLKUsingSingleDPPLuma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_THROUGHPUT_LUMA[k], + 1.0)); + + if ((mode_lib->vba.htaps[k] > 6 || mode_lib->vba.vtaps[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPLuma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPLuma = 2 * mode_lib->vba.PixelClock[k]; + } + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + locals->PSCL_THROUGHPUT_CHROMA[k] = 0.0; + locals->DPPCLKUsingSingleDPP[k] = + mode_lib->vba.DPPCLKUsingSingleDPPLuma; + } else { + if (mode_lib->vba.HRatio[k] > 1) { + locals->PSCL_THROUGHPUT_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_THROUGHPUT_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + mode_lib->vba.DPPCLKUsingSingleDPPChroma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4 + / locals->PSCL_THROUGHPUT_CHROMA[k], + 1.0)); + + if ((mode_lib->vba.HTAPsChroma[k] > 6 || mode_lib->vba.VTAPsChroma[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPChroma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPChroma = 2 + * mode_lib->vba.PixelClock[k]; + } + + locals->DPPCLKUsingSingleDPP[k] = dml_max( + mode_lib->vba.DPPCLKUsingSingleDPPLuma, + mode_lib->vba.DPPCLKUsingSingleDPPChroma); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] != k) + continue; + if (mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } else if (!mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } + } + + mode_lib->vba.DISPCLKWithRamping = dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.WritebackDISPCLK); + mode_lib->vba.DISPCLKWithoutRamping = dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.WritebackDISPCLK); + + ASSERT(mode_lib->vba.DISPCLKDPPCLKVCOSpeed != 0); + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDispclkRoundedToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.soc.clock_limits[mode_lib->vba.soc.num_states - 1].dispclk_mhz, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + if (mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity; + } else if (mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = mode_lib->vba.MaxDispclkRoundedToDFSGranularity; + } else { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity; + } + DTRACE(" dispclk_mhz (calculated) = %f", mode_lib->vba.DISPCLK_calculated); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.DPPCLK_calculated[k] = locals->DPPCLKUsingSingleDPP[k] + / mode_lib->vba.DPPPerPlane[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100); + mode_lib->vba.GlobalDPPCLK = dml_max( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DPPCLK_calculated[k]); + } + mode_lib->vba.GlobalDPPCLK = RoundToDFSGranularityUp( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.DPPCLK_calculated[k] = mode_lib->vba.GlobalDPPCLK / 255 + * dml_ceil( + mode_lib->vba.DPPCLK_calculated[k] * 255 + / mode_lib->vba.GlobalDPPCLK, + 1); + DTRACE(" dppclk_mhz[%i] (calculated) = %f", k, mode_lib->vba.DPPCLK_calculated[k]); + } + + // Urgent and B P-State/DRAM Clock Change Watermark + DTRACE(" dcfclk_mhz = %f", mode_lib->vba.DCFCLK); + DTRACE(" return_bw_to_dcn = %f", mode_lib->vba.ReturnBandwidthToDCN); + DTRACE(" return_bus_bw = %f", mode_lib->vba.ReturnBW); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourceScan[k] == dm_horz) + locals->SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportWidth[k]; + else + locals->SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportHeight[k]; + + if (mode_lib->vba.ODMCombineEnabled[k] == true) + MainPlaneDoesODMCombine = true; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[j] == true) + MainPlaneDoesODMCombine = true; + + if (MainPlaneDoesODMCombine == true) + locals->SwathWidthY[k] = dml_min( + (double) locals->SwathWidthSingleDPPY[k], + dml_round( + mode_lib->vba.HActive[k] / 2.0 + * mode_lib->vba.HRatio[k])); + else + locals->SwathWidthY[k] = locals->SwathWidthSingleDPPY[k] + / mode_lib->vba.DPPPerPlane[k]; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + locals->BytePerPixelDETY[k] = 8; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + locals->BytePerPixelDETY[k] = 4; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16 || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16) { + locals->BytePerPixelDETY[k] = 2; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8 || mode_lib->vba.SourcePixelFormat[k] == dm_mono_8) { + locals->BytePerPixelDETY[k] = 1; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + locals->BytePerPixelDETY[k] = 1; + locals->BytePerPixelDETC[k] = 2; + } else { // dm_420_10 + locals->BytePerPixelDETY[k] = 4.0 / 3.0; + locals->BytePerPixelDETC[k] = 8.0 / 3.0; + } + } + + mode_lib->vba.TotalDataReadBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + locals->ReadBandwidthPlaneLuma[k] = locals->SwathWidthSingleDPPY[k] + * dml_ceil(locals->BytePerPixelDETY[k], 1) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k]; + locals->ReadBandwidthPlaneChroma[k] = locals->SwathWidthSingleDPPY[k] + / 2 * dml_ceil(locals->BytePerPixelDETC[k], 2) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k] / 2; + DTRACE( + " read_bw[%i] = %fBps", + k, + locals->ReadBandwidthPlaneLuma[k] + + locals->ReadBandwidthPlaneChroma[k]); + mode_lib->vba.TotalDataReadBandwidth += locals->ReadBandwidthPlaneLuma[k] + + locals->ReadBandwidthPlaneChroma[k]; + } + + // DCFCLK Deep Sleep + CalculateDCFCLKDeepSleep( + mode_lib, + mode_lib->vba.NumberOfActivePlanes, + locals->BytePerPixelDETY, + locals->BytePerPixelDETC, + mode_lib->vba.VRatio, + locals->SwathWidthY, + mode_lib->vba.DPPPerPlane, + mode_lib->vba.HRatio, + mode_lib->vba.PixelClock, + locals->PSCL_THROUGHPUT_LUMA, + locals->PSCL_THROUGHPUT_CHROMA, + locals->DPPCLK, + &mode_lib->vba.DCFCLKDeepSleep); + + // DSCCLK + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if ((mode_lib->vba.BlendingAndTiming[k] != k) || !mode_lib->vba.DSCEnabled[k]) { + locals->DSCCLK_calculated[k] = 0.0; + } else { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] == dm_n422) + mode_lib->vba.DSCFormatFactor = 2; + else + mode_lib->vba.DSCFormatFactor = 1; + if (mode_lib->vba.ODMCombineEnabled[k]) + locals->DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 6 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + else + locals->DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 3 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + } + } + + // DSC Delay + // TODO + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + double bpp = mode_lib->vba.OutputBpp[k]; + unsigned int slices = mode_lib->vba.NumberOfDSCSlices[k]; + + if (mode_lib->vba.DSCEnabled[k] && bpp != 0) { + if (!mode_lib->vba.ODMCombineEnabled[k]) { + locals->DSCDelay[k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + locals->DSCDelay[k] = + 2 + * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices / 2.0, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k])); + } + locals->DSCDelay[k] = locals->DSCDelay[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.PixelClockBackEnd[k]; + } else { + locals->DSCDelay[k] = 0; + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) // NumberOfPlanes + if (j != k && mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.DSCEnabled[j]) + locals->DSCDelay[k] = locals->DSCDelay[j]; + + // Prefetch + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int PDEAndMetaPTEBytesFrameY; + unsigned int PixelPTEBytesPerRowY; + unsigned int MetaRowByteY; + unsigned int MetaRowByteC; + unsigned int PDEAndMetaPTEBytesFrameC; + unsigned int PixelPTEBytesPerRowC; + bool PTEBufferSizeNotExceededY; + bool PTEBufferSizeNotExceededC; + + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelDETY[k], 1), + dml_ceil(locals->BytePerPixelDETC[k], 2), + &locals->BlockHeight256BytesY[k], + &locals->BlockHeight256BytesC[k], + &locals->BlockWidth256BytesY[k], + &locals->BlockWidth256BytesC[k]); + + locals->PrefetchSourceLinesY[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.ViewportYStartY[k], + &locals->VInitPreFillY[k], + &locals->MaxNumSwathY[k]); + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_8)) { + PDEAndMetaPTEBytesFrameC = + CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->BlockHeight256BytesC[k], + locals->BlockWidth256BytesC[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil( + locals->BytePerPixelDETC[k], + 2), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2, + mode_lib->vba.ViewportHeight[k] / 2, + locals->SwathWidthY[k] / 2, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsChroma, + mode_lib->vba.PitchC[k], + mode_lib->vba.DCCMetaPitchC[k], + &locals->MacroTileWidthC[k], + &MetaRowByteC, + &PixelPTEBytesPerRowC, + &PTEBufferSizeNotExceededC, + &locals->dpte_row_width_chroma_ub[k], + &locals->dpte_row_height_chroma[k], + &locals->meta_req_width_chroma[k], + &locals->meta_req_height_chroma[k], + &locals->meta_row_width_chroma[k], + &locals->meta_row_height_chroma[k], + &locals->vm_group_bytes_chroma, + &locals->dpte_group_bytes_chroma, + &locals->PixelPTEReqWidthC[k], + &locals->PixelPTEReqHeightC[k], + &locals->PTERequestSizeC[k], + &locals->dpde0_bytes_per_frame_ub_c[k], + &locals->meta_pte_bytes_per_frame_ub_c[k]); + + locals->PrefetchSourceLinesC[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k] / 2, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightC[k], + mode_lib->vba.ViewportYStartC[k], + &locals->VInitPreFillC[k], + &locals->MaxNumSwathC[k]); + } else { + PixelPTEBytesPerRowC = 0; + PDEAndMetaPTEBytesFrameC = 0; + MetaRowByteC = 0; + locals->MaxNumSwathC[k] = 0; + locals->PrefetchSourceLinesC[k] = 0; + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma + mode_lib->vba.PTEBufferSizeInRequestsChroma; + } + + PDEAndMetaPTEBytesFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->BlockHeight256BytesY[k], + locals->BlockWidth256BytesY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelDETY[k], 1), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + locals->SwathWidthY[k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + locals->PTEBufferSizeInRequestsForLuma, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &locals->MacroTileWidthY[k], + &MetaRowByteY, + &PixelPTEBytesPerRowY, + &PTEBufferSizeNotExceededY, + &locals->dpte_row_width_luma_ub[k], + &locals->dpte_row_height[k], + &locals->meta_req_width[k], + &locals->meta_req_height[k], + &locals->meta_row_width[k], + &locals->meta_row_height[k], + &locals->vm_group_bytes[k], + &locals->dpte_group_bytes[k], + &locals->PixelPTEReqWidthY[k], + &locals->PixelPTEReqHeightY[k], + &locals->PTERequestSizeY[k], + &locals->dpde0_bytes_per_frame_ub_l[k], + &locals->meta_pte_bytes_per_frame_ub_l[k]); + + locals->PixelPTEBytesPerRow[k] = PixelPTEBytesPerRowY + PixelPTEBytesPerRowC; + locals->PDEAndMetaPTEBytesFrame[k] = PDEAndMetaPTEBytesFrameY + + PDEAndMetaPTEBytesFrameC; + locals->MetaRowByte[k] = MetaRowByteY + MetaRowByteC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + MetaRowByteY, + MetaRowByteC, + locals->meta_row_height[k], + locals->meta_row_height_chroma[k], + PixelPTEBytesPerRowY, + PixelPTEBytesPerRowC, + locals->dpte_row_height[k], + locals->dpte_row_height_chroma[k], + &locals->meta_row_bw[k], + &locals->dpte_row_bw[k]); + } + + mode_lib->vba.TotalDCCActiveDPP = 0; + mode_lib->vba.TotalActiveDPP = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalActiveDPP = mode_lib->vba.TotalActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + if (mode_lib->vba.DCCEnable[k]) + mode_lib->vba.TotalDCCActiveDPP = mode_lib->vba.TotalDCCActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + } + + mode_lib->vba.UrgentOutOfOrderReturnPerChannel = dml_max3( + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelMixedWithVMData, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelVMDataOnly); + + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency = + (mode_lib->vba.RoundTripPingLatencyCycles + 32) / mode_lib->vba.DCFCLK + + mode_lib->vba.UrgentOutOfOrderReturnPerChannel + * mode_lib->vba.NumberOfChannels + / mode_lib->vba.ReturnBW; + + mode_lib->vba.UrgentExtraLatency = CalculateExtraLatency( + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency, + mode_lib->vba.TotalActiveDPP, + mode_lib->vba.PixelChunkSizeInKByte, + mode_lib->vba.TotalDCCActiveDPP, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.ReturnBW, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.DPPPerPlane, + locals->dpte_group_bytes, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels); + + + mode_lib->vba.TCalc = 24.0 / mode_lib->vba.DCFCLKDeepSleep; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) + / mode_lib->vba.DISPCLK; + } else + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = 0; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[j] == k + && mode_lib->vba.WritebackEnable[j] == true) { + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = + dml_max( + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k], + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[j], + mode_lib->vba.WritebackHRatio[j], + mode_lib->vba.WritebackVRatio[j], + mode_lib->vba.WritebackLumaHTaps[j], + mode_lib->vba.WritebackLumaVTaps[j], + mode_lib->vba.WritebackChromaHTaps[j], + mode_lib->vba.WritebackChromaVTaps[j], + mode_lib->vba.WritebackDestinationWidth[j]) + / mode_lib->vba.DISPCLK); + } + } + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j) + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = + locals->WritebackDelay[mode_lib->vba.VoltageLevel][j]; + + mode_lib->vba.VStartupLines = 13; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + locals->MaxVStartupLines[k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] - dml_max(1.0, dml_ceil(locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]), 1)); + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + locals->MaximumMaxVStartupLines = dml_max(locals->MaximumMaxVStartupLines, locals->MaxVStartupLines[k]); + + // We don't really care to iterate between the various prefetch modes + //mode_lib->vba.PrefetchERROR = CalculateMinAndMaxPrefetchMode(mode_lib->vba.AllowDRAMSelfRefreshOrDRAMClockChangeInVblank, &mode_lib->vba.MinPrefetchMode, &mode_lib->vba.MaxPrefetchMode); + mode_lib->vba.UrgentLatency = dml_max3(mode_lib->vba.UrgentLatencyPixelDataOnly, mode_lib->vba.UrgentLatencyPixelMixedWithVMData, mode_lib->vba.UrgentLatencyVMDataOnly); + + do { + double MaxTotalRDBandwidth = 0; + double MaxTotalRDBandwidthNoUrgentBurst = 0; + bool DestinationLineTimesForPrefetchLessThan2 = false; + bool VRatioPrefetchMoreThan4 = false; + double TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + Pipe myPipe; + HostVM myHostVM; + + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthY[k], + dml_ceil( + locals->BytePerPixelDETY[k], + 1), + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0; + } + + myPipe.DPPCLK = locals->DPPCLK[k]; + myPipe.DISPCLK = mode_lib->vba.DISPCLK; + myPipe.PixelClock = mode_lib->vba.PixelClock[k]; + myPipe.DCFCLKDeepSleep = mode_lib->vba.DCFCLKDeepSleep; + myPipe.DPPPerPlane = mode_lib->vba.DPPPerPlane[k]; + myPipe.ScalerEnabled = mode_lib->vba.ScalerEnabled[k]; + myPipe.SourceScan = mode_lib->vba.SourceScan[k]; + myPipe.BlockWidth256BytesY = locals->BlockWidth256BytesY[k]; + myPipe.BlockHeight256BytesY = locals->BlockHeight256BytesY[k]; + myPipe.BlockWidth256BytesC = locals->BlockWidth256BytesC[k]; + myPipe.BlockHeight256BytesC = locals->BlockHeight256BytesC[k]; + myPipe.InterlaceEnable = mode_lib->vba.Interlace[k]; + myPipe.NumberOfCursors = mode_lib->vba.NumberOfCursors[k]; + myPipe.VBlank = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k]; + myPipe.HTotal = mode_lib->vba.HTotal[k]; + + + myHostVM.Enable = mode_lib->vba.HostVMEnable; + myHostVM.MaxPageTableLevels = mode_lib->vba.HostVMMaxPageTableLevels; + myHostVM.CachedPageTableLevels = mode_lib->vba.HostVMCachedPageTableLevels; + + mode_lib->vba.ErrorResult[k] = + CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + &myPipe, + locals->DSCDelay[k], + mode_lib->vba.DPPCLKDelaySubtotal, + mode_lib->vba.DPPCLKDelaySCL, + mode_lib->vba.DPPCLKDelaySCLLBOnly, + mode_lib->vba.DPPCLKDelayCNVCFormater, + mode_lib->vba.DPPCLKDelayCNVCCursor, + mode_lib->vba.DISPCLKDelaySubtotal, + (unsigned int) (locals->SwathWidthY[k] + / mode_lib->vba.HRatio[k]), + mode_lib->vba.OutputFormat[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + dml_min(mode_lib->vba.VStartupLines, locals->MaxVStartupLines[k]), + locals->MaxVStartupLines[k], + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + &myHostVM, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.TCalc, + locals->PDEAndMetaPTEBytesFrame[k], + locals->MetaRowByte[k], + locals->PixelPTEBytesPerRow[k], + locals->PrefetchSourceLinesY[k], + locals->SwathWidthY[k], + locals->BytePerPixelDETY[k], + locals->VInitPreFillY[k], + locals->MaxNumSwathY[k], + locals->PrefetchSourceLinesC[k], + locals->BytePerPixelDETC[k], + locals->VInitPreFillC[k], + locals->MaxNumSwathC[k], + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + &locals->DSTXAfterScaler[k], + &locals->DSTYAfterScaler[k], + &locals->DestinationLinesForPrefetch[k], + &locals->PrefetchBandwidth[k], + &locals->DestinationLinesToRequestVMInVBlank[k], + &locals->DestinationLinesToRequestRowInVBlank[k], + &locals->VRatioPrefetchY[k], + &locals->VRatioPrefetchC[k], + &locals->RequiredPrefetchPixDataBWLuma[k], + &locals->RequiredPrefetchPixDataBWChroma[k], + &locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + &locals->Tno_bw[k], + &locals->prefetch_vmrow_bw[k], + &locals->swath_width_luma_ub[k], + &locals->swath_width_chroma_ub[k], + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + if (mode_lib->vba.BlendingAndTiming[k] == k) { + locals->VStartup[k] = dml_min( + mode_lib->vba.VStartupLines, + locals->MaxVStartupLines[k]); + if (locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata + != 0) { + locals->VStartup[k] = + locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata; + } + } else { + locals->VStartup[k] = + dml_min( + mode_lib->vba.VStartupLines, + locals->MaxVStartupLines[mode_lib->vba.BlendingAndTiming[k]]); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int m; + + locals->cursor_bw[k] = 0; + locals->cursor_bw_pre[k] = 0; + for (m = 0; m < mode_lib->vba.NumberOfCursors[k]; m++) { + locals->cursor_bw[k] += mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->cursor_bw_pre[k] += mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * locals->VRatioPrefetchY[k]; + } + + CalculateUrgentBurstFactor( + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + locals->SwathWidthY[k], + mode_lib->vba.HTotal[k] / + mode_lib->vba.PixelClock[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.CursorBufferSize, + mode_lib->vba.CursorWidth[k][0] + mode_lib->vba.CursorWidth[k][1], + dml_max(mode_lib->vba.CursorBPP[k][0], mode_lib->vba.CursorBPP[k][1]), + mode_lib->vba.VRatio[k], + locals->VRatioPrefetchY[k], + locals->VRatioPrefetchC[k], + locals->BytePerPixelDETY[k], + locals->BytePerPixelDETC[k], + &locals->UrgentBurstFactorCursor[k], + &locals->UrgentBurstFactorCursorPre[k], + &locals->UrgentBurstFactorLuma[k], + &locals->UrgentBurstFactorLumaPre[k], + &locals->UrgentBurstFactorChroma[k], + &locals->UrgentBurstFactorChromaPre[k], + &locals->NotEnoughUrgentLatencyHiding, + &locals->NotEnoughUrgentLatencyHidingPre); + + if (mode_lib->vba.UseUrgentBurstBandwidth == false) { + locals->UrgentBurstFactorLuma[k] = 1; + locals->UrgentBurstFactorChroma[k] = 1; + locals->UrgentBurstFactorCursor[k] = 1; + locals->UrgentBurstFactorLumaPre[k] = 1; + locals->UrgentBurstFactorChromaPre[k] = 1; + locals->UrgentBurstFactorCursorPre[k] = 1; + } + + MaxTotalRDBandwidth = MaxTotalRDBandwidth + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->ReadBandwidthPlaneLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthPlaneChroma[k] * locals->UrgentBurstFactorChroma[k] + locals->cursor_bw[k] + * locals->UrgentBurstFactorCursor[k] + locals->meta_row_bw[k] + locals->dpte_row_bw[k], + locals->RequiredPrefetchPixDataBWLuma[k] * locals->UrgentBurstFactorLumaPre[k] + locals->RequiredPrefetchPixDataBWChroma[k] + * locals->UrgentBurstFactorChromaPre[k] + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + + MaxTotalRDBandwidthNoUrgentBurst = MaxTotalRDBandwidthNoUrgentBurst + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->ReadBandwidthPlaneLuma[k] + locals->ReadBandwidthPlaneChroma[k] + locals->cursor_bw[k] + + locals->meta_row_bw[k] + locals->dpte_row_bw[k], + locals->RequiredPrefetchPixDataBWLuma[k] + locals->RequiredPrefetchPixDataBWChroma[k] + locals->cursor_bw_pre[k]); + + if (locals->DestinationLinesForPrefetch[k] < 2) + DestinationLineTimesForPrefetchLessThan2 = true; + if (locals->VRatioPrefetchY[k] > 4 || locals->VRatioPrefetchC[k] > 4) + VRatioPrefetchMoreThan4 = true; + } + mode_lib->vba.FractionOfUrgentBandwidth = MaxTotalRDBandwidthNoUrgentBurst / mode_lib->vba.ReturnBW; + + if (MaxTotalRDBandwidth <= mode_lib->vba.ReturnBW && locals->NotEnoughUrgentLatencyHiding == 0 && locals->NotEnoughUrgentLatencyHidingPre == 0 && !VRatioPrefetchMoreThan4 + && !DestinationLineTimesForPrefetchLessThan2) + mode_lib->vba.PrefetchModeSupported = true; + else { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Bandwidth violation. Results are NOT valid\n"); + } + + if (mode_lib->vba.PrefetchModeSupported == true) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = mode_lib->vba.ReturnBW; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.BandwidthAvailableForImmediateFlip + - dml_max( + locals->ReadBandwidthPlaneLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthPlaneChroma[k] * locals->UrgentBurstFactorChroma[k] + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->RequiredPrefetchPixDataBWLuma[k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixDataBWChroma[k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + } + + mode_lib->vba.TotImmediateFlipBytes = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.TotImmediateFlipBytes = mode_lib->vba.TotImmediateFlipBytes + locals->PDEAndMetaPTEBytesFrame[k] + locals->MetaRowByte[k] + locals->PixelPTEBytesPerRow[k]; + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.GPUVMEnable, + locals->PDEAndMetaPTEBytesFrame[k], + locals->MetaRowByte[k], + locals->PixelPTEBytesPerRow[k], + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + locals->Tno_bw[k], + mode_lib->vba.DCCEnable[k], + locals->dpte_row_height[k], + locals->meta_row_height[k], + locals->dpte_row_height_chroma[k], + locals->meta_row_height_chroma[k], + &locals->DestinationLinesToRequestVMInImmediateFlip[k], + &locals->DestinationLinesToRequestRowInImmediateFlip[k], + &locals->final_flip_bw[k], + &locals->ImmediateFlipSupportedForPipe[k]); + } + mode_lib->vba.total_dcn_read_bw_with_flip = 0.0; + mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.total_dcn_read_bw_with_flip = + mode_lib->vba.total_dcn_read_bw_with_flip + dml_max3( + locals->prefetch_vmrow_bw[k], + locals->final_flip_bw[k] + locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthChroma[k] * locals->UrgentBurstFactorChroma[k] + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->final_flip_bw[k] + locals->RequiredPrefetchPixDataBWLuma[k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixDataBWChroma[k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst = + mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->final_flip_bw[k] + locals->ReadBandwidthPlaneLuma[k] + locals->ReadBandwidthPlaneChroma[k] + locals->cursor_bw[k], + locals->final_flip_bw[k] + locals->RequiredPrefetchPixDataBWLuma[k] + locals->RequiredPrefetchPixDataBWChroma[k] + locals->cursor_bw_pre[k]); + + } + mode_lib->vba.FractionOfUrgentBandwidthImmediateFlip = mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst / mode_lib->vba.ReturnBW; + + mode_lib->vba.ImmediateFlipSupported = true; + if (mode_lib->vba.total_dcn_read_bw_with_flip > mode_lib->vba.ReturnBW) { + mode_lib->vba.ImmediateFlipSupported = false; + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->ImmediateFlipSupportedForPipe[k] == false) { + mode_lib->vba.ImmediateFlipSupported = false; + } + } + } else { + mode_lib->vba.ImmediateFlipSupported = false; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ErrorResult[k]) { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Prefetch schedule violation. Results are NOT valid\n"); + } + } + + mode_lib->vba.VStartupLines = mode_lib->vba.VStartupLines + 1; + } while (!((mode_lib->vba.PrefetchModeSupported + && ((!mode_lib->vba.ImmediateFlipSupport && !mode_lib->vba.HostVMEnable) + || mode_lib->vba.ImmediateFlipSupported)) + || locals->MaximumMaxVStartupLines < mode_lib->vba.VStartupLines)); + + //Watermarks and NB P-State/DRAM Clock Change Support + { + enum clock_change_support DRAMClockChangeSupport; // dummy + CalculateWatermarksAndDRAMSpeedChangeSupport( + mode_lib, + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.MaxLineBufferLines, + mode_lib->vba.LineBufferSize, + mode_lib->vba.DPPOutputBufferPixels, + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.WritebackInterfaceLumaBufferSize, + mode_lib->vba.WritebackInterfaceChromaBufferSize, + mode_lib->vba.DCFCLK, + mode_lib->vba.UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels, + mode_lib->vba.ReturnBW, + mode_lib->vba.GPUVMEnable, + locals->dpte_group_bytes, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.UrgentLatency, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.WritebackLatency, + mode_lib->vba.WritebackChunkSize, + mode_lib->vba.SOCCLK, + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.SRExitTime, + mode_lib->vba.SREnterPlusExitTime, + mode_lib->vba.DCFCLKDeepSleep, + mode_lib->vba.DPPPerPlane, + mode_lib->vba.DCCEnable, + locals->DPPCLK, + locals->SwathWidthSingleDPPY, + mode_lib->vba.SwathHeightY, + locals->ReadBandwidthPlaneLuma, + mode_lib->vba.SwathHeightC, + locals->ReadBandwidthPlaneChroma, + mode_lib->vba.LBBitPerPixel, + locals->SwathWidthY, + mode_lib->vba.HRatio, + mode_lib->vba.vtaps, + mode_lib->vba.VTAPsChroma, + mode_lib->vba.VRatio, + mode_lib->vba.HTotal, + mode_lib->vba.PixelClock, + mode_lib->vba.BlendingAndTiming, + locals->BytePerPixelDETY, + locals->BytePerPixelDETC, + mode_lib->vba.WritebackEnable, + mode_lib->vba.WritebackPixelFormat, + mode_lib->vba.WritebackDestinationWidth, + mode_lib->vba.WritebackDestinationHeight, + mode_lib->vba.WritebackSourceHeight, + &DRAMClockChangeSupport, + &mode_lib->vba.UrgentWatermark, + &mode_lib->vba.WritebackUrgentWatermark, + &mode_lib->vba.DRAMClockChangeWatermark, + &mode_lib->vba.WritebackDRAMClockChangeWatermark, + &mode_lib->vba.StutterExitWatermark, + &mode_lib->vba.StutterEnterPlusExitWatermark, + &mode_lib->vba.MinActiveDRAMClockChangeLatencySupported); + } + + + //Display Pipeline Delivery Time in Prefetch, Groups + CalculatePixelDeliveryTimes( + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.VRatio, + locals->VRatioPrefetchY, + locals->VRatioPrefetchC, + locals->swath_width_luma_ub, + locals->swath_width_chroma_ub, + mode_lib->vba.DPPPerPlane, + mode_lib->vba.HRatio, + mode_lib->vba.PixelClock, + locals->PSCL_THROUGHPUT_LUMA, + locals->PSCL_THROUGHPUT_CHROMA, + locals->DPPCLK, + locals->BytePerPixelDETC, + mode_lib->vba.SourceScan, + locals->BlockWidth256BytesY, + locals->BlockHeight256BytesY, + locals->BlockWidth256BytesC, + locals->BlockHeight256BytesC, + locals->DisplayPipeLineDeliveryTimeLuma, + locals->DisplayPipeLineDeliveryTimeChroma, + locals->DisplayPipeLineDeliveryTimeLumaPrefetch, + locals->DisplayPipeLineDeliveryTimeChromaPrefetch, + locals->DisplayPipeRequestDeliveryTimeLuma, + locals->DisplayPipeRequestDeliveryTimeChroma, + locals->DisplayPipeRequestDeliveryTimeLumaPrefetch, + locals->DisplayPipeRequestDeliveryTimeChromaPrefetch); + + CalculateMetaAndPTETimes( + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.MinMetaChunkSizeBytes, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.HTotal, + mode_lib->vba.VRatio, + locals->VRatioPrefetchY, + locals->VRatioPrefetchC, + locals->DestinationLinesToRequestRowInVBlank, + locals->DestinationLinesToRequestRowInImmediateFlip, + locals->DestinationLinesToRequestVMInVBlank, + locals->DestinationLinesToRequestVMInImmediateFlip, + mode_lib->vba.DCCEnable, + mode_lib->vba.PixelClock, + locals->BytePerPixelDETY, + locals->BytePerPixelDETC, + mode_lib->vba.SourceScan, + locals->dpte_row_height, + locals->dpte_row_height_chroma, + locals->meta_row_width, + locals->meta_row_height, + locals->meta_req_width, + locals->meta_req_height, + locals->dpte_group_bytes, + locals->PTERequestSizeY, + locals->PTERequestSizeC, + locals->PixelPTEReqWidthY, + locals->PixelPTEReqHeightY, + locals->PixelPTEReqWidthC, + locals->PixelPTEReqHeightC, + locals->dpte_row_width_luma_ub, + locals->dpte_row_width_chroma_ub, + locals->vm_group_bytes, + locals->dpde0_bytes_per_frame_ub_l, + locals->dpde0_bytes_per_frame_ub_c, + locals->meta_pte_bytes_per_frame_ub_l, + locals->meta_pte_bytes_per_frame_ub_c, + locals->DST_Y_PER_PTE_ROW_NOM_L, + locals->DST_Y_PER_PTE_ROW_NOM_C, + locals->DST_Y_PER_META_ROW_NOM_L, + locals->TimePerMetaChunkNominal, + locals->TimePerMetaChunkVBlank, + locals->TimePerMetaChunkFlip, + locals->time_per_pte_group_nom_luma, + locals->time_per_pte_group_vblank_luma, + locals->time_per_pte_group_flip_luma, + locals->time_per_pte_group_nom_chroma, + locals->time_per_pte_group_vblank_chroma, + locals->time_per_pte_group_flip_chroma, + locals->TimePerVMGroupVBlank, + locals->TimePerVMGroupFlip, + locals->TimePerVMRequestVBlank, + locals->TimePerVMRequestFlip); + + + // Min TTUVBlank + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) { + locals->AllowDRAMClockChangeDuringVBlank[k] = true; + locals->AllowDRAMSelfRefreshDuringVBlank[k] = true; + locals->MinTTUVBlank[k] = dml_max( + mode_lib->vba.DRAMClockChangeWatermark, + dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark)); + } else if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 1) { + locals->AllowDRAMClockChangeDuringVBlank[k] = false; + locals->AllowDRAMSelfRefreshDuringVBlank[k] = true; + locals->MinTTUVBlank[k] = dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark); + } else { + locals->AllowDRAMClockChangeDuringVBlank[k] = false; + locals->AllowDRAMSelfRefreshDuringVBlank[k] = false; + locals->MinTTUVBlank[k] = mode_lib->vba.UrgentWatermark; + } + if (!mode_lib->vba.DynamicMetadataEnable[k]) + locals->MinTTUVBlank[k] = mode_lib->vba.TCalc + + locals->MinTTUVBlank[k]; + } + + // DCC Configuration + mode_lib->vba.ActiveDPPs = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + locals->MaximumDCCCompressionYSurface[k] = CalculateDCCConfiguration( + mode_lib->vba.DCCEnable[k], + false, // We should always know the direction DCCProgrammingAssumesScanDirectionUnknown, + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + mode_lib->vba.DETBufferSizeInKByte * 1024, + locals->BlockHeight256BytesY[k], + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SurfaceTiling[k], + locals->BytePerPixelDETY[k], + mode_lib->vba.SourceScan[k], + &locals->DCCYMaxUncompressedBlock[k], + &locals->DCCYMaxCompressedBlock[k], + &locals->DCCYIndependent64ByteBlock[k]); + } + + //XFC Parameters: + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.XFCEnabled[k] == true) { + double TWait; + + locals->XFCSlaveVUpdateOffset[k] = mode_lib->vba.XFCTSlvVupdateOffset; + locals->XFCSlaveVupdateWidth[k] = mode_lib->vba.XFCTSlvVupdateWidth; + locals->XFCSlaveVReadyOffset[k] = mode_lib->vba.XFCTSlvVreadyOffset; + TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + mode_lib->vba.XFCRemoteSurfaceFlipDelay = CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthY[k], + dml_ceil(locals->BytePerPixelDETY[k], 1), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + locals->XFCRemoteSurfaceFlipLatency[k] = + dml_floor( + mode_lib->vba.XFCRemoteSurfaceFlipDelay + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + locals->XFCTransferDelay[k] = + dml_ceil( + mode_lib->vba.XFCBusTransportTime + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + locals->XFCPrechargeDelay[k] = + dml_ceil( + (mode_lib->vba.XFCBusTransportTime + + mode_lib->vba.TInitXFill + + mode_lib->vba.TslvChk) + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.InitFillLevel = mode_lib->vba.XFCXBUFLatencyTolerance + * mode_lib->vba.SrcActiveDrainRate; + mode_lib->vba.FinalFillMargin = + (locals->DestinationLinesToRequestVMInVBlank[k] + + locals->DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k] + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.XFCFillConstant; + mode_lib->vba.FinalFillLevel = mode_lib->vba.XFCRemoteSurfaceFlipDelay + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.FinalFillMargin; + mode_lib->vba.RemainingFillLevel = dml_max( + 0.0, + mode_lib->vba.FinalFillLevel - mode_lib->vba.InitFillLevel); + mode_lib->vba.TFinalxFill = mode_lib->vba.RemainingFillLevel + / (mode_lib->vba.SrcActiveDrainRate + * mode_lib->vba.XFCFillBWOverhead / 100); + locals->XFCPrefetchMargin[k] = + mode_lib->vba.XFCRemoteSurfaceFlipDelay + + mode_lib->vba.TFinalxFill + + (locals->DestinationLinesToRequestVMInVBlank[k] + + locals->DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + locals->XFCSlaveVUpdateOffset[k] = 0; + locals->XFCSlaveVupdateWidth[k] = 0; + locals->XFCSlaveVReadyOffset[k] = 0; + locals->XFCRemoteSurfaceFlipLatency[k] = 0; + locals->XFCPrechargeDelay[k] = 0; + locals->XFCTransferDelay[k] = 0; + locals->XFCPrefetchMargin[k] = 0; + } + } + + // Stutter Efficiency + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + CalculateDETBufferSize( + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + &locals->DETBufferSizeY[k], + &locals->DETBufferSizeC[k]); + + locals->LinesInDETY[k] = locals->DETBufferSizeY[k] + / locals->BytePerPixelDETY[k] / locals->SwathWidthY[k]; + locals->LinesInDETYRoundedDownToSwath[k] = dml_floor( + locals->LinesInDETY[k], + mode_lib->vba.SwathHeightY[k]); + locals->FullDETBufferingTimeY[k] = + locals->LinesInDETYRoundedDownToSwath[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k]; + } + + mode_lib->vba.StutterPeriod = 999999.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->FullDETBufferingTimeY[k] < mode_lib->vba.StutterPeriod) { + mode_lib->vba.StutterPeriod = locals->FullDETBufferingTimeY[k]; + mode_lib->vba.FrameTimeForMinFullDETBufferingTime = + (double) mode_lib->vba.VTotal[k] * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + locals->BytePerPixelYCriticalPlane = dml_ceil(locals->BytePerPixelDETY[k], 1); + locals->SwathWidthYCriticalPlane = locals->SwathWidthY[k]; + locals->LinesToFinishSwathTransferStutterCriticalPlane = + mode_lib->vba.SwathHeightY[k] - (locals->LinesInDETY[k] - locals->LinesInDETYRoundedDownToSwath[k]); + } + } + + mode_lib->vba.AverageReadBandwidth = 0.0; + mode_lib->vba.TotalRowReadBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int DCCRateLimit; + + if (mode_lib->vba.DCCEnable[k]) { + if (locals->DCCYMaxCompressedBlock[k] == 256) + DCCRateLimit = 4; + else + DCCRateLimit = 2; + + mode_lib->vba.AverageReadBandwidth = + mode_lib->vba.AverageReadBandwidth + + (locals->ReadBandwidthPlaneLuma[k] + locals->ReadBandwidthPlaneChroma[k]) / + dml_min(mode_lib->vba.DCCRate[k], DCCRateLimit); + } else { + mode_lib->vba.AverageReadBandwidth = + mode_lib->vba.AverageReadBandwidth + + locals->ReadBandwidthPlaneLuma[k] + + locals->ReadBandwidthPlaneChroma[k]; + } + mode_lib->vba.TotalRowReadBandwidth = mode_lib->vba.TotalRowReadBandwidth + + locals->meta_row_bw[k] + locals->dpte_row_bw[k]; + } + + mode_lib->vba.AverageDCCCompressionRate = mode_lib->vba.TotalDataReadBandwidth / mode_lib->vba.AverageReadBandwidth; + + mode_lib->vba.PartOfBurstThatFitsInROB = + dml_min( + mode_lib->vba.StutterPeriod + * mode_lib->vba.TotalDataReadBandwidth, + mode_lib->vba.ROBBufferSizeInKByte * 1024 + * mode_lib->vba.AverageDCCCompressionRate); + mode_lib->vba.StutterBurstTime = mode_lib->vba.PartOfBurstThatFitsInROB + / mode_lib->vba.AverageDCCCompressionRate / mode_lib->vba.ReturnBW + + (mode_lib->vba.StutterPeriod * mode_lib->vba.TotalDataReadBandwidth + - mode_lib->vba.PartOfBurstThatFitsInROB) + / (mode_lib->vba.DCFCLK * 64) + + mode_lib->vba.StutterPeriod * mode_lib->vba.TotalRowReadBandwidth / mode_lib->vba.ReturnBW; + mode_lib->vba.StutterBurstTime = dml_max( + mode_lib->vba.StutterBurstTime, + (locals->LinesToFinishSwathTransferStutterCriticalPlane * locals->BytePerPixelYCriticalPlane * + locals->SwathWidthYCriticalPlane / mode_lib->vba.ReturnBW) + ); + + mode_lib->vba.TotalActiveWriteback = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.TotalActiveWriteback = mode_lib->vba.TotalActiveWriteback + 1; + } + } + + if (mode_lib->vba.TotalActiveWriteback == 0) { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = (1 + - (mode_lib->vba.SRExitTime + mode_lib->vba.StutterBurstTime) + / mode_lib->vba.StutterPeriod) * 100; + } else { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = 0; + } + + mode_lib->vba.SmallestVBlank = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { + mode_lib->vba.VBlankTime = (double) (mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k]) * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.VBlankTime = 0; + } + mode_lib->vba.SmallestVBlank = dml_min( + mode_lib->vba.SmallestVBlank, + mode_lib->vba.VBlankTime); + } + + mode_lib->vba.StutterEfficiency = (mode_lib->vba.StutterEfficiencyNotIncludingVBlank / 100 + * (mode_lib->vba.FrameTimeForMinFullDETBufferingTime + - mode_lib->vba.SmallestVBlank) + + mode_lib->vba.SmallestVBlank) + / mode_lib->vba.FrameTimeForMinFullDETBufferingTime * 100; +} + +static void DisplayPipeConfiguration(struct display_mode_lib *mode_lib) +{ + // Display Pipe Configuration + double BytePerPixDETY; + double BytePerPixDETC; + double Read256BytesBlockHeightY; + double Read256BytesBlockHeightC; + double Read256BytesBlockWidthY; + double Read256BytesBlockWidthC; + double MaximumSwathHeightY; + double MaximumSwathHeightC; + double MinimumSwathHeightY; + double MinimumSwathHeightC; + double SwathWidth; + double SwathWidthGranularityY; + double SwathWidthGranularityC; + double RoundedUpMaxSwathSizeBytesY; + double RoundedUpMaxSwathSizeBytesC; + unsigned int j, k; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + BytePerPixDETY = 8; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + BytePerPixDETY = 4; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + BytePerPixDETY = 2; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 2; + } else { + BytePerPixDETY = 4.0 / 3.0; + BytePerPixDETC = 8.0 / 3.0; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + Read256BytesBlockHeightY = 4; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + Read256BytesBlockHeightY = 8; + } else { + Read256BytesBlockHeightY = 16; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockHeightC = 0; + Read256BytesBlockWidthC = 0; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + Read256BytesBlockHeightC = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + Read256BytesBlockHeightY = 16; + Read256BytesBlockHeightC = 8; + } else { + Read256BytesBlockHeightY = 8; + Read256BytesBlockHeightC = 8; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockWidthC = 256 / dml_ceil(BytePerPixDETC, 2) + / Read256BytesBlockHeightC; + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + MaximumSwathHeightY = Read256BytesBlockHeightY; + MaximumSwathHeightC = Read256BytesBlockHeightC; + } else { + MaximumSwathHeightY = Read256BytesBlockWidthY; + MaximumSwathHeightC = Read256BytesBlockWidthC; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8 + && mode_lib->vba.SourceScan[k] != dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + } + MinimumSwathHeightC = MaximumSwathHeightC; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightC = MaximumSwathHeightC / 2.0; + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + SwathWidth = mode_lib->vba.ViewportWidth[k]; + } else { + SwathWidth = mode_lib->vba.ViewportHeight[k]; + } + + if (mode_lib->vba.ODMCombineEnabled[k] == true) { + MainPlaneDoesODMCombine = true; + } + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[j] == true) { + MainPlaneDoesODMCombine = true; + } + } + + if (MainPlaneDoesODMCombine == true) { + SwathWidth = dml_min( + SwathWidth, + mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k]); + } else { + SwathWidth = SwathWidth / mode_lib->vba.DPPPerPlane[k]; + } + + SwathWidthGranularityY = 256 / dml_ceil(BytePerPixDETY, 1) / MaximumSwathHeightY; + RoundedUpMaxSwathSizeBytesY = (dml_ceil( + (double) (SwathWidth - 1), + SwathWidthGranularityY) + SwathWidthGranularityY) * BytePerPixDETY + * MaximumSwathHeightY; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesY = dml_ceil(RoundedUpMaxSwathSizeBytesY, 256) + + 256; + } + if (MaximumSwathHeightC > 0) { + SwathWidthGranularityC = 256.0 / dml_ceil(BytePerPixDETC, 2) + / MaximumSwathHeightC; + RoundedUpMaxSwathSizeBytesC = (dml_ceil( + (double) (SwathWidth / 2.0 - 1), + SwathWidthGranularityC) + SwathWidthGranularityC) + * BytePerPixDETC * MaximumSwathHeightC; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesC = dml_ceil( + RoundedUpMaxSwathSizeBytesC, + 256) + 256; + } + } else + RoundedUpMaxSwathSizeBytesC = 0.0; + + if (RoundedUpMaxSwathSizeBytesY + RoundedUpMaxSwathSizeBytesC + <= mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0) { + mode_lib->vba.SwathHeightY[k] = MaximumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MaximumSwathHeightC; + } else { + mode_lib->vba.SwathHeightY[k] = MinimumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MinimumSwathHeightC; + } + + CalculateDETBufferSize( + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + &mode_lib->vba.DETBufferSizeY[k], + &mode_lib->vba.DETBufferSizeC[k]); + } +} + +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatency, + double SREnterPlusExitTime) +{ + if (PrefetchMode == 0) { + return dml_max( + DRAMClockChangeLatency + UrgentLatency, + dml_max(SREnterPlusExitTime, UrgentLatency)); + } else if (PrefetchMode == 1) { + return dml_max(SREnterPlusExitTime, UrgentLatency); + } else { + return UrgentLatency; + } +} + +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk) +{ + double TSlvSetup, AvgfillRate, result; + + *SrcActiveDrainRate = VRatio * SwathWidth * Bpp / LineTime; + TSlvSetup = XFCTSlvVupdateOffset + XFCTSlvVupdateWidth + XFCTSlvVreadyOffset; + *TInitXFill = XFCXBUFLatencyTolerance / (1 + XFCFillBWOverhead / 100); + AvgfillRate = *SrcActiveDrainRate * (1 + XFCFillBWOverhead / 100); + *TslvChk = XFCSlvChunkSize / AvgfillRate; + dml_print( + "DML::CalculateRemoteSurfaceFlipDelay: SrcActiveDrainRate: %f\n", + *SrcActiveDrainRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TSlvSetup: %f\n", TSlvSetup); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TInitXFill: %f\n", *TInitXFill); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: AvgfillRate: %f\n", AvgfillRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TslvChk: %f\n", *TslvChk); + result = 2 * XFCBusTransportTime + TSlvSetup + TCalc + TWait + *TslvChk + *TInitXFill; // TODO: This doesn't seem to match programming guide + dml_print("DML::CalculateRemoteSurfaceFlipDelay: RemoteSurfaceFlipDelay: %f\n", result); + return result; +} + +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth) +{ + double CalculateWriteBackDelay = + dml_max( + dml_ceil(WritebackLumaHTaps / 4.0, 1) / WritebackHRatio, + WritebackLumaVTaps * dml_ceil(1.0 / WritebackVRatio, 1) + * dml_ceil( + WritebackDestinationWidth + / 4.0, + 1) + + dml_ceil(1.0 / WritebackVRatio, 1) + * (dml_ceil( + WritebackLumaVTaps + / 4.0, + 1) + 4)); + + if (WritebackPixelFormat != dm_444_32) { + CalculateWriteBackDelay = + dml_max( + CalculateWriteBackDelay, + dml_max( + dml_ceil( + WritebackChromaHTaps + / 2.0, + 1) + / (2 + * WritebackHRatio), + WritebackChromaVTaps + * dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * dml_ceil( + WritebackDestinationWidth + / 2.0 + / 2.0, + 1) + + dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * (dml_ceil( + WritebackChromaVTaps + / 4.0, + 1) + + 4))); + } + return CalculateWriteBackDelay; +} + +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw) +{ + if (DCCEnable != true) { + *meta_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime) + + VRatio / 2 * MetaRowByteChroma + / (meta_row_height_chroma * LineTime); + } else { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime); + } + + if (GPUVMEnable != true) { + *dpte_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime) + + VRatio / 2 * PixelPTEBytesPerRowChroma + / (dpte_row_height_chroma * LineTime); + } else { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime); + } +} + +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + double UrgentExtraLatency, + double UrgentLatency, + unsigned int GPUVMMaxPageTableLevels, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + bool GPUVMEnable, + double PDEAndMetaPTEBytesPerFrame, + double MetaRowBytes, + double DPTEBytesPerRow, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + double LineTime, + double VRatio, + double Tno_bw, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + unsigned int dpte_row_height_chroma, + unsigned int meta_row_height_chroma, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe) +{ + double min_row_time = 0.0; + unsigned int HostVMDynamicLevels; + double TimeForFetchingMetaPTEImmediateFlip; + double TimeForFetchingRowInVBlankImmediateFlip; + double ImmediateFlipBW; + double HostVMInefficiencyFactor; + + if (GPUVMEnable == true && HostVMEnable == true) { + HostVMInefficiencyFactor = + PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData + / PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly; + HostVMDynamicLevels = HostVMMaxPageTableLevels - HostVMCachedPageTableLevels; + } else { + HostVMInefficiencyFactor = 1; + HostVMDynamicLevels = 0; + } + + ImmediateFlipBW = (PDEAndMetaPTEBytesPerFrame + MetaRowBytes + DPTEBytesPerRow) + * BandwidthAvailableForImmediateFlip / TotImmediateFlipBytes; + + if (GPUVMEnable == true) { + TimeForFetchingMetaPTEImmediateFlip = dml_max3( + Tno_bw + PDEAndMetaPTEBytesPerFrame * HostVMInefficiencyFactor / ImmediateFlipBW, + UrgentExtraLatency + UrgentLatency * (GPUVMMaxPageTableLevels * (HostVMDynamicLevels + 1) - 1), + LineTime / 4.0); + } else { + TimeForFetchingMetaPTEImmediateFlip = 0; + } + + *DestinationLinesToRequestVMInImmediateFlip = dml_ceil(4.0 * (TimeForFetchingMetaPTEImmediateFlip / LineTime), 1) / 4.0; + if ((GPUVMEnable == true || DCCEnable == true)) { + TimeForFetchingRowInVBlankImmediateFlip = dml_max3((MetaRowBytes + DPTEBytesPerRow) * HostVMInefficiencyFactor / ImmediateFlipBW, UrgentLatency * (HostVMDynamicLevels + 1), LineTime / 4); + } else { + TimeForFetchingRowInVBlankImmediateFlip = 0; + } + + *DestinationLinesToRequestRowInImmediateFlip = dml_ceil(4.0 * (TimeForFetchingRowInVBlankImmediateFlip / LineTime), 1) / 4.0; + *final_flip_bw = dml_max(PDEAndMetaPTEBytesPerFrame * HostVMInefficiencyFactor / (*DestinationLinesToRequestVMInImmediateFlip * LineTime), (MetaRowBytes + DPTEBytesPerRow) * HostVMInefficiencyFactor / (*DestinationLinesToRequestRowInImmediateFlip * LineTime)); + if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + if (GPUVMEnable == true && DCCEnable != true) { + min_row_time = dml_min( + dpte_row_height * LineTime / VRatio, + dpte_row_height_chroma * LineTime / (VRatio / 2)); + } else if (GPUVMEnable != true && DCCEnable == true) { + min_row_time = dml_min( + meta_row_height * LineTime / VRatio, + meta_row_height_chroma * LineTime / (VRatio / 2)); + } else { + min_row_time = dml_min4( + dpte_row_height * LineTime / VRatio, + meta_row_height * LineTime / VRatio, + dpte_row_height_chroma * LineTime / (VRatio / 2), + meta_row_height_chroma * LineTime / (VRatio / 2)); + } + } else { + if (GPUVMEnable == true && DCCEnable != true) { + min_row_time = dpte_row_height * LineTime / VRatio; + } else if (GPUVMEnable != true && DCCEnable == true) { + min_row_time = meta_row_height * LineTime / VRatio; + } else { + min_row_time = dml_min( + dpte_row_height * LineTime / VRatio, + meta_row_height * LineTime / VRatio); + } + } + + if (*DestinationLinesToRequestVMInImmediateFlip >= 32 + || *DestinationLinesToRequestRowInImmediateFlip >= 16 + || TimeForFetchingMetaPTEImmediateFlip + 2 * TimeForFetchingRowInVBlankImmediateFlip > min_row_time) { + *ImmediateFlipSupportedForPipe = false; + } else { + *ImmediateFlipSupportedForPipe = true; + } +} + +static unsigned int TruncToValidBPP( + double DecimalBPP, + double DesiredBPP, + bool DSCEnabled, + enum output_encoder_class Output, + enum output_format_class Format, + unsigned int DSCInputBitPerComponent) +{ + if (Output == dm_hdmi) { + if (Format == dm_420) { + if (DecimalBPP >= 18 && (DesiredBPP == 0 || DesiredBPP == 18)) + return 18; + else if (DecimalBPP >= 15 && (DesiredBPP == 0 || DesiredBPP == 15)) + return 15; + else if (DecimalBPP >= 12 && (DesiredBPP == 0 || DesiredBPP == 12)) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_444) { + if (DecimalBPP >= 36 && (DesiredBPP == 0 || DesiredBPP == 36)) + return 36; + else if (DecimalBPP >= 30 && (DesiredBPP == 0 || DesiredBPP == 30)) + return 30; + else if (DecimalBPP >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else if (DecimalBPP >= 18 && (DesiredBPP == 0 || DesiredBPP == 18)) + return 18; + else + return BPP_INVALID; + } else { + if (DecimalBPP / 1.5 >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else if (DecimalBPP / 1.5 >= 20 && (DesiredBPP == 0 || DesiredBPP == 20)) + return 20; + else if (DecimalBPP / 1.5 >= 16 && (DesiredBPP == 0 || DesiredBPP == 16)) + return 16; + else + return BPP_INVALID; + } + } else { + if (DSCEnabled) { + if (Format == dm_420) { + if (DesiredBPP == 0) { + if (DecimalBPP < 6) + return BPP_INVALID; + else if (DecimalBPP >= 1.5 * DSCInputBitPerComponent - 1.0 / 16.0) + return 1.5 * DSCInputBitPerComponent - 1.0 / 16.0; + else + return dml_floor(16 * DecimalBPP, 1) / 16.0; + } else { + if (DecimalBPP < 6 + || DesiredBPP < 6 + || DesiredBPP > 1.5 * DSCInputBitPerComponent - 1.0 / 16.0 + || DecimalBPP < DesiredBPP) { + return BPP_INVALID; + } else { + return DesiredBPP; + } + } + } else if (Format == dm_n422) { + if (DesiredBPP == 0) { + if (DecimalBPP < 7) + return BPP_INVALID; + else if (DecimalBPP >= 2 * DSCInputBitPerComponent - 1.0 / 16.0) + return 2 * DSCInputBitPerComponent - 1.0 / 16.0; + else + return dml_floor(16 * DecimalBPP, 1) / 16.0; + } else { + if (DecimalBPP < 7 + || DesiredBPP < 7 + || DesiredBPP > 2 * DSCInputBitPerComponent - 1.0 / 16.0 + || DecimalBPP < DesiredBPP) { + return BPP_INVALID; + } else { + return DesiredBPP; + } + } + } else { + if (DesiredBPP == 0) { + if (DecimalBPP < 8) + return BPP_INVALID; + else if (DecimalBPP >= 3 * DSCInputBitPerComponent - 1.0 / 16.0) + return 3 * DSCInputBitPerComponent - 1.0 / 16.0; + else + return dml_floor(16 * DecimalBPP, 1) / 16.0; + } else { + if (DecimalBPP < 8 + || DesiredBPP < 8 + || DesiredBPP > 3 * DSCInputBitPerComponent - 1.0 / 16.0 + || DecimalBPP < DesiredBPP) { + return BPP_INVALID; + } else { + return DesiredBPP; + } + } + } + } else if (Format == dm_420) { + if (DecimalBPP >= 18 && (DesiredBPP == 0 || DesiredBPP == 18)) + return 18; + else if (DecimalBPP >= 15 && (DesiredBPP == 0 || DesiredBPP == 15)) + return 15; + else if (DecimalBPP >= 12 && (DesiredBPP == 0 || DesiredBPP == 12)) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_s422 || Format == dm_n422) { + if (DecimalBPP >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else if (DecimalBPP >= 20 && (DesiredBPP == 0 || DesiredBPP == 20)) + return 20; + else if (DecimalBPP >= 16 && (DesiredBPP == 0 || DesiredBPP == 16)) + return 16; + else + return BPP_INVALID; + } else { + if (DecimalBPP >= 36 && (DesiredBPP == 0 || DesiredBPP == 36)) + return 36; + else if (DecimalBPP >= 30 && (DesiredBPP == 0 || DesiredBPP == 30)) + return 30; + else if (DecimalBPP >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else + return BPP_INVALID; + } + } +} + +void dml21_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) +{ + struct vba_vars_st *locals = &mode_lib->vba; + + int i; + unsigned int j, k, m; + + /*MODE SUPPORT, VOLTAGE STATE AND SOC CONFIGURATION*/ + + /*Scale Ratio, taps Support Check*/ + + mode_lib->vba.ScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.ScalerEnabled[k] == false + && ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) + || mode_lib->vba.HRatio[k] != 1.0 + || mode_lib->vba.htaps[k] != 1.0 + || mode_lib->vba.VRatio[k] != 1.0 + || mode_lib->vba.vtaps[k] != 1.0)) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } else if (mode_lib->vba.vtaps[k] < 1.0 || mode_lib->vba.vtaps[k] > 8.0 + || mode_lib->vba.htaps[k] < 1.0 || mode_lib->vba.htaps[k] > 8.0 + || (mode_lib->vba.htaps[k] > 1.0 + && (mode_lib->vba.htaps[k] % 2) == 1) + || mode_lib->vba.HRatio[k] > mode_lib->vba.MaxHSCLRatio + || mode_lib->vba.VRatio[k] > mode_lib->vba.MaxVSCLRatio + || mode_lib->vba.HRatio[k] > mode_lib->vba.htaps[k] + || mode_lib->vba.VRatio[k] > mode_lib->vba.vtaps[k] + || (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8 + && (mode_lib->vba.HRatio[k] / 2.0 + > mode_lib->vba.HTAPsChroma[k] + || mode_lib->vba.VRatio[k] / 2.0 + > mode_lib->vba.VTAPsChroma[k]))) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } + } + /*Source Format, Pixel Format and Scan Support Check*/ + + mode_lib->vba.SourceFormatPixelAndScanSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + && mode_lib->vba.SourceScan[k] != dm_horz) + || ((mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_t + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d_x) + && mode_lib->vba.SourcePixelFormat[k] != dm_444_64) + || (mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_r_x + && (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)) + || (((mode_lib->vba.SurfaceTiling[k] == dm_sw_gfx7_2d_thin_gl + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_gfx7_2d_thin_lvp) + && !((mode_lib->vba.SourcePixelFormat[k] + == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] + == dm_444_32) + && mode_lib->vba.SourceScan[k] + == dm_horz + && mode_lib->vba.SupportGFX7CompatibleTilingIn32bppAnd64bpp + == true + && mode_lib->vba.DCCEnable[k] + == false)) + || (mode_lib->vba.DCCEnable[k] == true + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_linear + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)))) { + mode_lib->vba.SourceFormatPixelAndScanSupport = false; + } + } + /*Bandwidth Support Check*/ + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + locals->BytePerPixelInDETY[k] = 8.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + locals->BytePerPixelInDETY[k] = 4.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16) { + locals->BytePerPixelInDETY[k] = 2.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 2.0; + } else { + locals->BytePerPixelInDETY[k] = 4.0 / 3; + locals->BytePerPixelInDETC[k] = 8.0 / 3; + } + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportWidth[k]; + } else { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportHeight[k]; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ReadBandwidthLuma[k] = locals->SwathWidthYSingleDPP[k] * dml_ceil(locals->BytePerPixelInDETY[k], 1.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->ReadBandwidthChroma[k] = locals->SwathWidthYSingleDPP[k] / 2 * dml_ceil(locals->BytePerPixelInDETC[k], 2.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k] / 2.0; + locals->ReadBandwidth[k] = locals->ReadBandwidthLuma[k] + locals->ReadBandwidthChroma[k]; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 4.0; + } else if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 3.0; + } else if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 1.5; + } else { + locals->WriteBandwidth[k] = 0.0; + } + } + mode_lib->vba.DCCEnabledInAnyPlane = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.DCCEnable[k] == true) { + mode_lib->vba.DCCEnabledInAnyPlane = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->IdealSDPPortBandwidthPerState[i] = dml_min3( + mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLKPerState[i], + mode_lib->vba.DRAMSpeedPerState[i] * mode_lib->vba.NumberOfChannels + * mode_lib->vba.DRAMChannelWidth, + mode_lib->vba.FabricClockPerState[i] + * mode_lib->vba.FabricDatapathToDCNDataReturn); + if (mode_lib->vba.HostVMEnable == false) { + locals->ReturnBWPerState[i] = locals->IdealSDPPortBandwidthPerState[i] + * mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly / 100.0; + } else { + locals->ReturnBWPerState[i] = locals->IdealSDPPortBandwidthPerState[i] + * mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData / 100.0; + } + } + /*Writeback Latency support check*/ + + mode_lib->vba.WritebackLatencySupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + if (locals->WriteBandwidth[k] + > (mode_lib->vba.WritebackInterfaceLumaBufferSize + + mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } else { + if (locals->WriteBandwidth[k] + > 1.5 + * dml_min( + mode_lib->vba.WritebackInterfaceLumaBufferSize, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } + } + } + /*Re-ordering Buffer Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i] = + (mode_lib->vba.RoundTripPingLatencyCycles + 32.0) / mode_lib->vba.DCFCLKPerState[i] + + dml_max3(mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelMixedWithVMData, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelVMDataOnly) + * mode_lib->vba.NumberOfChannels / locals->ReturnBWPerState[i]; + if ((mode_lib->vba.ROBBufferSizeInKByte - mode_lib->vba.PixelChunkSizeInKByte) * 1024.0 / locals->ReturnBWPerState[i] + > locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i]) { + locals->ROBSupport[i] = true; + } else { + locals->ROBSupport[i] = false; + } + } + /*Writeback Mode Support Check*/ + + mode_lib->vba.TotalNumberOfActiveWriteback = 0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.ActiveWritebacksPerPlane[k] == 0) + mode_lib->vba.ActiveWritebacksPerPlane[k] = 1; + mode_lib->vba.TotalNumberOfActiveWriteback = + mode_lib->vba.TotalNumberOfActiveWriteback + + mode_lib->vba.ActiveWritebacksPerPlane[k]; + } + } + mode_lib->vba.WritebackModeSupport = true; + if (mode_lib->vba.TotalNumberOfActiveWriteback > mode_lib->vba.MaxNumWriteback) { + mode_lib->vba.WritebackModeSupport = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.Writeback10bpc420Supported != true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + mode_lib->vba.WritebackModeSupport = false; + } + } + /*Writeback Scale Ratio and Taps Support Check*/ + + mode_lib->vba.WritebackScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackLumaAndChromaScalingSupported == false + && (mode_lib->vba.WritebackHRatio[k] != 1.0 + || mode_lib->vba.WritebackVRatio[k] != 1.0)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackHRatio[k] > mode_lib->vba.WritebackMaxHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackMaxVSCLRatio + || mode_lib->vba.WritebackHRatio[k] + < mode_lib->vba.WritebackMinHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + < mode_lib->vba.WritebackMinVSCLRatio + || mode_lib->vba.WritebackLumaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackLumaHTaps[k] + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackLumaVTaps[k] + || (mode_lib->vba.WritebackLumaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackLumaHTaps[k] % 2) + == 1)) + || (mode_lib->vba.WritebackPixelFormat[k] != dm_444_32 + && (mode_lib->vba.WritebackChromaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || 2.0 + * mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackChromaHTaps[k] + || 2.0 + * mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackChromaVTaps[k] + || (mode_lib->vba.WritebackChromaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackChromaHTaps[k] % 2) == 1))))) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackVRatio[k] < 1.0) { + mode_lib->vba.WritebackLumaVExtra = + dml_max(1.0 - 2.0 / dml_ceil(1.0 / mode_lib->vba.WritebackVRatio[k], 1.0), 0.0); + } else { + mode_lib->vba.WritebackLumaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_444_32 + && mode_lib->vba.WritebackLumaVTaps[k] + > (mode_lib->vba.WritebackLineBufferLumaBufferSize + + mode_lib->vba.WritebackLineBufferChromaBufferSize) + / 3.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (2.0 * mode_lib->vba.WritebackVRatio[k] < 1) { + mode_lib->vba.WritebackChromaVExtra = 0.0; + } else { + mode_lib->vba.WritebackChromaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + } + } + /*Maximum DISPCLK/DPPCLK Support check*/ + + mode_lib->vba.WritebackRequiredDISPCLK = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.WritebackRequiredDISPCLK = + dml_max( + mode_lib->vba.WritebackRequiredDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.HRatio[k] > 1.0) { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + if (locals->BytePerPixelInDETC[k] == 0.0) { + locals->PSCL_FACTOR_CHROMA[k] = 0.0; + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max3( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } else { + if (mode_lib->vba.HRatio[k] / 2.0 > 1.0) { + locals->PSCL_FACTOR_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2.0 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max5( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2.0), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4.0 + / locals->PSCL_FACTOR_CHROMA[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0 + || mode_lib->vba.HTAPsChroma[k] > 6.0 + || mode_lib->vba.VTAPsChroma[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + dml_ceil(locals->BytePerPixelInDETC[k], 2.0), + &locals->Read256BlockHeightY[k], + &locals->Read256BlockHeightC[k], + &locals->Read256BlockWidthY[k], + &locals->Read256BlockWidthC[k]); + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MaxSwathHeightY[k] = locals->Read256BlockHeightY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockHeightC[k]; + } else { + locals->MaxSwathHeightY[k] = locals->Read256BlockWidthY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockWidthC[k]; + } + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + } + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k] + / 2.0; + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } + } + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + mode_lib->vba.MaximumSwathWidthSupport = 8192.0; + } else { + mode_lib->vba.MaximumSwathWidthSupport = 5120.0; + } + mode_lib->vba.MaximumSwathWidthInDETBuffer = + dml_min( + mode_lib->vba.MaximumSwathWidthSupport, + mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0 + / (locals->BytePerPixelInDETY[k] + * locals->MinSwathHeightY[k] + + locals->BytePerPixelInDETC[k] + / 2.0 + * locals->MinSwathHeightC[k])); + if (locals->BytePerPixelInDETC[k] == 0.0) { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + mode_lib->vba.LineBufferSize + * dml_max(mode_lib->vba.HRatio[k], 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)); + } else { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + dml_min( + mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k], + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)), + 2.0 * mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k] + / 2.0, + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.VTAPsChroma[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k] + / 2.0, + 1.0) + - 2, + 0.0))); + } + locals->MaximumSwathWidth[k] = dml_min( + mode_lib->vba.MaximumSwathWidthInDETBuffer, + mode_lib->vba.MaximumSwathWidthInLineBuffer); + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDispclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDppclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = + mode_lib->vba.PixelClock[k] + * (1.0 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100.0) + * (1.0 + + mode_lib->vba.DISPCLKRampingMargin + / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = mode_lib->vba.PixelClock[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * (1 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + if (mode_lib->vba.ODMCapability == false || mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine <= mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) { + locals->ODMCombineEnablePerState[i][k] = false; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine; + } else { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } + if (locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) <= mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + && locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k] + && locals->ODMCombineEnablePerState[i][k] == false) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if ((locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity) + || (mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity)) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + if (j == 1) { + while (locals->TotalNumberOfActiveDPP[i][j] < mode_lib->vba.MaxNumDPP + && locals->TotalNumberOfActiveDPP[i][j] < 2 * mode_lib->vba.NumberOfActivePlanes) { + double BWOfNonSplitPlaneOfMaximumBandwidth; + unsigned int NumberOfNonSplitPlaneOfMaximumBandwidth; + + BWOfNonSplitPlaneOfMaximumBandwidth = 0; + NumberOfNonSplitPlaneOfMaximumBandwidth = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->ReadBandwidth[k] > BWOfNonSplitPlaneOfMaximumBandwidth && locals->NoOfDPP[i][j][k] == 1) { + BWOfNonSplitPlaneOfMaximumBandwidth = locals->ReadBandwidth[k]; + NumberOfNonSplitPlaneOfMaximumBandwidth = k; + } + } + locals->NoOfDPP[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = 2; + locals->RequiredDPPCLK[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = + locals->MinDPPCLKUsingSingleDPP[NumberOfNonSplitPlaneOfMaximumBandwidth] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100) / 2; + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + 1; + } + } + if (locals->TotalNumberOfActiveDPP[i][j] > mode_lib->vba.MaxNumDPP) { + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ODMCombineEnablePerState[i][k] = false; + if (locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k]) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + if (i != mode_lib->vba.soc.num_states) { + mode_lib->vba.PlaneRequiredDISPCLK = + mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + * (1.0 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + } else { + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if (locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + || mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.WritebackRequiredDISPCLK); + if (mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity + < mode_lib->vba.WritebackRequiredDISPCLK) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + } + /*Viewport Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->ViewportSizeSupport[i] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ODMCombineEnablePerState[i][k] == true) { + if (dml_min(locals->SwathWidthYSingleDPP[k], dml_round(mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k])) + > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i] = false; + } + } else { + if (locals->SwathWidthYSingleDPP[k] / 2.0 > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i] = false; + } + } + } + } + /*Total Available Pipes Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + if (locals->TotalNumberOfActiveDPP[i][j] <= mode_lib->vba.MaxNumDPP) + locals->TotalAvailablePipesSupport[i][j] = true; + else + locals->TotalAvailablePipesSupport[i][j] = false; + } + } + /*Total Available OTG Support Check*/ + + mode_lib->vba.TotalNumberOfActiveOTG = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.TotalNumberOfActiveOTG = mode_lib->vba.TotalNumberOfActiveOTG + + 1.0; + } + } + if (mode_lib->vba.TotalNumberOfActiveOTG <= mode_lib->vba.MaxNumOTG) { + mode_lib->vba.NumberOfOTGSupport = true; + } else { + mode_lib->vba.NumberOfOTGSupport = false; + } + /*Display IO and DSC Support Check*/ + + mode_lib->vba.NonsupportedDSCInputBPC = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (!(mode_lib->vba.DSCInputBitPerComponent[k] == 12.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 10.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 8.0)) { + mode_lib->vba.NonsupportedDSCInputBPC = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->RequiresDSC[i][k] = 0; + locals->RequiresFEC[i][k] = 0; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.Output[k] == dm_hdmi) { + locals->RequiresDSC[i][k] = 0; + locals->RequiresFEC[i][k] = 0; + locals->OutputBppPerState[i][k] = TruncToValidBPP( + dml_min(600.0, mode_lib->vba.PHYCLKPerState[i]) / mode_lib->vba.PixelClockBackEnd[k] * 24, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + } else if (mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp) { + if (mode_lib->vba.Output[k] == dm_edp) { + mode_lib->vba.EffectiveFECOverhead = 0.0; + } else { + mode_lib->vba.EffectiveFECOverhead = + mode_lib->vba.FECOverhead; + } + if (mode_lib->vba.PHYCLKPerState[i] >= 270.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID && mode_lib->vba.PHYCLKPerState[i] >= 540.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID + && mode_lib->vba.PHYCLKPerState[i] + >= 810.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true || mode_lib->vba.Outbpp == BPP_INVALID) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = + mode_lib->vba.Outbpp; + } + } + } else { + locals->OutputBppPerState[i][k] = BPP_BLENDED_PIPE; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->DIOSupport[i] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->OutputBppPerState[i][k] == BPP_INVALID + || (mode_lib->vba.OutputFormat[k] == dm_420 + && mode_lib->vba.Interlace[k] == true + && mode_lib->vba.ProgressiveToInterlaceUnitInOPP == true)) { + locals->DIOSupport[i] = false; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->DSCCLKRequiredMoreThanSupported[i] = false; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if ((mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp)) { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] + == dm_n422) { + mode_lib->vba.DSCFormatFactor = 2; + } else { + mode_lib->vba.DSCFormatFactor = 1; + } + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] + == true) { + if (mode_lib->vba.PixelClockBackEnd[k] / 6.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } else { + if (mode_lib->vba.PixelClockBackEnd[k] / 3.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } + } + } + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->NotEnoughDSCUnits[i] = false; + mode_lib->vba.TotalDSCUnitsRequired = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == true) { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 2.0; + } else { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 1.0; + } + } + } + if (mode_lib->vba.TotalDSCUnitsRequired > mode_lib->vba.NumberOfDSC) { + locals->NotEnoughDSCUnits[i] = true; + } + } + /*DSC Delay per state*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] != k) { + mode_lib->vba.slices = 0; + } else if (locals->RequiresDSC[i][k] == 0 + || locals->RequiresDSC[i][k] == false) { + mode_lib->vba.slices = 0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 3200.0) { + mode_lib->vba.slices = dml_ceil( + mode_lib->vba.PixelClockBackEnd[k] / 400.0, + 4.0); + } else if (mode_lib->vba.PixelClockBackEnd[k] > 1360.0) { + mode_lib->vba.slices = 8.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 680.0) { + mode_lib->vba.slices = 4.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 340.0) { + mode_lib->vba.slices = 2.0; + } else { + mode_lib->vba.slices = 1.0; + } + if (locals->OutputBppPerState[i][k] == BPP_BLENDED_PIPE + || locals->OutputBppPerState[i][k] == BPP_INVALID) { + mode_lib->vba.bpp = 0.0; + } else { + mode_lib->vba.bpp = locals->OutputBppPerState[i][k]; + } + if (locals->RequiresDSC[i][k] == true && mode_lib->vba.bpp != 0.0) { + if (locals->ODMCombineEnablePerState[i][k] == false) { + locals->DSCDelayPerState[i][k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil( + mode_lib->vba.HActive[k] + / mode_lib->vba.slices, + 1.0), + mode_lib->vba.slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + locals->DSCDelayPerState[i][k] = + 2.0 * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil(mode_lib->vba.HActive[k] / mode_lib->vba.slices, 1.0), + mode_lib->vba.slices / 2, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay(mode_lib->vba.OutputFormat[k])); + } + locals->DSCDelayPerState[i][k] = + locals->DSCDelayPerState[i][k] * mode_lib->vba.PixelClock[k] / mode_lib->vba.PixelClockBackEnd[k]; + } else { + locals->DSCDelayPerState[i][k] = 0.0; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + for (j = 0; j <= mode_lib->vba.NumberOfActivePlanes - 1; j++) { + if (mode_lib->vba.BlendingAndTiming[k] == m && locals->RequiresDSC[i][m] == true) + locals->DSCDelayPerState[i][k] = locals->DSCDelayPerState[i][m]; + } + } + } + } + + //Prefetch Check + for (i = 0; i <= mode_lib->vba.soc.num_states; ++i) { + for (j = 0; j <= 1; ++j) { + locals->TotalNumberOfDCCActiveDPP[i][j] = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.DCCEnable[k] == true) + locals->TotalNumberOfDCCActiveDPP[i][j] = locals->TotalNumberOfDCCActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + } + } + + mode_lib->vba.UrgentLatency = dml_max3( + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.UrgentLatencyPixelMixedWithVMData, + mode_lib->vba.UrgentLatencyVMDataOnly); + mode_lib->vba.PrefetchERROR = CalculateMinAndMaxPrefetchMode( + mode_lib->vba.AllowDRAMSelfRefreshOrDRAMClockChangeInVblank, + &mode_lib->vba.MinPrefetchMode, + &mode_lib->vba.MaxPrefetchMode); + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->RequiredDPPCLKThisState[k] = locals->RequiredDPPCLK[i][j][k]; + locals->NoOfDPPThisState[k] = locals->NoOfDPP[i][j][k]; + if (locals->ODMCombineEnablePerState[i][k] == true) { + locals->SwathWidthYThisState[k] = + dml_min(locals->SwathWidthYSingleDPP[k], dml_round(mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k])); + } else { + locals->SwathWidthYThisState[k] = locals->SwathWidthYSingleDPP[k] / locals->NoOfDPP[i][j][k]; + } + mode_lib->vba.SwathWidthGranularityY = 256.0 + / dml_ceil(locals->BytePerPixelInDETY[k], 1.0) + / locals->MaxSwathHeightY[k]; + mode_lib->vba.RoundedUpMaxSwathSizeBytesY = + (dml_ceil(locals->SwathWidthYThisState[k] - 1.0, mode_lib->vba.SwathWidthGranularityY) + + mode_lib->vba.SwathWidthGranularityY) * locals->BytePerPixelInDETY[k] * locals->MaxSwathHeightY[k]; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + mode_lib->vba.RoundedUpMaxSwathSizeBytesY = dml_ceil( + mode_lib->vba.RoundedUpMaxSwathSizeBytesY, + 256.0) + 256; + } + if (locals->MaxSwathHeightC[k] > 0.0) { + mode_lib->vba.SwathWidthGranularityC = 256.0 / dml_ceil(locals->BytePerPixelInDETC[k], 2.0) / locals->MaxSwathHeightC[k]; + mode_lib->vba.RoundedUpMaxSwathSizeBytesC = (dml_ceil(locals->SwathWidthYThisState[k] / 2.0 - 1.0, mode_lib->vba.SwathWidthGranularityC) + + mode_lib->vba.SwathWidthGranularityC) * locals->BytePerPixelInDETC[k] * locals->MaxSwathHeightC[k]; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + mode_lib->vba.RoundedUpMaxSwathSizeBytesC = dml_ceil(mode_lib->vba.RoundedUpMaxSwathSizeBytesC, 256.0) + 256; + } + } else { + mode_lib->vba.RoundedUpMaxSwathSizeBytesC = 0.0; + } + if (mode_lib->vba.RoundedUpMaxSwathSizeBytesY + mode_lib->vba.RoundedUpMaxSwathSizeBytesC + <= mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0) { + locals->SwathHeightYThisState[k] = locals->MaxSwathHeightY[k]; + locals->SwathHeightCThisState[k] = locals->MaxSwathHeightC[k]; + } else { + locals->SwathHeightYThisState[k] = + locals->MinSwathHeightY[k]; + locals->SwathHeightCThisState[k] = + locals->MinSwathHeightC[k]; + } + } + + CalculateDCFCLKDeepSleep( + mode_lib, + mode_lib->vba.NumberOfActivePlanes, + locals->BytePerPixelInDETY, + locals->BytePerPixelInDETC, + mode_lib->vba.VRatio, + locals->SwathWidthYThisState, + locals->NoOfDPPThisState, + mode_lib->vba.HRatio, + mode_lib->vba.PixelClock, + locals->PSCL_FACTOR, + locals->PSCL_FACTOR_CHROMA, + locals->RequiredDPPCLKThisState, + &mode_lib->vba.ProjectedDCFCLKDeepSleep); + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8)) { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->Read256BlockHeightC[k], + locals->Read256BlockWidthC[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETC[k], 2.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2.0, + mode_lib->vba.ViewportHeight[k] / 2.0, + locals->SwathWidthYThisState[k] / 2.0, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsChroma, + mode_lib->vba.PitchC[k], + 0.0, + &locals->MacroTileWidthC[k], + &mode_lib->vba.MetaRowBytesC, + &mode_lib->vba.DPTEBytesPerRowC, + &locals->PTEBufferSizeNotExceededC[i][j][k], + locals->dpte_row_width_chroma_ub, + &locals->dpte_row_height_chroma[k], + &locals->meta_req_width_chroma[k], + &locals->meta_req_height_chroma[k], + &locals->meta_row_width_chroma[k], + &locals->meta_row_height_chroma[k], + &locals->vm_group_bytes_chroma, + &locals->dpte_group_bytes_chroma, + locals->PixelPTEReqWidthC, + locals->PixelPTEReqHeightC, + locals->PTERequestSizeC, + locals->dpde0_bytes_per_frame_ub_c, + locals->meta_pte_bytes_per_frame_ub_c); + locals->PrefetchLinesC[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k]/2, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + locals->SwathHeightCThisState[k], + mode_lib->vba.ViewportYStartC[k], + &locals->PrefillC[k], + &locals->MaxNumSwC[k]); + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma; + } else { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = 0.0; + mode_lib->vba.MetaRowBytesC = 0.0; + mode_lib->vba.DPTEBytesPerRowC = 0.0; + locals->PrefetchLinesC[k] = 0.0; + locals->PTEBufferSizeNotExceededC[i][j][k] = true; + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma + mode_lib->vba.PTEBufferSizeInRequestsChroma; + } + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->Read256BlockHeightY[k], + locals->Read256BlockWidthY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + locals->SwathWidthYThisState[k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + locals->PTEBufferSizeInRequestsForLuma, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &locals->MacroTileWidthY[k], + &mode_lib->vba.MetaRowBytesY, + &mode_lib->vba.DPTEBytesPerRowY, + &locals->PTEBufferSizeNotExceededY[i][j][k], + locals->dpte_row_width_luma_ub, + &locals->dpte_row_height[k], + &locals->meta_req_width[k], + &locals->meta_req_height[k], + &locals->meta_row_width[k], + &locals->meta_row_height[k], + &locals->vm_group_bytes[k], + &locals->dpte_group_bytes[k], + locals->PixelPTEReqWidthY, + locals->PixelPTEReqHeightY, + locals->PTERequestSizeY, + locals->dpde0_bytes_per_frame_ub_l, + locals->meta_pte_bytes_per_frame_ub_l); + locals->PrefetchLinesY[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + locals->SwathHeightYThisState[k], + mode_lib->vba.ViewportYStartY[k], + &locals->PrefillY[k], + &locals->MaxNumSwY[k]); + locals->PDEAndMetaPTEBytesPerFrame[k] = + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC; + locals->MetaRowBytes[k] = mode_lib->vba.MetaRowBytesY + mode_lib->vba.MetaRowBytesC; + locals->DPTEBytesPerRow[k] = mode_lib->vba.DPTEBytesPerRowY + mode_lib->vba.DPTEBytesPerRowC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / + mode_lib->vba.PixelClock[k], + mode_lib->vba.MetaRowBytesY, + mode_lib->vba.MetaRowBytesC, + locals->meta_row_height[k], + locals->meta_row_height_chroma[k], + mode_lib->vba.DPTEBytesPerRowY, + mode_lib->vba.DPTEBytesPerRowC, + locals->dpte_row_height[k], + locals->dpte_row_height_chroma[k], + &locals->meta_row_bw[k], + &locals->dpte_row_bw[k]); + } + mode_lib->vba.ExtraLatency = CalculateExtraLatency( + locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i], + locals->TotalNumberOfActiveDPP[i][j], + mode_lib->vba.PixelChunkSizeInKByte, + locals->TotalNumberOfDCCActiveDPP[i][j], + mode_lib->vba.MetaChunkSize, + locals->ReturnBWPerState[i], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.NumberOfActivePlanes, + locals->NoOfDPPThisState, + locals->dpte_group_bytes, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels); + + mode_lib->vba.TimeCalc = 24.0 / mode_lib->vba.ProjectedDCFCLKDeepSleep; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WritebackDelay[i][k] = mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) / locals->RequiredDISPCLK[i][j]; + } else { + locals->WritebackDelay[i][k] = 0.0; + } + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[m] == k + && mode_lib->vba.WritebackEnable[m] + == true) { + locals->WritebackDelay[i][k] = dml_max(locals->WritebackDelay[i][k], + mode_lib->vba.WritebackLatency + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[m], + mode_lib->vba.WritebackHRatio[m], + mode_lib->vba.WritebackVRatio[m], + mode_lib->vba.WritebackLumaHTaps[m], + mode_lib->vba.WritebackLumaVTaps[m], + mode_lib->vba.WritebackChromaHTaps[m], + mode_lib->vba.WritebackChromaVTaps[m], + mode_lib->vba.WritebackDestinationWidth[m]) / locals->RequiredDISPCLK[i][j]); + } + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[k] == m) { + locals->WritebackDelay[i][k] = locals->WritebackDelay[i][m]; + } + } + } + mode_lib->vba.MaxMaxVStartup = 0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->MaximumVStartup[k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] + - dml_max(1.0, dml_ceil(locals->WritebackDelay[i][k] / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]), 1.0)); + mode_lib->vba.MaxMaxVStartup = dml_max(mode_lib->vba.MaxMaxVStartup, locals->MaximumVStartup[k]); + } + + mode_lib->vba.NextPrefetchMode = mode_lib->vba.MinPrefetchMode; + mode_lib->vba.NextMaxVStartup = mode_lib->vba.MaxMaxVStartup; + do { + mode_lib->vba.PrefetchMode[i][j] = mode_lib->vba.NextPrefetchMode; + mode_lib->vba.MaxVStartup = mode_lib->vba.NextMaxVStartup; + + mode_lib->vba.TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[i][j], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + Pipe myPipe; + HostVM myHostVM; + + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthYThisState[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TimeCalc, + mode_lib->vba.TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0.0; + } + + myPipe.DPPCLK = locals->RequiredDPPCLK[i][j][k]; + myPipe.DISPCLK = locals->RequiredDISPCLK[i][j]; + myPipe.PixelClock = mode_lib->vba.PixelClock[k]; + myPipe.DCFCLKDeepSleep = mode_lib->vba.ProjectedDCFCLKDeepSleep; + myPipe.DPPPerPlane = locals->NoOfDPP[i][j][k]; + myPipe.ScalerEnabled = mode_lib->vba.ScalerEnabled[k]; + myPipe.SourceScan = mode_lib->vba.SourceScan[k]; + myPipe.BlockWidth256BytesY = locals->Read256BlockWidthY[k]; + myPipe.BlockHeight256BytesY = locals->Read256BlockHeightY[k]; + myPipe.BlockWidth256BytesC = locals->Read256BlockWidthC[k]; + myPipe.BlockHeight256BytesC = locals->Read256BlockHeightC[k]; + myPipe.InterlaceEnable = mode_lib->vba.Interlace[k]; + myPipe.NumberOfCursors = mode_lib->vba.NumberOfCursors[k]; + myPipe.VBlank = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k]; + myPipe.HTotal = mode_lib->vba.HTotal[k]; + + + myHostVM.Enable = mode_lib->vba.HostVMEnable; + myHostVM.MaxPageTableLevels = mode_lib->vba.HostVMMaxPageTableLevels; + myHostVM.CachedPageTableLevels = mode_lib->vba.HostVMCachedPageTableLevels; + + + mode_lib->vba.IsErrorResult[i][j][k] = CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + &myPipe, + locals->DSCDelayPerState[i][k], + mode_lib->vba.DPPCLKDelaySubtotal, + mode_lib->vba.DPPCLKDelaySCL, + mode_lib->vba.DPPCLKDelaySCLLBOnly, + mode_lib->vba.DPPCLKDelayCNVCFormater, + mode_lib->vba.DPPCLKDelayCNVCCursor, + mode_lib->vba.DISPCLKDelaySubtotal, + locals->SwathWidthYThisState[k] / mode_lib->vba.HRatio[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + dml_min(mode_lib->vba.MaxVStartup, locals->MaximumVStartup[k]), + locals->MaximumVStartup[k], + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + &myHostVM, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.ExtraLatency, + mode_lib->vba.TimeCalc, + locals->PDEAndMetaPTEBytesPerFrame[k], + locals->MetaRowBytes[k], + locals->DPTEBytesPerRow[k], + locals->PrefetchLinesY[k], + locals->SwathWidthYThisState[k], + locals->BytePerPixelInDETY[k], + locals->PrefillY[k], + locals->MaxNumSwY[k], + locals->PrefetchLinesC[k], + locals->BytePerPixelInDETC[k], + locals->PrefillC[k], + locals->MaxNumSwC[k], + locals->SwathHeightYThisState[k], + locals->SwathHeightCThisState[k], + mode_lib->vba.TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + &locals->dst_x_after_scaler, + &locals->dst_y_after_scaler, + &locals->LineTimesForPrefetch[k], + &locals->PrefetchBW[k], + &locals->LinesForMetaPTE[k], + &locals->LinesForMetaAndDPTERow[k], + &locals->VRatioPreY[i][j][k], + &locals->VRatioPreC[i][j][k], + &locals->RequiredPrefetchPixelDataBWLuma[i][j][k], + &locals->RequiredPrefetchPixelDataBWChroma[i][j][k], + &locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + &locals->Tno_bw[k], + &locals->prefetch_vmrow_bw[k], + locals->swath_width_luma_ub, + locals->swath_width_chroma_ub, + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + } + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = 0.0; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + uint m; + + locals->cursor_bw[k] = 0; + locals->cursor_bw_pre[k] = 0; + for (m = 0; m < mode_lib->vba.NumberOfCursors[k]; m++) { + locals->cursor_bw[k] = mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] + / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->cursor_bw_pre[k] = mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] + / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * locals->VRatioPreY[i][j][k]; + } + + CalculateUrgentBurstFactor( + mode_lib->vba.DETBufferSizeInKByte, + locals->SwathHeightYThisState[k], + locals->SwathHeightCThisState[k], + locals->SwathWidthYThisState[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.CursorBufferSize, + mode_lib->vba.CursorWidth[k][0] + mode_lib->vba.CursorWidth[k][1], + dml_max(mode_lib->vba.CursorBPP[k][0], mode_lib->vba.CursorBPP[k][1]), + mode_lib->vba.VRatio[k], + locals->VRatioPreY[i][j][k], + locals->VRatioPreC[i][j][k], + locals->BytePerPixelInDETY[k], + locals->BytePerPixelInDETC[k], + &locals->UrgentBurstFactorCursor[k], + &locals->UrgentBurstFactorCursorPre[k], + &locals->UrgentBurstFactorLuma[k], + &locals->UrgentBurstFactorLumaPre[k], + &locals->UrgentBurstFactorChroma[k], + &locals->UrgentBurstFactorChromaPre[k], + &locals->NotEnoughUrgentLatencyHiding, + &locals->NotEnoughUrgentLatencyHidingPre); + + if (mode_lib->vba.UseUrgentBurstBandwidth == false) { + locals->UrgentBurstFactorCursor[k] = 1; + locals->UrgentBurstFactorCursorPre[k] = 1; + locals->UrgentBurstFactorLuma[k] = 1; + locals->UrgentBurstFactorLumaPre[k] = 1; + locals->UrgentBurstFactorChroma[k] = 1; + locals->UrgentBurstFactorChromaPre[k] = 1; + } + + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = mode_lib->vba.MaximumReadBandwidthWithoutPrefetch + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k] + locals->ReadBandwidthLuma[k] + * locals->UrgentBurstFactorLuma[k] + locals->ReadBandwidthChroma[k] + * locals->UrgentBurstFactorChroma[k] + locals->meta_row_bw[k] + locals->dpte_row_bw[k]; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = mode_lib->vba.MaximumReadBandwidthWithPrefetch + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + locals->ReadBandwidthChroma[k] + * locals->UrgentBurstFactorChroma[k] + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k] + + locals->meta_row_bw[k] + locals->dpte_row_bw[k], + locals->RequiredPrefetchPixelDataBWLuma[i][j][k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixelDataBWChroma[i][j][k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + } + locals->BandwidthWithoutPrefetchSupported[i] = true; + if (mode_lib->vba.MaximumReadBandwidthWithoutPrefetch > locals->ReturnBWPerState[i] + || locals->NotEnoughUrgentLatencyHiding == 1) { + locals->BandwidthWithoutPrefetchSupported[i] = false; + } + + locals->PrefetchSupported[i][j] = true; + if (mode_lib->vba.MaximumReadBandwidthWithPrefetch > locals->ReturnBWPerState[i] + || locals->NotEnoughUrgentLatencyHiding == 1 + || locals->NotEnoughUrgentLatencyHidingPre == 1) { + locals->PrefetchSupported[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->LineTimesForPrefetch[k] < 2.0 + || locals->LinesForMetaPTE[k] >= 32.0 + || locals->LinesForMetaAndDPTERow[k] >= 16.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->PrefetchSupported[i][j] = false; + } + } + locals->VRatioInPrefetchSupported[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->VRatioPreY[i][j][k] > 4.0 + || locals->VRatioPreC[i][j][k] > 4.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->VRatioInPrefetchSupported[i][j] = false; + } + } + mode_lib->vba.AnyLinesForVMOrRowTooLarge = false; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->LinesForMetaAndDPTERow[k] >= 16 || locals->LinesForMetaPTE[k] >= 32) { + mode_lib->vba.AnyLinesForVMOrRowTooLarge = true; + } + } + + if (mode_lib->vba.MaxVStartup <= 13 || mode_lib->vba.AnyLinesForVMOrRowTooLarge == false) { + mode_lib->vba.NextMaxVStartup = mode_lib->vba.MaxMaxVStartup; + mode_lib->vba.NextPrefetchMode = mode_lib->vba.NextPrefetchMode + 1; + } else { + mode_lib->vba.NextMaxVStartup = mode_lib->vba.NextMaxVStartup - 1; + } + } while ((locals->PrefetchSupported[i][j] != true || locals->VRatioInPrefetchSupported[i][j] != true) + && (mode_lib->vba.NextMaxVStartup != mode_lib->vba.MaxMaxVStartup + || mode_lib->vba.NextPrefetchMode < mode_lib->vba.MaxPrefetchMode)); + + if (locals->PrefetchSupported[i][j] == true && locals->VRatioInPrefetchSupported[i][j] == true) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = locals->ReturnBWPerState[i]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = mode_lib->vba.BandwidthAvailableForImmediateFlip + - dml_max(locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthChroma[k] * locals->UrgentBurstFactorChroma[k] + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->RequiredPrefetchPixelDataBWLuma[i][j][k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixelDataBWChroma[i][j][k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + } + mode_lib->vba.TotImmediateFlipBytes = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.TotImmediateFlipBytes = mode_lib->vba.TotImmediateFlipBytes + + locals->PDEAndMetaPTEBytesPerFrame[k] + locals->MetaRowBytes[k] + locals->DPTEBytesPerRow[k]; + } + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.ExtraLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.GPUVMEnable, + locals->PDEAndMetaPTEBytesPerFrame[k], + locals->MetaRowBytes[k], + locals->DPTEBytesPerRow[k], + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + locals->Tno_bw[k], + mode_lib->vba.DCCEnable[k], + locals->dpte_row_height[k], + locals->meta_row_height[k], + locals->dpte_row_height_chroma[k], + locals->meta_row_height_chroma[k], + &locals->DestinationLinesToRequestVMInImmediateFlip[k], + &locals->DestinationLinesToRequestRowInImmediateFlip[k], + &locals->final_flip_bw[k], + &locals->ImmediateFlipSupportedForPipe[k]); + } + mode_lib->vba.total_dcn_read_bw_with_flip = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.total_dcn_read_bw_with_flip = mode_lib->vba.total_dcn_read_bw_with_flip + dml_max3( + locals->prefetch_vmrow_bw[k], + locals->final_flip_bw[k] + locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthChroma[k] * locals->UrgentBurstFactorChroma[k] + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->final_flip_bw[k] + locals->RequiredPrefetchPixelDataBWLuma[i][j][k] + * locals->UrgentBurstFactorLumaPre[k] + locals->RequiredPrefetchPixelDataBWChroma[i][j][k] + * locals->UrgentBurstFactorChromaPre[k] + locals->cursor_bw_pre[k] + * locals->UrgentBurstFactorCursorPre[k]); + } + locals->ImmediateFlipSupportedForState[i][j] = true; + if (mode_lib->vba.total_dcn_read_bw_with_flip + > locals->ReturnBWPerState[i]) { + locals->ImmediateFlipSupportedForState[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ImmediateFlipSupportedForPipe[k] == false) { + locals->ImmediateFlipSupportedForState[i][j] = false; + } + } + } else { + locals->ImmediateFlipSupportedForState[i][j] = false; + } + mode_lib->vba.UrgentOutOfOrderReturnPerChannel = dml_max3( + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelMixedWithVMData, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelVMDataOnly); + CalculateWatermarksAndDRAMSpeedChangeSupport( + mode_lib, + mode_lib->vba.PrefetchMode[i][j], + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.MaxLineBufferLines, + mode_lib->vba.LineBufferSize, + mode_lib->vba.DPPOutputBufferPixels, + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.WritebackInterfaceLumaBufferSize, + mode_lib->vba.WritebackInterfaceChromaBufferSize, + mode_lib->vba.DCFCLKPerState[i], + mode_lib->vba.UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels, + locals->ReturnBWPerState[i], + mode_lib->vba.GPUVMEnable, + locals->dpte_group_bytes, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.UrgentLatency, + mode_lib->vba.ExtraLatency, + mode_lib->vba.WritebackLatency, + mode_lib->vba.WritebackChunkSize, + mode_lib->vba.SOCCLKPerState[i], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.SRExitTime, + mode_lib->vba.SREnterPlusExitTime, + mode_lib->vba.ProjectedDCFCLKDeepSleep, + locals->NoOfDPPThisState, + mode_lib->vba.DCCEnable, + locals->RequiredDPPCLKThisState, + locals->SwathWidthYSingleDPP, + locals->SwathHeightYThisState, + locals->ReadBandwidthLuma, + locals->SwathHeightCThisState, + locals->ReadBandwidthChroma, + mode_lib->vba.LBBitPerPixel, + locals->SwathWidthYThisState, + mode_lib->vba.HRatio, + mode_lib->vba.vtaps, + mode_lib->vba.VTAPsChroma, + mode_lib->vba.VRatio, + mode_lib->vba.HTotal, + mode_lib->vba.PixelClock, + mode_lib->vba.BlendingAndTiming, + locals->BytePerPixelInDETY, + locals->BytePerPixelInDETC, + mode_lib->vba.WritebackEnable, + mode_lib->vba.WritebackPixelFormat, + mode_lib->vba.WritebackDestinationWidth, + mode_lib->vba.WritebackDestinationHeight, + mode_lib->vba.WritebackSourceHeight, + &locals->DRAMClockChangeSupport[i][j], + &mode_lib->vba.UrgentWatermark, + &mode_lib->vba.WritebackUrgentWatermark, + &mode_lib->vba.DRAMClockChangeWatermark, + &mode_lib->vba.WritebackDRAMClockChangeWatermark, + &mode_lib->vba.StutterExitWatermark, + &mode_lib->vba.StutterEnterPlusExitWatermark, + &mode_lib->vba.MinActiveDRAMClockChangeLatencySupported); + } + } + + /*Vertical Active BW support*/ + { + double MaxTotalVActiveRDBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + MaxTotalVActiveRDBandwidth = MaxTotalVActiveRDBandwidth + locals->ReadBandwidth[k]; + } + for (i = 0; i <= mode_lib->vba.soc.num_states; ++i) { + locals->MaxTotalVerticalActiveAvailableBandwidth[i] = dml_min( + locals->IdealSDPPortBandwidthPerState[i] * + mode_lib->vba.MaxAveragePercentOfIdealSDPPortBWDisplayCanUseInNormalSystemOperation + / 100.0, mode_lib->vba.DRAMSpeedPerState[i] * + mode_lib->vba.NumberOfChannels * + mode_lib->vba.DRAMChannelWidth * + mode_lib->vba.MaxAveragePercentOfIdealDRAMBWDisplayCanUseInNormalSystemOperation + / 100.0); + + if (MaxTotalVActiveRDBandwidth <= locals->MaxTotalVerticalActiveAvailableBandwidth[i]) { + locals->TotalVerticalActiveBandwidthSupport[i] = true; + } else { + locals->TotalVerticalActiveBandwidthSupport[i] = false; + } + } + } + + /*PTE Buffer Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->PTEBufferSizeNotExceeded[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->PTEBufferSizeNotExceededY[i][j][k] == false + || locals->PTEBufferSizeNotExceededC[i][j][k] == false) { + locals->PTEBufferSizeNotExceeded[i][j] = false; + } + } + } + } + /*Cursor Support Check*/ + + mode_lib->vba.CursorSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.CursorWidth[k][0] > 0.0) { + for (m = 0; m < mode_lib->vba.NumberOfCursors[k]; m++) { + if (mode_lib->vba.CursorBPP[k][m] == 64 && mode_lib->vba.Cursor64BppSupport == false) { + mode_lib->vba.CursorSupport = false; + } + } + } + } + /*Valid Pitch Check*/ + + mode_lib->vba.PitchSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->AlignedYPitch[k] = dml_ceil( + dml_max(mode_lib->vba.PitchY[k], mode_lib->vba.ViewportWidth[k]), + locals->MacroTileWidthY[k]); + if (locals->AlignedYPitch[k] > mode_lib->vba.PitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.DCCEnable[k] == true) { + locals->AlignedDCCMetaPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.DCCMetaPitchY[k], + mode_lib->vba.ViewportWidth[k]), + 64.0 * locals->Read256BlockWidthY[k]); + } else { + locals->AlignedDCCMetaPitch[k] = mode_lib->vba.DCCMetaPitchY[k]; + } + if (locals->AlignedDCCMetaPitch[k] > mode_lib->vba.DCCMetaPitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) { + locals->AlignedCPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.PitchC[k], + mode_lib->vba.ViewportWidth[k] / 2.0), + locals->MacroTileWidthC[k]); + } else { + locals->AlignedCPitch[k] = mode_lib->vba.PitchC[k]; + } + if (locals->AlignedCPitch[k] > mode_lib->vba.PitchC[k]) { + mode_lib->vba.PitchSupport = false; + } + } + /*Mode Support, Voltage State and SOC Configuration*/ + + for (i = mode_lib->vba.soc.num_states; i >= 0; i--) { + for (j = 0; j < 2; j++) { + enum dm_validation_status status = DML_VALIDATION_OK; + + if (mode_lib->vba.ScaleRatioAndTapsSupport != true) { + status = DML_FAIL_SCALE_RATIO_TAP; + } else if (mode_lib->vba.SourceFormatPixelAndScanSupport != true) { + status = DML_FAIL_SOURCE_PIXEL_FORMAT; + } else if (locals->ViewportSizeSupport[i] != true) { + status = DML_FAIL_VIEWPORT_SIZE; + } else if (locals->DIOSupport[i] != true) { + status = DML_FAIL_DIO_SUPPORT; + } else if (locals->NotEnoughDSCUnits[i] != false) { + status = DML_FAIL_NOT_ENOUGH_DSC; + } else if (locals->DSCCLKRequiredMoreThanSupported[i] != false) { + status = DML_FAIL_DSC_CLK_REQUIRED; + } else if (locals->ROBSupport[i] != true) { + status = DML_FAIL_REORDERING_BUFFER; + } else if (locals->DISPCLK_DPPCLK_Support[i][j] != true) { + status = DML_FAIL_DISPCLK_DPPCLK; + } else if (locals->TotalAvailablePipesSupport[i][j] != true) { + status = DML_FAIL_TOTAL_AVAILABLE_PIPES; + } else if (mode_lib->vba.NumberOfOTGSupport != true) { + status = DML_FAIL_NUM_OTG; + } else if (mode_lib->vba.WritebackModeSupport != true) { + status = DML_FAIL_WRITEBACK_MODE; + } else if (mode_lib->vba.WritebackLatencySupport != true) { + status = DML_FAIL_WRITEBACK_LATENCY; + } else if (mode_lib->vba.WritebackScaleRatioAndTapsSupport != true) { + status = DML_FAIL_WRITEBACK_SCALE_RATIO_TAP; + } else if (mode_lib->vba.CursorSupport != true) { + status = DML_FAIL_CURSOR_SUPPORT; + } else if (mode_lib->vba.PitchSupport != true) { + status = DML_FAIL_PITCH_SUPPORT; + } else if (locals->TotalVerticalActiveBandwidthSupport[i] != true) { + status = DML_FAIL_TOTAL_V_ACTIVE_BW; + } else if (locals->PTEBufferSizeNotExceeded[i][j] != true) { + status = DML_FAIL_PTE_BUFFER_SIZE; + } else if (mode_lib->vba.NonsupportedDSCInputBPC != false) { + status = DML_FAIL_DSC_INPUT_BPC; + } else if ((mode_lib->vba.HostVMEnable != false + && locals->ImmediateFlipSupportedForState[i][j] != true)) { + status = DML_FAIL_HOST_VM_IMMEDIATE_FLIP; + } else if (locals->PrefetchSupported[i][j] != true) { + status = DML_FAIL_PREFETCH_SUPPORT; + } else if (locals->VRatioInPrefetchSupported[i][j] != true) { + status = DML_FAIL_V_RATIO_PREFETCH; + } + + if (status == DML_VALIDATION_OK) { + locals->ModeSupport[i][j] = true; + } else { + locals->ModeSupport[i][j] = false; + } + locals->ValidationStatus[i] = status; + } + } + { + unsigned int MaximumMPCCombine = 0; + mode_lib->vba.VoltageLevel = mode_lib->vba.soc.num_states + 1; + for (i = mode_lib->vba.VoltageOverrideLevel; i <= mode_lib->vba.soc.num_states; i++) { + if (locals->ModeSupport[i][0] == true || locals->ModeSupport[i][1] == true) { + mode_lib->vba.VoltageLevel = i; + if (locals->ModeSupport[i][1] == true && (locals->ModeSupport[i][0] == false + || mode_lib->vba.WhenToDoMPCCombine == dm_mpc_always_when_possible + || (mode_lib->vba.WhenToDoMPCCombine == dm_mpc_reduce_voltage_and_clocks + && ((locals->DRAMClockChangeSupport[i][1] == dm_dram_clock_change_vactive + && locals->DRAMClockChangeSupport[i][0] != dm_dram_clock_change_vactive) + || (locals->DRAMClockChangeSupport[i][1] == dm_dram_clock_change_vblank + && locals->DRAMClockChangeSupport[i][0] == dm_dram_clock_change_unsupported))))) { + MaximumMPCCombine = 1; + } else { + MaximumMPCCombine = 0; + } + break; + } + } + mode_lib->vba.ImmediateFlipSupport = + locals->ImmediateFlipSupportedForState[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.DPPPerPlane[k] = locals->NoOfDPP[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + locals->DPPCLK[k] = locals->RequiredDPPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + } + mode_lib->vba.DISPCLK = locals->RequiredDISPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + mode_lib->vba.maxMpcComb = MaximumMPCCombine; + } + mode_lib->vba.DCFCLK = mode_lib->vba.DCFCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.DRAMSpeed = mode_lib->vba.DRAMSpeedPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.FabricClock = mode_lib->vba.FabricClockPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.SOCCLK = mode_lib->vba.SOCCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.ReturnBW = locals->ReturnBWPerState[mode_lib->vba.VoltageLevel]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.ODMCombineEnabled[k] = + locals->ODMCombineEnablePerState[mode_lib->vba.VoltageLevel][k]; + } else { + mode_lib->vba.ODMCombineEnabled[k] = 0; + } + mode_lib->vba.DSCEnabled[k] = + locals->RequiresDSC[mode_lib->vba.VoltageLevel][k]; + mode_lib->vba.OutputBpp[k] = + locals->OutputBppPerState[mode_lib->vba.VoltageLevel][k]; + } +} + +static void CalculateWatermarksAndDRAMSpeedChangeSupport( + struct display_mode_lib *mode_lib, + unsigned int PrefetchMode, + unsigned int NumberOfActivePlanes, + unsigned int MaxLineBufferLines, + unsigned int LineBufferSize, + unsigned int DPPOutputBufferPixels, + double DETBufferSizeInKByte, + unsigned int WritebackInterfaceLumaBufferSize, + unsigned int WritebackInterfaceChromaBufferSize, + double DCFCLK, + double UrgentOutOfOrderReturn, + double ReturnBW, + bool GPUVMEnable, + long dpte_group_bytes[], + unsigned int MetaChunkSize, + double UrgentLatency, + double ExtraLatency, + double WritebackLatency, + double WritebackChunkSize, + double SOCCLK, + double DRAMClockChangeLatency, + double SRExitTime, + double SREnterPlusExitTime, + double DCFCLKDeepSleep, + int DPPPerPlane[], + bool DCCEnable[], + double DPPCLK[], + unsigned int SwathWidthSingleDPPY[], + unsigned int SwathHeightY[], + double ReadBandwidthPlaneLuma[], + unsigned int SwathHeightC[], + double ReadBandwidthPlaneChroma[], + unsigned int LBBitPerPixel[], + unsigned int SwathWidthY[], + double HRatio[], + unsigned int vtaps[], + unsigned int VTAPsChroma[], + double VRatio[], + unsigned int HTotal[], + double PixelClock[], + unsigned int BlendingAndTiming[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + bool WritebackEnable[], + enum source_format_class WritebackPixelFormat[], + double WritebackDestinationWidth[], + double WritebackDestinationHeight[], + double WritebackSourceHeight[], + enum clock_change_support *DRAMClockChangeSupport, + double *UrgentWatermark, + double *WritebackUrgentWatermark, + double *DRAMClockChangeWatermark, + double *WritebackDRAMClockChangeWatermark, + double *StutterExitWatermark, + double *StutterEnterPlusExitWatermark, + double *MinActiveDRAMClockChangeLatencySupported) +{ + double EffectiveLBLatencyHidingY; + double EffectiveLBLatencyHidingC; + double DPPOutputBufferLinesY; + double DPPOutputBufferLinesC; + double DETBufferSizeY; + double DETBufferSizeC; + double LinesInDETY[DC__NUM_DPP__MAX]; + double LinesInDETC; + unsigned int LinesInDETYRoundedDownToSwath[DC__NUM_DPP__MAX]; + unsigned int LinesInDETCRoundedDownToSwath; + double FullDETBufferingTimeY[DC__NUM_DPP__MAX]; + double FullDETBufferingTimeC; + double ActiveDRAMClockChangeLatencyMarginY; + double ActiveDRAMClockChangeLatencyMarginC; + double WritebackDRAMClockChangeLatencyMargin; + double PlaneWithMinActiveDRAMClockChangeMargin; + double SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank; + double FullDETBufferingTimeYStutterCriticalPlane = 0; + double TimeToFinishSwathTransferStutterCriticalPlane = 0; + uint k, j; + + mode_lib->vba.TotalActiveDPP = 0; + mode_lib->vba.TotalDCCActiveDPP = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalActiveDPP = mode_lib->vba.TotalActiveDPP + DPPPerPlane[k]; + if (DCCEnable[k] == true) { + mode_lib->vba.TotalDCCActiveDPP = mode_lib->vba.TotalDCCActiveDPP + DPPPerPlane[k]; + } + } + + mode_lib->vba.TotalDataReadBandwidth = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalDataReadBandwidth = mode_lib->vba.TotalDataReadBandwidth + + ReadBandwidthPlaneLuma[k] + ReadBandwidthPlaneChroma[k]; + } + + *UrgentWatermark = UrgentLatency + ExtraLatency; + + *DRAMClockChangeWatermark = DRAMClockChangeLatency + *UrgentWatermark; + + mode_lib->vba.TotalActiveWriteback = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (WritebackEnable[k] == true) { + mode_lib->vba.TotalActiveWriteback = mode_lib->vba.TotalActiveWriteback + 1; + } + } + + if (mode_lib->vba.TotalActiveWriteback <= 1) { + *WritebackUrgentWatermark = WritebackLatency; + } else { + *WritebackUrgentWatermark = WritebackLatency + + WritebackChunkSize * 1024.0 / 32.0 / SOCCLK; + } + + if (mode_lib->vba.TotalActiveWriteback <= 1) { + *WritebackDRAMClockChangeWatermark = DRAMClockChangeLatency + WritebackLatency; + } else { + *WritebackDRAMClockChangeWatermark = DRAMClockChangeLatency + WritebackLatency + + WritebackChunkSize * 1024.0 / 32.0 / SOCCLK; + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + + mode_lib->vba.LBLatencyHidingSourceLinesY = dml_min((double) MaxLineBufferLines, + dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthY[k] / dml_max(HRatio[k], 1.0)), 1)) + - (vtaps[k] - 1); + + mode_lib->vba.LBLatencyHidingSourceLinesC = dml_min((double) MaxLineBufferLines, + dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthY[k] / 2 / dml_max(HRatio[k] / 2, 1.0)), 1)) + - (VTAPsChroma[k] - 1); + + EffectiveLBLatencyHidingY = mode_lib->vba.LBLatencyHidingSourceLinesY / VRatio[k] + * (HTotal[k] / PixelClock[k]); + + EffectiveLBLatencyHidingC = mode_lib->vba.LBLatencyHidingSourceLinesC + / (VRatio[k] / 2) * (HTotal[k] / PixelClock[k]); + + if (SwathWidthY[k] > 2 * DPPOutputBufferPixels) { + DPPOutputBufferLinesY = (double) DPPOutputBufferPixels / SwathWidthY[k]; + } else if (SwathWidthY[k] > DPPOutputBufferPixels) { + DPPOutputBufferLinesY = 0.5; + } else { + DPPOutputBufferLinesY = 1; + } + + if (SwathWidthY[k] / 2.0 > 2 * DPPOutputBufferPixels) { + DPPOutputBufferLinesC = (double) DPPOutputBufferPixels + / (SwathWidthY[k] / 2.0); + } else if (SwathWidthY[k] / 2.0 > DPPOutputBufferPixels) { + DPPOutputBufferLinesC = 0.5; + } else { + DPPOutputBufferLinesC = 1; + } + + CalculateDETBufferSize( + DETBufferSizeInKByte, + SwathHeightY[k], + SwathHeightC[k], + &DETBufferSizeY, + &DETBufferSizeC); + + LinesInDETY[k] = DETBufferSizeY / BytePerPixelDETY[k] / SwathWidthY[k]; + LinesInDETYRoundedDownToSwath[k] = dml_floor(LinesInDETY[k], SwathHeightY[k]); + FullDETBufferingTimeY[k] = LinesInDETYRoundedDownToSwath[k] + * (HTotal[k] / PixelClock[k]) / VRatio[k]; + if (BytePerPixelDETC[k] > 0) { + LinesInDETC = DETBufferSizeC / BytePerPixelDETC[k] / (SwathWidthY[k] / 2.0); + LinesInDETCRoundedDownToSwath = dml_floor(LinesInDETC, SwathHeightC[k]); + FullDETBufferingTimeC = LinesInDETCRoundedDownToSwath + * (HTotal[k] / PixelClock[k]) / (VRatio[k] / 2); + } else { + LinesInDETC = 0; + FullDETBufferingTimeC = 999999; + } + + ActiveDRAMClockChangeLatencyMarginY = HTotal[k] / PixelClock[k] + * DPPOutputBufferLinesY + EffectiveLBLatencyHidingY + + FullDETBufferingTimeY[k] - *DRAMClockChangeWatermark; + + if (NumberOfActivePlanes > 1) { + ActiveDRAMClockChangeLatencyMarginY = ActiveDRAMClockChangeLatencyMarginY + - (1 - 1.0 / NumberOfActivePlanes) * SwathHeightY[k] * HTotal[k] / PixelClock[k] / VRatio[k]; + } + + if (BytePerPixelDETC[k] > 0) { + ActiveDRAMClockChangeLatencyMarginC = HTotal[k] / PixelClock[k] + * DPPOutputBufferLinesC + EffectiveLBLatencyHidingC + + FullDETBufferingTimeC - *DRAMClockChangeWatermark; + if (NumberOfActivePlanes > 1) { + ActiveDRAMClockChangeLatencyMarginC = ActiveDRAMClockChangeLatencyMarginC + - (1 - 1.0 / NumberOfActivePlanes) * SwathHeightC[k] * HTotal[k] / PixelClock[k] / (VRatio[k] / 2); + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + ActiveDRAMClockChangeLatencyMarginY, + ActiveDRAMClockChangeLatencyMarginC); + } else { + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = ActiveDRAMClockChangeLatencyMarginY; + } + + if (WritebackEnable[k] == true) { + if (WritebackPixelFormat[k] == dm_444_32) { + WritebackDRAMClockChangeLatencyMargin = (WritebackInterfaceLumaBufferSize + + WritebackInterfaceChromaBufferSize) / (WritebackDestinationWidth[k] + * WritebackDestinationHeight[k] / (WritebackSourceHeight[k] * HTotal[k] + / PixelClock[k]) * 4) - *WritebackDRAMClockChangeWatermark; + } else { + WritebackDRAMClockChangeLatencyMargin = dml_min( + WritebackInterfaceLumaBufferSize * 8.0 / 10, + 2 * WritebackInterfaceChromaBufferSize * 8.0 / 10) / (WritebackDestinationWidth[k] + * WritebackDestinationHeight[k] / (WritebackSourceHeight[k] * HTotal[k] / PixelClock[k])) + - *WritebackDRAMClockChangeWatermark; + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k], + WritebackDRAMClockChangeLatencyMargin); + } + } + + mode_lib->vba.MinActiveDRAMClockChangeMargin = 999999; + PlaneWithMinActiveDRAMClockChangeMargin = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] + < mode_lib->vba.MinActiveDRAMClockChangeMargin) { + mode_lib->vba.MinActiveDRAMClockChangeMargin = + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + if (BlendingAndTiming[k] == k) { + PlaneWithMinActiveDRAMClockChangeMargin = k; + } else { + for (j = 0; j < NumberOfActivePlanes; ++j) { + if (BlendingAndTiming[k] == j) { + PlaneWithMinActiveDRAMClockChangeMargin = j; + } + } + } + } + } + + *MinActiveDRAMClockChangeLatencySupported = mode_lib->vba.MinActiveDRAMClockChangeMargin + DRAMClockChangeLatency; + + SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank = 999999; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (!((k == PlaneWithMinActiveDRAMClockChangeMargin) && (BlendingAndTiming[k] == k)) + && !(BlendingAndTiming[k] == PlaneWithMinActiveDRAMClockChangeMargin) + && mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] + < SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank) { + SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank = + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + } + } + + mode_lib->vba.TotalNumberOfActiveOTG = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (BlendingAndTiming[k] == k) { + mode_lib->vba.TotalNumberOfActiveOTG = mode_lib->vba.TotalNumberOfActiveOTG + 1; + } + } + + if (mode_lib->vba.MinActiveDRAMClockChangeMargin > 0) { + *DRAMClockChangeSupport = dm_dram_clock_change_vactive; + } else if (((mode_lib->vba.SynchronizedVBlank == true + || mode_lib->vba.TotalNumberOfActiveOTG == 1 + || SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank > 0) + && PrefetchMode == 0)) { + *DRAMClockChangeSupport = dm_dram_clock_change_vblank; + } else { + *DRAMClockChangeSupport = dm_dram_clock_change_unsupported; + } + + FullDETBufferingTimeYStutterCriticalPlane = FullDETBufferingTimeY[0]; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (FullDETBufferingTimeY[k] <= FullDETBufferingTimeYStutterCriticalPlane) { + TimeToFinishSwathTransferStutterCriticalPlane = (SwathHeightY[k] + - (LinesInDETY[k] - LinesInDETYRoundedDownToSwath[k])) + * (HTotal[k] / PixelClock[k]) / VRatio[k]; + } + } + + *StutterExitWatermark = SRExitTime + mode_lib->vba.LastPixelOfLineExtraWatermark + + ExtraLatency + 10 / DCFCLKDeepSleep; + *StutterEnterPlusExitWatermark = dml_max( + SREnterPlusExitTime + mode_lib->vba.LastPixelOfLineExtraWatermark + + ExtraLatency + 10 / DCFCLKDeepSleep, + TimeToFinishSwathTransferStutterCriticalPlane); + +} + +static void CalculateDCFCLKDeepSleep( + struct display_mode_lib *mode_lib, + unsigned int NumberOfActivePlanes, + double BytePerPixelDETY[], + double BytePerPixelDETC[], + double VRatio[], + unsigned int SwathWidthY[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double *DCFCLKDeepSleep) +{ + uint k; + double DisplayPipeLineDeliveryTimeLuma; + double DisplayPipeLineDeliveryTimeChroma; + //double DCFCLKDeepSleepPerPlane[DC__NUM_DPP__MAX]; + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (VRatio[k] <= 1) { + DisplayPipeLineDeliveryTimeLuma = SwathWidthY[k] * DPPPerPlane[k] + / HRatio[k] / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeLuma = SwathWidthY[k] / PSCL_THROUGHPUT[k] + / DPPCLK[k]; + } + if (BytePerPixelDETC[k] == 0) { + DisplayPipeLineDeliveryTimeChroma = 0; + } else { + if (VRatio[k] / 2 <= 1) { + DisplayPipeLineDeliveryTimeChroma = SwathWidthY[k] / 2.0 + * DPPPerPlane[k] / (HRatio[k] / 2) / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeChroma = SwathWidthY[k] / 2.0 + / PSCL_THROUGHPUT_CHROMA[k] / DPPCLK[k]; + } + } + + if (BytePerPixelDETC[k] > 0) { + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = dml_max( + 1.1 * SwathWidthY[k] * dml_ceil(BytePerPixelDETY[k], 1) + / 32.0 / DisplayPipeLineDeliveryTimeLuma, + 1.1 * SwathWidthY[k] / 2.0 + * dml_ceil(BytePerPixelDETC[k], 2) / 32.0 + / DisplayPipeLineDeliveryTimeChroma); + } else { + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = 1.1 * SwathWidthY[k] + * dml_ceil(BytePerPixelDETY[k], 1) / 64.0 + / DisplayPipeLineDeliveryTimeLuma; + } + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = dml_max( + mode_lib->vba.DCFCLKDeepSleepPerPlane[k], + PixelClock[k] / 16); + + } + + *DCFCLKDeepSleep = 8; + for (k = 0; k < NumberOfActivePlanes; ++k) { + *DCFCLKDeepSleep = dml_max( + *DCFCLKDeepSleep, + mode_lib->vba.DCFCLKDeepSleepPerPlane[k]); + } +} + +static void CalculateDETBufferSize( + double DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double *DETBufferSizeY, + double *DETBufferSizeC) +{ + if (SwathHeightC == 0) { + *DETBufferSizeY = DETBufferSizeInKByte * 1024; + *DETBufferSizeC = 0; + } else if (SwathHeightY <= SwathHeightC) { + *DETBufferSizeY = DETBufferSizeInKByte * 1024 / 2; + *DETBufferSizeC = DETBufferSizeInKByte * 1024 / 2; + } else { + *DETBufferSizeY = DETBufferSizeInKByte * 1024 * 2 / 3; + *DETBufferSizeC = DETBufferSizeInKByte * 1024 / 3; + } +} + +static void CalculateUrgentBurstFactor( + unsigned int DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + unsigned int SwathWidthY, + double LineTime, + double UrgentLatency, + double CursorBufferSize, + unsigned int CursorWidth, + unsigned int CursorBPP, + double VRatio, + double VRatioPreY, + double VRatioPreC, + double BytePerPixelInDETY, + double BytePerPixelInDETC, + double *UrgentBurstFactorCursor, + double *UrgentBurstFactorCursorPre, + double *UrgentBurstFactorLuma, + double *UrgentBurstFactorLumaPre, + double *UrgentBurstFactorChroma, + double *UrgentBurstFactorChromaPre, + unsigned int *NotEnoughUrgentLatencyHiding, + unsigned int *NotEnoughUrgentLatencyHidingPre) +{ + double LinesInDETLuma; + double LinesInDETChroma; + unsigned int LinesInCursorBuffer; + double CursorBufferSizeInTime; + double CursorBufferSizeInTimePre; + double DETBufferSizeInTimeLuma; + double DETBufferSizeInTimeLumaPre; + double DETBufferSizeInTimeChroma; + double DETBufferSizeInTimeChromaPre; + double DETBufferSizeY; + double DETBufferSizeC; + + *NotEnoughUrgentLatencyHiding = 0; + *NotEnoughUrgentLatencyHidingPre = 0; + + if (CursorWidth > 0) { + LinesInCursorBuffer = 1 << (unsigned int) dml_floor( + dml_log2(CursorBufferSize * 1024.0 / (CursorWidth * CursorBPP / 8.0)), 1.0); + CursorBufferSizeInTime = LinesInCursorBuffer * LineTime / VRatio; + if (CursorBufferSizeInTime - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHiding = 1; + *UrgentBurstFactorCursor = 0; + } else { + *UrgentBurstFactorCursor = CursorBufferSizeInTime + / (CursorBufferSizeInTime - UrgentLatency); + } + if (VRatioPreY > 0) { + CursorBufferSizeInTimePre = LinesInCursorBuffer * LineTime / VRatioPreY; + if (CursorBufferSizeInTimePre - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHidingPre = 1; + *UrgentBurstFactorCursorPre = 0; + } else { + *UrgentBurstFactorCursorPre = CursorBufferSizeInTimePre + / (CursorBufferSizeInTimePre - UrgentLatency); + } + } else { + *UrgentBurstFactorCursorPre = 1; + } + } + + CalculateDETBufferSize( + DETBufferSizeInKByte, + SwathHeightY, + SwathHeightC, + &DETBufferSizeY, + &DETBufferSizeC); + + LinesInDETLuma = DETBufferSizeY / BytePerPixelInDETY / SwathWidthY; + DETBufferSizeInTimeLuma = dml_floor(LinesInDETLuma, SwathHeightY) * LineTime / VRatio; + if (DETBufferSizeInTimeLuma - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHiding = 1; + *UrgentBurstFactorLuma = 0; + } else { + *UrgentBurstFactorLuma = DETBufferSizeInTimeLuma + / (DETBufferSizeInTimeLuma - UrgentLatency); + } + if (VRatioPreY > 0) { + DETBufferSizeInTimeLumaPre = dml_floor(LinesInDETLuma, SwathHeightY) * LineTime + / VRatioPreY; + if (DETBufferSizeInTimeLumaPre - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHidingPre = 1; + *UrgentBurstFactorLumaPre = 0; + } else { + *UrgentBurstFactorLumaPre = DETBufferSizeInTimeLumaPre + / (DETBufferSizeInTimeLumaPre - UrgentLatency); + } + } else { + *UrgentBurstFactorLumaPre = 1; + } + + if (BytePerPixelInDETC > 0) { + LinesInDETChroma = DETBufferSizeC / BytePerPixelInDETC / (SwathWidthY / 2); + DETBufferSizeInTimeChroma = dml_floor(LinesInDETChroma, SwathHeightC) * LineTime + / (VRatio / 2); + if (DETBufferSizeInTimeChroma - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHiding = 1; + *UrgentBurstFactorChroma = 0; + } else { + *UrgentBurstFactorChroma = DETBufferSizeInTimeChroma + / (DETBufferSizeInTimeChroma - UrgentLatency); + } + if (VRatioPreC > 0) { + DETBufferSizeInTimeChromaPre = dml_floor(LinesInDETChroma, SwathHeightC) + * LineTime / VRatioPreC; + if (DETBufferSizeInTimeChromaPre - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHidingPre = 1; + *UrgentBurstFactorChromaPre = 0; + } else { + *UrgentBurstFactorChromaPre = DETBufferSizeInTimeChromaPre + / (DETBufferSizeInTimeChromaPre - UrgentLatency); + } + } else { + *UrgentBurstFactorChromaPre = 1; + } + } +} + +static void CalculatePixelDeliveryTimes( + unsigned int NumberOfActivePlanes, + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + unsigned int swath_width_luma_ub[], + unsigned int swath_width_chroma_ub[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int BlockWidth256BytesY[], + unsigned int BlockHeight256BytesY[], + unsigned int BlockWidth256BytesC[], + unsigned int BlockHeight256BytesC[], + double DisplayPipeLineDeliveryTimeLuma[], + double DisplayPipeLineDeliveryTimeChroma[], + double DisplayPipeLineDeliveryTimeLumaPrefetch[], + double DisplayPipeLineDeliveryTimeChromaPrefetch[], + double DisplayPipeRequestDeliveryTimeLuma[], + double DisplayPipeRequestDeliveryTimeChroma[], + double DisplayPipeRequestDeliveryTimeLumaPrefetch[], + double DisplayPipeRequestDeliveryTimeChromaPrefetch[]) +{ + double req_per_swath_ub; + uint k; + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (VRatio[k] <= 1) { + DisplayPipeLineDeliveryTimeLuma[k] = swath_width_luma_ub[k] * DPPPerPlane[k] + / HRatio[k] / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeLuma[k] = swath_width_luma_ub[k] + / PSCL_THROUGHPUT[k] / DPPCLK[k]; + } + + if (BytePerPixelDETC[k] == 0) { + DisplayPipeLineDeliveryTimeChroma[k] = 0; + } else { + if (VRatio[k] / 2 <= 1) { + DisplayPipeLineDeliveryTimeChroma[k] = swath_width_chroma_ub[k] + * DPPPerPlane[k] / (HRatio[k] / 2) / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeChroma[k] = swath_width_chroma_ub[k] + / PSCL_THROUGHPUT_CHROMA[k] / DPPCLK[k]; + } + } + + if (VRatioPrefetchY[k] <= 1) { + DisplayPipeLineDeliveryTimeLumaPrefetch[k] = swath_width_luma_ub[k] + * DPPPerPlane[k] / HRatio[k] / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeLumaPrefetch[k] = swath_width_luma_ub[k] + / PSCL_THROUGHPUT[k] / DPPCLK[k]; + } + + if (BytePerPixelDETC[k] == 0) { + DisplayPipeLineDeliveryTimeChromaPrefetch[k] = 0; + } else { + if (VRatioPrefetchC[k] <= 1) { + DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + swath_width_chroma_ub[k] * DPPPerPlane[k] + / (HRatio[k] / 2) / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + swath_width_chroma_ub[k] / PSCL_THROUGHPUT_CHROMA[k] / DPPCLK[k]; + } + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (SourceScan[k] == dm_horz) { + req_per_swath_ub = swath_width_luma_ub[k] / BlockWidth256BytesY[k]; + } else { + req_per_swath_ub = swath_width_luma_ub[k] / BlockHeight256BytesY[k]; + } + DisplayPipeRequestDeliveryTimeLuma[k] = DisplayPipeLineDeliveryTimeLuma[k] + / req_per_swath_ub; + DisplayPipeRequestDeliveryTimeLumaPrefetch[k] = + DisplayPipeLineDeliveryTimeLumaPrefetch[k] / req_per_swath_ub; + if (BytePerPixelDETC[k] == 0) { + DisplayPipeRequestDeliveryTimeChroma[k] = 0; + DisplayPipeRequestDeliveryTimeChromaPrefetch[k] = 0; + } else { + if (SourceScan[k] == dm_horz) { + req_per_swath_ub = swath_width_chroma_ub[k] + / BlockWidth256BytesC[k]; + } else { + req_per_swath_ub = swath_width_chroma_ub[k] + / BlockHeight256BytesC[k]; + } + DisplayPipeRequestDeliveryTimeChroma[k] = + DisplayPipeLineDeliveryTimeChroma[k] / req_per_swath_ub; + DisplayPipeRequestDeliveryTimeChromaPrefetch[k] = + DisplayPipeLineDeliveryTimeChromaPrefetch[k] / req_per_swath_ub; + } + } +} + +static void CalculateMetaAndPTETimes( + unsigned int NumberOfActivePlanes, + bool GPUVMEnable, + unsigned int MetaChunkSize, + unsigned int MinMetaChunkSizeBytes, + unsigned int GPUVMMaxPageTableLevels, + unsigned int HTotal[], + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + double DestinationLinesToRequestRowInVBlank[], + double DestinationLinesToRequestRowInImmediateFlip[], + double DestinationLinesToRequestVMInVBlank[], + double DestinationLinesToRequestVMInImmediateFlip[], + bool DCCEnable[], + double PixelClock[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int dpte_row_height[], + unsigned int dpte_row_height_chroma[], + unsigned int meta_row_width[], + unsigned int meta_row_height[], + unsigned int meta_req_width[], + unsigned int meta_req_height[], + long dpte_group_bytes[], + unsigned int PTERequestSizeY[], + unsigned int PTERequestSizeC[], + unsigned int PixelPTEReqWidthY[], + unsigned int PixelPTEReqHeightY[], + unsigned int PixelPTEReqWidthC[], + unsigned int PixelPTEReqHeightC[], + unsigned int dpte_row_width_luma_ub[], + unsigned int dpte_row_width_chroma_ub[], + unsigned int vm_group_bytes[], + unsigned int dpde0_bytes_per_frame_ub_l[], + unsigned int dpde0_bytes_per_frame_ub_c[], + unsigned int meta_pte_bytes_per_frame_ub_l[], + unsigned int meta_pte_bytes_per_frame_ub_c[], + double DST_Y_PER_PTE_ROW_NOM_L[], + double DST_Y_PER_PTE_ROW_NOM_C[], + double DST_Y_PER_META_ROW_NOM_L[], + double TimePerMetaChunkNominal[], + double TimePerMetaChunkVBlank[], + double TimePerMetaChunkFlip[], + double time_per_pte_group_nom_luma[], + double time_per_pte_group_vblank_luma[], + double time_per_pte_group_flip_luma[], + double time_per_pte_group_nom_chroma[], + double time_per_pte_group_vblank_chroma[], + double time_per_pte_group_flip_chroma[], + double TimePerVMGroupVBlank[], + double TimePerVMGroupFlip[], + double TimePerVMRequestVBlank[], + double TimePerVMRequestFlip[]) +{ + unsigned int meta_chunk_width; + unsigned int min_meta_chunk_width; + unsigned int meta_chunk_per_row_int; + unsigned int meta_row_remainder; + unsigned int meta_chunk_threshold; + unsigned int meta_chunks_per_row_ub; + unsigned int dpte_group_width_luma; + unsigned int dpte_group_width_chroma; + unsigned int dpte_groups_per_row_luma_ub; + unsigned int dpte_groups_per_row_chroma_ub; + unsigned int num_group_per_lower_vm_stage; + unsigned int num_req_per_lower_vm_stage; + uint k; + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (GPUVMEnable == true) { + DST_Y_PER_PTE_ROW_NOM_L[k] = dpte_row_height[k] / VRatio[k]; + if (BytePerPixelDETC[k] == 0) { + DST_Y_PER_PTE_ROW_NOM_C[k] = 0; + } else { + DST_Y_PER_PTE_ROW_NOM_C[k] = dpte_row_height_chroma[k] / (VRatio[k] / 2); + } + } else { + DST_Y_PER_PTE_ROW_NOM_L[k] = 0; + DST_Y_PER_PTE_ROW_NOM_C[k] = 0; + } + if (DCCEnable[k] == true) { + DST_Y_PER_META_ROW_NOM_L[k] = meta_row_height[k] / VRatio[k]; + } else { + DST_Y_PER_META_ROW_NOM_L[k] = 0; + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (DCCEnable[k] == true) { + meta_chunk_width = MetaChunkSize * 1024 * 256 + / dml_ceil(BytePerPixelDETY[k], 1) / meta_row_height[k]; + min_meta_chunk_width = MinMetaChunkSizeBytes * 256 + / dml_ceil(BytePerPixelDETY[k], 1) / meta_row_height[k]; + meta_chunk_per_row_int = meta_row_width[k] / meta_chunk_width; + meta_row_remainder = meta_row_width[k] % meta_chunk_width; + if (SourceScan[k] == dm_horz) { + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_width[k]; + } else { + meta_chunk_threshold = 2 * min_meta_chunk_width + - meta_req_height[k]; + } + if (meta_row_remainder <= meta_chunk_threshold) { + meta_chunks_per_row_ub = meta_chunk_per_row_int + 1; + } else { + meta_chunks_per_row_ub = meta_chunk_per_row_int + 2; + } + TimePerMetaChunkNominal[k] = meta_row_height[k] / VRatio[k] * HTotal[k] + / PixelClock[k] / meta_chunks_per_row_ub; + TimePerMetaChunkVBlank[k] = DestinationLinesToRequestRowInVBlank[k] + * HTotal[k] / PixelClock[k] / meta_chunks_per_row_ub; + TimePerMetaChunkFlip[k] = DestinationLinesToRequestRowInImmediateFlip[k] + * HTotal[k] / PixelClock[k] / meta_chunks_per_row_ub; + } else { + TimePerMetaChunkNominal[k] = 0; + TimePerMetaChunkVBlank[k] = 0; + TimePerMetaChunkFlip[k] = 0; + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (GPUVMEnable == true) { + if (SourceScan[k] == dm_horz) { + dpte_group_width_luma = dpte_group_bytes[k] / PTERequestSizeY[k] + * PixelPTEReqWidthY[k]; + } else { + dpte_group_width_luma = dpte_group_bytes[k] / PTERequestSizeY[k] + * PixelPTEReqHeightY[k]; + } + dpte_groups_per_row_luma_ub = dml_ceil( + dpte_row_width_luma_ub[k] / dpte_group_width_luma, + 1); + time_per_pte_group_nom_luma[k] = DST_Y_PER_PTE_ROW_NOM_L[k] * HTotal[k] + / PixelClock[k] / dpte_groups_per_row_luma_ub; + time_per_pte_group_vblank_luma[k] = DestinationLinesToRequestRowInVBlank[k] + * HTotal[k] / PixelClock[k] / dpte_groups_per_row_luma_ub; + time_per_pte_group_flip_luma[k] = + DestinationLinesToRequestRowInImmediateFlip[k] * HTotal[k] + / PixelClock[k] + / dpte_groups_per_row_luma_ub; + if (BytePerPixelDETC[k] == 0) { + time_per_pte_group_nom_chroma[k] = 0; + time_per_pte_group_vblank_chroma[k] = 0; + time_per_pte_group_flip_chroma[k] = 0; + } else { + if (SourceScan[k] == dm_horz) { + dpte_group_width_chroma = dpte_group_bytes[k] + / PTERequestSizeC[k] * PixelPTEReqWidthC[k]; + } else { + dpte_group_width_chroma = dpte_group_bytes[k] + / PTERequestSizeC[k] + * PixelPTEReqHeightC[k]; + } + dpte_groups_per_row_chroma_ub = dml_ceil( + dpte_row_width_chroma_ub[k] + / dpte_group_width_chroma, + 1); + time_per_pte_group_nom_chroma[k] = DST_Y_PER_PTE_ROW_NOM_C[k] + * HTotal[k] / PixelClock[k] + / dpte_groups_per_row_chroma_ub; + time_per_pte_group_vblank_chroma[k] = + DestinationLinesToRequestRowInVBlank[k] * HTotal[k] + / PixelClock[k] + / dpte_groups_per_row_chroma_ub; + time_per_pte_group_flip_chroma[k] = + DestinationLinesToRequestRowInImmediateFlip[k] + * HTotal[k] / PixelClock[k] + / dpte_groups_per_row_chroma_ub; + } + } else { + time_per_pte_group_nom_luma[k] = 0; + time_per_pte_group_vblank_luma[k] = 0; + time_per_pte_group_flip_luma[k] = 0; + time_per_pte_group_nom_chroma[k] = 0; + time_per_pte_group_vblank_chroma[k] = 0; + time_per_pte_group_flip_chroma[k] = 0; + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (GPUVMEnable == true && (DCCEnable[k] == true || GPUVMMaxPageTableLevels > 1)) { + if (DCCEnable[k] == false) { + if (BytePerPixelDETC[k] > 0) { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (dpde0_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1); + } else { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1); + } + } else { + if (GPUVMMaxPageTableLevels == 1) { + if (BytePerPixelDETC[k] > 0) { + num_group_per_lower_vm_stage = + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1); + } else { + num_group_per_lower_vm_stage = + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1); + } + } else { + if (BytePerPixelDETC[k] > 0) { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (dpde0_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1); + } else { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1); + } + } + } + + if (DCCEnable[k] == false) { + if (BytePerPixelDETC[k] > 0) { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] + / 64 + dpde0_bytes_per_frame_ub_c[k] / 64; + } else { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] + / 64; + } + } else { + if (GPUVMMaxPageTableLevels == 1) { + if (BytePerPixelDETC[k] > 0) { + num_req_per_lower_vm_stage = meta_pte_bytes_per_frame_ub_l[k] / 64 + + meta_pte_bytes_per_frame_ub_c[k] / 64; + } else { + num_req_per_lower_vm_stage = meta_pte_bytes_per_frame_ub_l[k] / 64; + } + } else { + if (BytePerPixelDETC[k] > 0) { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] / 64 + + dpde0_bytes_per_frame_ub_c[k] / 64 + + meta_pte_bytes_per_frame_ub_l[k] / 64 + + meta_pte_bytes_per_frame_ub_c[k] / 64; + } else { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] / 64 + + meta_pte_bytes_per_frame_ub_l[k] / 64; + } + } + } + + TimePerVMGroupVBlank[k] = DestinationLinesToRequestVMInVBlank[k] * HTotal[k] + / PixelClock[k] / num_group_per_lower_vm_stage; + TimePerVMGroupFlip[k] = DestinationLinesToRequestVMInImmediateFlip[k] + * HTotal[k] / PixelClock[k] / num_group_per_lower_vm_stage; + TimePerVMRequestVBlank[k] = DestinationLinesToRequestVMInVBlank[k] + * HTotal[k] / PixelClock[k] / num_req_per_lower_vm_stage; + TimePerVMRequestFlip[k] = DestinationLinesToRequestVMInImmediateFlip[k] + * HTotal[k] / PixelClock[k] / num_req_per_lower_vm_stage; + + if (GPUVMMaxPageTableLevels > 2) { + TimePerVMGroupVBlank[k] = TimePerVMGroupVBlank[k] / 2; + TimePerVMGroupFlip[k] = TimePerVMGroupFlip[k] / 2; + TimePerVMRequestVBlank[k] = TimePerVMRequestVBlank[k] / 2; + TimePerVMRequestFlip[k] = TimePerVMRequestFlip[k] / 2; + } + + } else { + TimePerVMGroupVBlank[k] = 0; + TimePerVMGroupFlip[k] = 0; + TimePerVMRequestVBlank[k] = 0; + TimePerVMRequestFlip[k] = 0; + } + } +} + +static double CalculateExtraLatency( + double UrgentRoundTripAndOutOfOrderLatency, + int TotalNumberOfActiveDPP, + int PixelChunkSizeInKByte, + int TotalNumberOfDCCActiveDPP, + int MetaChunkSize, + double ReturnBW, + bool GPUVMEnable, + bool HostVMEnable, + int NumberOfActivePlanes, + int NumberOfDPP[], + long dpte_group_bytes[], + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + int HostVMMaxPageTableLevels, + int HostVMCachedPageTableLevels) +{ + double CalculateExtraLatency; + double HostVMInefficiencyFactor; + int HostVMDynamicLevels; + + if (GPUVMEnable && HostVMEnable) { + HostVMInefficiencyFactor = + PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData + / PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly; + HostVMDynamicLevels = HostVMMaxPageTableLevels - HostVMCachedPageTableLevels; + } else { + HostVMInefficiencyFactor = 1; + HostVMDynamicLevels = 0; + } + + CalculateExtraLatency = UrgentRoundTripAndOutOfOrderLatency + + (TotalNumberOfActiveDPP * PixelChunkSizeInKByte + + TotalNumberOfDCCActiveDPP * MetaChunkSize) * 1024.0 + / ReturnBW; + + if (GPUVMEnable) { + int k; + + for (k = 0; k < NumberOfActivePlanes; k++) { + CalculateExtraLatency = CalculateExtraLatency + + NumberOfDPP[k] * dpte_group_bytes[k] + * (1 + 8 * HostVMDynamicLevels) + * HostVMInefficiencyFactor / ReturnBW; + } + } + return CalculateExtraLatency; +} + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.h b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.h new file mode 100644 index 000000000000..fb9548a2f894 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.h @@ -0,0 +1,32 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DML21_DISPLAY_MODE_VBA_H__ +#define __DML21_DISPLAY_MODE_VBA_H__ + +void dml21_recalculate(struct display_mode_lib *mode_lib); +void dml21_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib); + +#endif /* _DML21_DISPLAY_MODE_VBA_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c new file mode 100644 index 000000000000..a1f207cbb966 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c @@ -0,0 +1,1823 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifdef CONFIG_DRM_AMD_DC_DCN2_0 + +#include "../display_mode_lib.h" +#include "../display_mode_vba.h" +#include "../dml_inline_defs.h" +#include "display_rq_dlg_calc_21.h" + +/* + * NOTE: + * This file is gcc-parseable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ + +static void calculate_ttu_cursor( + struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp); + +static unsigned int get_bytes_per_element(enum source_format_class source_format, bool is_chroma) +{ + unsigned int ret_val = 0; + + if (source_format == dm_444_16) { + if (!is_chroma) + ret_val = 2; + } else if (source_format == dm_444_32) { + if (!is_chroma) + ret_val = 4; + } else if (source_format == dm_444_64) { + if (!is_chroma) + ret_val = 8; + } else if (source_format == dm_420_8) { + if (is_chroma) + ret_val = 2; + else + ret_val = 1; + } else if (source_format == dm_420_10) { + if (is_chroma) + ret_val = 4; + else + ret_val = 2; + } else if (source_format == dm_444_8) { + ret_val = 1; + } + return ret_val; +} + +static bool is_dual_plane(enum source_format_class source_format) +{ + bool ret_val = 0; + + if ((source_format == dm_420_8) || (source_format == dm_420_10)) + ret_val = 1; + + return ret_val; +} + +static double get_refcyc_per_delivery( + struct display_mode_lib *mode_lib, + double refclk_freq_in_mhz, + double pclk_freq_in_mhz, + bool odm_combine, + unsigned int recout_width, + unsigned int hactive, + double vratio, + double hscale_pixel_rate, + unsigned int delivery_width, + unsigned int req_per_swath_ub) +{ + double refcyc_per_delivery = 0.0; + + if (vratio <= 1.0) { + if (odm_combine) + refcyc_per_delivery = (double) refclk_freq_in_mhz + * dml_min((double) recout_width, (double) hactive / 2.0) + / pclk_freq_in_mhz / (double) req_per_swath_ub; + else + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) recout_width + / pclk_freq_in_mhz / (double) req_per_swath_ub; + } else { + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) delivery_width + / (double) hscale_pixel_rate / (double) req_per_swath_ub; + } + + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: recout_width = %d\n", __func__, recout_width); + dml_print("DML_DLG: %s: vratio = %3.2f\n", __func__, vratio); + dml_print("DML_DLG: %s: req_per_swath_ub = %d\n", __func__, req_per_swath_ub); + dml_print("DML_DLG: %s: refcyc_per_delivery= %3.2f\n", __func__, refcyc_per_delivery); + + return refcyc_per_delivery; + +} + +static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_size) +{ + if (tile_size == dm_256k_tile) + return (256 * 1024); + else if (tile_size == dm_64k_tile) + return (64 * 1024); + else + return (4 * 1024); +} + +static void extract_rq_sizing_regs( + struct display_mode_lib *mode_lib, + display_data_rq_regs_st *rq_regs, + const display_data_rq_sizing_params_st rq_sizing) +{ + dml_print("DML_DLG: %s: rq_sizing param\n", __func__); + print__data_rq_sizing_params_st(mode_lib, rq_sizing); + + rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10; + + if (rq_sizing.min_chunk_bytes == 0) + rq_regs->min_chunk_size = 0; + else + rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1; + + rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10; + if (rq_sizing.min_meta_chunk_bytes == 0) + rq_regs->min_meta_chunk_size = 0; + else + rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1; + + rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6; + rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6; +} + +static void extract_rq_regs( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_rq_params_st rq_param) +{ + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + unsigned int detile_buf_plane1_addr = 0; + + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l); + + rq_regs->rq_regs_l.pte_row_height_linear = dml_floor( + dml_log2(rq_param.dlg.rq_l.dpte_row_height), + 1) - 3; + + if (rq_param.yuv420) { + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c); + rq_regs->rq_regs_c.pte_row_height_linear = dml_floor( + dml_log2(rq_param.dlg.rq_c.dpte_row_height), + 1) - 3; + } + + rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height); + rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height); + + // FIXME: take the max between luma, chroma chunk size? + // okay for now, as we are setting chunk_bytes to 8kb anyways + if (rq_param.sizing.rq_l.chunk_bytes >= 32 * 1024) { //32kb + rq_regs->drq_expansion_mode = 0; + } else { + rq_regs->drq_expansion_mode = 2; + } + rq_regs->prq_expansion_mode = 1; + rq_regs->mrq_expansion_mode = 1; + rq_regs->crq_expansion_mode = 1; + + if (rq_param.yuv420) { + if ((double) rq_param.misc.rq_l.stored_swath_bytes + / (double) rq_param.misc.rq_c.stored_swath_bytes <= 1.5) { + detile_buf_plane1_addr = (detile_buf_size_in_bytes / 2.0 / 64.0); // half to chroma + } else { + detile_buf_plane1_addr = dml_round_to_multiple( + (unsigned int) ((2.0 * detile_buf_size_in_bytes) / 3.0), + 256, + 0) / 64.0; // 2/3 to chroma + } + } + rq_regs->plane1_base_address = detile_buf_plane1_addr; +} + +static void handle_det_buf_split( + struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_source_params_st pipe_src_param) +{ + unsigned int total_swath_bytes = 0; + unsigned int swath_bytes_l = 0; + unsigned int swath_bytes_c = 0; + unsigned int full_swath_bytes_packed_l = 0; + unsigned int full_swath_bytes_packed_c = 0; + bool req128_l = 0; + bool req128_c = 0; + bool surf_linear = (pipe_src_param.sw_mode == dm_sw_linear); + bool surf_vert = (pipe_src_param.source_scan == dm_vert); + unsigned int log2_swath_height_l = 0; + unsigned int log2_swath_height_c = 0; + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + + full_swath_bytes_packed_l = rq_param->misc.rq_l.full_swath_bytes; + full_swath_bytes_packed_c = rq_param->misc.rq_c.full_swath_bytes; + + if (rq_param->yuv420_10bpc) { + full_swath_bytes_packed_l = dml_round_to_multiple( + rq_param->misc.rq_l.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + full_swath_bytes_packed_c = dml_round_to_multiple( + rq_param->misc.rq_c.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + } + + if (rq_param->yuv420) { + total_swath_bytes = 2 * full_swath_bytes_packed_l + 2 * full_swath_bytes_packed_c; + + if (total_swath_bytes <= detile_buf_size_in_bytes) { //full 256b request + req128_l = 0; + req128_c = 0; + swath_bytes_l = full_swath_bytes_packed_l; + swath_bytes_c = full_swath_bytes_packed_c; + } else { //128b request (for luma only for yuv420 8bpc) + req128_l = 1; + req128_c = 0; + swath_bytes_l = full_swath_bytes_packed_l / 2; + swath_bytes_c = full_swath_bytes_packed_c; + } + // Note: assumption, the config that pass in will fit into + // the detiled buffer. + } else { + total_swath_bytes = 2 * full_swath_bytes_packed_l; + + if (total_swath_bytes <= detile_buf_size_in_bytes) + req128_l = 0; + else + req128_l = 1; + + swath_bytes_l = total_swath_bytes; + swath_bytes_c = 0; + } + rq_param->misc.rq_l.stored_swath_bytes = swath_bytes_l; + rq_param->misc.rq_c.stored_swath_bytes = swath_bytes_c; + + if (surf_linear) { + log2_swath_height_l = 0; + log2_swath_height_c = 0; + } else if (!surf_vert) { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_height) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_height) - req128_c; + } else { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_width) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_width) - req128_c; + } + rq_param->dlg.rq_l.swath_height = 1 << log2_swath_height_l; + rq_param->dlg.rq_c.swath_height = 1 << log2_swath_height_c; + + dml_print("DML_DLG: %s: req128_l = %0d\n", __func__, req128_l); + dml_print("DML_DLG: %s: req128_c = %0d\n", __func__, req128_c); + dml_print( + "DML_DLG: %s: full_swath_bytes_packed_l = %0d\n", + __func__, + full_swath_bytes_packed_l); + dml_print( + "DML_DLG: %s: full_swath_bytes_packed_c = %0d\n", + __func__, + full_swath_bytes_packed_c); +} + +static void get_meta_and_pte_attr( + struct display_mode_lib *mode_lib, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + display_data_rq_sizing_params_st *rq_sizing_param, + unsigned int vp_width, + unsigned int vp_height, + unsigned int data_pitch, + unsigned int meta_pitch, + unsigned int source_format, + unsigned int tiling, + unsigned int macro_tile_size, + unsigned int source_scan, + unsigned int hostvm_enable, + unsigned int is_chroma) +{ + bool surf_linear = (tiling == dm_sw_linear); + bool surf_vert = (source_scan == dm_vert); + + unsigned int bytes_per_element; + unsigned int bytes_per_element_y = get_bytes_per_element( + (enum source_format_class) (source_format), + false); + unsigned int bytes_per_element_c = get_bytes_per_element( + (enum source_format_class) (source_format), + true); + + unsigned int blk256_width = 0; + unsigned int blk256_height = 0; + + unsigned int blk256_width_y = 0; + unsigned int blk256_height_y = 0; + unsigned int blk256_width_c = 0; + unsigned int blk256_height_c = 0; + unsigned int log2_bytes_per_element; + unsigned int log2_blk256_width; + unsigned int log2_blk256_height; + unsigned int blk_bytes; + unsigned int log2_blk_bytes; + unsigned int log2_blk_height; + unsigned int log2_blk_width; + unsigned int log2_meta_req_bytes; + unsigned int log2_meta_req_height; + unsigned int log2_meta_req_width; + unsigned int meta_req_width; + unsigned int meta_req_height; + unsigned int log2_meta_row_height; + unsigned int meta_row_width_ub; + unsigned int log2_meta_chunk_bytes; + unsigned int log2_meta_chunk_height; + + //full sized meta chunk width in unit of data elements + unsigned int log2_meta_chunk_width; + unsigned int log2_min_meta_chunk_bytes; + unsigned int min_meta_chunk_width; + unsigned int meta_chunk_width; + unsigned int meta_chunk_per_row_int; + unsigned int meta_row_remainder; + unsigned int meta_chunk_threshold; + unsigned int meta_blk_bytes; + unsigned int meta_blk_height; + unsigned int meta_blk_width; + unsigned int meta_surface_bytes; + unsigned int vmpg_bytes; + unsigned int meta_pte_req_per_frame_ub; + unsigned int meta_pte_bytes_per_frame_ub; + const unsigned int log2_vmpg_bytes = dml_log2(mode_lib->soc.vmm_page_size_bytes); + const unsigned int dpte_buf_in_pte_reqs = + mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma + mode_lib->ip.dpte_buffer_size_in_pte_reqs_chroma; + const unsigned int pde_proc_buffer_size_64k_reqs = + mode_lib->ip.pde_proc_buffer_size_64k_reqs; + + unsigned int log2_vmpg_height = 0; + unsigned int log2_vmpg_width = 0; + unsigned int log2_dpte_req_height_ptes = 0; + unsigned int log2_dpte_req_height = 0; + unsigned int log2_dpte_req_width = 0; + unsigned int log2_dpte_row_height_linear = 0; + unsigned int log2_dpte_row_height = 0; + unsigned int log2_dpte_group_width = 0; + unsigned int dpte_row_width_ub = 0; + unsigned int dpte_req_height = 0; + unsigned int dpte_req_width = 0; + unsigned int dpte_group_width = 0; + unsigned int log2_dpte_group_bytes = 0; + unsigned int log2_dpte_group_length = 0; + unsigned int pde_buf_entries; + bool yuv420 = (source_format == dm_420_8 || source_format == dm_420_10); + + Calculate256BBlockSizes( + (enum source_format_class) (source_format), + (enum dm_swizzle_mode) (tiling), + bytes_per_element_y, + bytes_per_element_c, + &blk256_height_y, + &blk256_height_c, + &blk256_width_y, + &blk256_width_c); + + if (!is_chroma) { + blk256_width = blk256_width_y; + blk256_height = blk256_height_y; + bytes_per_element = bytes_per_element_y; + } else { + blk256_width = blk256_width_c; + blk256_height = blk256_height_c; + bytes_per_element = bytes_per_element_c; + } + + log2_bytes_per_element = dml_log2(bytes_per_element); + + dml_print("DML_DLG: %s: surf_linear = %d\n", __func__, surf_linear); + dml_print("DML_DLG: %s: surf_vert = %d\n", __func__, surf_vert); + dml_print("DML_DLG: %s: blk256_width = %d\n", __func__, blk256_width); + dml_print("DML_DLG: %s: blk256_height = %d\n", __func__, blk256_height); + + log2_blk256_width = dml_log2((double) blk256_width); + log2_blk256_height = dml_log2((double) blk256_height); + blk_bytes = surf_linear ? + 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); + log2_blk_bytes = dml_log2((double) blk_bytes); + log2_blk_height = 0; + log2_blk_width = 0; + + // remember log rule + // "+" in log is multiply + // "-" in log is divide + // "/2" is like square root + // blk is vertical biased + if (tiling != dm_sw_linear) + log2_blk_height = log2_blk256_height + + dml_ceil((double) (log2_blk_bytes - 8) / 2.0, 1); + else + log2_blk_height = 0; // blk height of 1 + + log2_blk_width = log2_blk_bytes - log2_bytes_per_element - log2_blk_height; + + if (!surf_vert) { + rq_dlg_param->swath_width_ub = dml_round_to_multiple(vp_width - 1, blk256_width, 1) + + blk256_width; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_width; + } else { + rq_dlg_param->swath_width_ub = dml_round_to_multiple( + vp_height - 1, + blk256_height, + 1) + blk256_height; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_height; + } + + if (!surf_vert) + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_height + * bytes_per_element; + else + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_width + * bytes_per_element; + + rq_misc_param->blk256_height = blk256_height; + rq_misc_param->blk256_width = blk256_width; + + // ------- + // meta + // ------- + log2_meta_req_bytes = 6; // meta request is 64b and is 8x8byte meta element + + // each 64b meta request for dcn is 8x8 meta elements and + // a meta element covers one 256b block of the the data surface. + log2_meta_req_height = log2_blk256_height + 3; // meta req is 8x8 byte, each byte represent 1 blk256 + log2_meta_req_width = log2_meta_req_bytes + 8 - log2_bytes_per_element + - log2_meta_req_height; + meta_req_width = 1 << log2_meta_req_width; + meta_req_height = 1 << log2_meta_req_height; + log2_meta_row_height = 0; + meta_row_width_ub = 0; + + // the dimensions of a meta row are meta_row_width x meta_row_height in elements. + // calculate upper bound of the meta_row_width + if (!surf_vert) { + log2_meta_row_height = log2_meta_req_height; + meta_row_width_ub = dml_round_to_multiple(vp_width - 1, meta_req_width, 1) + + meta_req_width; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_width; + } else { + log2_meta_row_height = log2_meta_req_width; + meta_row_width_ub = dml_round_to_multiple(vp_height - 1, meta_req_height, 1) + + meta_req_height; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_height; + } + rq_dlg_param->meta_bytes_per_row_ub = rq_dlg_param->meta_req_per_row_ub * 64; + + rq_dlg_param->meta_row_height = 1 << log2_meta_row_height; + + log2_meta_chunk_bytes = dml_log2(rq_sizing_param->meta_chunk_bytes); + log2_meta_chunk_height = log2_meta_row_height; + + //full sized meta chunk width in unit of data elements + log2_meta_chunk_width = log2_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height; + log2_min_meta_chunk_bytes = dml_log2(rq_sizing_param->min_meta_chunk_bytes); + min_meta_chunk_width = 1 + << (log2_min_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height); + meta_chunk_width = 1 << log2_meta_chunk_width; + meta_chunk_per_row_int = (unsigned int) (meta_row_width_ub / meta_chunk_width); + meta_row_remainder = meta_row_width_ub % meta_chunk_width; + meta_chunk_threshold = 0; + meta_blk_bytes = 4096; + meta_blk_height = blk256_height * 64; + meta_blk_width = meta_blk_bytes * 256 / bytes_per_element / meta_blk_height; + meta_surface_bytes = meta_pitch + * (dml_round_to_multiple(vp_height - 1, meta_blk_height, 1) + + meta_blk_height) * bytes_per_element / 256; + vmpg_bytes = mode_lib->soc.vmm_page_size_bytes; + meta_pte_req_per_frame_ub = (dml_round_to_multiple( + meta_surface_bytes - vmpg_bytes, + 8 * vmpg_bytes, + 1) + 8 * vmpg_bytes) / (8 * vmpg_bytes); + meta_pte_bytes_per_frame_ub = meta_pte_req_per_frame_ub * 64; //64B mpte request + rq_dlg_param->meta_pte_bytes_per_frame_ub = meta_pte_bytes_per_frame_ub; + + dml_print("DML_DLG: %s: meta_blk_height = %d\n", __func__, meta_blk_height); + dml_print("DML_DLG: %s: meta_blk_width = %d\n", __func__, meta_blk_width); + dml_print("DML_DLG: %s: meta_surface_bytes = %d\n", __func__, meta_surface_bytes); + dml_print( + "DML_DLG: %s: meta_pte_req_per_frame_ub = %d\n", + __func__, + meta_pte_req_per_frame_ub); + dml_print( + "DML_DLG: %s: meta_pte_bytes_per_frame_ub = %d\n", + __func__, + meta_pte_bytes_per_frame_ub); + + if (!surf_vert) + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_width; + else + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_height; + + if (meta_row_remainder <= meta_chunk_threshold) + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 1; + else + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 2; + + // ------ + // dpte + // ------ + if (surf_linear) { + log2_vmpg_height = 0; // one line high + } else { + log2_vmpg_height = (log2_vmpg_bytes - 8) / 2 + log2_blk256_height; + } + log2_vmpg_width = log2_vmpg_bytes - log2_bytes_per_element - log2_vmpg_height; + + // only 3 possible shapes for dpte request in dimensions of ptes: 8x1, 4x2, 2x4. + if (surf_linear) { //one 64B PTE request returns 8 PTEs + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_vmpg_width + 3; + log2_dpte_req_height = 0; + } else if (log2_blk_bytes == 12) { //4KB tile means 4kB page size + //one 64B req gives 8x1 PTEs for 4KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } else if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) { // tile block >= 64KB + //two 64B reqs of 2x4 PTEs give 16 PTEs to cover 64KB + log2_dpte_req_height_ptes = 4; + log2_dpte_req_width = log2_blk256_width + 4; // log2_64KB_width + log2_dpte_req_height = log2_blk256_height + 4; // log2_64KB_height + } else { //64KB page size and must 64KB tile block + //one 64B req gives 8x1 PTEs for 64KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } + + // The dpte request dimensions in data elements is dpte_req_width x dpte_req_height + // log2_vmpg_width is how much 1 pte represent, now calculating how much a 64b pte req represent + // That depends on the pte shape (i.e. 8x1, 4x2, 2x4) + //log2_dpte_req_height = log2_vmpg_height + log2_dpte_req_height_ptes; + //log2_dpte_req_width = log2_vmpg_width + log2_dpte_req_width_ptes; + dpte_req_height = 1 << log2_dpte_req_height; + dpte_req_width = 1 << log2_dpte_req_width; + + // calculate pitch dpte row buffer can hold + // round the result down to a power of two. + pde_buf_entries = + yuv420 ? (pde_proc_buffer_size_64k_reqs >> 1) : pde_proc_buffer_size_64k_reqs; + if (surf_linear) { + unsigned int dpte_row_height; + + log2_dpte_row_height_linear = dml_floor( + dml_log2( + dml_min( + 64 * 1024 * pde_buf_entries + / bytes_per_element, + dpte_buf_in_pte_reqs + * dpte_req_width) + / data_pitch), + 1); + + ASSERT(log2_dpte_row_height_linear >= 3); + + if (log2_dpte_row_height_linear > 7) + log2_dpte_row_height_linear = 7; + + log2_dpte_row_height = log2_dpte_row_height_linear; + // For linear, the dpte row is pitch dependent and the pte requests wrap at the pitch boundary. + // the dpte_row_width_ub is the upper bound of data_pitch*dpte_row_height in elements with this unique buffering. + dpte_row_height = 1 << log2_dpte_row_height; + dpte_row_width_ub = dml_round_to_multiple( + data_pitch * dpte_row_height - 1, + dpte_req_width, + 1) + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + // the upper bound of the dpte_row_width without dependency on viewport position follows. + // for tiled mode, row height is the same as req height and row store up to vp size upper bound + if (!surf_vert) { + log2_dpte_row_height = log2_dpte_req_height; + dpte_row_width_ub = dml_round_to_multiple(vp_width - 1, dpte_req_width, 1) + + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + log2_dpte_row_height = + (log2_blk_width < log2_dpte_req_width) ? + log2_blk_width : log2_dpte_req_width; + dpte_row_width_ub = dml_round_to_multiple(vp_height - 1, dpte_req_height, 1) + + dpte_req_height; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_height; + } + } + if (log2_blk_bytes >= 16 && log2_vmpg_bytes == 12) // tile block >= 64KB + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 128; //2*64B dpte request + else + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 64; //64B dpte request + + rq_dlg_param->dpte_row_height = 1 << log2_dpte_row_height; + + // the dpte_group_bytes is reduced for the specific case of vertical + // access of a tile surface that has dpte request of 8x1 ptes. + + if (hostvm_enable) + rq_sizing_param->dpte_group_bytes = 512; + else { + if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + rq_sizing_param->dpte_group_bytes = 512; + else + //full size + rq_sizing_param->dpte_group_bytes = 2048; + } + + //since pte request size is 64byte, the number of data pte requests per full sized group is as follows. + log2_dpte_group_bytes = dml_log2(rq_sizing_param->dpte_group_bytes); + log2_dpte_group_length = log2_dpte_group_bytes - 6; //length in 64b requests + + // full sized data pte group width in elements + if (!surf_vert) + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_width; + else + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_height; + + //But if the tile block >=64KB and the page size is 4KB, then each dPTE request is 2*64B + if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) // tile block >= 64KB + log2_dpte_group_width = log2_dpte_group_width - 1; + + dpte_group_width = 1 << log2_dpte_group_width; + + // since dpte groups are only aligned to dpte_req_width and not dpte_group_width, + // the upper bound for the dpte groups per row is as follows. + rq_dlg_param->dpte_groups_per_row_ub = dml_ceil( + (double) dpte_row_width_ub / dpte_group_width, + 1); +} + +static void get_surf_rq_param( + struct display_mode_lib *mode_lib, + display_data_rq_sizing_params_st *rq_sizing_param, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + const display_pipe_params_st pipe_param, + bool is_chroma) +{ + bool mode_422 = 0; + unsigned int vp_width = 0; + unsigned int vp_height = 0; + unsigned int data_pitch = 0; + unsigned int meta_pitch = 0; + unsigned int ppe = mode_422 ? 2 : 1; + + // FIXME check if ppe apply for both luma and chroma in 422 case + if (is_chroma) { + vp_width = pipe_param.src.viewport_width_c / ppe; + vp_height = pipe_param.src.viewport_height_c; + data_pitch = pipe_param.src.data_pitch_c; + meta_pitch = pipe_param.src.meta_pitch_c; + } else { + vp_width = pipe_param.src.viewport_width / ppe; + vp_height = pipe_param.src.viewport_height; + data_pitch = pipe_param.src.data_pitch; + meta_pitch = pipe_param.src.meta_pitch; + } + + if (pipe_param.dest.odm_combine) { + unsigned int access_dir; + unsigned int full_src_vp_width; + unsigned int hactive_half; + unsigned int src_hactive_half; + access_dir = (pipe_param.src.source_scan == dm_vert); // vp access direction: horizontal or vertical accessed + hactive_half = pipe_param.dest.hactive / 2; + if (is_chroma) { + full_src_vp_width = pipe_param.scale_ratio_depth.hscl_ratio_c * pipe_param.dest.full_recout_width; + src_hactive_half = pipe_param.scale_ratio_depth.hscl_ratio_c * hactive_half; + } else { + full_src_vp_width = pipe_param.scale_ratio_depth.hscl_ratio * pipe_param.dest.full_recout_width; + src_hactive_half = pipe_param.scale_ratio_depth.hscl_ratio * hactive_half; + } + + if (access_dir == 0) { + vp_width = dml_min(full_src_vp_width, src_hactive_half); + dml_print("DML_DLG: %s: vp_width = %d\n", __func__, vp_width); + } else { + vp_height = dml_min(full_src_vp_width, src_hactive_half); + dml_print("DML_DLG: %s: vp_height = %d\n", __func__, vp_height); + + } + dml_print("DML_DLG: %s: full_src_vp_width = %d\n", __func__, full_src_vp_width); + dml_print("DML_DLG: %s: hactive_half = %d\n", __func__, hactive_half); + dml_print("DML_DLG: %s: src_hactive_half = %d\n", __func__, src_hactive_half); + } + rq_sizing_param->chunk_bytes = 8192; + + if (rq_sizing_param->chunk_bytes == 64 * 1024) + rq_sizing_param->min_chunk_bytes = 0; + else + rq_sizing_param->min_chunk_bytes = 1024; + + rq_sizing_param->meta_chunk_bytes = 2048; + rq_sizing_param->min_meta_chunk_bytes = 256; + + if (pipe_param.src.hostvm) + rq_sizing_param->mpte_group_bytes = 512; + else + rq_sizing_param->mpte_group_bytes = 2048; + + get_meta_and_pte_attr( + mode_lib, + rq_dlg_param, + rq_misc_param, + rq_sizing_param, + vp_width, + vp_height, + data_pitch, + meta_pitch, + pipe_param.src.source_format, + pipe_param.src.sw_mode, + pipe_param.src.macro_tile_size, + pipe_param.src.source_scan, + pipe_param.src.hostvm, + is_chroma); +} + +static void dml_rq_dlg_get_rq_params( + struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_params_st pipe_param) +{ + // get param for luma surface + rq_param->yuv420 = pipe_param.src.source_format == dm_420_8 + || pipe_param.src.source_format == dm_420_10; + rq_param->yuv420_10bpc = pipe_param.src.source_format == dm_420_10; + + get_surf_rq_param( + mode_lib, + &(rq_param->sizing.rq_l), + &(rq_param->dlg.rq_l), + &(rq_param->misc.rq_l), + pipe_param, + 0); + + if (is_dual_plane((enum source_format_class) (pipe_param.src.source_format))) { + // get param for chroma surface + get_surf_rq_param( + mode_lib, + &(rq_param->sizing.rq_c), + &(rq_param->dlg.rq_c), + &(rq_param->misc.rq_c), + pipe_param, + 1); + } + + // calculate how to split the det buffer space between luma and chroma + handle_det_buf_split(mode_lib, rq_param, pipe_param.src); + print__rq_params_st(mode_lib, *rq_param); +} + +void dml21_rq_dlg_get_rq_reg( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param) +{ + display_rq_params_st rq_param = {0}; + + memset(rq_regs, 0, sizeof(*rq_regs)); + dml_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param); + extract_rq_regs(mode_lib, rq_regs, rq_param); + + print__rq_regs_st(mode_lib, *rq_regs); +} + +// Note: currently taken in as is. +// Nice to decouple code from hw register implement and extract code that are repeated for luma and chroma. +static void dml_rq_dlg_get_dlg_params( + struct display_mode_lib *mode_lib, + const display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + display_dlg_regs_st *disp_dlg_regs, + display_ttu_regs_st *disp_ttu_regs, + const display_rq_dlg_params_st rq_dlg_param, + const display_dlg_sys_params_st dlg_sys_param, + const bool cstate_en, + const bool pstate_en) +{ + const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src; + const display_pipe_dest_params_st *dst = &e2e_pipe_param[pipe_idx].pipe.dest; + const display_output_params_st *dout = &e2e_pipe_param[pipe_idx].dout; + const display_clocks_and_cfg_st *clks = &e2e_pipe_param[pipe_idx].clks_cfg; + const scaler_ratio_depth_st *scl = &e2e_pipe_param[pipe_idx].pipe.scale_ratio_depth; + const scaler_taps_st *taps = &e2e_pipe_param[pipe_idx].pipe.scale_taps; + + // ------------------------- + // Section 1.15.2.1: OTG dependent Params + // ------------------------- + // Timing + unsigned int htotal = dst->htotal; + // unsigned int hblank_start = dst.hblank_start; // TODO: Remove + unsigned int hblank_end = dst->hblank_end; + unsigned int vblank_start = dst->vblank_start; + unsigned int vblank_end = dst->vblank_end; + unsigned int min_vblank = mode_lib->ip.min_vblank_lines; + + double dppclk_freq_in_mhz = clks->dppclk_mhz; + double dispclk_freq_in_mhz = clks->dispclk_mhz; + double refclk_freq_in_mhz = clks->refclk_mhz; + double pclk_freq_in_mhz = dst->pixel_rate_mhz; + bool interlaced = dst->interlaced; + + double ref_freq_to_pix_freq = refclk_freq_in_mhz / pclk_freq_in_mhz; + + double min_dcfclk_mhz; + double t_calc_us; + double min_ttu_vblank; + + double min_dst_y_ttu_vblank; + unsigned int dlg_vblank_start; + bool dual_plane; + bool mode_422; + unsigned int access_dir; + unsigned int vp_height_l; + unsigned int vp_width_l; + unsigned int vp_height_c; + unsigned int vp_width_c; + + // Scaling + unsigned int htaps_l; + unsigned int htaps_c; + double hratio_l; + double hratio_c; + double vratio_l; + double vratio_c; + bool scl_enable; + + double line_time_in_us; + // double vinit_l; + // double vinit_c; + // double vinit_bot_l; + // double vinit_bot_c; + + // unsigned int swath_height_l; + unsigned int swath_width_ub_l; + // unsigned int dpte_bytes_per_row_ub_l; + unsigned int dpte_groups_per_row_ub_l; + // unsigned int meta_pte_bytes_per_frame_ub_l; + // unsigned int meta_bytes_per_row_ub_l; + + // unsigned int swath_height_c; + unsigned int swath_width_ub_c; + // unsigned int dpte_bytes_per_row_ub_c; + unsigned int dpte_groups_per_row_ub_c; + + unsigned int meta_chunks_per_row_ub_l; + unsigned int meta_chunks_per_row_ub_c; + unsigned int vupdate_offset; + unsigned int vupdate_width; + unsigned int vready_offset; + + unsigned int dppclk_delay_subtotal; + unsigned int dispclk_delay_subtotal; + unsigned int pixel_rate_delay_subtotal; + + unsigned int vstartup_start; + unsigned int dst_x_after_scaler; + unsigned int dst_y_after_scaler; + double line_wait; + double dst_y_prefetch; + double dst_y_per_vm_vblank; + double dst_y_per_row_vblank; + double dst_y_per_vm_flip; + double dst_y_per_row_flip; + double max_dst_y_per_vm_vblank; + double max_dst_y_per_row_vblank; + double lsw; + double vratio_pre_l; + double vratio_pre_c; + unsigned int req_per_swath_ub_l; + unsigned int req_per_swath_ub_c; + unsigned int meta_row_height_l; + unsigned int meta_row_height_c; + unsigned int swath_width_pixels_ub_l; + unsigned int swath_width_pixels_ub_c; + unsigned int scaler_rec_in_width_l; + unsigned int scaler_rec_in_width_c; + unsigned int dpte_row_height_l; + unsigned int dpte_row_height_c; + double hscale_pixel_rate_l; + double hscale_pixel_rate_c; + double min_hratio_fact_l; + double min_hratio_fact_c; + double refcyc_per_line_delivery_pre_l; + double refcyc_per_line_delivery_pre_c; + double refcyc_per_line_delivery_l; + double refcyc_per_line_delivery_c; + + double refcyc_per_req_delivery_pre_l; + double refcyc_per_req_delivery_pre_c; + double refcyc_per_req_delivery_l; + double refcyc_per_req_delivery_c; + + unsigned int full_recout_width; + double xfc_transfer_delay; + double xfc_precharge_delay; + double xfc_remote_surface_flip_latency; + double xfc_dst_y_delta_drq_limit; + double xfc_prefetch_margin; + double refcyc_per_req_delivery_pre_cur0; + double refcyc_per_req_delivery_cur0; + double refcyc_per_req_delivery_pre_cur1; + double refcyc_per_req_delivery_cur1; + + memset(disp_dlg_regs, 0, sizeof(*disp_dlg_regs)); + memset(disp_ttu_regs, 0, sizeof(*disp_ttu_regs)); + + dml_print("DML_DLG: %s: cstate_en = %d\n", __func__, cstate_en); + dml_print("DML_DLG: %s: pstate_en = %d\n", __func__, pstate_en); + + dml_print("DML_DLG: %s: dppclk_freq_in_mhz = %3.2f\n", __func__, dppclk_freq_in_mhz); + dml_print("DML_DLG: %s: dispclk_freq_in_mhz = %3.2f\n", __func__, dispclk_freq_in_mhz); + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: interlaced = %d\n", __func__, interlaced); + ASSERT(ref_freq_to_pix_freq < 4.0); + + disp_dlg_regs->ref_freq_to_pix_freq = + (unsigned int) (ref_freq_to_pix_freq * dml_pow(2, 19)); + disp_dlg_regs->refcyc_per_htotal = (unsigned int) (ref_freq_to_pix_freq * (double) htotal + * dml_pow(2, 8)); + disp_dlg_regs->dlg_vblank_end = interlaced ? (vblank_end / 2) : vblank_end; // 15 bits + disp_dlg_regs->refcyc_h_blank_end = (unsigned int) ((double) hblank_end + * (double) ref_freq_to_pix_freq); + ASSERT(disp_dlg_regs->refcyc_h_blank_end < (unsigned int)dml_pow(2, 13)); + + min_dcfclk_mhz = dlg_sys_param.deepsleep_dcfclk_mhz; + t_calc_us = get_tcalc(mode_lib, e2e_pipe_param, num_pipes); + min_ttu_vblank = get_min_ttu_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + min_dst_y_ttu_vblank = min_ttu_vblank * pclk_freq_in_mhz / (double) htotal; + dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start; + + disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start) * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)dml_pow(2, 18)); + + dml_print( + "DML_DLG: %s: min_dcfclk_mhz = %3.2f\n", + __func__, + min_dcfclk_mhz); + dml_print( + "DML_DLG: %s: min_ttu_vblank = %3.2f\n", + __func__, + min_ttu_vblank); + dml_print( + "DML_DLG: %s: min_dst_y_ttu_vblank = %3.2f\n", + __func__, + min_dst_y_ttu_vblank); + dml_print( + "DML_DLG: %s: t_calc_us = %3.2f\n", + __func__, + t_calc_us); + dml_print( + "DML_DLG: %s: disp_dlg_regs->min_dst_y_next_start = 0x%0x\n", + __func__, + disp_dlg_regs->min_dst_y_next_start); + dml_print( + "DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", + __func__, + ref_freq_to_pix_freq); + + // ------------------------- + // Section 1.15.2.2: Prefetch, Active and TTU + // ------------------------- + // Prefetch Calc + // Source + // dcc_en = src.dcc; + dual_plane = is_dual_plane((enum source_format_class) (src->source_format)); + mode_422 = 0; // FIXME + access_dir = (src->source_scan == dm_vert); // vp access direction: horizontal or vertical accessed + // bytes_per_element_l = get_bytes_per_element(source_format_class(src.source_format), 0); + // bytes_per_element_c = get_bytes_per_element(source_format_class(src.source_format), 1); + vp_height_l = src->viewport_height; + vp_width_l = src->viewport_width; + vp_height_c = src->viewport_height_c; + vp_width_c = src->viewport_width_c; + + // Scaling + htaps_l = taps->htaps; + htaps_c = taps->htaps_c; + hratio_l = scl->hscl_ratio; + hratio_c = scl->hscl_ratio_c; + vratio_l = scl->vscl_ratio; + vratio_c = scl->vscl_ratio_c; + scl_enable = scl->scl_enable; + + line_time_in_us = (htotal / pclk_freq_in_mhz); + swath_width_ub_l = rq_dlg_param.rq_l.swath_width_ub; + dpte_groups_per_row_ub_l = rq_dlg_param.rq_l.dpte_groups_per_row_ub; + swath_width_ub_c = rq_dlg_param.rq_c.swath_width_ub; + dpte_groups_per_row_ub_c = rq_dlg_param.rq_c.dpte_groups_per_row_ub; + + meta_chunks_per_row_ub_l = rq_dlg_param.rq_l.meta_chunks_per_row_ub; + meta_chunks_per_row_ub_c = rq_dlg_param.rq_c.meta_chunks_per_row_ub; + vupdate_offset = dst->vupdate_offset; + vupdate_width = dst->vupdate_width; + vready_offset = dst->vready_offset; + + dppclk_delay_subtotal = mode_lib->ip.dppclk_delay_subtotal; + dispclk_delay_subtotal = mode_lib->ip.dispclk_delay_subtotal; + + if (scl_enable) + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl; + else + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl_lb_only; + + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_cnvc_formatter + + src->num_cursors * mode_lib->ip.dppclk_delay_cnvc_cursor; + + if (dout->dsc_enable) { + double dsc_delay = get_dsc_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dispclk_delay_subtotal += dsc_delay; + } + + pixel_rate_delay_subtotal = dppclk_delay_subtotal * pclk_freq_in_mhz / dppclk_freq_in_mhz + + dispclk_delay_subtotal * pclk_freq_in_mhz / dispclk_freq_in_mhz; + + vstartup_start = dst->vstartup_start; + if (interlaced) { + if (vstartup_start / 2.0 + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end / 2.0) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } else { + if (vstartup_start + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } + + // TODO: Where is this coming from? + if (interlaced) + vstartup_start = vstartup_start / 2; + + // TODO: What if this min_vblank doesn't match the value in the dml_config_settings.cpp? + if (vstartup_start >= min_vblank) { + dml_print( + "WARNING: DML_DLG: %s: vblank_start=%d vblank_end=%d\n", + __func__, + vblank_start, + vblank_end); + dml_print( + "WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + min_vblank = vstartup_start + 1; + dml_print( + "WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + } + + dst_x_after_scaler = get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_after_scaler = get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: htotal = %d\n", __func__, htotal); + dml_print( + "DML_DLG: %s: pixel_rate_delay_subtotal = %d\n", + __func__, + pixel_rate_delay_subtotal); + dml_print( + "DML_DLG: %s: dst_x_after_scaler = %d\n", + __func__, + dst_x_after_scaler); + dml_print( + "DML_DLG: %s: dst_y_after_scaler = %d\n", + __func__, + dst_y_after_scaler); + + // Lwait + // TODO: Should this be urgent_latency_pixel_mixed_with_vm_data_us? + line_wait = mode_lib->soc.urgent_latency_pixel_data_only_us; + if (cstate_en) + line_wait = dml_max(mode_lib->soc.sr_enter_plus_exit_time_us, line_wait); + if (pstate_en) + line_wait = dml_max( + mode_lib->soc.dram_clock_change_latency_us + + mode_lib->soc.urgent_latency_pixel_data_only_us, // TODO: Should this be urgent_latency_pixel_mixed_with_vm_data_us? + line_wait); + line_wait = line_wait / line_time_in_us; + + dst_y_prefetch = get_dst_y_prefetch(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dml_print("DML_DLG: %s: dst_y_prefetch (after rnd) = %3.2f\n", __func__, dst_y_prefetch); + + dst_y_per_vm_vblank = get_dst_y_per_vm_vblank( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_row_vblank = get_dst_y_per_row_vblank( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_vm_flip = get_dst_y_per_vm_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_per_row_flip = get_dst_y_per_row_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + max_dst_y_per_vm_vblank = 32.0; + max_dst_y_per_row_vblank = 16.0; + + // magic! + if (htotal <= 75) { + min_vblank = 300; + max_dst_y_per_vm_vblank = 100.0; + max_dst_y_per_row_vblank = 100.0; + } + + dml_print("DML_DLG: %s: dst_y_per_vm_flip = %3.2f\n", __func__, dst_y_per_vm_flip); + dml_print("DML_DLG: %s: dst_y_per_row_flip = %3.2f\n", __func__, dst_y_per_row_flip); + dml_print("DML_DLG: %s: dst_y_per_vm_vblank = %3.2f\n", __func__, dst_y_per_vm_vblank); + dml_print("DML_DLG: %s: dst_y_per_row_vblank = %3.2f\n", __func__, dst_y_per_row_vblank); + + ASSERT(dst_y_per_vm_vblank < max_dst_y_per_vm_vblank); + ASSERT(dst_y_per_row_vblank < max_dst_y_per_row_vblank); + + ASSERT(dst_y_prefetch > (dst_y_per_vm_vblank + dst_y_per_row_vblank)); + lsw = dst_y_prefetch - (dst_y_per_vm_vblank + dst_y_per_row_vblank); + + dml_print("DML_DLG: %s: lsw = %3.2f\n", __func__, lsw); + + vratio_pre_l = get_vratio_prefetch_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + vratio_pre_c = get_vratio_prefetch_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: vratio_pre_l=%3.2f\n", __func__, vratio_pre_l); + dml_print("DML_DLG: %s: vratio_pre_c=%3.2f\n", __func__, vratio_pre_c); + + // Active + req_per_swath_ub_l = rq_dlg_param.rq_l.req_per_swath_ub; + req_per_swath_ub_c = rq_dlg_param.rq_c.req_per_swath_ub; + meta_row_height_l = rq_dlg_param.rq_l.meta_row_height; + meta_row_height_c = rq_dlg_param.rq_c.meta_row_height; + swath_width_pixels_ub_l = 0; + swath_width_pixels_ub_c = 0; + scaler_rec_in_width_l = 0; + scaler_rec_in_width_c = 0; + dpte_row_height_l = rq_dlg_param.rq_l.dpte_row_height; + dpte_row_height_c = rq_dlg_param.rq_c.dpte_row_height; + + if (mode_422) { + swath_width_pixels_ub_l = swath_width_ub_l * 2; // *2 for 2 pixel per element + swath_width_pixels_ub_c = swath_width_ub_c * 2; + } else { + swath_width_pixels_ub_l = swath_width_ub_l * 1; + swath_width_pixels_ub_c = swath_width_ub_c * 1; + } + + hscale_pixel_rate_l = 0.; + hscale_pixel_rate_c = 0.; + min_hratio_fact_l = 1.0; + min_hratio_fact_c = 1.0; + + if (htaps_l <= 1) + min_hratio_fact_l = 2.0; + else if (htaps_l <= 6) { + if ((hratio_l * 2.0) > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l * 2.0; + } else { + if (hratio_l > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l; + } + + hscale_pixel_rate_l = min_hratio_fact_l * dppclk_freq_in_mhz; + + if (htaps_c <= 1) + min_hratio_fact_c = 2.0; + else if (htaps_c <= 6) { + if ((hratio_c * 2.0) > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c * 2.0; + } else { + if (hratio_c > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c; + } + + hscale_pixel_rate_c = min_hratio_fact_c * dppclk_freq_in_mhz; + + refcyc_per_line_delivery_pre_l = 0.; + refcyc_per_line_delivery_pre_c = 0.; + refcyc_per_line_delivery_l = 0.; + refcyc_per_line_delivery_c = 0.; + + refcyc_per_req_delivery_pre_l = 0.; + refcyc_per_req_delivery_pre_c = 0.; + refcyc_per_req_delivery_l = 0.; + refcyc_per_req_delivery_c = 0.; + + full_recout_width = 0; + // In ODM + if (src->is_hsplit) { + // This "hack" is only allowed (and valid) for MPC combine. In ODM + // combine, you MUST specify the full_recout_width...according to Oswin + if (dst->full_recout_width == 0 && !dst->odm_combine) { + dml_print( + "DML_DLG: %s: Warning: full_recout_width not set in hsplit mode\n", + __func__); + full_recout_width = dst->recout_width * 2; // assume half split for dcn1 + } else + full_recout_width = dst->full_recout_width; + } else + full_recout_width = dst->recout_width; + + // As of DCN2, mpc_combine and odm_combine are mutually exclusive + refcyc_per_line_delivery_pre_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + refcyc_per_line_delivery_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + dml_print("DML_DLG: %s: full_recout_width = %d\n", __func__, full_recout_width); + dml_print( + "DML_DLG: %s: hscale_pixel_rate_l = %3.2f\n", + __func__, + hscale_pixel_rate_l); + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_l); + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_l); + + if (dual_plane) { + refcyc_per_line_delivery_pre_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + refcyc_per_line_delivery_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_c); + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_c); + } + + // TTU - Luma / Chroma + if (access_dir) { // vertical access + scaler_rec_in_width_l = vp_height_l; + scaler_rec_in_width_c = vp_height_c; + } else { + scaler_rec_in_width_l = vp_width_l; + scaler_rec_in_width_c = vp_width_c; + } + + refcyc_per_req_delivery_pre_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + refcyc_per_req_delivery_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_l); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_l); + + ASSERT(refcyc_per_req_delivery_pre_l < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_l < dml_pow(2, 13)); + + if (dual_plane) { + refcyc_per_req_delivery_pre_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + refcyc_per_req_delivery_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_c); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_c); + + ASSERT(refcyc_per_req_delivery_pre_c < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_c < dml_pow(2, 13)); + } + + // XFC + xfc_transfer_delay = get_xfc_transfer_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + xfc_precharge_delay = get_xfc_precharge_delay( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_remote_surface_flip_latency = get_xfc_remote_surface_flip_latency( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_dst_y_delta_drq_limit = xfc_remote_surface_flip_latency; + xfc_prefetch_margin = get_xfc_prefetch_margin( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + + // TTU - Cursor + refcyc_per_req_delivery_pre_cur0 = 0.0; + refcyc_per_req_delivery_cur0 = 0.0; + if (src->num_cursors > 0) { + calculate_ttu_cursor( + mode_lib, + &refcyc_per_req_delivery_pre_cur0, + &refcyc_per_req_delivery_cur0, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur0_src_width, + (enum cursor_bpp) (src->cur0_bpp)); + } + + refcyc_per_req_delivery_pre_cur1 = 0.0; + refcyc_per_req_delivery_cur1 = 0.0; + if (src->num_cursors > 1) { + calculate_ttu_cursor( + mode_lib, + &refcyc_per_req_delivery_pre_cur1, + &refcyc_per_req_delivery_cur1, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur1_src_width, + (enum cursor_bpp) (src->cur1_bpp)); + } + + // TTU - Misc + // all hard-coded + + // Assignment to register structures + disp_dlg_regs->dst_y_after_scaler = dst_y_after_scaler; // in terms of line + disp_dlg_regs->refcyc_x_after_scaler = dst_x_after_scaler * ref_freq_to_pix_freq; // in terms of refclk + ASSERT(disp_dlg_regs->refcyc_x_after_scaler < (unsigned int)dml_pow(2, 13)); + disp_dlg_regs->dst_y_prefetch = (unsigned int) (dst_y_prefetch * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_vblank = (unsigned int) (dst_y_per_vm_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_vblank = (unsigned int) (dst_y_per_row_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_flip = (unsigned int) (dst_y_per_vm_flip * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_flip = (unsigned int) (dst_y_per_row_flip * dml_pow(2, 2)); + + disp_dlg_regs->vratio_prefetch = (unsigned int) (vratio_pre_l * dml_pow(2, 19)); + disp_dlg_regs->vratio_prefetch_c = (unsigned int) (vratio_pre_c * dml_pow(2, 19)); + + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_vblank); + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_vblank); + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip); + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip); + + disp_dlg_regs->refcyc_per_pte_group_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13)); + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_vblank_c = (unsigned int) (dst_y_per_row_vblank + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c + < (unsigned int)dml_pow(2, 13)); + } + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) meta_chunks_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_meta_chunk_vblank_l < (unsigned int)dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_c = + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l; // dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + disp_dlg_regs->refcyc_per_pte_group_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_l; + disp_dlg_regs->refcyc_per_meta_chunk_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_l; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_c; + disp_dlg_regs->refcyc_per_meta_chunk_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_c; + } + + disp_dlg_regs->refcyc_per_vm_group_vblank = get_refcyc_per_vm_group_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; + disp_dlg_regs->refcyc_per_vm_group_flip = get_refcyc_per_vm_group_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; + disp_dlg_regs->refcyc_per_vm_req_vblank = get_refcyc_per_vm_req_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;; + disp_dlg_regs->refcyc_per_vm_req_flip = get_refcyc_per_vm_req_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;; + + // Clamp to max for now + if (disp_dlg_regs->refcyc_per_vm_group_vblank >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_group_vblank = dml_pow(2, 23) - 1; + + if (disp_dlg_regs->refcyc_per_vm_group_flip >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_group_flip = dml_pow(2, 23) - 1; + + if (disp_dlg_regs->refcyc_per_vm_req_vblank >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_req_vblank = dml_pow(2, 23) - 1; + + if (disp_dlg_regs->refcyc_per_vm_req_flip >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_req_flip = dml_pow(2, 23) - 1; + disp_dlg_regs->dst_y_per_pte_row_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_pte_row_nom_l < (unsigned int)dml_pow(2, 17)); + + if (dual_plane) { + disp_dlg_regs->dst_y_per_pte_row_nom_c = (unsigned int) ((double) dpte_row_height_c + / (double) vratio_c * dml_pow(2, 2)); + if (disp_dlg_regs->dst_y_per_pte_row_nom_c >= (unsigned int) dml_pow(2, 17)) { + dml_print( + "DML_DLG: %s: Warning dst_y_per_pte_row_nom_c %u larger than supported by register format U15.2 %u\n", + __func__, + disp_dlg_regs->dst_y_per_pte_row_nom_c, + (unsigned int)dml_pow(2, 17) - 1); + } + } + + disp_dlg_regs->dst_y_per_meta_row_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_meta_row_nom_l < (unsigned int)dml_pow(2, 17)); + + disp_dlg_regs->dst_y_per_meta_row_nom_c = disp_dlg_regs->dst_y_per_meta_row_nom_l; // TODO: dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + dml_print( + "DML: Trow: %fus\n", + line_time_in_us * (double)dpte_row_height_l / (double)vratio_l); + + disp_dlg_regs->refcyc_per_pte_group_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_pte_group_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_l = dml_pow(2, 23) - 1; + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = dml_pow(2, 23) - 1; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_nom_c = + (unsigned int) ((double) dpte_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_pte_group_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_c = dml_pow(2, 23) - 1; + + // TODO: Is this the right calculation? Does htotal need to be halved? + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = + (unsigned int) ((double) meta_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = dml_pow(2, 23) - 1; + } + + disp_dlg_regs->refcyc_per_line_delivery_pre_l = (unsigned int) dml_floor( + refcyc_per_line_delivery_pre_l, 1); + disp_dlg_regs->refcyc_per_line_delivery_l = (unsigned int) dml_floor( + refcyc_per_line_delivery_l, 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_l < (unsigned int)dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_l < (unsigned int)dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_line_delivery_pre_c = (unsigned int) dml_floor( + refcyc_per_line_delivery_pre_c, 1); + disp_dlg_regs->refcyc_per_line_delivery_c = (unsigned int) dml_floor( + refcyc_per_line_delivery_c, 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_c < (unsigned int)dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_c < (unsigned int)dml_pow(2, 13)); + + disp_dlg_regs->chunk_hdl_adjust_cur0 = 3; + disp_dlg_regs->dst_y_offset_cur0 = 0; + disp_dlg_regs->chunk_hdl_adjust_cur1 = 3; + disp_dlg_regs->dst_y_offset_cur1 = 0; + + disp_dlg_regs->xfc_reg_transfer_delay = xfc_transfer_delay; + disp_dlg_regs->xfc_reg_precharge_delay = xfc_precharge_delay; + disp_dlg_regs->xfc_reg_remote_surface_flip_latency = xfc_remote_surface_flip_latency; + disp_dlg_regs->xfc_reg_prefetch_margin = dml_ceil( + xfc_prefetch_margin * refclk_freq_in_mhz, 1); + + // slave has to have this value also set to off + if (src->xfc_enable && !src->xfc_slave) + disp_dlg_regs->dst_y_delta_drq_limit = dml_ceil(xfc_dst_y_delta_drq_limit, 1); + else + disp_dlg_regs->dst_y_delta_drq_limit = 0x7fff; // off + + disp_ttu_regs->refcyc_per_req_delivery_pre_l = (unsigned int) (refcyc_per_req_delivery_pre_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_l = (unsigned int) (refcyc_per_req_delivery_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_c = (unsigned int) (refcyc_per_req_delivery_pre_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_c = (unsigned int) (refcyc_per_req_delivery_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur0 = + (unsigned int) (refcyc_per_req_delivery_pre_cur0 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur0 = (unsigned int) (refcyc_per_req_delivery_cur0 + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur1 = + (unsigned int) (refcyc_per_req_delivery_pre_cur1 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur1 = (unsigned int) (refcyc_per_req_delivery_cur1 + * dml_pow(2, 10)); + disp_ttu_regs->qos_level_low_wm = 0; + ASSERT(disp_ttu_regs->qos_level_low_wm < dml_pow(2, 14)); + disp_ttu_regs->qos_level_high_wm = (unsigned int) (4.0 * (double) htotal + * ref_freq_to_pix_freq); + ASSERT(disp_ttu_regs->qos_level_high_wm < dml_pow(2, 14)); + + disp_ttu_regs->qos_level_flip = 14; + disp_ttu_regs->qos_level_fixed_l = 8; + disp_ttu_regs->qos_level_fixed_c = 8; + disp_ttu_regs->qos_level_fixed_cur0 = 8; + disp_ttu_regs->qos_ramp_disable_l = 0; + disp_ttu_regs->qos_ramp_disable_c = 0; + disp_ttu_regs->qos_ramp_disable_cur0 = 0; + + disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz; + ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24)); + + print__ttu_regs_st(mode_lib, *disp_ttu_regs); + print__dlg_regs_st(mode_lib, *disp_dlg_regs); +} + +void dml21_rq_dlg_get_dlg_reg( + struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support) +{ + display_rq_params_st rq_param = {0}; + display_dlg_sys_params_st dlg_sys_param = {0}; + + // Get watermark and Tex. + dlg_sys_param.t_urg_wm_us = get_wm_urgent(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.deepsleep_dcfclk_mhz = get_clk_dcf_deepsleep( + mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_extra_us = get_urgent_extra_latency(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.mem_trip_us = get_wm_memory_trip(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_mclk_wm_us = get_wm_dram_clock_change(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_sr_wm_us = get_wm_stutter_enter_exit(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.total_flip_bw = get_total_immediate_flip_bw( + mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.total_flip_bytes = get_total_immediate_flip_bytes( + mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency + / dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated + + print__dlg_sys_params_st(mode_lib, dlg_sys_param); + + // system parameter calculation done + + dml_print("DML_DLG: Calculation for pipe[%d] start\n\n", pipe_idx); + dml_rq_dlg_get_rq_params(mode_lib, &rq_param, e2e_pipe_param[pipe_idx].pipe); + dml_rq_dlg_get_dlg_params( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx, + dlg_regs, + ttu_regs, + rq_param.dlg, + dlg_sys_param, + cstate_en, + pstate_en); + dml_print("DML_DLG: Calculation for pipe[%d] end\n", pipe_idx); +} + +void dml_rq_dlg_get_arb_params(struct display_mode_lib *mode_lib, display_arb_params_st *arb_param) +{ + memset(arb_param, 0, sizeof(*arb_param)); + arb_param->max_req_outstanding = 256; + arb_param->min_req_outstanding = 68; + arb_param->sat_level_us = 60; +} + +static void calculate_ttu_cursor( + struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp) +{ + unsigned int cur_src_width = cur_width; + unsigned int cur_req_size = 0; + unsigned int cur_req_width = 0; + double cur_width_ub = 0.0; + double cur_req_per_width = 0.0; + double hactive_cur = 0.0; + + ASSERT(cur_src_width <= 256); + + *refcyc_per_req_delivery_pre_cur = 0.0; + *refcyc_per_req_delivery_cur = 0.0; + if (cur_src_width > 0) { + unsigned int cur_bit_per_pixel = 0; + + if (cur_bpp == dm_cur_2bit) { + cur_req_size = 64; // byte + cur_bit_per_pixel = 2; + } else { // 32bit + cur_bit_per_pixel = 32; + if (cur_src_width >= 1 && cur_src_width <= 16) + cur_req_size = 64; + else if (cur_src_width >= 17 && cur_src_width <= 31) + cur_req_size = 128; + else + cur_req_size = 256; + } + + cur_req_width = (double) cur_req_size / ((double) cur_bit_per_pixel / 8.0); + cur_width_ub = dml_ceil((double) cur_src_width / (double) cur_req_width, 1) + * (double) cur_req_width; + cur_req_per_width = cur_width_ub / (double) cur_req_width; + hactive_cur = (double) cur_src_width / hscl_ratio; // FIXME: oswin to think about what to do for cursor + + if (vratio_pre_l <= 1.0) { + *refcyc_per_req_delivery_pre_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_pre_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + ASSERT(*refcyc_per_req_delivery_pre_cur < dml_pow(2, 13)); + + if (vratio_l <= 1.0) { + *refcyc_per_req_delivery_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + dml_print( + "DML_DLG: %s: cur_req_width = %d\n", + __func__, + cur_req_width); + dml_print( + "DML_DLG: %s: cur_width_ub = %3.2f\n", + __func__, + cur_width_ub); + dml_print( + "DML_DLG: %s: cur_req_per_width = %3.2f\n", + __func__, + cur_req_per_width); + dml_print( + "DML_DLG: %s: hactive_cur = %3.2f\n", + __func__, + hactive_cur); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_pre_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_pre_cur); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_cur); + + ASSERT(*refcyc_per_req_delivery_cur < dml_pow(2, 13)); + } +} + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h new file mode 100644 index 000000000000..83e95f8cbff2 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h @@ -0,0 +1,73 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DML21_DISPLAY_RQ_DLG_CALC_H__ +#define __DML21_DISPLAY_RQ_DLG_CALC_H__ + +#include "../dml_common_defs.h" +#include "../display_rq_dlg_helpers.h" + +struct display_mode_lib; + + +// Function: dml_rq_dlg_get_rq_reg +// Main entry point for test to get the register values out of this DML class. +// This function calls <get_rq_param> and <extract_rq_regs> functions to calculate +// and then populate the rq_regs struct +// Input: +// pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.) +// Output: +// rq_regs - struct that holds all the RQ registers field value. +// See also: <display_rq_regs_st> +void dml21_rq_dlg_get_rq_reg( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param); + +// Function: dml_rq_dlg_get_dlg_reg +// Calculate and return DLG and TTU register struct given the system setting +// Output: +// dlg_regs - output DLG register struct +// ttu_regs - output DLG TTU register struct +// Input: +// e2e_pipe_param - "compacted" array of e2e pipe param struct +// num_pipes - num of active "pipe" or "route" +// pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg +// cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered. +// Added for legacy or unrealistic timing tests. +void dml21_rq_dlg_get_dlg_reg( + struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h index 0c2fab1e93b6..1c97083b8d0b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h @@ -37,11 +37,14 @@ enum source_format_class { dm_444_64 = 2, dm_420_8 = 3, dm_420_10 = 4, - dm_422_8 = 5, - dm_422_10 = 6, - dm_444_8 = 7, + dm_420_12 = 5, + dm_422_8 = 6, + dm_422_10 = 7, + dm_444_8 = 8, dm_mono_8 = dm_444_8, - dm_mono_16 = dm_444_16 + dm_mono_16 = dm_444_16, + dm_rgbe = 9, + dm_rgbe_alpha = 10, }; enum output_bpc_class { dm_out_6 = 0, dm_out_8 = 1, dm_out_10 = 2, dm_out_12 = 3, dm_out_16 = 4 @@ -83,7 +86,7 @@ enum dm_swizzle_mode { dm_sw_var_d_x = 30, dm_sw_64kb_r_x, dm_sw_gfx7_2d_thin_lvp, - dm_sw_gfx7_2d_thin_gl + dm_sw_gfx7_2d_thin_gl, }; enum lb_depth { dm_lb_10 = 0, dm_lb_8 = 1, dm_lb_6 = 2, dm_lb_12 = 3, dm_lb_16 = 4, @@ -112,7 +115,8 @@ enum output_standard { enum mpc_combine_affinity { dm_mpc_always_when_possible, dm_mpc_reduce_voltage, - dm_mpc_reduce_voltage_and_clocks + dm_mpc_reduce_voltage_and_clocks, + dm_mpc_never }; enum self_refresh_affinity { @@ -157,4 +161,10 @@ enum writeback_config { dm_whole_buffer_for_single_stream_interleave, }; +enum odm_combine_mode { + dm_odm_combine_mode_disabled, + dm_odm_combine_mode_2to1, + dm_odm_combine_mode_4to1, +}; + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c index 91810c7d5cf5..704efefdcba8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c @@ -28,6 +28,12 @@ #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/display_mode_vba_20.h" #include "dcn20/display_rq_dlg_calc_20.h" +#include "dcn20/display_mode_vba_20v2.h" +#include "dcn20/display_rq_dlg_calc_20v2.h" +#endif +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 +#include "dcn21/display_mode_vba_21.h" +#include "dcn21/display_rq_dlg_calc_21.h" #endif #if defined(CONFIG_DRM_AMD_DC_DCN2_0) @@ -37,6 +43,22 @@ const struct dml_funcs dml20_funcs = { .rq_dlg_get_dlg_reg = dml20_rq_dlg_get_dlg_reg, .rq_dlg_get_rq_reg = dml20_rq_dlg_get_rq_reg }; + +const struct dml_funcs dml20v2_funcs = { + .validate = dml20v2_ModeSupportAndSystemConfigurationFull, + .recalculate = dml20v2_recalculate, + .rq_dlg_get_dlg_reg = dml20v2_rq_dlg_get_dlg_reg, + .rq_dlg_get_rq_reg = dml20v2_rq_dlg_get_rq_reg +}; +#endif + +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 +const struct dml_funcs dml21_funcs = { + .validate = dml21_ModeSupportAndSystemConfigurationFull, + .recalculate = dml21_recalculate, + .rq_dlg_get_dlg_reg = dml21_rq_dlg_get_dlg_reg, + .rq_dlg_get_rq_reg = dml21_rq_dlg_get_rq_reg +}; #endif void dml_init_instance(struct display_mode_lib *lib, @@ -52,7 +74,16 @@ void dml_init_instance(struct display_mode_lib *lib, case DML_PROJECT_NAVI10: lib->funcs = dml20_funcs; break; + case DML_PROJECT_NAVI10v2: + lib->funcs = dml20v2_funcs; + break; #endif +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 + case DML_PROJECT_DCN21: + lib->funcs = dml21_funcs; + break; +#endif + default: break; } diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h index 5bf13d67f289..d8c59aa356b6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h @@ -36,6 +36,10 @@ enum dml_project { DML_PROJECT_RAVEN1, #ifdef CONFIG_DRM_AMD_DC_DCN2_0 DML_PROJECT_NAVI10, + DML_PROJECT_NAVI10v2, +#endif +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 + DML_PROJECT_DCN21, #endif }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index 5678472546ab..f4c1ef9046bf 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -100,6 +100,7 @@ struct _vcs_dpi_soc_bounding_box_st { unsigned int vmm_page_size_bytes; unsigned int hostvm_min_page_size_bytes; double dram_clock_change_latency_us; + double dummy_pstate_latency_us; double writeback_dram_clock_change_latency_us; unsigned int return_bus_width_bytes; unsigned int voltage_override; @@ -108,6 +109,9 @@ struct _vcs_dpi_soc_bounding_box_st { int use_urgent_burst_bw; unsigned int num_states; struct _vcs_dpi_voltage_scaling_st clock_limits[MAX_CLOCK_LIMIT_STATES]; + bool do_urgent_latency_adjustment; + double urgent_latency_adjustment_fabric_clock_component_us; + double urgent_latency_adjustment_fabric_clock_reference_mhz; }; struct _vcs_dpi_ip_params_st { diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c index 4d2a1262d9db..65cf4edddaff 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c @@ -262,6 +262,13 @@ static void fetch_socbb_params(struct display_mode_lib *mode_lib) //mode_lib->vba.DRAMSpeedPerState[i] = soc->clock_limits[i].dram_speed_mhz; mode_lib->vba.MaxDispclk[i] = soc->clock_limits[i].dispclk_mhz; } + + mode_lib->vba.DoUrgentLatencyAdjustment = + soc->do_urgent_latency_adjustment; + mode_lib->vba.UrgentLatencyAdjustmentFabricClockComponent = + soc->urgent_latency_adjustment_fabric_clock_component_us; + mode_lib->vba.UrgentLatencyAdjustmentFabricClockReference = + soc->urgent_latency_adjustment_fabric_clock_reference_mhz; } static void fetch_ip_params(struct display_mode_lib *mode_lib) @@ -385,8 +392,10 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) src->viewport_y_c; mode_lib->vba.PitchY[mode_lib->vba.NumberOfActivePlanes] = src->data_pitch; mode_lib->vba.SurfaceHeightY[mode_lib->vba.NumberOfActivePlanes] = src->viewport_height; + mode_lib->vba.SurfaceWidthY[mode_lib->vba.NumberOfActivePlanes] = src->viewport_width; mode_lib->vba.PitchC[mode_lib->vba.NumberOfActivePlanes] = src->data_pitch_c; mode_lib->vba.SurfaceHeightC[mode_lib->vba.NumberOfActivePlanes] = src->viewport_height_c; + mode_lib->vba.SurfaceWidthC[mode_lib->vba.NumberOfActivePlanes] = src->viewport_width_c; mode_lib->vba.DCCMetaPitchY[mode_lib->vba.NumberOfActivePlanes] = src->meta_pitch; mode_lib->vba.DCCMetaPitchC[mode_lib->vba.NumberOfActivePlanes] = src->meta_pitch_c; mode_lib->vba.HRatio[mode_lib->vba.NumberOfActivePlanes] = scl->hscl_ratio; @@ -457,6 +466,10 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) dout->wb.wb_dst_width; mode_lib->vba.WritebackDestinationHeight[mode_lib->vba.NumberOfActivePlanes] = dout->wb.wb_dst_height; + mode_lib->vba.WritebackHRatio[mode_lib->vba.NumberOfActivePlanes] = + dout->wb.wb_hratio; + mode_lib->vba.WritebackVRatio[mode_lib->vba.NumberOfActivePlanes] = + dout->wb.wb_vratio; mode_lib->vba.WritebackPixelFormat[mode_lib->vba.NumberOfActivePlanes] = (enum source_format_class) (dout->wb.wb_pixel_format); mode_lib->vba.WritebackHTaps[mode_lib->vba.NumberOfActivePlanes] = @@ -568,6 +581,7 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) if (src->is_hsplit) { for (k = j + 1; k < mode_lib->vba.cache_num_pipes; ++k) { display_pipe_source_params_st *src_k = &pipes[k].pipe.src; + display_pipe_dest_params_st *dst_k = &pipes[k].pipe.dest; if (src_k->is_hsplit && !visited[k] && src->hsplit_grp == src_k->hsplit_grp) { @@ -575,12 +589,15 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.NumberOfActivePlanes; mode_lib->vba.DPPPerPlane[mode_lib->vba.NumberOfActivePlanes]++; if (mode_lib->vba.SourceScan[mode_lib->vba.NumberOfActivePlanes] - == dm_horz) + == dm_horz) { mode_lib->vba.ViewportWidth[mode_lib->vba.NumberOfActivePlanes] += src_k->viewport_width; - else + mode_lib->vba.ScalerRecoutWidth[mode_lib->vba.NumberOfActivePlanes] += + dst_k->recout_width; + } else { mode_lib->vba.ViewportHeight[mode_lib->vba.NumberOfActivePlanes] += src_k->viewport_height; + } visited[k] = true; } diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h index 0347f74cda3a..91decac50557 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h @@ -290,6 +290,7 @@ struct vba_vars_st { double PixelClock[DC__NUM_DPP__MAX]; double PixelClockBackEnd[DC__NUM_DPP__MAX]; bool DCCEnable[DC__NUM_DPP__MAX]; + bool FECEnable[DC__NUM_DPP__MAX]; unsigned int DCCMetaPitchY[DC__NUM_DPP__MAX]; unsigned int DCCMetaPitchC[DC__NUM_DPP__MAX]; enum scan_direction_class SourceScan[DC__NUM_DPP__MAX]; @@ -317,6 +318,7 @@ struct vba_vars_st { double DCCRate[DC__NUM_DPP__MAX]; double AverageDCCCompressionRate; bool ODMCombineEnabled[DC__NUM_DPP__MAX]; + enum odm_combine_mode ODMCombineTypeEnabled[DC__NUM_DPP__MAX]; double OutputBpp[DC__NUM_DPP__MAX]; bool DSCEnabled[DC__NUM_DPP__MAX]; unsigned int DSCInputBitPerComponent[DC__NUM_DPP__MAX]; @@ -395,6 +397,7 @@ struct vba_vars_st { double FabricClockPerState[DC__VOLTAGE_STATES + 1]; double SOCCLKPerState[DC__VOLTAGE_STATES + 1]; double PHYCLKPerState[DC__VOLTAGE_STATES + 1]; + double DTBCLKPerState[DC__VOLTAGE_STATES + 1]; double MaxDppclk[DC__VOLTAGE_STATES + 1]; double MaxDSCCLK[DC__VOLTAGE_STATES + 1]; double DRAMSpeedPerState[DC__VOLTAGE_STATES + 1]; @@ -488,6 +491,7 @@ struct vba_vars_st { unsigned int NoOfDPP[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; int NoOfDPPThisState[DC__NUM_DPP__MAX]; bool ODMCombineEnablePerState[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; + enum odm_combine_mode ODMCombineTypeEnablePerState[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; unsigned int SwathWidthYThisState[DC__NUM_DPP__MAX]; unsigned int SwathHeightCPerState[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; unsigned int SwathHeightYThisState[DC__NUM_DPP__MAX]; @@ -513,6 +517,7 @@ struct vba_vars_st { bool DIOSupport[DC__VOLTAGE_STATES + 1]; bool NotEnoughDSCUnits[DC__VOLTAGE_STATES + 1]; bool DSCCLKRequiredMoreThanSupported[DC__VOLTAGE_STATES + 1]; + bool DTBCLKRequiredMoreThanSupported[DC__VOLTAGE_STATES + 1]; double UrgentRoundTripAndOutOfOrderLatencyPerState[DC__VOLTAGE_STATES + 1]; bool ROBSupport[DC__VOLTAGE_STATES + 1]; bool PTEBufferSizeNotExceeded[DC__VOLTAGE_STATES + 1][2]; @@ -605,6 +610,7 @@ struct vba_vars_st { double MaximumSwathWidthLuma[DC__NUM_DPP__MAX]; double MaximumSwathWidthChroma[DC__NUM_DPP__MAX]; bool odm_combine_dummy[DC__NUM_DPP__MAX]; + enum odm_combine_mode odm_combine_mode_dummy[DC__NUM_DPP__MAX]; double dummy1[DC__NUM_DPP__MAX]; double dummy2[DC__NUM_DPP__MAX]; double dummy3[DC__NUM_DPP__MAX]; @@ -625,6 +631,11 @@ struct vba_vars_st { unsigned int dummyinteger10; unsigned int dummyinteger11; unsigned int dummyinteger12; + unsigned int dummyintegerarr1[DC__NUM_DPP__MAX]; + unsigned int dummyintegerarr2[DC__NUM_DPP__MAX]; + unsigned int dummyintegerarr3[DC__NUM_DPP__MAX]; + unsigned int dummyintegerarr4[DC__NUM_DPP__MAX]; + long dummylongarr1[DC__NUM_DPP__MAX]; bool dummysinglestring; bool SingleDPPViewportSizeSupportPerPlane[DC__NUM_DPP__MAX]; double PlaneRequiredDISPCLKWithODMCombine2To1; @@ -633,6 +644,7 @@ struct vba_vars_st { bool LinkDSCEnable; bool ODMCombine4To1SupportCheckOK[DC__VOLTAGE_STATES + 1]; bool ODMCombineEnableThisState[DC__NUM_DPP__MAX]; + enum odm_combine_mode ODMCombineEnableTypeThisState[DC__NUM_DPP__MAX]; unsigned int SwathWidthCThisState[DC__NUM_DPP__MAX]; bool ViewportSizeSupportPerPlane[DC__NUM_DPP__MAX]; double AlignedDCCMetaPitchY[DC__NUM_DPP__MAX]; @@ -641,6 +653,7 @@ struct vba_vars_st { unsigned int NotEnoughUrgentLatencyHiding; unsigned int NotEnoughUrgentLatencyHidingPre; long PTEBufferSizeInRequestsForLuma; + long PTEBufferSizeInRequestsForChroma; // Missing from VBA long dpte_group_bytes_chroma; @@ -787,6 +800,9 @@ struct vba_vars_st { unsigned int PDEProcessingBufIn64KBReqs; double MaxTotalVActiveRDBandwidth; + bool DoUrgentLatencyAdjustment; + double UrgentLatencyAdjustmentFabricClockComponent; + double UrgentLatencyAdjustmentFabricClockReference; double MinUrgentLatencySupportUs; double MinFullDETBufferingTime; double AverageReadBandwidthGBytePerSecond; @@ -801,6 +817,8 @@ struct vba_vars_st { bool ModeIsSupported; bool ODMCombine4To1Supported; + unsigned int SurfaceWidthY[DC__NUM_DPP__MAX]; + unsigned int SurfaceWidthC[DC__NUM_DPP__MAX]; unsigned int SurfaceHeightY[DC__NUM_DPP__MAX]; unsigned int SurfaceHeightC[DC__NUM_DPP__MAX]; unsigned int WritebackHTaps[DC__NUM_DPP__MAX]; diff --git a/drivers/gpu/drm/amd/display/dc/dsc/Makefile b/drivers/gpu/drm/amd/display/dc/dsc/Makefile index e019cd9447e8..b456cd23c6fa 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile @@ -9,10 +9,13 @@ endif dsc_ccflags := -mhard-float -msse $(cc_stack_align) -CFLAGS_rc_calc.o := $(dsc_ccflags) -CFLAGS_rc_calc_dpi.o := $(dsc_ccflags) -CFLAGS_codec_main_amd.o := $(dsc_ccflags) -CFLAGS_dc_dsc.o := $(dsc_ccflags) +ifdef CONFIG_CC_IS_CLANG +dsc_ccflags += -msse2 +endif + +CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc_dpi.o := $(dsc_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dsc/dc_dsc.o := $(dsc_ccflags) DSC = dc_dsc.o rc_calc.o rc_calc_dpi.o diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c index ef5f84a144c3..5995bcdfed54 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -28,6 +28,23 @@ #include "dsc.h" #include <drm/drm_dp_helper.h> +struct dc_dsc_policy { + bool use_min_slices_h; + int max_slices_h; // Maximum available if 0 + int min_sice_height; // Must not be less than 8 + int max_target_bpp; + int min_target_bpp; // Minimum target bits per pixel +}; + +const struct dc_dsc_policy dsc_policy = { + .use_min_slices_h = true, // DSC Policy: Use minimum number of slices that fits the pixel clock + .max_slices_h = 0, // DSC Policy: Use max available slices (in our case 4 for or 8, depending on the mode) + .min_sice_height = 108, // DSC Policy: Use slice height recommended by VESA DSC Spreadsheet user guide + .max_target_bpp = 16, + .min_target_bpp = 8, +}; + + /* This module's internal functions */ static bool dsc_buff_block_size_from_dpcd(int dpcd_buff_block_size, int *buff_block_size) @@ -241,14 +258,6 @@ static bool intersect_dsc_caps( return true; } -struct dc_dsc_policy { - bool use_min_slices_h; - int max_slices_h; // Maximum available if 0 - int num_slices_v; - int max_target_bpp; - int min_target_bpp; // Minimum target bits per pixel -}; - static inline uint32_t dsc_div_by_10_round_up(uint32_t value) { return (value + 9) / 10; @@ -270,19 +279,6 @@ static inline uint32_t calc_dsc_bpp_x16(uint32_t stream_bandwidth_kbps, uint32_t return dsc_target_bpp_x16; } -const struct dc_dsc_policy dsc_policy = { - .use_min_slices_h = true, // DSC Policy: Use minimum number of slices that fits the pixel clock - .max_slices_h = 0, // DSC Policy: Use max available slices (in our case 4 for or 8, depending on the mode) - /* DSC Policy: Number of vertical slices set to 2 for no particular reason. - * Seems small enough to not affect the quality too much, while still providing some error - * propagation control (which may also help debugging). - */ - .num_slices_v = 16, - .max_target_bpp = 16, - .min_target_bpp = 8, -}; - - /* Get DSC bandwidth range based on [min_bpp, max_bpp] target bitrate range, and timing's pixel clock * and uncompressed bandwidth. */ @@ -528,8 +524,8 @@ static bool setup_dsc_config( int sink_per_slice_throughput_mps; int branch_max_throughput_mps = 0; bool is_dsc_possible = false; - int num_slices_v; int pic_height; + int slice_height; memset(dsc_cfg, 0, sizeof(struct dc_dsc_config)); @@ -615,7 +611,7 @@ static bool setup_dsc_config( if (!is_dsc_possible) goto done; - // DSC slicing + // Slice width (i.e. number of slices per line) max_slices_h = get_max_dsc_slices(dsc_common_caps.slice_caps); while (max_slices_h > 0) { @@ -678,29 +674,26 @@ static bool setup_dsc_config( dsc_cfg->num_slices_h = num_slices_h; slice_width = pic_width / num_slices_h; - // Vertical number of slices: start from policy and pick the first one that height is divisible by. + is_dsc_possible = slice_width <= dsc_common_caps.max_slice_width; + if (!is_dsc_possible) + goto done; + + // Slice height (i.e. number of slices per column): start with policy and pick the first one that height is divisible by. // For 4:2:0 make sure the slice height is divisible by 2 as well. - num_slices_v = dsc_policy.num_slices_v; - if (num_slices_v < 1) - num_slices_v = 1; - - while (num_slices_v >= 1) { - if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) { - int slice_height = pic_height / num_slices_v; - if (pic_height % num_slices_v == 0 && slice_height % 2 == 0) - break; - } else if (pic_height % num_slices_v == 0) - break; + slice_height = min(dsc_policy.min_sice_height, pic_height); - num_slices_v--; - } + while (slice_height < pic_height && (pic_height % slice_height != 0 || + (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420 && slice_height % 2 != 0))) + slice_height++; - dsc_cfg->num_slices_v = num_slices_v; + if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) // For the case when pic_height < dsc_policy.min_sice_height + is_dsc_possible = (slice_height % 2 == 0); - is_dsc_possible = slice_width <= dsc_common_caps.max_slice_width; if (!is_dsc_possible) goto done; + dsc_cfg->num_slices_v = pic_height/slice_height; + // Final decission: can we do DSC or not? if (is_dsc_possible) { // Fill out the rest of DSC settings diff --git a/drivers/gpu/drm/amd/display/dc/dsc/drm_dsc_dc.c b/drivers/gpu/drm/amd/display/dc/dsc/drm_dsc_dc.c deleted file mode 100644 index 340ef4d41ebd..000000000000 --- a/drivers/gpu/drm/amd/display/dc/dsc/drm_dsc_dc.c +++ /dev/null @@ -1,388 +0,0 @@ -// SPDX-License-Identifier: MIT -/* - * Copyright © 2018 Intel Corp - * - * Author: - * Manasi Navare <manasi.d.navare@intel.com> - */ - -/* DC versions of linux includes */ -#include <include/drm_dsc_dc.h> - -#define EXPORT_SYMBOL(symbol) /* nothing */ -#define BUILD_BUG_ON(cond) /* nothing */ -#define DIV_ROUND_UP(a, b) (((b) + (a) - 1) / (b)) -#define ERANGE -1 -#define DRM_DEBUG_KMS(msg) /* nothing */ -#define cpu_to_be16(__x) little_to_big(__x) - -static unsigned short little_to_big(int data) -{ - /* Swap lower and upper byte. DMCU uses big endian format. */ - return (0xff & (data >> 8)) + ((data & 0xff) << 8); -} - -/* - * Everything below this comment was copied directly from drm_dsc.c. - * Only the functions needed in DC are included. - * Please keep this file synced with upstream. - */ - -/** - * DOC: dsc helpers - * - * These functions contain some common logic and helpers to deal with VESA - * Display Stream Compression standard required for DSC on Display Port/eDP or - * MIPI display interfaces. - */ - -/** - * drm_dsc_pps_payload_pack() - Populates the DSC PPS - * - * @pps_payload: - * Bitwise struct for DSC Picture Parameter Set. This is defined - * by &struct drm_dsc_picture_parameter_set - * @dsc_cfg: - * DSC Configuration data filled by driver as defined by - * &struct drm_dsc_config - * - * DSC source device sends a picture parameter set (PPS) containing the - * information required by the sink to decode the compressed frame. Driver - * populates the DSC PPS struct using the DSC configuration parameters in - * the order expected by the DSC Display Sink device. For the DSC, the sink - * device expects the PPS payload in big endian format for fields - * that span more than 1 byte. - */ -void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_payload, - const struct drm_dsc_config *dsc_cfg) -{ - int i; - - /* Protect against someone accidently changing struct size */ - BUILD_BUG_ON(sizeof(*pps_payload) != - DP_SDP_PPS_HEADER_PAYLOAD_BYTES_MINUS_1 + 1); - - memset(pps_payload, 0, sizeof(*pps_payload)); - - /* PPS 0 */ - pps_payload->dsc_version = - dsc_cfg->dsc_version_minor | - dsc_cfg->dsc_version_major << DSC_PPS_VERSION_MAJOR_SHIFT; - - /* PPS 1, 2 is 0 */ - - /* PPS 3 */ - pps_payload->pps_3 = - dsc_cfg->line_buf_depth | - dsc_cfg->bits_per_component << DSC_PPS_BPC_SHIFT; - - /* PPS 4 */ - pps_payload->pps_4 = - ((dsc_cfg->bits_per_pixel & DSC_PPS_BPP_HIGH_MASK) >> - DSC_PPS_MSB_SHIFT) | - dsc_cfg->vbr_enable << DSC_PPS_VBR_EN_SHIFT | - dsc_cfg->simple_422 << DSC_PPS_SIMPLE422_SHIFT | - dsc_cfg->convert_rgb << DSC_PPS_CONVERT_RGB_SHIFT | - dsc_cfg->block_pred_enable << DSC_PPS_BLOCK_PRED_EN_SHIFT; - - /* PPS 5 */ - pps_payload->bits_per_pixel_low = - (dsc_cfg->bits_per_pixel & DSC_PPS_LSB_MASK); - - /* - * The DSC panel expects the PPS packet to have big endian format - * for data spanning 2 bytes. Use a macro cpu_to_be16() to convert - * to big endian format. If format is little endian, it will swap - * bytes to convert to Big endian else keep it unchanged. - */ - - /* PPS 6, 7 */ - pps_payload->pic_height = cpu_to_be16(dsc_cfg->pic_height); - - /* PPS 8, 9 */ - pps_payload->pic_width = cpu_to_be16(dsc_cfg->pic_width); - - /* PPS 10, 11 */ - pps_payload->slice_height = cpu_to_be16(dsc_cfg->slice_height); - - /* PPS 12, 13 */ - pps_payload->slice_width = cpu_to_be16(dsc_cfg->slice_width); - - /* PPS 14, 15 */ - pps_payload->chunk_size = cpu_to_be16(dsc_cfg->slice_chunk_size); - - /* PPS 16 */ - pps_payload->initial_xmit_delay_high = - ((dsc_cfg->initial_xmit_delay & - DSC_PPS_INIT_XMIT_DELAY_HIGH_MASK) >> - DSC_PPS_MSB_SHIFT); - - /* PPS 17 */ - pps_payload->initial_xmit_delay_low = - (dsc_cfg->initial_xmit_delay & DSC_PPS_LSB_MASK); - - /* PPS 18, 19 */ - pps_payload->initial_dec_delay = - cpu_to_be16(dsc_cfg->initial_dec_delay); - - /* PPS 20 is 0 */ - - /* PPS 21 */ - pps_payload->initial_scale_value = - dsc_cfg->initial_scale_value; - - /* PPS 22, 23 */ - pps_payload->scale_increment_interval = - cpu_to_be16(dsc_cfg->scale_increment_interval); - - /* PPS 24 */ - pps_payload->scale_decrement_interval_high = - ((dsc_cfg->scale_decrement_interval & - DSC_PPS_SCALE_DEC_INT_HIGH_MASK) >> - DSC_PPS_MSB_SHIFT); - - /* PPS 25 */ - pps_payload->scale_decrement_interval_low = - (dsc_cfg->scale_decrement_interval & DSC_PPS_LSB_MASK); - - /* PPS 26[7:0], PPS 27[7:5] RESERVED */ - - /* PPS 27 */ - pps_payload->first_line_bpg_offset = - dsc_cfg->first_line_bpg_offset; - - /* PPS 28, 29 */ - pps_payload->nfl_bpg_offset = - cpu_to_be16(dsc_cfg->nfl_bpg_offset); - - /* PPS 30, 31 */ - pps_payload->slice_bpg_offset = - cpu_to_be16(dsc_cfg->slice_bpg_offset); - - /* PPS 32, 33 */ - pps_payload->initial_offset = - cpu_to_be16(dsc_cfg->initial_offset); - - /* PPS 34, 35 */ - pps_payload->final_offset = cpu_to_be16(dsc_cfg->final_offset); - - /* PPS 36 */ - pps_payload->flatness_min_qp = dsc_cfg->flatness_min_qp; - - /* PPS 37 */ - pps_payload->flatness_max_qp = dsc_cfg->flatness_max_qp; - - /* PPS 38, 39 */ - pps_payload->rc_model_size = - cpu_to_be16(DSC_RC_MODEL_SIZE_CONST); - - /* PPS 40 */ - pps_payload->rc_edge_factor = DSC_RC_EDGE_FACTOR_CONST; - - /* PPS 41 */ - pps_payload->rc_quant_incr_limit0 = - dsc_cfg->rc_quant_incr_limit0; - - /* PPS 42 */ - pps_payload->rc_quant_incr_limit1 = - dsc_cfg->rc_quant_incr_limit1; - - /* PPS 43 */ - pps_payload->rc_tgt_offset = DSC_RC_TGT_OFFSET_LO_CONST | - DSC_RC_TGT_OFFSET_HI_CONST << DSC_PPS_RC_TGT_OFFSET_HI_SHIFT; - - /* PPS 44 - 57 */ - for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++) - pps_payload->rc_buf_thresh[i] = - dsc_cfg->rc_buf_thresh[i]; - - /* PPS 58 - 87 */ - /* - * For DSC sink programming the RC Range parameter fields - * are as follows: Min_qp[15:11], max_qp[10:6], offset[5:0] - */ - for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { - pps_payload->rc_range_parameters[i] = - ((dsc_cfg->rc_range_params[i].range_min_qp << - DSC_PPS_RC_RANGE_MINQP_SHIFT) | - (dsc_cfg->rc_range_params[i].range_max_qp << - DSC_PPS_RC_RANGE_MAXQP_SHIFT) | - (dsc_cfg->rc_range_params[i].range_bpg_offset)); - pps_payload->rc_range_parameters[i] = - cpu_to_be16(pps_payload->rc_range_parameters[i]); - } - - /* PPS 88 */ - pps_payload->native_422_420 = dsc_cfg->native_422 | - dsc_cfg->native_420 << DSC_PPS_NATIVE_420_SHIFT; - - /* PPS 89 */ - pps_payload->second_line_bpg_offset = - dsc_cfg->second_line_bpg_offset; - - /* PPS 90, 91 */ - pps_payload->nsl_bpg_offset = - cpu_to_be16(dsc_cfg->nsl_bpg_offset); - - /* PPS 92, 93 */ - pps_payload->second_line_offset_adj = - cpu_to_be16(dsc_cfg->second_line_offset_adj); - - /* PPS 94 - 127 are O */ -} -EXPORT_SYMBOL(drm_dsc_pps_payload_pack); - -/** - * drm_dsc_compute_rc_parameters() - Write rate control - * parameters to the dsc configuration defined in - * &struct drm_dsc_config in accordance with the DSC 1.2 - * specification. Some configuration fields must be present - * beforehand. - * - * @vdsc_cfg: - * DSC Configuration data partially filled by driver - */ -int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg) -{ - unsigned long groups_per_line = 0; - unsigned long groups_total = 0; - unsigned long num_extra_mux_bits = 0; - unsigned long slice_bits = 0; - unsigned long hrd_delay = 0; - unsigned long final_scale = 0; - unsigned long rbs_min = 0; - - if (vdsc_cfg->native_420 || vdsc_cfg->native_422) { - /* Number of groups used to code each line of a slice */ - groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width / 2, - DSC_RC_PIXELS_PER_GROUP); - - /* chunksize in Bytes */ - vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width / 2 * - vdsc_cfg->bits_per_pixel, - (8 * 16)); - } else { - /* Number of groups used to code each line of a slice */ - groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width, - DSC_RC_PIXELS_PER_GROUP); - - /* chunksize in Bytes */ - vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width * - vdsc_cfg->bits_per_pixel, - (8 * 16)); - } - - if (vdsc_cfg->convert_rgb) - num_extra_mux_bits = 3 * (vdsc_cfg->mux_word_size + - (4 * vdsc_cfg->bits_per_component + 4) - - 2); - else if (vdsc_cfg->native_422) - num_extra_mux_bits = 4 * vdsc_cfg->mux_word_size + - (4 * vdsc_cfg->bits_per_component + 4) + - 3 * (4 * vdsc_cfg->bits_per_component) - 2; - else - num_extra_mux_bits = 3 * vdsc_cfg->mux_word_size + - (4 * vdsc_cfg->bits_per_component + 4) + - 2 * (4 * vdsc_cfg->bits_per_component) - 2; - /* Number of bits in one Slice */ - slice_bits = 8 * vdsc_cfg->slice_chunk_size * vdsc_cfg->slice_height; - - while ((num_extra_mux_bits > 0) && - ((slice_bits - num_extra_mux_bits) % vdsc_cfg->mux_word_size)) - num_extra_mux_bits--; - - if (groups_per_line < vdsc_cfg->initial_scale_value - 8) - vdsc_cfg->initial_scale_value = groups_per_line + 8; - - /* scale_decrement_interval calculation according to DSC spec 1.11 */ - if (vdsc_cfg->initial_scale_value > 8) - vdsc_cfg->scale_decrement_interval = groups_per_line / - (vdsc_cfg->initial_scale_value - 8); - else - vdsc_cfg->scale_decrement_interval = DSC_SCALE_DECREMENT_INTERVAL_MAX; - - vdsc_cfg->final_offset = vdsc_cfg->rc_model_size - - (vdsc_cfg->initial_xmit_delay * - vdsc_cfg->bits_per_pixel + 8) / 16 + num_extra_mux_bits; - - if (vdsc_cfg->final_offset >= vdsc_cfg->rc_model_size) { - DRM_DEBUG_KMS("FinalOfs < RcModelSze for this InitialXmitDelay\n"); - return -ERANGE; - } - - final_scale = (vdsc_cfg->rc_model_size * 8) / - (vdsc_cfg->rc_model_size - vdsc_cfg->final_offset); - if (vdsc_cfg->slice_height > 1) - /* - * NflBpgOffset is 16 bit value with 11 fractional bits - * hence we multiply by 2^11 for preserving the - * fractional part - */ - vdsc_cfg->nfl_bpg_offset = DIV_ROUND_UP((vdsc_cfg->first_line_bpg_offset << 11), - (vdsc_cfg->slice_height - 1)); - else - vdsc_cfg->nfl_bpg_offset = 0; - - /* 2^16 - 1 */ - if (vdsc_cfg->nfl_bpg_offset > 65535) { - DRM_DEBUG_KMS("NflBpgOffset is too large for this slice height\n"); - return -ERANGE; - } - - /* Number of groups used to code the entire slice */ - groups_total = groups_per_line * vdsc_cfg->slice_height; - - /* slice_bpg_offset is 16 bit value with 11 fractional bits */ - vdsc_cfg->slice_bpg_offset = DIV_ROUND_UP(((vdsc_cfg->rc_model_size - - vdsc_cfg->initial_offset + - num_extra_mux_bits) << 11), - groups_total); - - if (final_scale > 9) { - /* - * ScaleIncrementInterval = - * finaloffset/((NflBpgOffset + SliceBpgOffset)*8(finalscale - 1.125)) - * as (NflBpgOffset + SliceBpgOffset) has 11 bit fractional value, - * we need divide by 2^11 from pstDscCfg values - */ - vdsc_cfg->scale_increment_interval = - (vdsc_cfg->final_offset * (1 << 11)) / - ((vdsc_cfg->nfl_bpg_offset + - vdsc_cfg->slice_bpg_offset) * - (final_scale - 9)); - } else { - /* - * If finalScaleValue is less than or equal to 9, a value of 0 should - * be used to disable the scale increment at the end of the slice - */ - vdsc_cfg->scale_increment_interval = 0; - } - - if (vdsc_cfg->scale_increment_interval > 65535) { - DRM_DEBUG_KMS("ScaleIncrementInterval is large for slice height\n"); - return -ERANGE; - } - - /* - * DSC spec mentions that bits_per_pixel specifies the target - * bits/pixel (bpp) rate that is used by the encoder, - * in steps of 1/16 of a bit per pixel - */ - rbs_min = vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset + - DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay * - vdsc_cfg->bits_per_pixel, 16) + - groups_per_line * vdsc_cfg->first_line_bpg_offset; - - hrd_delay = DIV_ROUND_UP((rbs_min * 16), vdsc_cfg->bits_per_pixel); - vdsc_cfg->rc_bits = (hrd_delay * vdsc_cfg->bits_per_pixel) / 16; - vdsc_cfg->initial_dec_delay = hrd_delay - vdsc_cfg->initial_xmit_delay; - - /* As per DSC spec v1.2a recommendation: */ - if (vdsc_cfg->native_420) - vdsc_cfg->second_line_offset_adj = 512; - else - vdsc_cfg->second_line_offset_adj = 0; - - return 0; -} -EXPORT_SYMBOL(drm_dsc_compute_rc_parameters); diff --git a/drivers/gpu/drm/amd/display/dc/gpio/Makefile b/drivers/gpu/drm/amd/display/dc/gpio/Makefile index c3d92878875d..b3062275711e 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/Makefile +++ b/drivers/gpu/drm/amd/display/dc/gpio/Makefile @@ -24,7 +24,7 @@ # It provides the control and status of HW GPIO pins. GPIO = gpio_base.o gpio_service.o hw_factory.o \ - hw_gpio.o hw_hpd.o hw_ddc.o hw_translate.o + hw_gpio.o hw_hpd.o hw_ddc.o hw_generic.o hw_translate.o AMD_DAL_GPIO = $(addprefix $(AMDDALPATH)/dc/gpio/,$(GPIO)) @@ -80,6 +80,13 @@ AMD_DAL_GPIO_DCN20 = $(addprefix $(AMDDALPATH)/dc/gpio/dcn20/,$(GPIO_DCN20)) AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCN20) endif +ifdef CONFIG_DRM_AMD_DC_DCN2_1 +GPIO_DCN21 = hw_translate_dcn21.o hw_factory_dcn21.o + +AMD_DAL_GPIO_DCN21 = $(addprefix $(AMDDALPATH)/dc/gpio/dcn21/,$(GPIO_DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCN21) +endif ############################################################################### # Diagnostics on FPGA ############################################################################### diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c b/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c index 20d81bca119c..66e4841f41e4 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c @@ -24,9 +24,15 @@ */ #include "dm_services.h" + #include "include/gpio_types.h" #include "../hw_factory.h" +#include "../hw_gpio.h" +#include "../hw_ddc.h" +#include "../hw_hpd.h" +#include "../hw_generic.h" + #include "hw_factory_dce110.h" #include "dce/dce_11_0_d.h" @@ -143,12 +149,12 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) } static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = NULL, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = NULL, .define_hpd_registers = define_hpd_registers, .define_ddc_registers = define_ddc_registers }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c b/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c index ea3f888e5c65..cf98aa827a9a 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c @@ -27,10 +27,10 @@ #include "include/gpio_types.h" #include "../hw_factory.h" - #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "hw_factory_dce120.h" @@ -164,12 +164,12 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) /* fucntion table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = NULL, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = NULL, .define_hpd_registers = define_hpd_registers, .define_ddc_registers = define_ddc_registers }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c b/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c index 48b67866377e..496d3ffb74bb 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c @@ -32,10 +32,12 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "dce/dce_8_0_d.h" #include "dce/dce_8_0_sh_mask.h" + #define REG(reg_name)\ mm ## reg_name @@ -147,12 +149,12 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) } static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = NULL, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = NULL, .define_hpd_registers = define_hpd_registers, .define_ddc_registers = define_ddc_registers }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c index 32aa47a04a0d..b38c96c9fed3 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c @@ -31,6 +31,7 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "hw_factory_dcn10.h" @@ -121,6 +122,42 @@ static const struct ddc_sh_mask ddc_mask = { DDC_MASK_SH_LIST(_MASK) }; +#include "../generic_regs.h" + +/* set field name */ +#define SF_GENERIC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define generic_regs(id) \ +{\ + GENERIC_REG_LIST(id)\ +} + +static const struct generic_registers generic_regs[] = { + generic_regs(A), + generic_regs(B), +}; + +static const struct generic_sh_mask generic_shift[] = { + GENERIC_MASK_SH_LIST(__SHIFT, A), + GENERIC_MASK_SH_LIST(__SHIFT, B), +}; + +static const struct generic_sh_mask generic_mask[] = { + GENERIC_MASK_SH_LIST(_MASK, A), + GENERIC_MASK_SH_LIST(_MASK, B), +}; + +static void define_generic_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(pin); + + generic->regs = &generic_regs[en]; + generic->shifts = &generic_shift[en]; + generic->masks = &generic_mask[en]; + generic->base.regs = &generic_regs[en].gpio; +} + static void define_ddc_registers( struct hw_gpio_pin *pin, uint32_t en) @@ -159,14 +196,15 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) /* fucntion table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = dal_hw_generic_init, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = dal_hw_generic_get_pin, .define_hpd_registers = define_hpd_registers, - .define_ddc_registers = define_ddc_registers + .define_ddc_registers = define_ddc_registers, + .define_generic_registers = define_generic_registers }; /* * dal_hw_factory_dcn10_init diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c index abd76d855375..43a440385b43 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c @@ -31,6 +31,7 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "hw_factory_dcn20.h" @@ -138,6 +139,32 @@ static const struct ddc_sh_mask ddc_mask[] = { DDC_MASK_SH_LIST_DCN2(_MASK, 6) }; +#include "../generic_regs.h" + +/* set field name */ +#define SF_GENERIC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define generic_regs(id) \ +{\ + GENERIC_REG_LIST(id)\ +} + +static const struct generic_registers generic_regs[] = { + generic_regs(A), + generic_regs(B), +}; + +static const struct generic_sh_mask generic_shift[] = { + GENERIC_MASK_SH_LIST(__SHIFT, A), + GENERIC_MASK_SH_LIST(__SHIFT, B), +}; + +static const struct generic_sh_mask generic_mask[] = { + GENERIC_MASK_SH_LIST(_MASK, A), + GENERIC_MASK_SH_LIST(_MASK, B), +}; + static void define_ddc_registers( struct hw_gpio_pin *pin, uint32_t en) @@ -173,17 +200,27 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) hpd->base.regs = &hpd_regs[en].gpio; } +static void define_generic_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(pin); + + generic->regs = &generic_regs[en]; + generic->shifts = &generic_shift[en]; + generic->masks = &generic_mask[en]; + generic->base.regs = &generic_regs[en].gpio; +} /* fucntion table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = dal_hw_generic_init, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = dal_hw_generic_get_pin, .define_hpd_registers = define_hpd_registers, - .define_ddc_registers = define_ddc_registers + .define_ddc_registers = define_ddc_registers, + .define_generic_registers = define_generic_registers, }; /* * dal_hw_factory_dcn10_init diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c index b393cc13298a..915e896e0e91 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c @@ -71,7 +71,7 @@ static bool offset_to_id( { switch (offset) { /* GENERIC */ - case REG(DC_GENERICA): + case REG(DC_GPIO_GENERIC_A): *id = GPIO_ID_GENERIC; switch (mask) { case DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK: diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c new file mode 100644 index 000000000000..8572678f8d4f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c @@ -0,0 +1,242 @@ +/* + * Copyright 2013-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#if defined(CONFIG_DRM_AMD_DC_DCN2_0) +#include "dm_services.h" +#include "include/gpio_types.h" +#include "../hw_factory.h" + + +#include "../hw_gpio.h" +#include "../hw_ddc.h" +#include "../hw_hpd.h" +#include "../hw_generic.h" + +#include "hw_factory_dcn21.h" + +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" +#include "renoir_ip_offset.h" + +#include "reg_helper.h" +#include "../hpd_regs.h" +/* begin ********************* + * macros to expend register list macro defined in HW object header file */ + +/* DCN */ +#define block HPD +#define reg_num 0 + +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +#define BASE(seg) BASE_INNER(seg) + + + +#define REG(reg_name)\ + BASE(mm ## reg_name ## _BASE_IDX) + mm ## reg_name + +#define SF_HPD(reg_name, field_name, post_fix)\ + .field_name = HPD0_ ## reg_name ## __ ## field_name ## post_fix + +#define REGI(reg_name, block, id)\ + BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define SF(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +/* macros to expend register list macro defined in HW object header file + * end *********************/ + + + +#define hpd_regs(id) \ +{\ + HPD_REG_LIST(id)\ +} + +static const struct hpd_registers hpd_regs[] = { + hpd_regs(0), + hpd_regs(1), + hpd_regs(2), + hpd_regs(3), + hpd_regs(4), +}; + +static const struct hpd_sh_mask hpd_shift = { + HPD_MASK_SH_LIST(__SHIFT) +}; + +static const struct hpd_sh_mask hpd_mask = { + HPD_MASK_SH_LIST(_MASK) +}; + +#include "../ddc_regs.h" + + /* set field name */ +#define SF_DDC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +static const struct ddc_registers ddc_data_regs_dcn[] = { + ddc_data_regs_dcn2(1), + ddc_data_regs_dcn2(2), + ddc_data_regs_dcn2(3), + ddc_data_regs_dcn2(4), + ddc_data_regs_dcn2(5), +}; + +static const struct ddc_registers ddc_clk_regs_dcn[] = { + ddc_clk_regs_dcn2(1), + ddc_clk_regs_dcn2(2), + ddc_clk_regs_dcn2(3), + ddc_clk_regs_dcn2(4), + ddc_clk_regs_dcn2(5), +}; + +static const struct ddc_sh_mask ddc_shift[] = { + DDC_MASK_SH_LIST_DCN2(__SHIFT, 1), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 2), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 3), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 4), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 5), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 6) +}; + +static const struct ddc_sh_mask ddc_mask[] = { + DDC_MASK_SH_LIST_DCN2(_MASK, 1), + DDC_MASK_SH_LIST_DCN2(_MASK, 2), + DDC_MASK_SH_LIST_DCN2(_MASK, 3), + DDC_MASK_SH_LIST_DCN2(_MASK, 4), + DDC_MASK_SH_LIST_DCN2(_MASK, 5), + DDC_MASK_SH_LIST_DCN2(_MASK, 6) +}; + +#include "../generic_regs.h" + +/* set field name */ +#define SF_GENERIC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define generic_regs(id) \ +{\ + GENERIC_REG_LIST(id)\ +} + +static const struct generic_registers generic_regs[] = { + generic_regs(A), +}; + +static const struct generic_sh_mask generic_shift[] = { + GENERIC_MASK_SH_LIST(__SHIFT, A), +}; + +static const struct generic_sh_mask generic_mask[] = { + GENERIC_MASK_SH_LIST(_MASK, A), +}; + +static void define_generic_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(pin); + + generic->regs = &generic_regs[en]; + generic->shifts = &generic_shift[en]; + generic->masks = &generic_mask[en]; + generic->base.regs = &generic_regs[en].gpio; +} + +static void define_ddc_registers( + struct hw_gpio_pin *pin, + uint32_t en) +{ + struct hw_ddc *ddc = HW_DDC_FROM_BASE(pin); + + switch (pin->id) { + case GPIO_ID_DDC_DATA: + ddc->regs = &ddc_data_regs_dcn[en]; + ddc->base.regs = &ddc_data_regs_dcn[en].gpio; + break; + case GPIO_ID_DDC_CLOCK: + ddc->regs = &ddc_clk_regs_dcn[en]; + ddc->base.regs = &ddc_clk_regs_dcn[en].gpio; + break; + default: + ASSERT_CRITICAL(false); + return; + } + + ddc->shifts = &ddc_shift[en]; + ddc->masks = &ddc_mask[en]; + +} + +static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_hpd *hpd = HW_HPD_FROM_BASE(pin); + + hpd->regs = &hpd_regs[en]; + hpd->shifts = &hpd_shift; + hpd->masks = &hpd_mask; + hpd->base.regs = &hpd_regs[en].gpio; +} + + +/* fucntion table */ +static const struct hw_factory_funcs funcs = { + .init_ddc_data = dal_hw_ddc_init, + .init_generic = dal_hw_generic_init, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = dal_hw_generic_get_pin, + .define_hpd_registers = define_hpd_registers, + .define_ddc_registers = define_ddc_registers, + .define_generic_registers = define_generic_registers +}; +/* + * dal_hw_factory_dcn10_init + * + * @brief + * Initialize HW factory function pointers and pin info + * + * @param + * struct hw_factory *factory - [out] struct of function pointers + */ +void dal_hw_factory_dcn21_init(struct hw_factory *factory) +{ + /*TODO check ASIC CAPs*/ + factory->number_of_pins[GPIO_ID_DDC_DATA] = 8; + factory->number_of_pins[GPIO_ID_DDC_CLOCK] = 8; + factory->number_of_pins[GPIO_ID_GENERIC] = 4; + factory->number_of_pins[GPIO_ID_HPD] = 6; + factory->number_of_pins[GPIO_ID_GPIO_PAD] = 28; + factory->number_of_pins[GPIO_ID_VIP_PAD] = 0; + factory->number_of_pins[GPIO_ID_SYNC] = 0; + factory->number_of_pins[GPIO_ID_GSL] = 0;/*add this*/ + + factory->funcs = &funcs; +} + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.h b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.h new file mode 100644 index 000000000000..2443f9e7afbf --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.h @@ -0,0 +1,33 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#ifndef __DAL_HW_FACTORY_DCN21_H__ +#define __DAL_HW_FACTORY_DCN21_H__ + +/* Initialize HW factory function pointers and pin info */ +void dal_hw_factory_dcn21_init(struct hw_factory *factory); + +#endif /* __DAL_HW_FACTORY_DCN20_H__ */ +#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.c new file mode 100644 index 000000000000..fbb58fb8c318 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.c @@ -0,0 +1,385 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +/* + * Pre-requisites: headers required by header of this unit + */ +#if defined(CONFIG_DRM_AMD_DC_DCN2_0) +#include "hw_translate_dcn21.h" + +#include "dm_services.h" +#include "include/gpio_types.h" +#include "../hw_translate.h" + +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" +#include "renoir_ip_offset.h" + + + + +/* begin ********************* + * macros to expend register list macro defined in HW object header file */ + +/* DCN */ +#define block HPD +#define reg_num 0 + +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +#define BASE(seg) BASE_INNER(seg) + +#undef REG +#define REG(reg_name)\ + BASE(mm ## reg_name ## _BASE_IDX) + mm ## reg_name +#define SF_HPD(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +/* macros to expend register list macro defined in HW object header file + * end *********************/ + + +static bool offset_to_id( + uint32_t offset, + uint32_t mask, + enum gpio_id *id, + uint32_t *en) +{ + switch (offset) { + /* GENERIC */ + case REG(DC_GPIO_GENERIC_A): + *id = GPIO_ID_GENERIC; + switch (mask) { + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK: + *en = GPIO_GENERIC_A; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICB_A_MASK: + *en = GPIO_GENERIC_B; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICC_A_MASK: + *en = GPIO_GENERIC_C; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICD_A_MASK: + *en = GPIO_GENERIC_D; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICE_A_MASK: + *en = GPIO_GENERIC_E; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICF_A_MASK: + *en = GPIO_GENERIC_F; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICG_A_MASK: + *en = GPIO_GENERIC_G; + return true; + default: + ASSERT_CRITICAL(false); +#ifdef PALLADIUM_SUPPORTED + *en = GPIO_DDC_LINE_DDC1; + return true; +#endif + return false; + } + break; + /* HPD */ + case REG(DC_GPIO_HPD_A): + *id = GPIO_ID_HPD; + switch (mask) { + case DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK: + *en = GPIO_HPD_1; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD2_A_MASK: + *en = GPIO_HPD_2; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD3_A_MASK: + *en = GPIO_HPD_3; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD4_A_MASK: + *en = GPIO_HPD_4; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD5_A_MASK: + *en = GPIO_HPD_5; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD6_A_MASK: + *en = GPIO_HPD_6; + return true; + default: + ASSERT_CRITICAL(false); + return false; + } + break; + /* REG(DC_GPIO_GENLK_MASK */ + case REG(DC_GPIO_GENLK_A): + *id = GPIO_ID_GSL; + switch (mask) { + case DC_GPIO_GENLK_A__DC_GPIO_GENLK_CLK_A_MASK: + *en = GPIO_GSL_GENLOCK_CLOCK; + return true; + case DC_GPIO_GENLK_A__DC_GPIO_GENLK_VSYNC_A_MASK: + *en = GPIO_GSL_GENLOCK_VSYNC; + return true; + case DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_A_A_MASK: + *en = GPIO_GSL_SWAPLOCK_A; + return true; + case DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_B_A_MASK: + *en = GPIO_GSL_SWAPLOCK_B; + return true; + default: + ASSERT_CRITICAL(false); + return false; + } + break; + /* DDC */ + /* we don't care about the GPIO_ID for DDC + * in DdcHandle it will use GPIO_ID_DDC_DATA/GPIO_ID_DDC_CLOCK + * directly in the create method */ + case REG(DC_GPIO_DDC1_A): + *en = GPIO_DDC_LINE_DDC1; + return true; + case REG(DC_GPIO_DDC2_A): + *en = GPIO_DDC_LINE_DDC2; + return true; + case REG(DC_GPIO_DDC3_A): + *en = GPIO_DDC_LINE_DDC3; + return true; + case REG(DC_GPIO_DDC4_A): + *en = GPIO_DDC_LINE_DDC4; + return true; + case REG(DC_GPIO_DDC5_A): + *en = GPIO_DDC_LINE_DDC5; + return true; + case REG(DC_GPIO_DDCVGA_A): + *en = GPIO_DDC_LINE_DDC_VGA; + return true; + +// case REG(DC_GPIO_I2CPAD_A): not exit +// case REG(DC_GPIO_PWRSEQ_A): +// case REG(DC_GPIO_PAD_STRENGTH_1): +// case REG(DC_GPIO_PAD_STRENGTH_2): +// case REG(DC_GPIO_DEBUG): + /* UNEXPECTED */ + default: +// case REG(DC_GPIO_SYNCA_A): not exist +#ifdef PALLADIUM_SUPPORTED + *id = GPIO_ID_HPD; + *en = GPIO_DDC_LINE_DDC1; + return true; +#endif + ASSERT_CRITICAL(false); + return false; + } +} + +static bool id_to_offset( + enum gpio_id id, + uint32_t en, + struct gpio_pin_info *info) +{ + bool result = true; + + switch (id) { + case GPIO_ID_DDC_DATA: + info->mask = DC_GPIO_DDC5_A__DC_GPIO_DDC5DATA_A_MASK; + switch (en) { + case GPIO_DDC_LINE_DDC1: + info->offset = REG(DC_GPIO_DDC1_A); + break; + case GPIO_DDC_LINE_DDC2: + info->offset = REG(DC_GPIO_DDC2_A); + break; + case GPIO_DDC_LINE_DDC3: + info->offset = REG(DC_GPIO_DDC3_A); + break; + case GPIO_DDC_LINE_DDC4: + info->offset = REG(DC_GPIO_DDC4_A); + break; + case GPIO_DDC_LINE_DDC5: + info->offset = REG(DC_GPIO_DDC5_A); + break; + case GPIO_DDC_LINE_DDC_VGA: + info->offset = REG(DC_GPIO_DDCVGA_A); + break; + case GPIO_DDC_LINE_I2C_PAD: + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_DDC_CLOCK: + info->mask = DC_GPIO_DDC5_A__DC_GPIO_DDC5CLK_A_MASK; + switch (en) { + case GPIO_DDC_LINE_DDC1: + info->offset = REG(DC_GPIO_DDC1_A); + break; + case GPIO_DDC_LINE_DDC2: + info->offset = REG(DC_GPIO_DDC2_A); + break; + case GPIO_DDC_LINE_DDC3: + info->offset = REG(DC_GPIO_DDC3_A); + break; + case GPIO_DDC_LINE_DDC4: + info->offset = REG(DC_GPIO_DDC4_A); + break; + case GPIO_DDC_LINE_DDC5: + info->offset = REG(DC_GPIO_DDC5_A); + break; + case GPIO_DDC_LINE_DDC_VGA: + info->offset = REG(DC_GPIO_DDCVGA_A); + break; + case GPIO_DDC_LINE_I2C_PAD: + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_GENERIC: + info->offset = REG(DC_GPIO_GENERIC_A); + switch (en) { + case GPIO_GENERIC_A: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK; + break; + case GPIO_GENERIC_B: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICB_A_MASK; + break; + case GPIO_GENERIC_C: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICC_A_MASK; + break; + case GPIO_GENERIC_D: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICD_A_MASK; + break; + case GPIO_GENERIC_E: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICE_A_MASK; + break; + case GPIO_GENERIC_F: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICF_A_MASK; + break; + case GPIO_GENERIC_G: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICG_A_MASK; + break; + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_HPD: + info->offset = REG(DC_GPIO_HPD_A); + switch (en) { + case GPIO_HPD_1: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK; + break; + case GPIO_HPD_2: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD2_A_MASK; + break; + case GPIO_HPD_3: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD3_A_MASK; + break; + case GPIO_HPD_4: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD4_A_MASK; + break; + case GPIO_HPD_5: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD5_A_MASK; + break; + case GPIO_HPD_6: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD6_A_MASK; + break; + default: + ASSERT_CRITICAL(false); +#ifdef PALLADIUM_SUPPORTED + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK; + result = true; +#endif + result = false; + } + break; + case GPIO_ID_GSL: + switch (en) { + case GPIO_GSL_GENLOCK_CLOCK: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + break; + case GPIO_GSL_GENLOCK_VSYNC: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + break; + case GPIO_GSL_SWAPLOCK_A: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + break; + case GPIO_GSL_SWAPLOCK_B: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + + break; + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_SYNC: + case GPIO_ID_VIP_PAD: + default: + ASSERT_CRITICAL(false); + result = false; + } + + if (result) { + info->offset_y = info->offset + 2; + info->offset_en = info->offset + 1; + info->offset_mask = info->offset - 1; + + info->mask_y = info->mask; + info->mask_en = info->mask; + info->mask_mask = info->mask; + } + + return result; +} + +/* function table */ +static const struct hw_translate_funcs funcs = { + .offset_to_id = offset_to_id, + .id_to_offset = id_to_offset, +}; + +/* + * dal_hw_translate_dcn10_init + * + * @brief + * Initialize Hw translate function pointers. + * + * @param + * struct hw_translate *tr - [out] struct of function pointers + * + */ +void dal_hw_translate_dcn21_init(struct hw_translate *tr) +{ + tr->funcs = &funcs; +} + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.h b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.h new file mode 100644 index 000000000000..2bfaac24c574 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.h @@ -0,0 +1,35 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#ifndef __DAL_HW_TRANSLATE_DCN21_H__ +#define __DAL_HW_TRANSLATE_DCN21_H__ + +struct hw_translate; + +/* Initialize Hw translate function pointers */ +void dal_hw_translate_dcn21_init(struct hw_translate *tr); + +#endif /* __DAL_HW_TRANSLATE_DCN21_H__ */ +#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c b/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c index 26695b963c58..df68430aeb0c 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c @@ -38,15 +38,13 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" /* function table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = NULL, - .create_ddc_clock = NULL, - .create_generic = NULL, - .create_hpd = NULL, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = NULL, + .init_generic = NULL, + .init_hpd = NULL, }; void dal_hw_factory_diag_fpga_init(struct hw_factory *factory) diff --git a/drivers/gpu/drm/amd/display/dc/gpio/generic_regs.h b/drivers/gpu/drm/amd/display/dc/gpio/generic_regs.h new file mode 100644 index 000000000000..8c05295c05c2 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/generic_regs.h @@ -0,0 +1,66 @@ +/* + * Copyright 2012-16 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef DRIVERS_GPU_DRM_AMD_DC_DEV_DC_GPIO_GENERIC_REGS_H_ +#define DRIVERS_GPU_DRM_AMD_DC_DEV_DC_GPIO_GENERIC_REGS_H_ + +#include "gpio_regs.h" + +#define GENERIC_GPIO_REG_LIST_ENTRY(type, cd, id) \ + .type ## _reg = REG(DC_GPIO_GENERIC_## type),\ + .type ## _mask = DC_GPIO_GENERIC_ ## type ## __DC_GPIO_GENERIC ## id ## _ ## type ## _MASK,\ + .type ## _shift = DC_GPIO_GENERIC_ ## type ## __DC_GPIO_GENERIC ## id ## _ ## type ## __SHIFT + +#define GENERIC_GPIO_REG_LIST(id) \ + {\ + GENERIC_GPIO_REG_LIST_ENTRY(MASK, cd, id),\ + GENERIC_GPIO_REG_LIST_ENTRY(A, cd, id),\ + GENERIC_GPIO_REG_LIST_ENTRY(EN, cd, id),\ + GENERIC_GPIO_REG_LIST_ENTRY(Y, cd, id)\ + } + +#define GENERIC_REG_LIST(id) \ + GENERIC_GPIO_REG_LIST(id), \ + .mux = REG(DC_GENERIC ## id),\ + +#define GENERIC_MASK_SH_LIST(mask_sh, cd) \ + {(DC_GENERIC ## cd ##__GENERIC ## cd ##_EN## mask_sh),\ + (DC_GENERIC ## cd ##__GENERIC ## cd ##_SEL## mask_sh)} + +struct generic_registers { + struct gpio_registers gpio; + uint32_t mux; +}; + +struct generic_sh_mask { + /* enable */ + uint32_t GENERIC_EN; + /* select */ + uint32_t GENERIC_SEL; + +}; + + +#endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_GPIO_GENERIC_REGS_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c index d03b38e80d9b..f8f85490e77e 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c @@ -67,10 +67,14 @@ enum gpio_result dal_gpio_open_ex( return GPIO_RESULT_ALREADY_OPENED; } + // No action if allocation failed during gpio construct + if (!gpio->hw_container.ddc) { + ASSERT_CRITICAL(false); + return GPIO_RESULT_NON_SPECIFIC_ERROR; + } gpio->mode = mode; - return dal_gpio_service_open( - gpio->service, gpio->id, gpio->en, mode, &gpio->pin); + return dal_gpio_service_open(gpio); } enum gpio_result dal_gpio_get_value( @@ -231,6 +235,21 @@ enum gpio_pin_output_state dal_gpio_get_output_state( return gpio->output_state; } +struct hw_ddc *dal_gpio_get_ddc(struct gpio *gpio) +{ + return gpio->hw_container.ddc; +} + +struct hw_hpd *dal_gpio_get_hpd(struct gpio *gpio) +{ + return gpio->hw_container.hpd; +} + +struct hw_generic *dal_gpio_get_generic(struct gpio *gpio) +{ + return gpio->hw_container.generic; +} + void dal_gpio_close( struct gpio *gpio) { @@ -267,6 +286,30 @@ struct gpio *dal_gpio_create( gpio->mode = GPIO_MODE_UNKNOWN; gpio->output_state = output_state; + //initialize hw_container union based on id + switch (gpio->id) { + case GPIO_ID_DDC_DATA: + gpio->service->factory.funcs->init_ddc_data(&gpio->hw_container.ddc, service->ctx, id, en); + break; + case GPIO_ID_DDC_CLOCK: + gpio->service->factory.funcs->init_ddc_data(&gpio->hw_container.ddc, service->ctx, id, en); + break; + case GPIO_ID_GENERIC: + gpio->service->factory.funcs->init_generic(&gpio->hw_container.generic, service->ctx, id, en); + break; + case GPIO_ID_HPD: + gpio->service->factory.funcs->init_hpd(&gpio->hw_container.hpd, service->ctx, id, en); + break; + // TODO: currently gpio for sync and gsl does not get created, might need it later + case GPIO_ID_SYNC: + break; + case GPIO_ID_GSL: + break; + default: + ASSERT_CRITICAL(false); + gpio->pin = NULL; + } + return gpio; } @@ -280,6 +323,33 @@ void dal_gpio_destroy( dal_gpio_close(*gpio); + switch ((*gpio)->id) { + case GPIO_ID_DDC_DATA: + kfree((*gpio)->hw_container.ddc); + (*gpio)->hw_container.ddc = NULL; + break; + case GPIO_ID_DDC_CLOCK: + //TODO: might want to change it to init_ddc_clock + kfree((*gpio)->hw_container.ddc); + (*gpio)->hw_container.ddc = NULL; + break; + case GPIO_ID_GENERIC: + kfree((*gpio)->hw_container.generic); + (*gpio)->hw_container.generic = NULL; + break; + case GPIO_ID_HPD: + kfree((*gpio)->hw_container.hpd); + (*gpio)->hw_container.hpd = NULL; + break; + // TODO: currently gpio for sync and gsl does not get created, might need it later + case GPIO_ID_SYNC: + break; + case GPIO_ID_GSL: + break; + default: + break; + } + kfree(*gpio); *gpio = NULL; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c index a7fab44f66b6..d03165e71dc6 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c @@ -141,6 +141,58 @@ struct gpio *dal_gpio_service_create_irq( return dal_gpio_create_irq(service, id, en); } +struct gpio *dal_gpio_service_create_generic_mux( + struct gpio_service *service, + uint32_t offset, + uint32_t mask) +{ + enum gpio_id id; + uint32_t en; + struct gpio *generic; + + if (!service->translate.funcs->offset_to_id(offset, mask, &id, &en)) { + ASSERT_CRITICAL(false); + return NULL; + } + + generic = dal_gpio_create( + service, id, en, GPIO_PIN_OUTPUT_STATE_DEFAULT); + + return generic; +} + +void dal_gpio_destroy_generic_mux( + struct gpio **mux) +{ + if (!mux || !*mux) { + ASSERT_CRITICAL(false); + return; + } + + dal_gpio_close(*mux); + dal_gpio_destroy(mux); + kfree(*mux); + + *mux = NULL; +} + +struct gpio_pin_info dal_gpio_get_generic_pin_info( + struct gpio_service *service, + enum gpio_id id, + uint32_t en) +{ + struct gpio_pin_info pin; + + if (service->translate.funcs->id_to_offset) { + service->translate.funcs->id_to_offset(id, en, &pin); + } else { + pin.mask = 0xFFFFFFFF; + pin.offset = 0xFFFFFFFF; + } + + return pin; +} + void dal_gpio_service_destroy( struct gpio_service **ptr) { @@ -165,6 +217,21 @@ void dal_gpio_service_destroy( *ptr = NULL; } +enum gpio_result dal_mux_setup_config( + struct gpio *mux, + struct gpio_generic_mux_config *config) +{ + struct gpio_config_data config_data; + + if (!config) + return GPIO_RESULT_INVALID_DATA; + + config_data.config.generic_mux = *config; + config_data.type = GPIO_CONFIG_TYPE_GENERIC_MUX; + + return dal_gpio_set_config(mux, &config_data); +} + /* * @brief * Private API. @@ -223,13 +290,15 @@ enum gpio_result dal_gpio_service_unlock( } enum gpio_result dal_gpio_service_open( - struct gpio_service *service, - enum gpio_id id, - uint32_t en, - enum gpio_mode mode, - struct hw_gpio_pin **ptr) + struct gpio *gpio) { - struct hw_gpio_pin *pin; + struct gpio_service *service = gpio->service; + enum gpio_id id = gpio->id; + uint32_t en = gpio->en; + enum gpio_mode mode = gpio->mode; + + struct hw_gpio_pin **pin = &gpio->pin; + if (!service->busyness[id]) { ASSERT_CRITICAL(false); @@ -243,50 +312,43 @@ enum gpio_result dal_gpio_service_open( switch (id) { case GPIO_ID_DDC_DATA: - pin = service->factory.funcs->create_ddc_data( - service->ctx, id, en); - service->factory.funcs->define_ddc_registers(pin, en); + *pin = service->factory.funcs->get_ddc_pin(gpio); + service->factory.funcs->define_ddc_registers(*pin, en); break; case GPIO_ID_DDC_CLOCK: - pin = service->factory.funcs->create_ddc_clock( - service->ctx, id, en); - service->factory.funcs->define_ddc_registers(pin, en); + *pin = service->factory.funcs->get_ddc_pin(gpio); + service->factory.funcs->define_ddc_registers(*pin, en); break; case GPIO_ID_GENERIC: - pin = service->factory.funcs->create_generic( - service->ctx, id, en); + *pin = service->factory.funcs->get_generic_pin(gpio); + service->factory.funcs->define_generic_registers(*pin, en); break; case GPIO_ID_HPD: - pin = service->factory.funcs->create_hpd( - service->ctx, id, en); - service->factory.funcs->define_hpd_registers(pin, en); + *pin = service->factory.funcs->get_hpd_pin(gpio); + service->factory.funcs->define_hpd_registers(*pin, en); break; + + //TODO: gsl and sync support? create_sync and create_gsl are NULL case GPIO_ID_SYNC: - pin = service->factory.funcs->create_sync( - service->ctx, id, en); - break; case GPIO_ID_GSL: - pin = service->factory.funcs->create_gsl( - service->ctx, id, en); break; default: ASSERT_CRITICAL(false); return GPIO_RESULT_NON_SPECIFIC_ERROR; } - if (!pin) { + if (!*pin) { ASSERT_CRITICAL(false); return GPIO_RESULT_NON_SPECIFIC_ERROR; } - if (!pin->funcs->open(pin, mode)) { + if (!(*pin)->funcs->open(*pin, mode)) { ASSERT_CRITICAL(false); - dal_gpio_service_close(service, &pin); + dal_gpio_service_close(service, pin); return GPIO_RESULT_OPEN_FAILED; } set_pin_busy(service, id, en); - *ptr = pin; return GPIO_RESULT_OK; } @@ -308,11 +370,10 @@ void dal_gpio_service_close( pin->funcs->close(pin); - pin->funcs->destroy(ptr); + *ptr = NULL; } } - enum dc_irq_source dal_irq_get_source( const struct gpio *irq) { diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h index 0c678af75331..b9775a131ecd 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h @@ -42,11 +42,7 @@ struct gpio_service { }; enum gpio_result dal_gpio_service_open( - struct gpio_service *service, - enum gpio_id id, - uint32_t en, - enum gpio_mode mode, - struct hw_gpio_pin **ptr); + struct gpio *gpio); void dal_gpio_service_close( struct gpio_service *service, diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c index 408857d19c84..1c12961f6472 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c @@ -28,6 +28,7 @@ #include "dm_services.h" +#include "include/gpio_interface.h" #include "include/gpio_types.h" #include "hw_gpio.h" #include "hw_ddc.h" @@ -45,6 +46,8 @@ #define REG(reg)\ (ddc->regs->reg) +struct gpio; + static void destruct( struct hw_ddc *pin) { @@ -227,24 +230,29 @@ static void construct( ddc->base.base.funcs = &funcs; } -struct hw_gpio_pin *dal_hw_ddc_create( +void dal_hw_ddc_init( + struct hw_ddc **hw_ddc, struct dc_context *ctx, enum gpio_id id, uint32_t en) { - struct hw_ddc *pin; - if ((en < GPIO_DDC_LINE_MIN) || (en > GPIO_DDC_LINE_MAX)) { ASSERT_CRITICAL(false); - return NULL; + *hw_ddc = NULL; } - pin = kzalloc(sizeof(struct hw_ddc), GFP_KERNEL); - if (!pin) { + *hw_ddc = kzalloc(sizeof(struct hw_ddc), GFP_KERNEL); + if (!*hw_ddc) { ASSERT_CRITICAL(false); - return NULL; + return; } - construct(pin, id, en, ctx); - return &pin->base.base; + construct(*hw_ddc, id, en, ctx); +} + +struct hw_gpio_pin *dal_hw_ddc_get_pin(struct gpio *gpio) +{ + struct hw_ddc *hw_ddc = dal_gpio_get_ddc(gpio); + + return &hw_ddc->base.base; } diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h index 9690e2a885d7..cc30e65df431 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h @@ -38,9 +38,12 @@ struct hw_ddc { #define HW_DDC_FROM_BASE(hw_gpio) \ container_of((HW_GPIO_FROM_BASE(hw_gpio)), struct hw_ddc, base) -struct hw_gpio_pin *dal_hw_ddc_create( +void dal_hw_ddc_init( + struct hw_ddc **hw_ddc, struct dc_context *ctx, enum gpio_id id, uint32_t en); +struct hw_gpio_pin *dal_hw_ddc_get_pin(struct gpio *gpio); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c index 78f528f92907..fa9f1d055ec8 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c @@ -51,6 +51,9 @@ #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/hw_factory_dcn20.h" #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#include "dcn21/hw_factory_dcn21.h" +#endif #include "diagnostics/hw_factory_diag.h" @@ -99,6 +102,11 @@ bool dal_hw_factory_init( dal_hw_factory_dcn20_init(factory); return true; #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + case DCN_VERSION_2_1: + dal_hw_factory_dcn21_init(factory); + return true; +#endif default: ASSERT_CRITICAL(false); diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h index 6e4dd3521935..e15b037f3bcd 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h @@ -28,41 +28,44 @@ struct hw_gpio_pin; struct hw_hpd; +struct hw_ddc; +struct hw_generic; +struct gpio; struct hw_factory { uint32_t number_of_pins[GPIO_ID_COUNT]; const struct hw_factory_funcs { - struct hw_gpio_pin *(*create_ddc_data)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_ddc_clock)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_generic)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_hpd)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_sync)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_gsl)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); + void (*init_ddc_data)( + struct hw_ddc **hw_ddc, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + void (*init_generic)( + struct hw_generic **hw_generic, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + void (*init_hpd)( + struct hw_hpd **hw_hpd, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + struct hw_gpio_pin *(*get_hpd_pin)( + struct gpio *gpio); + struct hw_gpio_pin *(*get_ddc_pin)( + struct gpio *gpio); + struct hw_gpio_pin *(*get_generic_pin)( + struct gpio *gpio); void (*define_hpd_registers)( struct hw_gpio_pin *pin, uint32_t en); void (*define_ddc_registers)( struct hw_gpio_pin *pin, uint32_t en); + void (*define_generic_registers)( + struct hw_gpio_pin *pin, + uint32_t en); } *funcs; }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c new file mode 100644 index 000000000000..69b899741f6d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c @@ -0,0 +1,138 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/slab.h> + +#include "dm_services.h" + +#include "include/gpio_interface.h" +#include "include/gpio_types.h" +#include "hw_gpio.h" +#include "hw_generic.h" + +#include "reg_helper.h" +#include "generic_regs.h" + +#undef FN +#define FN(reg_name, field_name) \ + generic->shifts->field_name, generic->masks->field_name + +#define CTX \ + generic->base.base.ctx +#define REG(reg)\ + (generic->regs->reg) + +struct gpio; + +static void dal_hw_generic_construct( + struct hw_generic *pin, + enum gpio_id id, + uint32_t en, + struct dc_context *ctx) +{ + dal_hw_gpio_construct(&pin->base, id, en, ctx); +} + +static void dal_hw_generic_destruct( + struct hw_generic *pin) +{ + dal_hw_gpio_destruct(&pin->base); +} + +static void destroy( + struct hw_gpio_pin **ptr) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(*ptr); + + dal_hw_generic_destruct(generic); + + kfree(generic); + + *ptr = NULL; +} + +static enum gpio_result set_config( + struct hw_gpio_pin *ptr, + const struct gpio_config_data *config_data) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(ptr); + + if (!config_data) + return GPIO_RESULT_INVALID_DATA; + + REG_UPDATE_2(mux, + GENERIC_EN, config_data->config.generic_mux.enable_output_from_mux, + GENERIC_SEL, config_data->config.generic_mux.mux_select); + + return GPIO_RESULT_OK; +} + +static const struct hw_gpio_pin_funcs funcs = { + .destroy = destroy, + .open = dal_hw_gpio_open, + .get_value = dal_hw_gpio_get_value, + .set_value = dal_hw_gpio_set_value, + .set_config = set_config, + .change_mode = dal_hw_gpio_change_mode, + .close = dal_hw_gpio_close, +}; + +static void construct( + struct hw_generic *generic, + enum gpio_id id, + uint32_t en, + struct dc_context *ctx) +{ + dal_hw_generic_construct(generic, id, en, ctx); + generic->base.base.funcs = &funcs; +} + +void dal_hw_generic_init( + struct hw_generic **hw_generic, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en) +{ + if ((en < GPIO_DDC_LINE_MIN) || (en > GPIO_DDC_LINE_MAX)) { + ASSERT_CRITICAL(false); + *hw_generic = NULL; + } + + *hw_generic = kzalloc(sizeof(struct hw_generic), GFP_KERNEL); + if (!*hw_generic) { + ASSERT_CRITICAL(false); + return; + } + + construct(*hw_generic, id, en, ctx); +} + + +struct hw_gpio_pin *dal_hw_generic_get_pin(struct gpio *gpio) +{ + struct hw_generic *hw_generic = dal_gpio_get_generic(gpio); + + return &hw_generic->base.base; +} diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h new file mode 100644 index 000000000000..bd6ffeb5e9df --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h @@ -0,0 +1,50 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DAL_HW_generic_H__ +#define __DAL_HW_generic_H__ + +#include "generic_regs.h" +#include "hw_gpio.h" + +struct hw_generic { + struct hw_gpio base; + const struct generic_registers *regs; + const struct generic_sh_mask *shifts; + const struct generic_sh_mask *masks; +}; + +#define HW_GENERIC_FROM_BASE(hw_gpio) \ + container_of((HW_GPIO_FROM_BASE(hw_gpio)), struct hw_generic, base) + +void dal_hw_generic_init( + struct hw_generic **hw_generic, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + +struct hw_gpio_pin *dal_hw_generic_get_pin(struct gpio *gpio); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c index 5e11d748e6f3..00c9bcf660a3 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c @@ -27,6 +27,7 @@ #include "dm_services.h" +#include "include/gpio_interface.h" #include "include/gpio_types.h" #include "hw_gpio.h" #include "hw_hpd.h" @@ -43,6 +44,8 @@ #define REG(reg)\ (hpd->regs->reg) +struct gpio; + static void dal_hw_hpd_construct( struct hw_hpd *pin, enum gpio_id id, @@ -136,29 +139,29 @@ static void construct( hpd->base.base.funcs = &funcs; } -struct hw_gpio_pin *dal_hw_hpd_create( +void dal_hw_hpd_init( + struct hw_hpd **hw_hpd, struct dc_context *ctx, enum gpio_id id, uint32_t en) { - struct hw_hpd *hpd; - - if (id != GPIO_ID_HPD) { + if ((en < GPIO_DDC_LINE_MIN) || (en > GPIO_DDC_LINE_MAX)) { ASSERT_CRITICAL(false); - return NULL; + *hw_hpd = NULL; } - if ((en < GPIO_HPD_MIN) || (en > GPIO_HPD_MAX)) { + *hw_hpd = kzalloc(sizeof(struct hw_hpd), GFP_KERNEL); + if (!*hw_hpd) { ASSERT_CRITICAL(false); - return NULL; + return; } - hpd = kzalloc(sizeof(struct hw_hpd), GFP_KERNEL); - if (!hpd) { - ASSERT_CRITICAL(false); - return NULL; - } + construct(*hw_hpd, id, en, ctx); +} + +struct hw_gpio_pin *dal_hw_hpd_get_pin(struct gpio *gpio) +{ + struct hw_hpd *hw_hpd = dal_gpio_get_hpd(gpio); - construct(hpd, id, en, ctx); - return &hpd->base.base; + return &hw_hpd->base.base; } diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h index 4ab7a208f781..e7d8b3bb016c 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h @@ -38,9 +38,12 @@ struct hw_hpd { #define HW_HPD_FROM_BASE(hw_gpio) \ container_of((HW_GPIO_FROM_BASE(hw_gpio)), struct hw_hpd, base) -struct hw_gpio_pin *dal_hw_hpd_create( +void dal_hw_hpd_init( + struct hw_hpd **hw_hpd, struct dc_context *ctx, enum gpio_id id, uint32_t en); +struct hw_gpio_pin *dal_hw_hpd_get_pin(struct gpio *gpio); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c index c35fe201d335..f2046f55d6a8 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c @@ -49,6 +49,9 @@ #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/hw_translate_dcn20.h" #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#include "dcn21/hw_translate_dcn21.h" +#endif #include "diagnostics/hw_translate_diag.h" @@ -94,6 +97,11 @@ bool dal_hw_translate_init( dal_hw_translate_dcn20_init(translate); return true; #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + case DCN_VERSION_2_1: + dal_hw_translate_dcn21_init(translate); + return true; +#endif default: BREAK_TO_DEBUGGER(); diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_status.h b/drivers/gpu/drm/amd/display/dc/inc/core_status.h index 0a094d7c9380..fd39e2abe2ed 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_status.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_status.h @@ -48,6 +48,9 @@ enum dc_status { DC_NO_DSC_RESOURCE = 17, #endif DC_FAIL_UNSUPPORTED_1 = 18, + DC_FAIL_CLK_EXCEED_MAX = 21, + DC_FAIL_CLK_BELOW_MIN = 22, /*THIS IS MIN PER IP*/ + DC_FAIL_CLK_BELOW_CFG_REQUIRED = 23, /*THIS IS hard_min in PPLIB*/ DC_ERROR_UNEXPECTED = -1 }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index a148ffde8b12..f189307750ab 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -63,11 +63,6 @@ struct link_init_data { TODO: remove it when DC is complete. */ }; -enum { - FREE_ACQUIRED_RESOURCE = 0, - KEEP_ACQUIRED_RESOURCE = 1, -}; - struct dc_link *link_create(const struct link_init_data *init_params); void link_destroy(struct dc_link **link); @@ -82,7 +77,7 @@ void core_link_enable_stream( struct dc_state *state, struct pipe_ctx *pipe_ctx); -void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option); +void core_link_disable_stream(struct pipe_ctx *pipe_ctx); void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable); /********** DAL Core*********************/ @@ -92,6 +87,9 @@ void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable); struct resource_pool; struct dc_state; struct resource_context; +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +struct clk_bw_params; +#endif struct resource_funcs { void (*destroy)(struct resource_pool **pool); @@ -147,6 +145,11 @@ struct resource_funcs { display_e2e_pipe_params_st *pipes, int pipe_cnt); #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + void (*update_bw_bounding_box)( + struct dc *dc, + struct clk_bw_params *bw_params); +#endif }; @@ -228,14 +231,12 @@ struct resource_pool { struct dcn_fe_bandwidth { int dppclk_khz; - }; struct stream_resource { struct output_pixel_processor *opp; #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct display_stream_compressor *dsc; - int dscclk_khz; #endif struct timing_generator *tg; struct stream_encoder *stream_enc; @@ -299,6 +300,8 @@ struct pipe_ctx { struct pipe_ctx *top_pipe; struct pipe_ctx *bottom_pipe; + struct pipe_ctx *next_odm_pipe; + struct pipe_ctx *prev_odm_pipe; #ifdef CONFIG_DRM_AMD_DC_DCN1_0 struct _vcs_dpi_display_dlg_regs_st dlg_regs; diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h index 2d95eff94239..08a4df2c61a8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h @@ -38,6 +38,11 @@ bool dp_verify_link_cap( struct dc_link_settings *known_limit_link_setting, int *fail_count); +bool dp_verify_link_cap_with_retries( + struct dc_link *link, + struct dc_link_settings *known_limit_link_setting, + int attempts); + bool dp_validate_mode_timing( struct dc_link *link, const struct dc_crtc_timing *timing); @@ -62,10 +67,15 @@ bool is_dp_active_dongle(const struct dc_link *link); void dp_enable_mst_on_sink(struct dc_link *link, bool enable); +enum dp_panel_mode dp_get_panel_mode(struct dc_link *link); +void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode); + #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT void dp_set_fec_ready(struct dc_link *link, bool ready); void dp_set_fec_enable(struct dc_link *link, bool enable); bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable); +bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable); +void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable); bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx); #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index 36ebd5bc7863..76f9ad1b23df 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -28,6 +28,131 @@ #include "dc.h" +#define DCN_MINIMUM_DISPCLK_Khz 100000 +#define DCN_MINIMUM_DPPCLK_Khz 100000 + +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 +/* Constants */ +#define DDR4_DRAM_WIDTH 64 +#define WM_A 0 +#define WM_B 1 +#define WM_C 2 +#define WM_D 3 +#define WM_SET_COUNT 4 +#endif + +#define DCN_MINIMUM_DISPCLK_Khz 100000 +#define DCN_MINIMUM_DPPCLK_Khz 100000 + +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 +/* Will these bw structures be ASIC specific? */ + +#define MAX_NUM_DPM_LVL 4 +#define WM_SET_COUNT 4 + + +struct clk_limit_table_entry { + unsigned int voltage; /* milivolts withh 2 fractional bits */ + unsigned int dcfclk_mhz; + unsigned int fclk_mhz; + unsigned int memclk_mhz; + unsigned int socclk_mhz; +}; + +/* This table is contiguous */ +struct clk_limit_table { + struct clk_limit_table_entry entries[MAX_NUM_DPM_LVL]; + unsigned int num_entries; +}; + +struct wm_range_table_entry { + unsigned int wm_inst; + unsigned int wm_type; + double pstate_latency_us; + bool valid; +}; + + +struct clk_log_info { + bool enabled; + char *pBuf; + unsigned int bufSize; + unsigned int *sum_chars_printed; +}; + +struct clk_state_registers_and_bypass { + uint32_t dcfclk; + uint32_t dcf_deep_sleep_divider; + uint32_t dcf_deep_sleep_allow; + uint32_t dprefclk; + uint32_t dispclk; + uint32_t dppclk; + + uint32_t dppclk_bypass; + uint32_t dcfclk_bypass; + uint32_t dprefclk_bypass; + uint32_t dispclk_bypass; +}; + +struct rv1_clk_internal { + uint32_t CLK0_CLK8_CURRENT_CNT; //dcfclk + uint32_t CLK0_CLK8_DS_CNTL; //dcf_deep_sleep_divider + uint32_t CLK0_CLK8_ALLOW_DS; //dcf_deep_sleep_allow + uint32_t CLK0_CLK10_CURRENT_CNT; //dprefclk + uint32_t CLK0_CLK11_CURRENT_CNT; //dispclk + + uint32_t CLK0_CLK8_BYPASS_CNTL; //dcfclk bypass + uint32_t CLK0_CLK10_BYPASS_CNTL; //dprefclk bypass + uint32_t CLK0_CLK11_BYPASS_CNTL; //dispclk bypass +}; + +struct rn_clk_internal { + uint32_t CLK1_CLK0_CURRENT_CNT; //dispclk + uint32_t CLK1_CLK1_CURRENT_CNT; //dppclk + uint32_t CLK1_CLK2_CURRENT_CNT; //dprefclk + uint32_t CLK1_CLK3_CURRENT_CNT; //dcfclk + uint32_t CLK1_CLK3_DS_CNTL; //dcf_deep_sleep_divider + uint32_t CLK1_CLK3_ALLOW_DS; //dcf_deep_sleep_allow + + uint32_t CLK1_CLK0_BYPASS_CNTL; //dispclk bypass + uint32_t CLK1_CLK1_BYPASS_CNTL; //dppclk bypass + uint32_t CLK1_CLK2_BYPASS_CNTL; //dprefclk bypass + uint32_t CLK1_CLK3_BYPASS_CNTL; //dcfclk bypass + +}; + +/* For dtn logging and debugging */ +struct clk_state_registers { + uint32_t CLK0_CLK8_CURRENT_CNT; //dcfclk + uint32_t CLK0_CLK8_DS_CNTL; //dcf_deep_sleep_divider + uint32_t CLK0_CLK8_ALLOW_DS; //dcf_deep_sleep_allow + uint32_t CLK0_CLK10_CURRENT_CNT; //dprefclk + uint32_t CLK0_CLK11_CURRENT_CNT; //dispclk +}; + +/* TODO: combine this with the above */ +struct clk_bypass { + uint32_t dcfclk_bypass; + uint32_t dispclk_pypass; + uint32_t dprefclk_bypass; +}; +/* + * This table is not contiguous, can have holes, each + * entry correspond to one set of WM. For example if + * we have 2 DPM and LPDDR, we will WM set A, B and + * D occupied, C will be emptry. + */ +struct wm_table { + struct wm_range_table_entry entries[WM_SET_COUNT]; +}; + +struct clk_bw_params { + unsigned int vram_type; + unsigned int num_channels; + struct clk_limit_table clk_table; + struct wm_table wm_table; +}; +#endif /* Public interfaces */ struct clk_states { @@ -51,6 +176,10 @@ struct clk_mgr_funcs { void (*init_clocks)(struct clk_mgr *clk_mgr); void (*enable_pme_wa) (struct clk_mgr *clk_mgr); + void (*get_clock)(struct clk_mgr *clk_mgr, + struct dc_state *context, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg); }; struct clk_mgr { @@ -58,6 +187,9 @@ struct clk_mgr { struct clk_mgr_funcs *funcs; struct dc_clocks clks; int dprefclk_khz; // Used by program pixel clock in clock source funcs, need to figureout where this goes +#ifdef CONFIG_DRM_AMD_DC_DCN2_1 + struct clk_bw_params *bw_params; +#endif }; /* forward declarations */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h index 0835ac041acf..7dd46eb96d67 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h @@ -64,6 +64,8 @@ enum dentist_divider_range { *************************************************************************************** */ +/* Macros */ + #define TO_CLK_MGR_INTERNAL(clk_mgr)\ container_of(clk_mgr, struct clk_mgr_internal, base) @@ -189,6 +191,7 @@ struct state_dependent_clocks { struct clk_mgr_internal { struct clk_mgr base; + int smu_ver; struct pp_smu_funcs *pp_smu; struct clk_mgr_internal_funcs *funcs; @@ -213,6 +216,8 @@ struct clk_mgr_internal { bool dfs_bypass_enabled; /* True if the DFS-bypass feature is enabled and active. */ bool dfs_bypass_active; + + uint32_t dfs_ref_freq_khz; /* * Cache the display clock returned by VBIOS if DFS-bypass is enabled. * This is basically "Crystal Frequency In KHz" (XTALIN) frequency @@ -276,8 +281,14 @@ static inline bool should_set_clock(bool safe_to_lower, int calc_clk, int cur_cl static inline bool should_update_pstate_support(bool safe_to_lower, bool calc_support, bool cur_support) { - // Whenever we are transitioning pstate support, we always want to notify prior to committing state - return (calc_support != cur_support) ? !safe_to_lower : false; + if (cur_support != calc_support) { + if (calc_support == true && safe_to_lower) + return true; + else if (calc_support == false && !safe_to_lower) + return true; + } + + return false; } int clk_mgr_helper_get_active_display_cnt( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index 05ee5295d2c1..d8e744f366e5 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -38,7 +38,8 @@ struct dccg { struct dccg_funcs { void (*update_dpp_dto)(struct dccg *dccg, int dpp_inst, - int req_dppclk); + int req_dppclk, + bool reduce_divider_only); void (*get_dccg_ref_freq)(struct dccg *dccg, unsigned int xtalin_freq_inKhz, unsigned int *dccg_ref_freq_inKhz); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index 9502478c4a1b..a6297219d7fc 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -80,6 +80,8 @@ struct dcn_hubbub_phys_addr_config { uint64_t page_table_end_addr; uint64_t page_table_base_addr; } gart_config; + + uint64_t page_table_default_page_addr; }; struct dcn_hubbub_virt_addr_config { @@ -141,6 +143,10 @@ struct hubbub_funcs { struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower); + + bool (*is_allow_self_refresh_enabled)(struct hubbub *hubbub); + void (*allow_self_refresh_control)(struct hubbub *hubbub, bool allow); + }; struct hubbub { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h index 60c671fcf186..474c7194a9f8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h @@ -42,6 +42,7 @@ struct dpp { #if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct pwl_params shaper_params; + bool cm_bypass_mode; #endif }; @@ -200,7 +201,7 @@ struct dpp_funcs { void (*set_cursor_attributes)( struct dpp *dpp_base, - enum dc_cursor_color_format color_format); + struct dc_cursor_attributes *cursor_attributes); void (*set_cursor_position)( struct dpp *dpp_base, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h index c905d020b59e..1ddb1c6fa149 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h @@ -92,7 +92,9 @@ struct dsc_funcs { void (*dsc_read_state)(struct display_stream_compressor *dsc, struct dcn_dsc_state *s); bool (*dsc_validate_stream)(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg); void (*dsc_set_config)(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, - struct dsc_optc_config *dsc_optc_cfg, uint8_t *dsc_packed_pps); + struct dsc_optc_config *dsc_optc_cfg); + bool (*dsc_get_packed_pps)(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, + uint8_t *dsc_packed_pps); void (*dsc_enable)(struct display_stream_compressor *dsc, int opp_pipe); void (*dsc_disable)(struct display_stream_compressor *dsc); }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h index a3409294ae0c..ff1a07b35c85 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h @@ -45,22 +45,10 @@ enum dwb_source { dwb_src_scl = 0, /* for DCE7x/9x, DCN won't support. */ dwb_src_blnd, /* for DCE7x/9x */ dwb_src_fmt, /* for DCE7x/9x */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) dwb_src_otg0 = 0x100, /* for DCN1.x/DCN2.x, register: mmDWB_SOURCE_SELECT */ dwb_src_otg1, /* for DCN1.x/DCN2.x */ dwb_src_otg2, /* for DCN1.x/DCN2.x */ dwb_src_otg3, /* for DCN1.x/DCN2.x */ -#else - dwb_src_otg0 = 0x100, /* for DCN1.x, register: mmDWB_SOURCE_SELECT */ - dwb_src_otg1, /* for DCN1.x */ - dwb_src_otg2, /* for DCN1.x */ - dwb_src_otg3, /* for DCN1.x */ -#endif - dwb_src_mpc0 = 0x200, /* for DCN2, register: mmMPC_DWB0_MUX, mmMPC_DWB1_MUX, mmMPC_DWB2_MUX */ - dwb_src_mpc1, /* for DCN2 */ - dwb_src_mpc2, /* for DCN2 */ - dwb_src_mpc3, /* for DCN2 */ - dwb_src_mpc4, /* for DCN2 */ }; #if defined(CONFIG_DRM_AMD_DC_DCN2_0) diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h b/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h index 90d0148430fb..5253dc8b15f8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h @@ -28,12 +28,22 @@ #include "gpio_types.h" + +union gpio_hw_container { + struct hw_ddc *ddc; + struct hw_generic *generic; + struct hw_hpd *hpd; +}; + struct gpio { struct gpio_service *service; struct hw_gpio_pin *pin; enum gpio_id id; uint32_t en; + + union gpio_hw_container hw_container; enum gpio_mode mode; + /* when GPIO comes from VBIOS, it has defined output state */ enum gpio_pin_output_state output_state; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 51bff8717cc9..809b62b51a43 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -47,6 +47,11 @@ enum cursor_lines_per_chunk { CURSOR_LINE_PER_CHUNK_16 }; +enum hubp_ind_block_size { + hubp_ind_block_unconstrained = 0, + hubp_ind_block_64b, +}; + struct hubp { const struct hubp_funcs *funcs; struct dc_context *ctx; @@ -74,7 +79,8 @@ struct hubp_funcs { struct _vcs_dpi_display_ttu_regs_st *ttu_regs); void (*dcc_control)(struct hubp *hubp, bool enable, - bool independent_64b_blks); + enum hubp_ind_block_size blk_size); + void (*mem_program_viewport)( struct hubp *hubp, const struct rect *viewport, @@ -103,7 +109,7 @@ struct hubp_funcs { struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, @@ -111,9 +117,6 @@ struct hubp_funcs { bool (*hubp_is_flip_pending)(struct hubp *hubp); - void (*hubp_update_dchub)(struct hubp *hubp, - struct dchub_init_data *dh_data); - void (*set_blank)(struct hubp *hubp, bool blank); void (*set_hubp_blank_en)(struct hubp *hubp, bool blank); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h index e5e8640a9ef3..abb4e4237fb6 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h @@ -183,6 +183,9 @@ struct link_encoder_funcs { bool (*fec_is_active)(struct link_encoder *enc); #endif + bool (*is_in_alt_mode) (struct link_encoder *enc); + enum signal_type (*get_dig_mode)( + struct link_encoder *enc); }; #endif /* LINK_ENCODER_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h index da89c2edb07c..e8668388581b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h @@ -40,6 +40,10 @@ struct cstate_pstate_watermarks_st { struct dcn_watermarks { uint32_t pte_meta_urgent_ns; uint32_t urgent_ns; +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + uint32_t frac_urg_bw_nom; + uint32_t frac_urg_bw_flip; +#endif struct cstate_pstate_watermarks_st cstate_pstate; }; @@ -149,7 +153,7 @@ struct mem_input_funcs { struct mem_input *mem_input, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 45b94e319cd4..58826be81395 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -128,6 +128,7 @@ struct mpc { struct mpcc mpcc_array[MAX_MPCC]; #if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct pwl_params blender_params; + bool cm_bypass_mode; #endif }; @@ -198,6 +199,9 @@ struct mpc_funcs { * Return: void */ void (*mpc_init)(struct mpc *mpc); + void (*mpc_init_single_inst)( + struct mpc *mpc, + unsigned int mpcc_id); /* * Update the blending configuration for a specified MPCC. @@ -250,6 +254,10 @@ struct mpc_funcs { struct mpc *mpc, int mpcc_id, const struct pwl_params *params); + void (*power_on_mpc_mem_pwr)( + struct mpc *mpc, + int mpcc_id, + bool power_on); #endif }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h index 5d8a7bcccc6f..957e9047381a 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h @@ -316,11 +316,6 @@ struct opp_funcs { bool (*dpg_is_blanked)( struct output_pixel_processor *opp); - void (*opp_convert_pti)( - struct output_pixel_processor *opp, - bool enable, - bool polarity); - void (*opp_dpg_set_blank_color)( struct output_pixel_processor *opp, const struct tg_color *color); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h index ed7d9588b309..fe9b7a10a1c3 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h @@ -91,7 +91,7 @@ struct encoder_unblank_param { struct dc_link_settings link_settings; struct dc_crtc_timing timing; #ifdef CONFIG_DRM_AMD_DC_DCN2_0 - bool odm; + int opp_cnt; #endif }; @@ -122,9 +122,6 @@ struct enc_state { #endif struct stream_encoder_funcs { - #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - void (*enc_read_state)(struct stream_encoder *enc, struct enc_state *s); - #endif void (*dp_set_stream_attribute)( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, @@ -211,14 +208,25 @@ struct stream_encoder_funcs { struct stream_encoder *enc, int tg_inst); + void (*hdmi_reset_stream_attribute)( + struct stream_encoder *enc); + + unsigned int (*dig_source_otg)( + struct stream_encoder *enc); + #if defined(CONFIG_DRM_AMD_DC_DCN2_0) #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + void (*enc_read_state)(struct stream_encoder *enc, struct enc_state *s); + void (*dp_set_dsc_config)( struct stream_encoder *enc, enum optc_dsc_mode dsc_mode, uint32_t dsc_bytes_per_pixel, - uint32_t dsc_slice_width, - uint8_t *dsc_packed_pps); + uint32_t dsc_slice_width); + + void (*dp_set_dsc_pps_info_packet)(struct stream_encoder *enc, + bool enable, + uint8_t *dsc_packed_pps); #endif void (*set_dynamic_metadata)(struct stream_encoder *enc, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index 5e93bc0e8ff9..6196cc32356e 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -59,6 +59,8 @@ struct gsl_params { struct drr_params { uint32_t vertical_total_min; uint32_t vertical_total_max; + uint32_t vertical_total_mid; + uint32_t vertical_total_mid_frame_num; bool immediate_flip; }; @@ -96,6 +98,11 @@ enum crc_selection { INTERSECT_WINDOW_NOT_A_NOT_B, }; +enum h_timing_div_mode { + H_TIMING_NO_DIV, + H_TIMING_DIV_BY2, +}; + struct crc_params { /* Regions used to calculate CRC*/ uint16_t windowa_x_start; @@ -184,10 +191,8 @@ struct timing_generator_funcs { bool (*did_triggered_reset_occur)(struct timing_generator *tg); void (*setup_global_swap_lock)(struct timing_generator *tg, const struct dcp_gsl_params *gsl_params); - void (*setup_global_lock)(struct timing_generator *tg); void (*unlock)(struct timing_generator *tg); void (*lock)(struct timing_generator *tg); - void (*lock_global)(struct timing_generator *tg); void (*lock_doublebuffer_disable)(struct timing_generator *tg); void (*lock_doublebuffer_enable)(struct timing_generator *tg); #if defined(CONFIG_DRM_AMD_DC_DCN2_0) @@ -267,9 +272,9 @@ struct timing_generator_funcs { uint32_t dsc_bytes_per_pixel, uint32_t dsc_slice_width); #endif - void (*set_odm_bypass)(struct timing_generator *tg, const struct dc_crtc_timing *dc_crtc_timing); - void (*set_odm_combine)(struct timing_generator *tg, int combine_opp_id, - int mpcc_hactive, enum dc_pixel_encoding pixel_encoding); + void (*set_odm_bypass)(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); + void (*set_odm_combine)(struct timing_generator *optc, int *opp_id, int opp_cnt, + struct dc_crtc_timing *timing); void (*set_gsl)(struct timing_generator *optc, const struct gsl_params *params); void (*set_gsl_source_select)(struct timing_generator *optc, int group_idx, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index 4d56d48a3179..3a938cd414ea 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -48,6 +48,7 @@ struct dce_hwseq_wa { bool DEGVIDCN10_253; bool false_optc_underflow; bool DEGVIDCN10_254; + bool DEGVIDCN21; }; struct hwseq_wa_state { @@ -78,6 +79,8 @@ struct stream_resource; struct dc_phy_addr_space_config; struct dc_virtual_addr_space_config; #endif +struct hubp; +struct dpp; struct hw_sequencer_funcs { @@ -194,8 +197,7 @@ struct hw_sequencer_funcs { void (*enable_stream)(struct pipe_ctx *pipe_ctx); - void (*disable_stream)(struct pipe_ctx *pipe_ctx, - int option); + void (*disable_stream)(struct pipe_ctx *pipe_ctx); void (*unblank_stream)(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); @@ -204,7 +206,7 @@ struct hw_sequencer_funcs { void (*enable_audio_stream)(struct pipe_ctx *pipe_ctx); - void (*disable_audio_stream)(struct pipe_ctx *pipe_ctx, int option); + void (*disable_audio_stream)(struct pipe_ctx *pipe_ctx); void (*pipe_control_lock)( struct dc *dc, @@ -231,11 +233,13 @@ struct hw_sequencer_funcs { bool (*update_bandwidth)( struct dc *dc, struct dc_state *context); + void (*program_dmdata_engine)(struct pipe_ctx *pipe_ctx); bool (*dmdata_status_done)(struct pipe_ctx *pipe_ctx); #endif void (*set_drr)(struct pipe_ctx **pipe_ctx, int num_pipes, - int vmin, int vmax); + unsigned int vmin, unsigned int vmax, + unsigned int vmid, unsigned int vmid_frame_number); void (*get_position)(struct pipe_ctx **pipe_ctx, int num_pipes, struct crtc_position *position); @@ -279,6 +283,36 @@ struct hw_sequencer_funcs { void (*setup_vupdate_interrupt)(struct pipe_ctx *pipe_ctx); bool (*did_underflow_occur)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*init_blank)(struct dc *dc, struct timing_generator *tg); + void (*disable_vga)(struct dce_hwseq *hws); + void (*bios_golden_init)(struct dc *dc); + void (*plane_atomic_power_down)(struct dc *dc, + struct dpp *dpp, + struct hubp *hubp); + + void (*plane_atomic_disable)( + struct dc *dc, struct pipe_ctx *pipe_ctx); + + void (*enable_power_gating_plane)( + struct dce_hwseq *hws, + bool enable); + + void (*dpp_pg_control)( + struct dce_hwseq *hws, + unsigned int dpp_inst, + bool power_on); + + void (*hubp_pg_control)( + struct dce_hwseq *hws, + unsigned int hubp_inst, + bool power_on); + + void (*dsc_pg_control)( + struct dce_hwseq *hws, + unsigned int dsc_inst, + bool power_on); + + #if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*update_odm)(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx); void (*program_all_writeback_pipes_in_tree)( @@ -294,6 +328,15 @@ struct hw_sequencer_funcs { void (*disable_writeback)(struct dc *dc, unsigned int dwb_pipe_inst); #endif + enum dc_status (*set_clock)(struct dc *dc, + enum dc_clock_type clock_type, + uint32_t clk_khz, + uint32_t stepping); + + void (*get_clock)(struct dc *dc, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg); + }; void color_space_to_black_color( diff --git a/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h b/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h index 30be7bb4a01a..4eff5d38a2f9 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h @@ -60,7 +60,7 @@ void dp_disable_link_phy_mst(struct dc_link *link, enum signal_type signal); bool dp_set_hw_training_pattern( struct dc_link *link, - enum hw_dp_training_pattern pattern); + enum dc_dp_training_pattern pattern); void dp_set_hw_lane_settings( struct dc_link *link, @@ -72,8 +72,6 @@ void dp_set_hw_test_pattern( uint8_t *custom_pattern, uint32_t custom_pattern_size); -enum dp_panel_mode dp_get_panel_mode(struct dc_link *link); - void dp_retrain_link_dp_test(struct dc_link *link, struct dc_link_settings *link_setting, bool skip_video_pattern); diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index 47f81072d7e9..bef224bf803e 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -31,6 +31,8 @@ #include "dm_pp_smu.h" #define MEMORY_TYPE_MULTIPLIER_CZ 4 +#define MEMORY_TYPE_HBM 2 + enum dce_version resource_parse_asic_id( struct hw_asic_id asic_id); @@ -179,7 +181,4 @@ void update_audio_usage( unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format); -struct pipe_ctx *dc_res_get_odm_bottom_pipe(struct pipe_ctx *pipe_ctx); -bool dc_res_is_odm_head_pipe(struct pipe_ctx *pipe_ctx); - #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/irq/Makefile b/drivers/gpu/drm/amd/display/dc/irq/Makefile index ad87c2f093e2..ea75420fc876 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/Makefile +++ b/drivers/gpu/drm/amd/display/dc/irq/Makefile @@ -77,3 +77,13 @@ AMD_DAL_IRQ_DCN2 = $(addprefix $(AMDDALPATH)/dc/irq/dcn20/,$(IRQ_DCN2)) AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCN2) endif +############################################################################### +# DCN 21 +############################################################################### +ifdef CONFIG_DRM_AMD_DC_DCN2_1 +IRQ_DCN21 = irq_service_dcn21.o + +AMD_DAL_IRQ_DCN21= $(addprefix $(AMDDALPATH)/dc/irq/dcn21/,$(IRQ_DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCN21) +endif diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c b/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c index 3cc0f2a1f77c..5db29bf582d3 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c @@ -167,6 +167,11 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = { .ack = NULL }; +static const struct irq_source_info_funcs vupdate_no_lock_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + #undef BASE_INNER #define BASE_INNER(seg) DCN_BASE__INST0_SEG ## seg @@ -221,12 +226,15 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = { .funcs = &pflip_irq_info_funcs\ } -#define vupdate_int_entry(reg_num)\ +/* vupdate_no_lock_int_entry maps to DC_IRQ_SOURCE_VUPDATEx, to match semantic + * of DCE's DC_IRQ_SOURCE_VUPDATEx. + */ +#define vupdate_no_lock_int_entry(reg_num)\ [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\ IRQ_REG_ENTRY(OTG, reg_num,\ - OTG_GLOBAL_SYNC_STATUS, VUPDATE_INT_EN,\ - OTG_GLOBAL_SYNC_STATUS, VUPDATE_EVENT_CLEAR),\ - .funcs = &vblank_irq_info_funcs\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_NO_LOCK_INT_EN,\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_NO_LOCK_EVENT_CLEAR),\ + .funcs = &vupdate_no_lock_irq_info_funcs\ } #define vblank_int_entry(reg_num)\ @@ -333,12 +341,12 @@ irq_source_info_dcn20[DAL_IRQ_SOURCES_NUMBER] = { dc_underflow_int_entry(6), [DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(), [DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(), - vupdate_int_entry(0), - vupdate_int_entry(1), - vupdate_int_entry(2), - vupdate_int_entry(3), - vupdate_int_entry(4), - vupdate_int_entry(5), + vupdate_no_lock_int_entry(0), + vupdate_no_lock_int_entry(1), + vupdate_no_lock_int_entry(2), + vupdate_no_lock_int_entry(3), + vupdate_no_lock_int_entry(4), + vupdate_no_lock_int_entry(5), vblank_int_entry(0), vblank_int_entry(1), vblank_int_entry(2), diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c new file mode 100644 index 000000000000..cbe7818529bb --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c @@ -0,0 +1,374 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/slab.h> + +#include "dm_services.h" + +#include "include/logger_interface.h" + +#include "../dce110/irq_service_dce110.h" + +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" +#include "renoir_ip_offset.h" + + +#include "irq_service_dcn21.h" + +#include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" + +enum dc_irq_source to_dal_irq_source_dcn21( + struct irq_service *irq_service, + uint32_t src_id, + uint32_t ext_id) +{ + switch (src_id) { + case DCN_1_0__SRCID__DC_D1_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK1; + case DCN_1_0__SRCID__DC_D2_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK2; + case DCN_1_0__SRCID__DC_D3_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK3; + case DCN_1_0__SRCID__DC_D4_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK4; + case DCN_1_0__SRCID__DC_D5_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK5; + case DCN_1_0__SRCID__DC_D6_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK6; + case DCN_1_0__SRCID__HUBP0_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP1; + case DCN_1_0__SRCID__HUBP1_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP2; + case DCN_1_0__SRCID__HUBP2_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP3; + case DCN_1_0__SRCID__HUBP3_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP4; + case DCN_1_0__SRCID__HUBP4_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP5; + case DCN_1_0__SRCID__HUBP5_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP6; + case DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE1; + case DCN_1_0__SRCID__OTG1_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE2; + case DCN_1_0__SRCID__OTG2_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE3; + case DCN_1_0__SRCID__OTG3_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE4; + case DCN_1_0__SRCID__OTG4_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE5; + case DCN_1_0__SRCID__OTG5_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE6; + + case DCN_1_0__SRCID__DC_HPD1_INT: + /* generic src_id for all HPD and HPDRX interrupts */ + switch (ext_id) { + case DCN_1_0__CTXID__DC_HPD1_INT: + return DC_IRQ_SOURCE_HPD1; + case DCN_1_0__CTXID__DC_HPD2_INT: + return DC_IRQ_SOURCE_HPD2; + case DCN_1_0__CTXID__DC_HPD3_INT: + return DC_IRQ_SOURCE_HPD3; + case DCN_1_0__CTXID__DC_HPD4_INT: + return DC_IRQ_SOURCE_HPD4; + case DCN_1_0__CTXID__DC_HPD5_INT: + return DC_IRQ_SOURCE_HPD5; + case DCN_1_0__CTXID__DC_HPD6_INT: + return DC_IRQ_SOURCE_HPD6; + case DCN_1_0__CTXID__DC_HPD1_RX_INT: + return DC_IRQ_SOURCE_HPD1RX; + case DCN_1_0__CTXID__DC_HPD2_RX_INT: + return DC_IRQ_SOURCE_HPD2RX; + case DCN_1_0__CTXID__DC_HPD3_RX_INT: + return DC_IRQ_SOURCE_HPD3RX; + case DCN_1_0__CTXID__DC_HPD4_RX_INT: + return DC_IRQ_SOURCE_HPD4RX; + case DCN_1_0__CTXID__DC_HPD5_RX_INT: + return DC_IRQ_SOURCE_HPD5RX; + case DCN_1_0__CTXID__DC_HPD6_RX_INT: + return DC_IRQ_SOURCE_HPD6RX; + default: + return DC_IRQ_SOURCE_INVALID; + } + break; + + default: + break; + } + return DC_IRQ_SOURCE_INVALID; +} + +static bool hpd_ack( + struct irq_service *irq_service, + const struct irq_source_info *info) +{ + uint32_t addr = info->status_reg; + uint32_t value = dm_read_reg(irq_service->ctx, addr); + uint32_t current_status = + get_reg_field_value( + value, + HPD0_DC_HPD_INT_STATUS, + DC_HPD_SENSE_DELAYED); + + dal_irq_service_ack_generic(irq_service, info); + + value = dm_read_reg(irq_service->ctx, info->enable_reg); + + set_reg_field_value( + value, + current_status ? 0 : 1, + HPD0_DC_HPD_INT_CONTROL, + DC_HPD_INT_POLARITY); + + dm_write_reg(irq_service->ctx, info->enable_reg, value); + + return true; +} + +static const struct irq_source_info_funcs hpd_irq_info_funcs = { + .set = NULL, + .ack = hpd_ack +}; + +static const struct irq_source_info_funcs hpd_rx_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + +static const struct irq_source_info_funcs pflip_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + +static const struct irq_source_info_funcs vblank_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +/* compile time expand base address. */ +#define BASE(seg) \ + BASE_INNER(seg) + + +#define SRI(reg_name, block, id)\ + BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + + +#define IRQ_REG_ENTRY(block, reg_num, reg1, mask1, reg2, mask2)\ + .enable_reg = SRI(reg1, block, reg_num),\ + .enable_mask = \ + block ## reg_num ## _ ## reg1 ## __ ## mask1 ## _MASK,\ + .enable_value = {\ + block ## reg_num ## _ ## reg1 ## __ ## mask1 ## _MASK,\ + ~block ## reg_num ## _ ## reg1 ## __ ## mask1 ## _MASK \ + },\ + .ack_reg = SRI(reg2, block, reg_num),\ + .ack_mask = \ + block ## reg_num ## _ ## reg2 ## __ ## mask2 ## _MASK,\ + .ack_value = \ + block ## reg_num ## _ ## reg2 ## __ ## mask2 ## _MASK \ + + + +#define hpd_int_entry(reg_num)\ + [DC_IRQ_SOURCE_HPD1 + reg_num] = {\ + IRQ_REG_ENTRY(HPD, reg_num,\ + DC_HPD_INT_CONTROL, DC_HPD_INT_EN,\ + DC_HPD_INT_CONTROL, DC_HPD_INT_ACK),\ + .status_reg = SRI(DC_HPD_INT_STATUS, HPD, reg_num),\ + .funcs = &hpd_irq_info_funcs\ + } + +#define hpd_rx_int_entry(reg_num)\ + [DC_IRQ_SOURCE_HPD1RX + reg_num] = {\ + IRQ_REG_ENTRY(HPD, reg_num,\ + DC_HPD_INT_CONTROL, DC_HPD_RX_INT_EN,\ + DC_HPD_INT_CONTROL, DC_HPD_RX_INT_ACK),\ + .status_reg = SRI(DC_HPD_INT_STATUS, HPD, reg_num),\ + .funcs = &hpd_rx_irq_info_funcs\ + } +#define pflip_int_entry(reg_num)\ + [DC_IRQ_SOURCE_PFLIP1 + reg_num] = {\ + IRQ_REG_ENTRY(HUBPREQ, reg_num,\ + DCSURF_SURFACE_FLIP_INTERRUPT, SURFACE_FLIP_INT_MASK,\ + DCSURF_SURFACE_FLIP_INTERRUPT, SURFACE_FLIP_CLEAR),\ + .funcs = &pflip_irq_info_funcs\ + } + +#define vupdate_int_entry(reg_num)\ + [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\ + IRQ_REG_ENTRY(OTG, reg_num,\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_INT_EN,\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_EVENT_CLEAR),\ + .funcs = &vblank_irq_info_funcs\ + } + +#define vblank_int_entry(reg_num)\ + [DC_IRQ_SOURCE_VBLANK1 + reg_num] = {\ + IRQ_REG_ENTRY(OTG, reg_num,\ + OTG_GLOBAL_SYNC_STATUS, VSTARTUP_INT_EN,\ + OTG_GLOBAL_SYNC_STATUS, VSTARTUP_EVENT_CLEAR),\ + .funcs = &vblank_irq_info_funcs\ + } + +#define dummy_irq_entry() \ + {\ + .funcs = &dummy_irq_info_funcs\ + } + +#define i2c_int_entry(reg_num) \ + [DC_IRQ_SOURCE_I2C_DDC ## reg_num] = dummy_irq_entry() + +#define dp_sink_int_entry(reg_num) \ + [DC_IRQ_SOURCE_DPSINK ## reg_num] = dummy_irq_entry() + +#define gpio_pad_int_entry(reg_num) \ + [DC_IRQ_SOURCE_GPIOPAD ## reg_num] = dummy_irq_entry() + +#define dc_underflow_int_entry(reg_num) \ + [DC_IRQ_SOURCE_DC ## reg_num ## UNDERFLOW] = dummy_irq_entry() + +static const struct irq_source_info_funcs dummy_irq_info_funcs = { + .set = dal_irq_service_dummy_set, + .ack = dal_irq_service_dummy_ack +}; + +static const struct irq_source_info +irq_source_info_dcn21[DAL_IRQ_SOURCES_NUMBER] = { + [DC_IRQ_SOURCE_INVALID] = dummy_irq_entry(), + hpd_int_entry(0), + hpd_int_entry(1), + hpd_int_entry(2), + hpd_int_entry(3), + hpd_int_entry(4), + hpd_rx_int_entry(0), + hpd_rx_int_entry(1), + hpd_rx_int_entry(2), + hpd_rx_int_entry(3), + hpd_rx_int_entry(4), + i2c_int_entry(1), + i2c_int_entry(2), + i2c_int_entry(3), + i2c_int_entry(4), + i2c_int_entry(5), + i2c_int_entry(6), + dp_sink_int_entry(1), + dp_sink_int_entry(2), + dp_sink_int_entry(3), + dp_sink_int_entry(4), + dp_sink_int_entry(5), + dp_sink_int_entry(6), + [DC_IRQ_SOURCE_TIMER] = dummy_irq_entry(), + pflip_int_entry(0), + pflip_int_entry(1), + pflip_int_entry(2), + pflip_int_entry(3), + [DC_IRQ_SOURCE_PFLIP5] = dummy_irq_entry(), + [DC_IRQ_SOURCE_PFLIP6] = dummy_irq_entry(), + [DC_IRQ_SOURCE_PFLIP_UNDERLAY0] = dummy_irq_entry(), + gpio_pad_int_entry(0), + gpio_pad_int_entry(1), + gpio_pad_int_entry(2), + gpio_pad_int_entry(3), + gpio_pad_int_entry(4), + gpio_pad_int_entry(5), + gpio_pad_int_entry(6), + gpio_pad_int_entry(7), + gpio_pad_int_entry(8), + gpio_pad_int_entry(9), + gpio_pad_int_entry(10), + gpio_pad_int_entry(11), + gpio_pad_int_entry(12), + gpio_pad_int_entry(13), + gpio_pad_int_entry(14), + gpio_pad_int_entry(15), + gpio_pad_int_entry(16), + gpio_pad_int_entry(17), + gpio_pad_int_entry(18), + gpio_pad_int_entry(19), + gpio_pad_int_entry(20), + gpio_pad_int_entry(21), + gpio_pad_int_entry(22), + gpio_pad_int_entry(23), + gpio_pad_int_entry(24), + gpio_pad_int_entry(25), + gpio_pad_int_entry(26), + gpio_pad_int_entry(27), + gpio_pad_int_entry(28), + gpio_pad_int_entry(29), + gpio_pad_int_entry(30), + dc_underflow_int_entry(1), + dc_underflow_int_entry(2), + dc_underflow_int_entry(3), + dc_underflow_int_entry(4), + dc_underflow_int_entry(5), + dc_underflow_int_entry(6), + [DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(), + [DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(), + vupdate_int_entry(0), + vupdate_int_entry(1), + vupdate_int_entry(2), + vupdate_int_entry(3), + vupdate_int_entry(4), + vupdate_int_entry(5), + vblank_int_entry(0), + vblank_int_entry(1), + vblank_int_entry(2), + vblank_int_entry(3), + vblank_int_entry(4), + vblank_int_entry(5), +}; + +static const struct irq_service_funcs irq_service_funcs_dcn21 = { + .to_dal_irq_source = to_dal_irq_source_dcn21 +}; + +static void construct( + struct irq_service *irq_service, + struct irq_service_init_data *init_data) +{ + dal_irq_service_construct(irq_service, init_data); + + irq_service->info = irq_source_info_dcn21; + irq_service->funcs = &irq_service_funcs_dcn21; +} + +struct irq_service *dal_irq_service_dcn21_create( + struct irq_service_init_data *init_data) +{ + struct irq_service *irq_service = kzalloc(sizeof(*irq_service), + GFP_KERNEL); + + if (!irq_service) + return NULL; + + construct(irq_service, init_data); + return irq_service; +} diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.h b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.h new file mode 100644 index 000000000000..da2bd0e93d7a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.h @@ -0,0 +1,34 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DAL_IRQ_SERVICE_DCN21_H__ +#define __DAL_IRQ_SERVICE_DCN21_H__ + +#include "../irq_service.h" + +struct irq_service *dal_irq_service_dcn21_create( + struct irq_service_init_data *init_data); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c index c9a6dd878d9b..ff664bdb1482 100644 --- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c @@ -77,6 +77,10 @@ static void virtual_audio_mute_control( struct stream_encoder *enc, bool mute) {} +static void virtual_stream_encoder_reset_hdmi_stream_attribute( + struct stream_encoder *enc) +{} + #ifdef CONFIG_DRM_AMD_DC_DCN2_0 #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT static void virtual_enc_dp_set_odm_combine( @@ -116,6 +120,7 @@ static const struct stream_encoder_funcs virtual_str_enc_funcs = { .audio_mute_control = virtual_audio_mute_control, .set_avmute = virtual_stream_encoder_set_avmute, + .hdmi_reset_stream_attribute = virtual_stream_encoder_reset_hdmi_stream_attribute, }; bool virtual_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/include/audio_types.h b/drivers/gpu/drm/amd/display/include/audio_types.h index 6364fbc24cfe..66a54da0641c 100644 --- a/drivers/gpu/drm/amd/display/include/audio_types.h +++ b/drivers/gpu/drm/amd/display/include/audio_types.h @@ -38,8 +38,8 @@ struct audio_crtc_info { uint32_t h_active; uint32_t v_active; uint32_t pixel_repetition; - uint32_t requested_pixel_clock; /* in KHz */ - uint32_t calculated_pixel_clock; /* in KHz */ + uint32_t requested_pixel_clock_100Hz; /* in 100Hz */ + uint32_t calculated_pixel_clock_100Hz; /* in 100Hz */ uint32_t refresh_rate; enum dc_color_depth color_depth; bool interlaced; diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index 887e6a8597c4..1be6c44fd32f 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -137,10 +137,13 @@ #define RAVEN1_F0 0xF0 #define RAVEN_UNKNOWN 0xFF +#define PICASSO_15D8_REV_E3 0xE3 +#define PICASSO_15D8_REV_E4 0xE4 + #define ASICREV_IS_RAVEN(eChipRev) ((eChipRev >= RAVEN_A0) && eChipRev < RAVEN_UNKNOWN) #define ASICREV_IS_PICASSO(eChipRev) ((eChipRev >= PICASSO_A0) && (eChipRev < RAVEN2_A0)) -#define ASICREV_IS_RAVEN2(eChipRev) ((eChipRev >= RAVEN2_A0) && (eChipRev < 0xF0)) - +#define ASICREV_IS_RAVEN2(eChipRev) ((eChipRev >= RAVEN2_A0) && (eChipRev < PICASSO_15D8_REV_E3)) +#define ASICREV_IS_DALI(eChipRev) ((eChipRev >= PICASSO_15D8_REV_E3) && (eChipRev < RAVEN1_F0)) #define ASICREV_IS_RV1_F0(eChipRev) ((eChipRev >= RAVEN1_F0) && (eChipRev < RAVEN_UNKNOWN)) @@ -151,6 +154,21 @@ #define FAMILY_NV 143 /* DCN 2*/ +enum { + NV_NAVI10_P_A0 = 1, + NV_NAVI12_P_A0 = 10, + NV_NAVI14_M_A0 = 20, + NV_UNKNOWN = 0xFF +}; + +#define ASICREV_IS_NAVI10_P(eChipRev) (eChipRev < NV_NAVI12_P_A0) +#define ASICREV_IS_NAVI12_P(eChipRev) ((eChipRev >= NV_NAVI12_P_A0) && (eChipRev < NV_NAVI14_M_A0)) +#define ASICREV_IS_NAVI14_M(eChipRev) ((eChipRev >= NV_NAVI14_M_A0) && (eChipRev < NV_UNKNOWN)) +#endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) +#define RENOIR_A0 0x91 +#define DEVICE_ID_RENOIR_1636 0x1636 // Renoir +#define ASICREV_IS_RENOIR(eChipRev) ((eChipRev >= RENOIR_A0) && (eChipRev < 0xFF)) #endif /* diff --git a/drivers/gpu/drm/amd/display/include/dal_types.h b/drivers/gpu/drm/amd/display/include/dal_types.h index 1e3ce4d847ae..fcc42372b6cf 100644 --- a/drivers/gpu/drm/amd/display/include/dal_types.h +++ b/drivers/gpu/drm/amd/display/include/dal_types.h @@ -49,6 +49,9 @@ enum dce_version { #if defined(CONFIG_DRM_AMD_DC_DCN2_0) DCN_VERSION_2_0, #endif +#if defined(CONFIG_DRM_AMD_DC_DCN2_1) + DCN_VERSION_2_1, +#endif DCN_VERSION_MAX }; diff --git a/drivers/gpu/drm/amd/display/include/ddc_service_types.h b/drivers/gpu/drm/amd/display/include/ddc_service_types.h index d968956a10cd..18961707db23 100644 --- a/drivers/gpu/drm/amd/display/include/ddc_service_types.h +++ b/drivers/gpu/drm/amd/display/include/ddc_service_types.h @@ -25,10 +25,12 @@ #ifndef __DAL_DDC_SERVICE_TYPES_H__ #define __DAL_DDC_SERVICE_TYPES_H__ -#define DP_BRANCH_DEVICE_ID_1 0x0010FA -#define DP_BRANCH_DEVICE_ID_2 0x0022B9 -#define DP_BRANCH_DEVICE_ID_3 0x00001A -#define DP_BRANCH_DEVICE_ID_4 0x0080e1 +/* 0010FA dongles (ST Micro) external converter chip id */ +#define DP_BRANCH_DEVICE_ID_0010FA 0x0010FA +/* 0022B9 external converter chip id */ +#define DP_BRANCH_DEVICE_ID_0022B9 0x0022B9 +#define DP_BRANCH_DEVICE_ID_00001A 0x00001A +#define DP_BRANCH_DEVICE_ID_0080E1 0x0080e1 enum ddc_result { DDC_RESULT_UNKNOWN = 0, diff --git a/drivers/gpu/drm/amd/display/include/gpio_interface.h b/drivers/gpu/drm/amd/display/include/gpio_interface.h index 7de64195dc33..5e888a093c16 100644 --- a/drivers/gpu/drm/amd/display/include/gpio_interface.h +++ b/drivers/gpu/drm/amd/display/include/gpio_interface.h @@ -93,8 +93,17 @@ enum sync_source dal_gpio_get_sync_source( enum gpio_pin_output_state dal_gpio_get_output_state( const struct gpio *gpio); +struct hw_ddc *dal_gpio_get_ddc(struct gpio *gpio); + +struct hw_hpd *dal_gpio_get_hpd(struct gpio *gpio); + +struct hw_generic *dal_gpio_get_generic(struct gpio *gpio); + /* Close the handle */ void dal_gpio_close( struct gpio *gpio); + + + #endif diff --git a/drivers/gpu/drm/amd/display/include/gpio_service_interface.h b/drivers/gpu/drm/amd/display/include/gpio_service_interface.h index f40259bade40..9c55d247227e 100644 --- a/drivers/gpu/drm/amd/display/include/gpio_service_interface.h +++ b/drivers/gpu/drm/amd/display/include/gpio_service_interface.h @@ -51,13 +51,29 @@ struct gpio *dal_gpio_service_create_irq( uint32_t offset, uint32_t mask); +struct gpio *dal_gpio_service_create_generic_mux( + struct gpio_service *service, + uint32_t offset, + uint32_t mask); + +void dal_gpio_destroy_generic_mux( + struct gpio **mux); + +enum gpio_result dal_mux_setup_config( + struct gpio *mux, + struct gpio_generic_mux_config *config); + +struct gpio_pin_info dal_gpio_get_generic_pin_info( + struct gpio_service *service, + enum gpio_id id, + uint32_t en); + struct ddc *dal_gpio_create_ddc( struct gpio_service *service, uint32_t offset, uint32_t mask, struct gpio_ddc_hw_info *info); - void dal_gpio_destroy_ddc( struct ddc **ddc); diff --git a/drivers/gpu/drm/amd/display/include/link_service_types.h b/drivers/gpu/drm/amd/display/include/link_service_types.h index 80f0d93cfd94..876b0b3e1a9c 100644 --- a/drivers/gpu/drm/amd/display/include/link_service_types.h +++ b/drivers/gpu/drm/amd/display/include/link_service_types.h @@ -71,14 +71,17 @@ enum link_training_result { struct link_training_settings { struct dc_link_settings link_settings; struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]; - bool allow_invalid_msa_timing_param; -}; -enum hw_dp_training_pattern { - HW_DP_TRAINING_PATTERN_1 = 0, - HW_DP_TRAINING_PATTERN_2, - HW_DP_TRAINING_PATTERN_3, - HW_DP_TRAINING_PATTERN_4 + enum dc_voltage_swing *voltage_swing; + enum dc_pre_emphasis *pre_emphasis; + enum dc_post_cursor2 *post_cursor2; + + uint16_t cr_pattern_time; + uint16_t eq_pattern_time; + enum dc_dp_training_pattern pattern_for_eq; + + bool enhanced_framing; + bool allow_invalid_msa_timing_param; }; /*TODO: Move this enum test harness*/ diff --git a/drivers/gpu/drm/amd/display/include/logger_interface.h b/drivers/gpu/drm/amd/display/include/logger_interface.h index a0b68c266dab..6e008de25629 100644 --- a/drivers/gpu/drm/amd/display/include/logger_interface.h +++ b/drivers/gpu/drm/amd/display/include/logger_interface.h @@ -155,4 +155,6 @@ void context_clock_trace( #define DISPLAY_STATS_END(entry) (void)(entry) +#define LOG_GAMMA_WRITE(msg, ...) + #endif /* __DAL_LOGGER_INTERFACE_H__ */ diff --git a/drivers/gpu/drm/amd/display/include/logger_types.h b/drivers/gpu/drm/amd/display/include/logger_types.h index ea8d445816b8..2b219cdb13ad 100644 --- a/drivers/gpu/drm/amd/display/include/logger_types.h +++ b/drivers/gpu/drm/amd/display/include/logger_types.h @@ -63,6 +63,9 @@ #define DC_LOG_IF_TRACE(...) pr_debug("[IF_TRACE]:"__VA_ARGS__) #define DC_LOG_PERF_TRACE(...) DRM_DEBUG_KMS(__VA_ARGS__) #define DC_LOG_RETIMER_REDRIVER(...) DRM_DEBUG_KMS(__VA_ARGS__) +#define DC_LOG_GAMMA(...) pr_debug("[GAMMA]:"__VA_ARGS__) +#define DC_LOG_ALL_GAMMA(...) pr_debug("[GAMMA]:"__VA_ARGS__) +#define DC_LOG_ALL_TF_CHANNELS(...) pr_debug("[GAMMA]:"__VA_ARGS__) #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #define DC_LOG_DSC(...) DRM_DEBUG_KMS(__VA_ARGS__) #endif @@ -117,6 +120,10 @@ enum dc_log_type { LOG_DSC, #endif LOG_DWB, + LOG_GAMMA_DEBUG, + LOG_MAX_HW_POINTS, + LOG_ALL_TF_CHANNELS, + LOG_SAMPLE_1DLUT, LOG_SECTION_TOTAL_COUNT }; diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c index 88898935a5e6..2d8f14b69117 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c @@ -30,7 +30,6 @@ #include "opp.h" #include "color_gamma.h" - #define NUM_PTS_IN_REGION 16 #define NUM_REGIONS 32 #define MAX_HW_POINTS (NUM_PTS_IN_REGION*NUM_REGIONS) @@ -40,6 +39,33 @@ static struct hw_x_point coordinates_x[MAX_HW_POINTS + 2]; static struct fixed31_32 pq_table[MAX_HW_POINTS + 2]; static struct fixed31_32 de_pq_table[MAX_HW_POINTS + 2]; +// these are helpers for calculations to reduce stack usage +// do not depend on these being preserved across calls +static struct fixed31_32 scratch_1; +static struct fixed31_32 scratch_2; +static struct translate_from_linear_space_args scratch_gamma_args; + +/* Helper to optimize gamma calculation, only use in translate_from_linear, in + * particular the dc_fixpt_pow function which is very expensive + * The idea is that our regions for X points are exponential and currently they all use + * the same number of points (NUM_PTS_IN_REGION) and in each region every point + * is exactly 2x the one at the same index in the previous region. In other words + * X[i] = 2 * X[i-NUM_PTS_IN_REGION] for i>=16 + * The other fact is that (2x)^gamma = 2^gamma * x^gamma + * So we compute and save x^gamma for the first 16 regions, and for every next region + * just multiply with 2^gamma which can be computed once, and save the result so we + * recursively compute all the values. + */ +static struct fixed31_32 pow_buffer[NUM_PTS_IN_REGION]; +static struct fixed31_32 gamma_of_2; // 2^gamma +int pow_buffer_ptr = -1; + /*sRGB 709 2.2 2.4 P3*/ +static const int32_t gamma_numerator01[] = { 31308, 180000, 0, 0, 0}; +static const int32_t gamma_numerator02[] = { 12920, 4500, 0, 0, 0}; +static const int32_t gamma_numerator03[] = { 55, 99, 0, 0, 0}; +static const int32_t gamma_numerator04[] = { 55, 99, 0, 0, 0}; +static const int32_t gamma_numerator05[] = { 2400, 2200, 2200, 2400, 2600}; + static bool pq_initialized; /* = false; */ static bool de_pq_initialized; /* = false; */ @@ -71,6 +97,18 @@ void setup_x_points_distribution(void) } } +void log_x_points_distribution(struct dal_logger *logger) +{ + int i = 0; + + if (logger != NULL) { + LOG_GAMMA_WRITE("Log X Distribution\n"); + + for (i = 0; i < MAX_HW_POINTS; i++) + LOG_GAMMA_WRITE("%llu\n", coordinates_x[i].x.value); + } +} + static void compute_pq(struct fixed31_32 in_x, struct fixed31_32 *out_y) { /* consts for PQ gamma formula. */ @@ -135,59 +173,68 @@ static void compute_de_pq(struct fixed31_32 in_x, struct fixed31_32 *out_y) } + /*de gamma, none linear to linear*/ -static void compute_hlg_oetf(struct fixed31_32 in_x, bool is_light0_12, struct fixed31_32 *out_y) +static void compute_hlg_eotf(struct fixed31_32 in_x, + struct fixed31_32 *out_y, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { struct fixed31_32 a; struct fixed31_32 b; struct fixed31_32 c; struct fixed31_32 threshold; - struct fixed31_32 reference_white_level; + struct fixed31_32 x; + struct fixed31_32 scaling_factor = + dc_fixpt_from_fraction(max_luminance_nits, sdr_white_level); a = dc_fixpt_from_fraction(17883277, 100000000); - if (is_light0_12) { - /*light 0-12*/ - b = dc_fixpt_from_fraction(28466892, 100000000); - c = dc_fixpt_from_fraction(55991073, 100000000); - threshold = dc_fixpt_one; - reference_white_level = dc_fixpt_half; + b = dc_fixpt_from_fraction(28466892, 100000000); + c = dc_fixpt_from_fraction(55991073, 100000000); + threshold = dc_fixpt_from_fraction(1, 2); + + if (dc_fixpt_lt(in_x, threshold)) { + x = dc_fixpt_mul(in_x, in_x); + x = dc_fixpt_div_int(x, 3); } else { - /*light 0-1*/ - b = dc_fixpt_from_fraction(2372241, 100000000); - c = dc_fixpt_add(dc_fixpt_one, dc_fixpt_from_fraction(429347, 100000000)); - threshold = dc_fixpt_from_fraction(1, 12); - reference_white_level = dc_fixpt_pow(dc_fixpt_from_fraction(3, 1), dc_fixpt_half); + x = dc_fixpt_sub(in_x, c); + x = dc_fixpt_div(x, a); + x = dc_fixpt_exp(x); + x = dc_fixpt_add(x, b); + x = dc_fixpt_div_int(x, 12); } - if (dc_fixpt_lt(threshold, in_x)) - *out_y = dc_fixpt_add(c, dc_fixpt_mul(a, dc_fixpt_log(dc_fixpt_sub(in_x, b)))); - else - *out_y = dc_fixpt_mul(dc_fixpt_pow(in_x, dc_fixpt_half), reference_white_level); + *out_y = dc_fixpt_mul(x, scaling_factor); + } /*re gamma, linear to none linear*/ -static void compute_hlg_eotf(struct fixed31_32 in_x, bool is_light0_12, struct fixed31_32 *out_y) +static void compute_hlg_oetf(struct fixed31_32 in_x, struct fixed31_32 *out_y, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { struct fixed31_32 a; struct fixed31_32 b; struct fixed31_32 c; - struct fixed31_32 reference_white_level; + struct fixed31_32 threshold; + struct fixed31_32 x; + struct fixed31_32 scaling_factor = + dc_fixpt_from_fraction(sdr_white_level, max_luminance_nits); a = dc_fixpt_from_fraction(17883277, 100000000); - if (is_light0_12) { - /*light 0-12*/ - b = dc_fixpt_from_fraction(28466892, 100000000); - c = dc_fixpt_from_fraction(55991073, 100000000); - reference_white_level = dc_fixpt_from_fraction(4, 1); + b = dc_fixpt_from_fraction(28466892, 100000000); + c = dc_fixpt_from_fraction(55991073, 100000000); + threshold = dc_fixpt_from_fraction(1, 12); + x = dc_fixpt_mul(in_x, scaling_factor); + + + if (dc_fixpt_lt(x, threshold)) { + x = dc_fixpt_mul(x, dc_fixpt_from_fraction(3, 1)); + *out_y = dc_fixpt_pow(x, dc_fixpt_half); } else { - /*light 0-1*/ - b = dc_fixpt_from_fraction(2372241, 100000000); - c = dc_fixpt_add(dc_fixpt_one, dc_fixpt_from_fraction(429347, 100000000)); - reference_white_level = dc_fixpt_from_fraction(1, 3); + x = dc_fixpt_mul(x, dc_fixpt_from_fraction(12, 1)); + x = dc_fixpt_sub(x, b); + x = dc_fixpt_log(x); + x = dc_fixpt_mul(a, x); + *out_y = dc_fixpt_add(x, c); } - if (dc_fixpt_lt(dc_fixpt_half, in_x)) - *out_y = dc_fixpt_add(dc_fixpt_exp(dc_fixpt_div(dc_fixpt_sub(in_x, c), a)), b); - else - *out_y = dc_fixpt_mul(dc_fixpt_pow(in_x, dc_fixpt_from_fraction(2, 1)), reference_white_level); } @@ -243,93 +290,101 @@ struct dividers { struct fixed31_32 divider3; }; -enum gamma_type_index { - gamma_type_index_2_4, - gamma_type_index_2_2, - gamma_type_index_2_2_flat -}; -static void build_coefficients(struct gamma_coefficients *coefficients, enum gamma_type_index type) +static bool build_coefficients(struct gamma_coefficients *coefficients, enum dc_transfer_func_predefined type) { - static const int32_t numerator01[] = { 31308, 180000, 0}; - static const int32_t numerator02[] = { 12920, 4500, 0}; - static const int32_t numerator03[] = { 55, 99, 0}; - static const int32_t numerator04[] = { 55, 99, 0}; - static const int32_t numerator05[] = { 2400, 2200, 2200}; uint32_t i = 0; uint32_t index = 0; + bool ret = true; - if (type == gamma_type_index_2_2) + if (type == TRANSFER_FUNCTION_SRGB) + index = 0; + else if (type == TRANSFER_FUNCTION_BT709) index = 1; - else if (type == gamma_type_index_2_2_flat) + else if (type == TRANSFER_FUNCTION_GAMMA22) index = 2; + else if (type == TRANSFER_FUNCTION_GAMMA24) + index = 3; + else if (type == TRANSFER_FUNCTION_GAMMA26) + index = 4; + else { + ret = false; + goto release; + } do { coefficients->a0[i] = dc_fixpt_from_fraction( - numerator01[index], 10000000); + gamma_numerator01[index], 10000000); coefficients->a1[i] = dc_fixpt_from_fraction( - numerator02[index], 1000); + gamma_numerator02[index], 1000); coefficients->a2[i] = dc_fixpt_from_fraction( - numerator03[index], 1000); + gamma_numerator03[index], 1000); coefficients->a3[i] = dc_fixpt_from_fraction( - numerator04[index], 1000); + gamma_numerator04[index], 1000); coefficients->user_gamma[i] = dc_fixpt_from_fraction( - numerator05[index], 1000); + gamma_numerator05[index], 1000); ++i; } while (i != ARRAY_SIZE(coefficients->a0)); +release: + return ret; } static struct fixed31_32 translate_from_linear_space( - struct fixed31_32 arg, - struct fixed31_32 a0, - struct fixed31_32 a1, - struct fixed31_32 a2, - struct fixed31_32 a3, - struct fixed31_32 gamma) + struct translate_from_linear_space_args *args) { const struct fixed31_32 one = dc_fixpt_from_int(1); - if (dc_fixpt_lt(one, arg)) + if (dc_fixpt_le(one, args->arg)) return one; - if (dc_fixpt_le(arg, dc_fixpt_neg(a0))) - return dc_fixpt_sub( - a2, - dc_fixpt_mul( - dc_fixpt_add( - one, - a3), - dc_fixpt_pow( - dc_fixpt_neg(arg), - dc_fixpt_recip(gamma)))); - else if (dc_fixpt_le(a0, arg)) - return dc_fixpt_sub( - dc_fixpt_mul( - dc_fixpt_add( - one, - a3), - dc_fixpt_pow( - arg, - dc_fixpt_recip(gamma))), - a2); + if (dc_fixpt_le(args->arg, dc_fixpt_neg(args->a0))) { + scratch_1 = dc_fixpt_add(one, args->a3); + scratch_2 = dc_fixpt_pow( + dc_fixpt_neg(args->arg), + dc_fixpt_recip(args->gamma)); + scratch_1 = dc_fixpt_mul(scratch_1, scratch_2); + scratch_1 = dc_fixpt_sub(args->a2, scratch_1); + + return scratch_1; + } else if (dc_fixpt_le(args->a0, args->arg)) { + if (pow_buffer_ptr == 0) { + gamma_of_2 = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_recip(args->gamma)); + } + scratch_1 = dc_fixpt_add(one, args->a3); + if (pow_buffer_ptr < 16) + scratch_2 = dc_fixpt_pow(args->arg, + dc_fixpt_recip(args->gamma)); + else + scratch_2 = dc_fixpt_mul(gamma_of_2, + pow_buffer[pow_buffer_ptr%16]); + + pow_buffer[pow_buffer_ptr%16] = scratch_2; + pow_buffer_ptr++; + + scratch_1 = dc_fixpt_mul(scratch_1, scratch_2); + scratch_1 = dc_fixpt_sub(scratch_1, args->a2); + + return scratch_1; + } else - return dc_fixpt_mul( - arg, - a1); + return dc_fixpt_mul(args->arg, args->a1); } static struct fixed31_32 calculate_gamma22(struct fixed31_32 arg) { struct fixed31_32 gamma = dc_fixpt_from_fraction(22, 10); - return translate_from_linear_space(arg, - dc_fixpt_zero, - dc_fixpt_zero, - dc_fixpt_zero, - dc_fixpt_zero, - gamma); + scratch_gamma_args.arg = arg; + scratch_gamma_args.a0 = dc_fixpt_zero; + scratch_gamma_args.a1 = dc_fixpt_zero; + scratch_gamma_args.a2 = dc_fixpt_zero; + scratch_gamma_args.a3 = dc_fixpt_zero; + scratch_gamma_args.gamma = gamma; + + return translate_from_linear_space(&scratch_gamma_args); } static struct fixed31_32 translate_to_linear_space( @@ -365,18 +420,19 @@ static struct fixed31_32 translate_to_linear_space( return linear; } -static inline struct fixed31_32 translate_from_linear_space_ex( +static struct fixed31_32 translate_from_linear_space_ex( struct fixed31_32 arg, struct gamma_coefficients *coeff, uint32_t color_index) { - return translate_from_linear_space( - arg, - coeff->a0[color_index], - coeff->a1[color_index], - coeff->a2[color_index], - coeff->a3[color_index], - coeff->user_gamma[color_index]); + scratch_gamma_args.arg = arg; + scratch_gamma_args.a0 = coeff->a0[color_index]; + scratch_gamma_args.a1 = coeff->a1[color_index]; + scratch_gamma_args.a2 = coeff->a2[color_index]; + scratch_gamma_args.a3 = coeff->a3[color_index]; + scratch_gamma_args.gamma = coeff->user_gamma[color_index]; + + return translate_from_linear_space(&scratch_gamma_args); } @@ -709,30 +765,42 @@ static void build_de_pq(struct pwl_float_data_ex *de_pq, } } -static void build_regamma(struct pwl_float_data_ex *rgb_regamma, +static bool build_regamma(struct pwl_float_data_ex *rgb_regamma, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, enum gamma_type_index type) + const struct hw_x_point *coordinate_x, enum dc_transfer_func_predefined type) { uint32_t i; + bool ret = false; - struct gamma_coefficients coeff; + struct gamma_coefficients *coeff; struct pwl_float_data_ex *rgb = rgb_regamma; const struct hw_x_point *coord_x = coordinate_x; - build_coefficients(&coeff, type); + coeff = kvzalloc(sizeof(*coeff), GFP_KERNEL); + if (!coeff) + goto release; - i = 0; + if (!build_coefficients(coeff, type)) + goto release; - while (i != hw_points_num + 1) { + memset(pow_buffer, 0, NUM_PTS_IN_REGION * sizeof(struct fixed31_32)); + pow_buffer_ptr = 0; // see variable definition for more info + i = 0; + while (i <= hw_points_num) { /*TODO use y vs r,g,b*/ rgb->r = translate_from_linear_space_ex( - coord_x->x, &coeff, 0); + coord_x->x, coeff, 0); rgb->g = rgb->r; rgb->b = rgb->r; ++coord_x; ++rgb; ++i; } + pow_buffer_ptr = -1; // reset back to no optimize + ret = true; +release: + kfree(coeff); + return ret; } static void hermite_spline_eetf(struct fixed31_32 input_x, @@ -862,6 +930,8 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, else max_content = max_display; + if (!use_eetf) + pow_buffer_ptr = 0; // see var definition for more info rgb += 32; // first 32 points have problems with fixed point, too small coord_x += 32; for (i = 32; i <= hw_points_num; i++) { @@ -900,19 +970,23 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, ++coord_x; ++rgb; } + pow_buffer_ptr = -1; return true; } -static void build_degamma(struct pwl_float_data_ex *curve, +static bool build_degamma(struct pwl_float_data_ex *curve, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, enum gamma_type_index type) + const struct hw_x_point *coordinate_x, enum dc_transfer_func_predefined type) { uint32_t i; struct gamma_coefficients coeff; uint32_t begin_index, end_index; + bool ret = false; + + if (!build_coefficients(&coeff, type)) + goto release; - build_coefficients(&coeff, type); i = 0; /* X points is 2^-25 to 2^7 @@ -941,11 +1015,19 @@ static void build_degamma(struct pwl_float_data_ex *curve, curve[i].b = dc_fixpt_one; i++; } + ret = true; +release: + return ret; } + + + + static void build_hlg_degamma(struct pwl_float_data_ex *degamma, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, bool is_light0_12) + const struct hw_x_point *coordinate_x, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { uint32_t i; @@ -953,9 +1035,9 @@ static void build_hlg_degamma(struct pwl_float_data_ex *degamma, const struct hw_x_point *coord_x = coordinate_x; i = 0; - + //check when i == 434 while (i != hw_points_num + 1) { - compute_hlg_oetf(coord_x->x, is_light0_12, &rgb->r); + compute_hlg_eotf(coord_x->x, &rgb->r, sdr_white_level, max_luminance_nits); rgb->g = rgb->r; rgb->b = rgb->r; ++coord_x; @@ -964,9 +1046,11 @@ static void build_hlg_degamma(struct pwl_float_data_ex *degamma, } } + static void build_hlg_regamma(struct pwl_float_data_ex *regamma, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, bool is_light0_12) + const struct hw_x_point *coordinate_x, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { uint32_t i; @@ -975,8 +1059,9 @@ static void build_hlg_regamma(struct pwl_float_data_ex *regamma, i = 0; + //when i == 471 while (i != hw_points_num + 1) { - compute_hlg_eotf(coord_x->x, is_light0_12, &rgb->r); + compute_hlg_oetf(coord_x->x, &rgb->r, sdr_white_level, max_luminance_nits); rgb->g = rgb->r; rgb->b = rgb->r; ++coord_x; @@ -1572,14 +1657,15 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, output_tf->tf == TRANSFER_FUNCTION_SRGB) { if (ramp == NULL) return true; - if ((ramp->is_logical_identity) || + if ((ramp->is_identity && ramp->type != GAMMA_CS_TFM_1D) || (!mapUserRamp && ramp->type == GAMMA_RGB_256)) return true; } output_tf->type = TF_TYPE_DISTRIBUTED_POINTS; - if (ramp && (mapUserRamp || ramp->type != GAMMA_RGB_256)) { + if (ramp && ramp->type != GAMMA_CS_TFM_1D && + (mapUserRamp || ramp->type != GAMMA_RGB_256)) { rgb_user = kvcalloc(ramp->num_entries + _EXTRA_POINTS, sizeof(*rgb_user), GFP_KERNEL); @@ -1634,6 +1720,12 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, MAX_HW_POINTS, coordinates_x, fs_params); + } else if (tf == TRANSFER_FUNCTION_HLG) { + build_freesync_hdr(rgb_regamma, + MAX_HW_POINTS, + coordinates_x, + fs_params); + } else { tf_pts->end_exponent = 0; tf_pts->x_point_at_y1_red = 1; @@ -1642,9 +1734,7 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, build_regamma(rgb_regamma, MAX_HW_POINTS, - coordinates_x, tf == TRANSFER_FUNCTION_SRGB ? gamma_type_index_2_4 : - tf == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); + coordinates_x, tf); } map_regamma_hw_to_x_user(ramp, coeff, rgb_user, coordinates_x, axis_x, rgb_regamma, @@ -1845,13 +1935,19 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, MAX_HW_POINTS, coordinates_x); else if (tf == TRANSFER_FUNCTION_SRGB || - tf == TRANSFER_FUNCTION_BT709) + tf == TRANSFER_FUNCTION_BT709 || + tf == TRANSFER_FUNCTION_GAMMA22 || + tf == TRANSFER_FUNCTION_GAMMA24 || + tf == TRANSFER_FUNCTION_GAMMA26) build_degamma(curve, MAX_HW_POINTS, coordinates_x, - tf == TRANSFER_FUNCTION_SRGB ? - gamma_type_index_2_4 : tf == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); + tf); + else if (tf == TRANSFER_FUNCTION_HLG) + build_hlg_degamma(curve, + MAX_HW_POINTS, + coordinates_x, + 80, 1000); else if (tf == TRANSFER_FUNCTION_LINEAR) { // just copy coordinates_x into curve i = 0; @@ -1938,7 +2034,10 @@ bool mod_color_calculate_curve(enum dc_transfer_func_predefined trans, kvfree(rgb_regamma); } else if (trans == TRANSFER_FUNCTION_SRGB || - trans == TRANSFER_FUNCTION_BT709) { + trans == TRANSFER_FUNCTION_BT709 || + trans == TRANSFER_FUNCTION_GAMMA22 || + trans == TRANSFER_FUNCTION_GAMMA24 || + trans == TRANSFER_FUNCTION_GAMMA26) { rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*rgb_regamma), GFP_KERNEL); @@ -1952,9 +2051,7 @@ bool mod_color_calculate_curve(enum dc_transfer_func_predefined trans, build_regamma(rgb_regamma, MAX_HW_POINTS, coordinates_x, - trans == TRANSFER_FUNCTION_SRGB ? - gamma_type_index_2_4 : trans == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); + trans); for (i = 0; i <= MAX_HW_POINTS ; i++) { points->red[i] = rgb_regamma[i].r; points->green[i] = rgb_regamma[i].g; @@ -1963,18 +2060,21 @@ bool mod_color_calculate_curve(enum dc_transfer_func_predefined trans, ret = true; kvfree(rgb_regamma); - } else if (trans == TRANSFER_FUNCTION_HLG || - trans == TRANSFER_FUNCTION_HLG12) { + } else if (trans == TRANSFER_FUNCTION_HLG) { rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*rgb_regamma), GFP_KERNEL); if (!rgb_regamma) goto rgb_regamma_alloc_fail; + points->end_exponent = 4; + points->x_point_at_y1_red = 12; + points->x_point_at_y1_green = 12; + points->x_point_at_y1_blue = 12; build_hlg_regamma(rgb_regamma, MAX_HW_POINTS, coordinates_x, - trans == TRANSFER_FUNCTION_HLG12 ? true:false); + 80, 1000); for (i = 0; i <= MAX_HW_POINTS ; i++) { points->red[i] = rgb_regamma[i].r; points->green[i] = rgb_regamma[i].g; @@ -2024,8 +2124,10 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, kvfree(rgb_degamma); } else if (trans == TRANSFER_FUNCTION_SRGB || - trans == TRANSFER_FUNCTION_BT709 || - trans == TRANSFER_FUNCTION_GAMMA22) { + trans == TRANSFER_FUNCTION_BT709 || + trans == TRANSFER_FUNCTION_GAMMA22 || + trans == TRANSFER_FUNCTION_GAMMA24 || + trans == TRANSFER_FUNCTION_GAMMA26) { rgb_degamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*rgb_degamma), GFP_KERNEL); @@ -2035,9 +2137,7 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, build_degamma(rgb_degamma, MAX_HW_POINTS, coordinates_x, - trans == TRANSFER_FUNCTION_SRGB ? - gamma_type_index_2_4 : trans == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); + trans); for (i = 0; i <= MAX_HW_POINTS ; i++) { points->red[i] = rgb_degamma[i].r; points->green[i] = rgb_degamma[i].g; @@ -2046,8 +2146,7 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, ret = true; kvfree(rgb_degamma); - } else if (trans == TRANSFER_FUNCTION_HLG || - trans == TRANSFER_FUNCTION_HLG12) { + } else if (trans == TRANSFER_FUNCTION_HLG) { rgb_degamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*rgb_degamma), GFP_KERNEL); @@ -2057,7 +2156,7 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, build_hlg_degamma(rgb_degamma, MAX_HW_POINTS, coordinates_x, - trans == TRANSFER_FUNCTION_HLG12 ? true:false); + 80, 1000); for (i = 0; i <= MAX_HW_POINTS ; i++) { points->red[i] = rgb_degamma[i].r; points->green[i] = rgb_degamma[i].g; diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h index 369953fafadf..44ddea58523a 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h @@ -82,7 +82,17 @@ struct freesync_hdr_tf_params { unsigned int skip_tm; // skip tm }; +struct translate_from_linear_space_args { + struct fixed31_32 arg; + struct fixed31_32 a0; + struct fixed31_32 a1; + struct fixed31_32 a2; + struct fixed31_32 a3; + struct fixed31_32 gamma; +}; + void setup_x_points_distribution(void); +void log_x_points_distribution(struct dal_logger *logger); void precompute_pq(void); void precompute_de_pq(void); diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 7c20171a3b6d..ec70c9b12e1a 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -52,93 +52,6 @@ struct core_freesync { struct dc *dc; }; -void setFieldWithMask(unsigned char *dest, unsigned int mask, unsigned int value) -{ - unsigned int shift = 0; - - if (!mask || !dest) - return; - - while (!((mask >> shift) & 1)) - shift++; - - //reset - *dest = *dest & ~mask; - //set - //dont let value span past mask - value = value & (mask >> shift); - //insert value - *dest = *dest | (value << shift); -} - -// VTEM Byte Offset -#define VRR_VTEM_PB0 0 -#define VRR_VTEM_PB1 1 -#define VRR_VTEM_PB2 2 -#define VRR_VTEM_PB3 3 -#define VRR_VTEM_PB4 4 -#define VRR_VTEM_PB5 5 -#define VRR_VTEM_PB6 6 - -#define VRR_VTEM_MD0 7 -#define VRR_VTEM_MD1 8 -#define VRR_VTEM_MD2 9 -#define VRR_VTEM_MD3 10 - - -// VTEM Byte Masks -//PB0 -#define MASK__VRR_VTEM_PB0__RESERVED0 0x01 -#define MASK__VRR_VTEM_PB0__SYNC 0x02 -#define MASK__VRR_VTEM_PB0__VFR 0x04 -#define MASK__VRR_VTEM_PB0__AFR 0x08 -#define MASK__VRR_VTEM_PB0__DS_TYPE 0x30 - //0: Periodic pseudo-static EM Data Set - //1: Periodic dynamic EM Data Set - //2: Unique EM Data Set - //3: Reserved -#define MASK__VRR_VTEM_PB0__END 0x40 -#define MASK__VRR_VTEM_PB0__NEW 0x80 - -//PB1 -#define MASK__VRR_VTEM_PB1__RESERVED1 0xFF - -//PB2 -#define MASK__VRR_VTEM_PB2__ORGANIZATION_ID 0xFF - //0: This is a Vendor Specific EM Data Set - //1: This EM Data Set is defined by This Specification (HDMI 2.1 r102.clean) - //2: This EM Data Set is defined by CTA-861-G - //3: This EM Data Set is defined by VESA -//PB3 -#define MASK__VRR_VTEM_PB3__DATA_SET_TAG_MSB 0xFF -//PB4 -#define MASK__VRR_VTEM_PB4__DATA_SET_TAG_LSB 0xFF -//PB5 -#define MASK__VRR_VTEM_PB5__DATA_SET_LENGTH_MSB 0xFF -//PB6 -#define MASK__VRR_VTEM_PB6__DATA_SET_LENGTH_LSB 0xFF - - - -//PB7-27 (20 bytes): -//PB7 = MD0 -#define MASK__VRR_VTEM_MD0__VRR_EN 0x01 -#define MASK__VRR_VTEM_MD0__M_CONST 0x02 -#define MASK__VRR_VTEM_MD0__RESERVED2 0x0C -#define MASK__VRR_VTEM_MD0__FVA_FACTOR_M1 0xF0 - -//MD1 -#define MASK__VRR_VTEM_MD1__BASE_VFRONT 0xFF - -//MD2 -#define MASK__VRR_VTEM_MD2__BASE_REFRESH_RATE_98 0x03 -#define MASK__VRR_VTEM_MD2__RB 0x04 -#define MASK__VRR_VTEM_MD2__RESERVED3 0xF8 - -//MD3 -#define MASK__VRR_VTEM_MD3__BASE_REFRESH_RATE_07 0xFF - - #define MOD_FREESYNC_TO_CORE(mod_freesync)\ container_of(mod_freesync, struct core_freesync, public) @@ -435,6 +348,12 @@ static void apply_below_the_range(struct core_freesync *core_freesync, /* Either we've calculated the number of frames to insert, * or we need to insert min duration frames */ + if (last_render_time_in_us / frames_to_insert < + in_out_vrr->min_duration_in_us){ + frames_to_insert -= (frames_to_insert > 1) ? + 1 : 0; + } + if (frames_to_insert > 0) inserted_frame_duration_in_us = last_render_time_in_us / frames_to_insert; @@ -568,22 +487,64 @@ bool mod_freesync_get_v_position(struct mod_freesync *mod_freesync, return false; } -static void build_vrr_infopacket_header_vtem(enum signal_type signal, +static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, struct dc_info_packet *infopacket) { - // HEADER - - // HB0, HB1, HB2 indicates PacketType VTEMPacket - infopacket->hb0 = 0x7F; - infopacket->hb1 = 0xC0; - infopacket->hb2 = 0x00; //sequence_index - - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB0], MASK__VRR_VTEM_PB0__VFR, 1); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB2], MASK__VRR_VTEM_PB2__ORGANIZATION_ID, 1); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB3], MASK__VRR_VTEM_PB3__DATA_SET_TAG_MSB, 0); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB4], MASK__VRR_VTEM_PB4__DATA_SET_TAG_LSB, 1); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB5], MASK__VRR_VTEM_PB5__DATA_SET_LENGTH_MSB, 0); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB6], MASK__VRR_VTEM_PB6__DATA_SET_LENGTH_LSB, 4); + /* PB1 = 0x1A (24bit AMD IEEE OUI (0x00001A) - Byte 0) */ + infopacket->sb[1] = 0x1A; + + /* PB2 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 1) */ + infopacket->sb[2] = 0x00; + + /* PB3 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 2) */ + infopacket->sb[3] = 0x00; + + /* PB4 = Reserved */ + + /* PB5 = Reserved */ + + /* PB6 = [Bits 7:3 = Reserved] */ + + /* PB6 = [Bit 0 = FreeSync Supported] */ + if (vrr->state != VRR_STATE_UNSUPPORTED) + infopacket->sb[6] |= 0x01; + + /* PB6 = [Bit 1 = FreeSync Enabled] */ + if (vrr->state != VRR_STATE_DISABLED && + vrr->state != VRR_STATE_UNSUPPORTED) + infopacket->sb[6] |= 0x02; + + /* PB6 = [Bit 2 = FreeSync Active] */ + if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || + vrr->state == VRR_STATE_ACTIVE_FIXED) + infopacket->sb[6] |= 0x04; + + /* PB7 = FreeSync Minimum refresh rate (Hz) */ + infopacket->sb[7] = (unsigned char)(vrr->min_refresh_in_uhz / 1000000); + + /* PB8 = FreeSync Maximum refresh rate (Hz) + * Note: We should never go above the field rate of the mode timing set. + */ + infopacket->sb[8] = (unsigned char)(vrr->max_refresh_in_uhz / 1000000); + + + //FreeSync HDR + infopacket->sb[9] = 0; + infopacket->sb[10] = 0; +} + +static void build_vrr_infopacket_fs2_data(enum color_transfer_func app_tf, + struct dc_info_packet *infopacket) +{ + if (app_tf != TRANSFER_FUNC_UNKNOWN) { + infopacket->valid = true; + + infopacket->sb[6] |= 0x08; // PB6 = [Bit 3 = Native Color Active] + + if (app_tf == TRANSFER_FUNC_GAMMA_22) { + infopacket->sb[9] |= 0x04; // PB6 = [Bit 2 = Gamma 2.2 EOTF Active] + } + } } static void build_vrr_infopacket_header_v1(enum signal_type signal, @@ -684,105 +645,6 @@ static void build_vrr_infopacket_header_v2(enum signal_type signal, } } -static void build_vrr_vtem_infopacket_data(const struct dc_stream_state *stream, - const struct mod_vrr_params *vrr, - struct dc_info_packet *infopacket) -{ - unsigned int fieldRateInHz; - - if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || - vrr->state == VRR_STATE_ACTIVE_FIXED) { - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD0], MASK__VRR_VTEM_MD0__VRR_EN, 1); - } else { - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD0], MASK__VRR_VTEM_MD0__VRR_EN, 0); - } - - if (!stream->timing.vic) { - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD1], MASK__VRR_VTEM_MD1__BASE_VFRONT, - stream->timing.v_front_porch); - - - /* TODO: In dal2, we check mode flags for a reduced blanking timing. - * Need a way to relay that information to this function. - * if("ReducedBlanking") - * { - * setFieldWithMask(&infopacket->sb[VRR_VTEM_MD2], MASK__VRR_VTEM_MD2__RB, 1; - * } - */ - - //TODO: DAL2 does FixPoint and rounding. Here we might need to account for that - fieldRateInHz = (stream->timing.pix_clk_100hz * 100)/ - (stream->timing.h_total * stream->timing.v_total); - - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD2], MASK__VRR_VTEM_MD2__BASE_REFRESH_RATE_98, - fieldRateInHz >> 8); - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD3], MASK__VRR_VTEM_MD3__BASE_REFRESH_RATE_07, - fieldRateInHz); - - } - infopacket->valid = true; -} - -static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, - struct dc_info_packet *infopacket) -{ - /* PB1 = 0x1A (24bit AMD IEEE OUI (0x00001A) - Byte 0) */ - infopacket->sb[1] = 0x1A; - - /* PB2 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 1) */ - infopacket->sb[2] = 0x00; - - /* PB3 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 2) */ - infopacket->sb[3] = 0x00; - - /* PB4 = Reserved */ - - /* PB5 = Reserved */ - - /* PB6 = [Bits 7:3 = Reserved] */ - - /* PB6 = [Bit 0 = FreeSync Supported] */ - if (vrr->state != VRR_STATE_UNSUPPORTED) - infopacket->sb[6] |= 0x01; - - /* PB6 = [Bit 1 = FreeSync Enabled] */ - if (vrr->state != VRR_STATE_DISABLED && - vrr->state != VRR_STATE_UNSUPPORTED) - infopacket->sb[6] |= 0x02; - - /* PB6 = [Bit 2 = FreeSync Active] */ - if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || - vrr->state == VRR_STATE_ACTIVE_FIXED) - infopacket->sb[6] |= 0x04; - - /* PB7 = FreeSync Minimum refresh rate (Hz) */ - infopacket->sb[7] = (unsigned char)(vrr->min_refresh_in_uhz / 1000000); - - /* PB8 = FreeSync Maximum refresh rate (Hz) - * Note: We should never go above the field rate of the mode timing set. - */ - infopacket->sb[8] = (unsigned char)(vrr->max_refresh_in_uhz / 1000000); - - - //FreeSync HDR - infopacket->sb[9] = 0; - infopacket->sb[10] = 0; -} - -static void build_vrr_infopacket_fs2_data(enum color_transfer_func app_tf, - struct dc_info_packet *infopacket) -{ - if (app_tf != TRANSFER_FUNC_UNKNOWN) { - infopacket->valid = true; - - infopacket->sb[6] |= 0x08; // PB6 = [Bit 3 = Native Color Active] - - if (app_tf == TRANSFER_FUNC_GAMMA_22) { - infopacket->sb[9] |= 0x04; // PB6 = [Bit 2 = Gamma 2.2 EOTF Active] - } - } -} - static void build_vrr_infopacket_checksum(unsigned int *payload_size, struct dc_info_packet *infopacket) { @@ -835,21 +697,6 @@ static void build_vrr_infopacket_v2(enum signal_type signal, infopacket->valid = true; } -static void build_vrr_infopacket_vtem(const struct dc_stream_state *stream, - const struct mod_vrr_params *vrr, - struct dc_info_packet *infopacket) -{ - //VTEM info packet for HdmiVrr - - memset(infopacket, 0, sizeof(struct dc_info_packet)); - - //VTEM Packet is structured differently - build_vrr_infopacket_header_vtem(stream->signal, infopacket); - build_vrr_vtem_infopacket_data(stream, vrr, infopacket); - - infopacket->valid = true; -} - void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync, const struct dc_stream_state *stream, const struct mod_vrr_params *vrr, @@ -862,16 +709,13 @@ void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync, * Check if Freesync is supported. Return if false. If true, * set the corresponding bit in the info packet */ - if (!vrr->supported || (!vrr->send_info_frame && packet_type != PACKET_TYPE_VTEM)) + if (!vrr->supported || (!vrr->send_info_frame)) return; switch (packet_type) { case PACKET_TYPE_FS2: build_vrr_infopacket_v2(stream->signal, vrr, app_tf, infopacket); break; - case PACKET_TYPE_VTEM: - build_vrr_infopacket_vtem(stream, vrr, infopacket); - break; case PACKET_TYPE_VRR: case PACKET_TYPE_FS1: default: @@ -887,8 +731,8 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, struct core_freesync *core_freesync = NULL; unsigned long long nominal_field_rate_in_uhz = 0; unsigned int refresh_range = 0; - unsigned int min_refresh_in_uhz = 0; - unsigned int max_refresh_in_uhz = 0; + unsigned long long min_refresh_in_uhz = 0; + unsigned long long max_refresh_in_uhz = 0; if (mod_freesync == NULL) return; @@ -915,7 +759,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, min_refresh_in_uhz = nominal_field_rate_in_uhz; if (!vrr_settings_require_update(core_freesync, - in_config, min_refresh_in_uhz, max_refresh_in_uhz, + in_config, (unsigned int)min_refresh_in_uhz, (unsigned int)max_refresh_in_uhz, in_out_vrr)) return; @@ -931,15 +775,15 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, return; } else { - in_out_vrr->min_refresh_in_uhz = min_refresh_in_uhz; + in_out_vrr->min_refresh_in_uhz = (unsigned int)min_refresh_in_uhz; in_out_vrr->max_duration_in_us = calc_duration_in_us_from_refresh_in_uhz( - min_refresh_in_uhz); + (unsigned int)min_refresh_in_uhz); - in_out_vrr->max_refresh_in_uhz = max_refresh_in_uhz; + in_out_vrr->max_refresh_in_uhz = (unsigned int)max_refresh_in_uhz; in_out_vrr->min_duration_in_us = calc_duration_in_us_from_refresh_in_uhz( - max_refresh_in_uhz); + (unsigned int)max_refresh_in_uhz); refresh_range = in_out_vrr->max_refresh_in_uhz - in_out_vrr->min_refresh_in_uhz; @@ -950,17 +794,18 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, in_out_vrr->fixed.ramping_active = in_config->ramping; in_out_vrr->btr.btr_enabled = in_config->btr; + if (in_out_vrr->max_refresh_in_uhz < 2 * in_out_vrr->min_refresh_in_uhz) in_out_vrr->btr.btr_enabled = false; + in_out_vrr->btr.btr_active = false; in_out_vrr->btr.inserted_duration_in_us = 0; in_out_vrr->btr.frames_to_insert = 0; in_out_vrr->btr.frame_counter = 0; in_out_vrr->btr.mid_point_in_us = - in_out_vrr->min_duration_in_us + - (in_out_vrr->max_duration_in_us - - in_out_vrr->min_duration_in_us) / 2; + (in_out_vrr->min_duration_in_us + + in_out_vrr->max_duration_in_us) / 2; if (in_out_vrr->state == VRR_STATE_UNSUPPORTED) { in_out_vrr->adjust.v_total_min = stream->timing.v_total; diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h index dcef85994c45..dc187844d10b 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h @@ -173,4 +173,6 @@ bool mod_freesync_is_valid_range(struct mod_freesync *mod_freesync, uint32_t min_refresh_request_in_uhz, uint32_t max_refresh_request_in_uhz); + + #endif diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h index 5b1c9a4c7643..d930bdecb117 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h @@ -27,10 +27,10 @@ #define MOD_INFO_PACKET_H_ #include "mod_shared.h" - //Forward Declarations struct dc_stream_state; struct dc_info_packet; +struct mod_vrr_params; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, struct dc_info_packet *info_packet); diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index bc13c552797f..d885d642ed7f 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -27,9 +27,78 @@ #include "core_types.h" #include "dc_types.h" #include "mod_shared.h" +#include "mod_freesync.h" +#include "dc.h" #define HDMI_INFOFRAME_TYPE_VENDOR 0x81 +// VTEM Byte Offset +#define VTEM_PB0 0 +#define VTEM_PB1 1 +#define VTEM_PB2 2 +#define VTEM_PB3 3 +#define VTEM_PB4 4 +#define VTEM_PB5 5 +#define VTEM_PB6 6 + +#define VTEM_MD0 7 +#define VTEM_MD1 8 +#define VTEM_MD2 9 +#define VTEM_MD3 10 + + +// VTEM Byte Masks +//PB0 +#define MASK_VTEM_PB0__RESERVED0 0x01 +#define MASK_VTEM_PB0__SYNC 0x02 +#define MASK_VTEM_PB0__VFR 0x04 +#define MASK_VTEM_PB0__AFR 0x08 +#define MASK_VTEM_PB0__DS_TYPE 0x30 + //0: Periodic pseudo-static EM Data Set + //1: Periodic dynamic EM Data Set + //2: Unique EM Data Set + //3: Reserved +#define MASK_VTEM_PB0__END 0x40 +#define MASK_VTEM_PB0__NEW 0x80 + +//PB1 +#define MASK_VTEM_PB1__RESERVED1 0xFF + +//PB2 +#define MASK_VTEM_PB2__ORGANIZATION_ID 0xFF + //0: This is a Vendor Specific EM Data Set + //1: This EM Data Set is defined by This Specification (HDMI 2.1 r102.clean) + //2: This EM Data Set is defined by CTA-861-G + //3: This EM Data Set is defined by VESA +//PB3 +#define MASK_VTEM_PB3__DATA_SET_TAG_MSB 0xFF +//PB4 +#define MASK_VTEM_PB4__DATA_SET_TAG_LSB 0xFF +//PB5 +#define MASK_VTEM_PB5__DATA_SET_LENGTH_MSB 0xFF +//PB6 +#define MASK_VTEM_PB6__DATA_SET_LENGTH_LSB 0xFF + + + +//PB7-27 (20 bytes): +//PB7 = MD0 +#define MASK_VTEM_MD0__VRR_EN 0x01 +#define MASK_VTEM_MD0__M_CONST 0x02 +#define MASK_VTEM_MD0__RESERVED2 0x0C +#define MASK_VTEM_MD0__FVA_FACTOR_M1 0xF0 + +//MD1 +#define MASK_VTEM_MD1__BASE_VFRONT 0xFF + +//MD2 +#define MASK_VTEM_MD2__BASE_REFRESH_RATE_98 0x03 +#define MASK_VTEM_MD2__RB 0x04 +#define MASK_VTEM_MD2__RESERVED3 0xF8 + +//MD3 +#define MASK_VTEM_MD3__BASE_REFRESH_RATE_07 0xFF + enum ColorimetryRGBDP { ColorimetryRGB_DP_sRGB = 0, ColorimetryRGB_DP_AdobeRGB = 3, diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c index b3810b864676..05e2be856037 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -66,6 +66,39 @@ static const unsigned char abm_config[abm_defines_max_config][abm_defines_max_le { 3, 6, 10, 12 }, /* Alt #3 - Super aggressiveness */ }; +struct abm_parameters { + unsigned char min_reduction; + unsigned char max_reduction; + unsigned char bright_pos_gain; + unsigned char dark_pos_gain; + unsigned char brightness_gain; + unsigned char contrast_factor; + unsigned char deviation_gain; + unsigned char min_knee; + unsigned char max_knee; +}; + +static const struct abm_parameters abm_settings_config0[abm_defines_max_level] = { +// min_red max_red bright_pos dark_pos brightness_gain contrast deviation min_knee max_knee + {0xff, 0xbf, 0x20, 0x00, 0xff, 0x99, 0xb3, 0x40, 0xE0}, + {0xff, 0x85, 0x20, 0x00, 0xff, 0x90, 0xa8, 0x40, 0xE0}, + {0xff, 0x40, 0x20, 0x00, 0xff, 0x90, 0x68, 0x40, 0xE0}, + {0x82, 0x4d, 0x20, 0x00, 0x00, 0x90, 0xb3, 0x70, 0x70}, +}; + +static const struct abm_parameters abm_settings_config1[abm_defines_max_level] = { +// min_red max_red bright_pos dark_pos brightness_gain contrast deviation min_knee max_knee + {0xf0, 0xd9, 0x20, 0x00, 0x00, 0xa8, 0xb3, 0x70, 0x70}, + {0xcd, 0xa5, 0x20, 0x00, 0x00, 0xa8, 0xb3, 0x70, 0x70}, + {0x99, 0x65, 0x20, 0x00, 0x00, 0xa8, 0xb3, 0x70, 0x70}, + {0x82, 0x4d, 0x20, 0x00, 0x00, 0xa8, 0xb3, 0x70, 0x70}, +}; + +static const struct abm_parameters * const abm_settings[] = { + abm_settings_config0, + abm_settings_config1, +}; + #define NUM_AMBI_LEVEL 5 #define NUM_AGGR_LEVEL 4 #define NUM_POWER_FN_SEGS 8 @@ -131,11 +164,13 @@ struct iram_table_v_2_2 { uint8_t max_reduction[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x16 U0.8 */ uint8_t bright_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x2a U2.6 */ uint8_t dark_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x3e U2.6 */ - uint8_t hybridFactor[NUM_AGGR_LEVEL]; /* 0x52 U0.8 */ - uint8_t contrastFactor[NUM_AGGR_LEVEL]; /* 0x56 U0.8 */ + uint8_t hybrid_factor[NUM_AGGR_LEVEL]; /* 0x52 U0.8 */ + uint8_t contrast_factor[NUM_AGGR_LEVEL]; /* 0x56 U0.8 */ uint8_t deviation_gain[NUM_AGGR_LEVEL]; /* 0x5a U0.8 */ uint8_t iir_curve[NUM_AMBI_LEVEL]; /* 0x5e U0.8 */ - uint8_t pad[29]; /* 0x63 U0.8 */ + uint8_t min_knee[NUM_AGGR_LEVEL]; /* 0x63 U0.8 */ + uint8_t max_knee[NUM_AGGR_LEVEL]; /* 0x67 U0.8 */ + uint8_t pad[21]; /* 0x6b U0.8 */ /* parameters for crgb conversion */ uint16_t crgb_thresh[NUM_POWER_FN_SEGS]; /* 0x80 U3.13 */ @@ -501,15 +536,72 @@ void fill_iram_v_2_2(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parame ram_table->dark_pos_gain[4][2] = 0x00; ram_table->dark_pos_gain[4][3] = 0x00; - ram_table->hybridFactor[0] = 0xff; - ram_table->hybridFactor[1] = 0xff; - ram_table->hybridFactor[2] = 0xff; - ram_table->hybridFactor[3] = 0xc0; + ram_table->hybrid_factor[0] = 0xff; + ram_table->hybrid_factor[1] = 0xff; + ram_table->hybrid_factor[2] = 0xff; + ram_table->hybrid_factor[3] = 0xc0; - ram_table->contrastFactor[0] = 0x99; - ram_table->contrastFactor[1] = 0x99; - ram_table->contrastFactor[2] = 0x90; - ram_table->contrastFactor[3] = 0x80; + ram_table->contrast_factor[0] = 0x99; + ram_table->contrast_factor[1] = 0x99; + ram_table->contrast_factor[2] = 0x90; + ram_table->contrast_factor[3] = 0x80; + + ram_table->iir_curve[0] = 0x65; + ram_table->iir_curve[1] = 0x65; + ram_table->iir_curve[2] = 0x65; + ram_table->iir_curve[3] = 0x65; + ram_table->iir_curve[4] = 0x65; + + //Gamma 2.2 + ram_table->crgb_thresh[0] = cpu_to_be16(0x127c); + ram_table->crgb_thresh[1] = cpu_to_be16(0x151b); + ram_table->crgb_thresh[2] = cpu_to_be16(0x17d5); + ram_table->crgb_thresh[3] = cpu_to_be16(0x1a56); + ram_table->crgb_thresh[4] = cpu_to_be16(0x1c83); + ram_table->crgb_thresh[5] = cpu_to_be16(0x1e72); + ram_table->crgb_thresh[6] = cpu_to_be16(0x20f0); + ram_table->crgb_thresh[7] = cpu_to_be16(0x232b); + ram_table->crgb_offset[0] = cpu_to_be16(0x2999); + ram_table->crgb_offset[1] = cpu_to_be16(0x3999); + ram_table->crgb_offset[2] = cpu_to_be16(0x4666); + ram_table->crgb_offset[3] = cpu_to_be16(0x5999); + ram_table->crgb_offset[4] = cpu_to_be16(0x6333); + ram_table->crgb_offset[5] = cpu_to_be16(0x7800); + ram_table->crgb_offset[6] = cpu_to_be16(0x8c00); + ram_table->crgb_offset[7] = cpu_to_be16(0xa000); + ram_table->crgb_slope[0] = cpu_to_be16(0x3609); + ram_table->crgb_slope[1] = cpu_to_be16(0x2dfa); + ram_table->crgb_slope[2] = cpu_to_be16(0x27ea); + ram_table->crgb_slope[3] = cpu_to_be16(0x235d); + ram_table->crgb_slope[4] = cpu_to_be16(0x2042); + ram_table->crgb_slope[5] = cpu_to_be16(0x1dc3); + ram_table->crgb_slope[6] = cpu_to_be16(0x1b1a); + ram_table->crgb_slope[7] = cpu_to_be16(0x1910); + + fill_backlight_transform_table_v_2_2( + params, ram_table); +} + +void fill_iram_v_2_3(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parameters params) +{ + unsigned int i, j; + unsigned int set = params.set; + + ram_table->flags = 0x0; + for (i = 0; i < NUM_AGGR_LEVEL; i++) { + ram_table->hybrid_factor[i] = abm_settings[set][i].brightness_gain; + ram_table->contrast_factor[i] = abm_settings[set][i].contrast_factor; + ram_table->deviation_gain[i] = abm_settings[set][i].deviation_gain; + ram_table->min_knee[i] = abm_settings[set][i].min_knee; + ram_table->max_knee[i] = abm_settings[set][i].max_knee; + + for (j = 0; j < NUM_AMBI_LEVEL; j++) { + ram_table->min_reduction[j][i] = abm_settings[set][i].min_reduction; + ram_table->max_reduction[j][i] = abm_settings[set][i].max_reduction; + ram_table->bright_pos_gain[j][i] = abm_settings[set][i].bright_pos_gain; + ram_table->dark_pos_gain[j][i] = abm_settings[set][i].dark_pos_gain; + } + } ram_table->iir_curve[0] = 0x65; ram_table->iir_curve[1] = 0x65; @@ -561,7 +653,12 @@ bool dmcu_load_iram(struct dmcu *dmcu, memset(&ram_table, 0, sizeof(ram_table)); - if (dmcu->dmcu_version.abm_version == 0x22) { + if (dmcu->dmcu_version.abm_version == 0x23) { + fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params); + + result = dmcu->funcs->load_iram( + dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2); + } else if (dmcu->dmcu_version.abm_version == 0x22) { fill_iram_v_2_2((struct iram_table_v_2_2 *)ram_table, params); result = dmcu->funcs->load_iram( |