diff options
Diffstat (limited to 'drivers/gpu/drm/amd')
224 files changed, 8040 insertions, 3423 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 8afa0bceb460..28d76bd95051 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -100,7 +100,8 @@ amdgpu-y += \ amdgpu_psp.o \ psp_v3_1.o \ psp_v10_0.o \ - psp_v11_0.o + psp_v11_0.o \ + psp_v12_0.o # add SMC block amdgpu-y += \ @@ -154,6 +155,7 @@ amdgpu-y += \ # add ATHUB block amdgpu-y += \ + athub_v1_0.o \ athub_v2_0.o # add amdkfd interfaces diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index f85e7174babb..b2c801fb4ab5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -87,6 +87,7 @@ #include "amdgpu_discovery.h" #include "amdgpu_mes.h" #include "amdgpu_umc.h" +#include "amdgpu_mmhub.h" #define MAX_GPU_INSTANCE 16 @@ -788,7 +789,6 @@ struct amdgpu_device { int usec_timeout; const struct amdgpu_asic_funcs *asic_funcs; bool shutdown; - bool need_dma32; bool need_swiotlb; bool accel_working; struct notifier_block acpi_nb; @@ -976,6 +976,7 @@ struct amdgpu_device { const struct amdgpu_nbio_funcs *nbio_funcs; const struct amdgpu_df_funcs *df_funcs; + const struct amdgpu_mmhub_funcs *mmhub_funcs; /* delayed work_func for deferring clockgating during resume */ struct delayed_work delayed_init_work; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c index 7c03a7fcd011..d10f483f5e27 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c @@ -801,42 +801,6 @@ static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd, return reg & ATC_VMID0_PASID_MAPPING__PASID_MASK; } -static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid) -{ - struct amdgpu_device *adev = (struct amdgpu_device *) kgd; - uint32_t req = (1 << vmid) | - (0 << GCVM_INVALIDATE_ENG0_REQ__FLUSH_TYPE__SHIFT) |/* legacy */ - GCVM_INVALIDATE_ENG0_REQ__INVALIDATE_L2_PTES_MASK | - GCVM_INVALIDATE_ENG0_REQ__INVALIDATE_L2_PDE0_MASK | - GCVM_INVALIDATE_ENG0_REQ__INVALIDATE_L2_PDE1_MASK | - GCVM_INVALIDATE_ENG0_REQ__INVALIDATE_L2_PDE2_MASK | - GCVM_INVALIDATE_ENG0_REQ__INVALIDATE_L1_PTES_MASK; - - mutex_lock(&adev->srbm_mutex); - - /* Use light weight invalidation. - * - * TODO 1: agree on the right set of invalidation registers for - * KFD use. Use the last one for now. Invalidate only GCHUB as - * SDMA is now moved to GCHUB - * - * TODO 2: support range-based invalidation, requires kfg2kgd - * interface change - */ - WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_INVALIDATE_ENG0_ADDR_RANGE_LO32), - 0xffffffff); - WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_INVALIDATE_ENG0_ADDR_RANGE_HI32), - 0x0000001f); - - WREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_INVALIDATE_ENG0_REQ), req); - - while (!(RREG32(SOC15_REG_OFFSET(GC, 0, mmGCVM_INVALIDATE_ENG0_ACK)) & - (1 << vmid))) - cpu_relax(); - - mutex_unlock(&adev->srbm_mutex); -} - static int invalidate_tlbs_with_kiq(struct amdgpu_device *adev, uint16_t pasid) { signed long r; @@ -877,7 +841,8 @@ static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid) if (get_atc_vmid_pasid_mapping_valid(kgd, vmid)) { if (get_atc_vmid_pasid_mapping_pasid(kgd, vmid) == pasid) { - write_vmid_invalidate_request(kgd, vmid); + amdgpu_gmc_flush_gpu_tlb(adev, vmid, + AMDGPU_GFXHUB_0, 0); break; } } @@ -895,7 +860,7 @@ static int invalidate_tlbs_vmid(struct kgd_dev *kgd, uint16_t vmid) return 0; } - write_vmid_invalidate_request(kgd, vmid); + amdgpu_gmc_flush_gpu_tlb(adev, vmid, AMDGPU_GFXHUB_0, 0); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 9d153cf39581..e262f2ac07a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -670,7 +670,7 @@ static int invalidate_tlbs_with_kiq(struct amdgpu_device *adev, uint16_t pasid, int kgd_gfx_v9_invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid) { struct amdgpu_device *adev = (struct amdgpu_device *) kgd; - int vmid; + int vmid, i; struct amdgpu_ring *ring = &adev->gfx.kiq.ring; uint32_t flush_type = 0; @@ -689,8 +689,9 @@ int kgd_gfx_v9_invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid) if (kgd_gfx_v9_get_atc_vmid_pasid_mapping_valid(kgd, vmid)) { if (kgd_gfx_v9_get_atc_vmid_pasid_mapping_pasid(kgd, vmid) == pasid) { - amdgpu_gmc_flush_gpu_tlb(adev, vmid, - flush_type); + for (i = 0; i < adev->num_vmhubs; i++) + amdgpu_gmc_flush_gpu_tlb(adev, vmid, + i, flush_type); break; } } @@ -702,6 +703,7 @@ int kgd_gfx_v9_invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid) int kgd_gfx_v9_invalidate_tlbs_vmid(struct kgd_dev *kgd, uint16_t vmid) { struct amdgpu_device *adev = (struct amdgpu_device *) kgd; + int i; if (!amdgpu_amdkfd_is_kfd_vmid(adev, vmid)) { pr_err("non kfd vmid %d\n", vmid); @@ -723,7 +725,9 @@ int kgd_gfx_v9_invalidate_tlbs_vmid(struct kgd_dev *kgd, uint16_t vmid) * TODO 2: support range-based invalidation, requires kfg2kgd * interface change */ - amdgpu_gmc_flush_gpu_tlb(adev, vmid, 0); + for (i = 0; i < adev->num_vmhubs; i++) + amdgpu_gmc_flush_gpu_tlb(adev, vmid, i, 0); + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 8c50be56f458..2e53feed40e2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -1143,6 +1143,9 @@ static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p, num_deps = chunk->length_dw * 4 / sizeof(struct drm_amdgpu_cs_chunk_sem); + if (p->post_deps) + return -EINVAL; + p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps), GFP_KERNEL); p->num_post_deps = 0; @@ -1166,8 +1169,7 @@ static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p, static int amdgpu_cs_process_syncobj_timeline_out_dep(struct amdgpu_cs_parser *p, - struct amdgpu_cs_chunk - *chunk) + struct amdgpu_cs_chunk *chunk) { struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps; unsigned num_deps; @@ -1177,6 +1179,9 @@ static int amdgpu_cs_process_syncobj_timeline_out_dep(struct amdgpu_cs_parser *p num_deps = chunk->length_dw * 4 / sizeof(struct drm_amdgpu_cs_chunk_syncobj); + if (p->post_deps) + return -EINVAL; + p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps), GFP_KERNEL); p->num_post_deps = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index ec311de86fba..6614d8a6f4c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -42,7 +42,7 @@ const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM] = { [AMDGPU_HW_IP_VCN_JPEG] = 1, }; -static int amdgput_ctx_total_num_entities(void) +static int amdgpu_ctx_total_num_entities(void) { unsigned i, num_entities = 0; @@ -73,7 +73,7 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev, struct drm_file *filp, struct amdgpu_ctx *ctx) { - unsigned num_entities = amdgput_ctx_total_num_entities(); + unsigned num_entities = amdgpu_ctx_total_num_entities(); unsigned i, j, k; int r; @@ -207,7 +207,7 @@ error_free_fences: static void amdgpu_ctx_fini(struct kref *ref) { struct amdgpu_ctx *ctx = container_of(ref, struct amdgpu_ctx, refcount); - unsigned num_entities = amdgput_ctx_total_num_entities(); + unsigned num_entities = amdgpu_ctx_total_num_entities(); struct amdgpu_device *adev = ctx->adev; unsigned i, j; @@ -289,10 +289,7 @@ static void amdgpu_ctx_do_release(struct kref *ref) ctx = container_of(ref, struct amdgpu_ctx, refcount); - num_entities = 0; - for (i = 0; i < AMDGPU_HW_IP_NUM; i++) - num_entities += amdgpu_ctx_num_entities[i]; - + num_entities = amdgpu_ctx_total_num_entities(); for (i = 0; i < num_entities; i++) drm_sched_entity_destroy(&ctx->entities[0][i].entity); @@ -354,7 +351,7 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev, { struct amdgpu_ctx *ctx; struct amdgpu_ctx_mgr *mgr; - uint32_t ras_counter; + unsigned long ras_counter; if (!fpriv) return -EINVAL; @@ -524,7 +521,7 @@ struct dma_fence *amdgpu_ctx_get_fence(struct amdgpu_ctx *ctx, void amdgpu_ctx_priority_override(struct amdgpu_ctx *ctx, enum drm_sched_priority priority) { - unsigned num_entities = amdgput_ctx_total_num_entities(); + unsigned num_entities = amdgpu_ctx_total_num_entities(); enum drm_sched_priority ctx_prio; unsigned i; @@ -544,21 +541,24 @@ int amdgpu_ctx_wait_prev_fence(struct amdgpu_ctx *ctx, struct drm_sched_entity *entity) { struct amdgpu_ctx_entity *centity = to_amdgpu_ctx_entity(entity); - unsigned idx = centity->sequence & (amdgpu_sched_jobs - 1); - struct dma_fence *other = centity->fences[idx]; + struct dma_fence *other; + unsigned idx; + long r; - if (other) { - signed long r; - r = dma_fence_wait(other, true); - if (r < 0) { - if (r != -ERESTARTSYS) - DRM_ERROR("Error (%ld) waiting for fence!\n", r); + spin_lock(&ctx->ring_lock); + idx = centity->sequence & (amdgpu_sched_jobs - 1); + other = dma_fence_get(centity->fences[idx]); + spin_unlock(&ctx->ring_lock); - return r; - } - } + if (!other) + return 0; - return 0; + r = dma_fence_wait(other, true); + if (r < 0 && r != -ERESTARTSYS) + DRM_ERROR("Error (%ld) waiting for fence!\n", r); + + dma_fence_put(other); + return r; } void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr) @@ -569,7 +569,7 @@ void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr) long amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr, long timeout) { - unsigned num_entities = amdgput_ctx_total_num_entities(); + unsigned num_entities = amdgpu_ctx_total_num_entities(); struct amdgpu_ctx *ctx; struct idr *idp; uint32_t id, i; @@ -591,7 +591,7 @@ long amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr, long timeout) void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr) { - unsigned num_entities = amdgput_ctx_total_num_entities(); + unsigned num_entities = amdgpu_ctx_total_num_entities(); struct amdgpu_ctx *ctx; struct idr *idp; uint32_t id, i; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h index 5f1b54c9bcdb..da808633732b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h @@ -49,8 +49,8 @@ struct amdgpu_ctx { enum drm_sched_priority override_priority; struct mutex lock; atomic_t guilty; - uint32_t ras_counter_ce; - uint32_t ras_counter_ue; + unsigned long ras_counter_ce; + unsigned long ras_counter_ue; }; struct amdgpu_ctx_mgr { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 682833f90fdd..2f884699eaef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -71,6 +71,7 @@ MODULE_FIRMWARE("amdgpu/raven_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/picasso_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/raven2_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/arcturus_gpu_info.bin"); +MODULE_FIRMWARE("amdgpu/renoir_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/navi10_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/navi14_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/navi12_gpu_info.bin"); @@ -102,6 +103,7 @@ static const char *amdgpu_asic_name[] = { "VEGA20", "RAVEN", "ARCTURUS", + "RENOIR", "NAVI10", "NAVI14", "NAVI12", @@ -1427,6 +1429,9 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) case CHIP_ARCTURUS: chip_name = "arcturus"; break; + case CHIP_RENOIR: + chip_name = "renoir"; + break; case CHIP_NAVI10: chip_name = "navi10"; break; @@ -1579,7 +1584,9 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) case CHIP_VEGA20: case CHIP_RAVEN: case CHIP_ARCTURUS: - if (adev->asic_type == CHIP_RAVEN) + case CHIP_RENOIR: + if (adev->asic_type == CHIP_RAVEN || + adev->asic_type == CHIP_RENOIR) adev->family = AMDGPU_FAMILY_RV; else adev->family = AMDGPU_FAMILY_AI; @@ -3518,6 +3525,7 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev) case CHIP_VEGA20: case CHIP_VEGA10: case CHIP_VEGA12: + case CHIP_RAVEN: break; default: goto disabled; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index e9046922fe94..48a2070e72f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -79,9 +79,10 @@ * - 3.31.0 - Add support for per-flip tiling attribute changes with DC * - 3.32.0 - Add syncobj timeline support to AMDGPU_CS. * - 3.33.0 - Fixes for GDS ENOMEM failures in AMDGPU_CS. + * - 3.34.0 - Non-DC can flip correctly between buffers with different pitches */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 33 +#define KMS_DRIVER_MINOR 34 #define KMS_DRIVER_PATCHLEVEL 0 #define AMDGPU_MAX_TIMEOUT_PARAM_LENTH 256 @@ -142,7 +143,7 @@ int amdgpu_async_gfx_ring = 1; int amdgpu_mcbp = 0; int amdgpu_discovery = -1; int amdgpu_mes = 0; -int amdgpu_noretry; +int amdgpu_noretry = 1; struct amdgpu_mgpu_info mgpu_info = { .mutex = __MUTEX_INITIALIZER(mgpu_info.mutex), @@ -610,7 +611,7 @@ MODULE_PARM_DESC(mes, module_param_named(mes, amdgpu_mes, int, 0444); MODULE_PARM_DESC(noretry, - "Disable retry faults (0 = retry enabled (default), 1 = retry disabled)"); + "Disable retry faults (0 = retry enabled, 1 = retry disabled (default))"); module_param_named(noretry, amdgpu_noretry, int, 0644); #ifdef CONFIG_HSA_AMD @@ -1000,6 +1001,7 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x738C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ARCTURUS|AMD_EXP_HW_SUPPORT}, {0x1002, 0x7388, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ARCTURUS|AMD_EXP_HW_SUPPORT}, {0x1002, 0x738E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ARCTURUS|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x7390, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ARCTURUS|AMD_EXP_HW_SUPPORT}, /* Navi10 */ {0x1002, 0x7310, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, {0x1002, 0x7312, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, @@ -1008,6 +1010,11 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x731A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, {0x1002, 0x731B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, {0x1002, 0x731F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, + /* Navi14 */ + {0x1002, 0x7340, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI14}, + + /* Renoir */ + {0x1002, 0x1636, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RENOIR|AMD_IS_APU|AMD_EXP_HW_SUPPORT}, {0, 0, 0} }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index d79ab1da9e07..5e8bdded265f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -251,7 +251,9 @@ int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, } mb(); amdgpu_asic_flush_hdp(adev, NULL); - amdgpu_gmc_flush_gpu_tlb(adev, 0, 0); + for (i = 0; i < adev->num_vmhubs; i++) + amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); + return 0; } @@ -310,9 +312,9 @@ int amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset, uint64_t flags) { #ifdef CONFIG_DRM_AMDGPU_GART_DEBUGFS - unsigned i,t,p; + unsigned t,p; #endif - int r; + int r, i; if (!adev->gart.ready) { WARN(1, "trying to bind memory to uninitialized GART !\n"); @@ -336,7 +338,8 @@ int amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset, mb(); amdgpu_asic_flush_hdp(adev, NULL); - amdgpu_gmc_flush_gpu_tlb(adev, 0, 0); + for (i = 0; i < adev->num_vmhubs; i++) + amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h index df8a23554831..f6ac1e9548f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h @@ -32,7 +32,6 @@ struct amdgpu_gds { uint32_t gws_size; uint32_t oa_size; uint32_t gds_compute_max_wave_id; - uint32_t vgt_gs_max_wave_id; }; struct amdgpu_gds_reg_offset { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 924d83e711ef..5790db61fa2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -220,6 +220,14 @@ void amdgpu_gmc_agp_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc) const uint64_t sixteen_gb_mask = ~(sixteen_gb - 1); u64 size_af, size_bf; + if (amdgpu_sriov_vf(adev)) { + mc->agp_start = 0xffffffff; + mc->agp_end = 0x0; + mc->agp_size = 0; + + return; + } + if (mc->fb_start > mc->gart_start) { size_bf = (mc->fb_start & sixteen_gb_mask) - ALIGN(mc->gart_end + 1, sixteen_gb); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 071145ac67b5..b6e1d98ef01e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -89,8 +89,8 @@ struct amdgpu_vmhub { */ struct amdgpu_gmc_funcs { /* flush the vm tlb via mmio */ - void (*flush_gpu_tlb)(struct amdgpu_device *adev, - uint32_t vmid, uint32_t flush_type); + void (*flush_gpu_tlb)(struct amdgpu_device *adev, uint32_t vmid, + uint32_t vmhub, uint32_t flush_type); /* flush the vm tlb via ring */ uint64_t (*emit_flush_gpu_tlb)(struct amdgpu_ring *ring, unsigned vmid, uint64_t pd_addr); @@ -177,10 +177,11 @@ struct amdgpu_gmc { struct amdgpu_xgmi xgmi; struct amdgpu_irq_src ecc_irq; - struct ras_common_if *ras_if; + struct ras_common_if *umc_ras_if; + struct ras_common_if *mmhub_ras_if; }; -#define amdgpu_gmc_flush_gpu_tlb(adev, vmid, type) (adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (type)) +#define amdgpu_gmc_flush_gpu_tlb(adev, vmid, vmhub, type) ((adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (vmhub), (type))) #define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr)) #define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid)) #define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h new file mode 100644 index 000000000000..2d75ecfa199b --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h @@ -0,0 +1,31 @@ +/* + * Copyright (C) 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) 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. + */ +#ifndef __AMDGPU_MMHUB_H__ +#define __AMDGPU_MMHUB_H__ + +struct amdgpu_mmhub_funcs { + void (*ras_init)(struct amdgpu_device *adev); + void (*query_ras_error_count)(struct amdgpu_device *adev, + void *ras_error_status); +}; + +#endif + diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 6ebe61e14f29..1fead0e8b890 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -246,8 +246,9 @@ int amdgpu_bo_create_reserved(struct amdgpu_device *adev, bp.size = size; bp.byte_align = align; bp.domain = domain; - bp.flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED | - AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS; + bp.flags = cpu_addr ? AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED + : AMDGPU_GEM_CREATE_NO_CPU_ACCESS; + bp.flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS; bp.type = ttm_bo_type_kernel; bp.resv = NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index 39998f203b49..2fccaf412e03 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -2828,10 +2828,12 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) DRM_ERROR("failed to create device file pp_dpm_socclk\n"); return ret; } - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_dcefclk); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_dcefclk\n"); - return ret; + if (adev->asic_type != CHIP_ARCTURUS) { + ret = device_create_file(adev->dev, &dev_attr_pp_dpm_dcefclk); + if (ret) { + DRM_ERROR("failed to create device file pp_dpm_dcefclk\n"); + return ret; + } } } if (adev->asic_type >= CHIP_VEGA20) { @@ -2841,10 +2843,12 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) return ret; } } - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_pcie); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_pcie\n"); - return ret; + if (adev->asic_type != CHIP_ARCTURUS) { + ret = device_create_file(adev->dev, &dev_attr_pp_dpm_pcie); + if (ret) { + DRM_ERROR("failed to create device file pp_dpm_pcie\n"); + return ret; + } } ret = device_create_file(adev->dev, &dev_attr_pp_sclk_od); if (ret) { @@ -2948,9 +2952,11 @@ void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev) device_remove_file(adev->dev, &dev_attr_pp_dpm_mclk); if (adev->asic_type >= CHIP_VEGA10) { device_remove_file(adev->dev, &dev_attr_pp_dpm_socclk); - device_remove_file(adev->dev, &dev_attr_pp_dpm_dcefclk); + if (adev->asic_type != CHIP_ARCTURUS) + device_remove_file(adev->dev, &dev_attr_pp_dpm_dcefclk); } - device_remove_file(adev->dev, &dev_attr_pp_dpm_pcie); + if (adev->asic_type != CHIP_ARCTURUS) + device_remove_file(adev->dev, &dev_attr_pp_dpm_pcie); if (adev->asic_type >= CHIP_VEGA20) device_remove_file(adev->dev, &dev_attr_pp_dpm_fclk); device_remove_file(adev->dev, &dev_attr_pp_sclk_od); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 51fb890e2d3b..f06f5ef0ca4b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -32,6 +32,7 @@ #include "psp_v3_1.h" #include "psp_v10_0.h" #include "psp_v11_0.h" +#include "psp_v12_0.h" static void psp_set_funcs(struct amdgpu_device *adev); @@ -63,6 +64,9 @@ static int psp_early_init(void *handle) psp_v11_0_set_psp_funcs(psp); psp->autoload_supported = true; break; + case CHIP_RENOIR: + psp_v12_0_set_psp_funcs(psp); + break; default: return -EINVAL; } @@ -140,8 +144,7 @@ psp_cmd_submit_buf(struct psp_context *psp, memcpy(psp->cmd_buf_mem, cmd, sizeof(struct psp_gfx_cmd_resp)); index = atomic_inc_return(&psp->fence_value); - ret = psp_cmd_submit(psp, ucode, psp->cmd_buf_mc_addr, - fence_mc_addr, index); + ret = psp_cmd_submit(psp, psp->cmd_buf_mc_addr, fence_mc_addr, index); if (ret) { atomic_dec(&psp->fence_value); mutex_unlock(&psp->mutex); @@ -260,7 +263,7 @@ static int psp_tmr_init(struct psp_context *psp) ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_SIZE, AMDGPU_GEM_DOMAIN_VRAM, - &psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf); + &psp->tmr_bo, &psp->tmr_mc_addr, NULL); return ret; } @@ -940,6 +943,60 @@ static int psp_get_fw_type(struct amdgpu_firmware_info *ucode, return 0; } +static void psp_print_fw_hdr(struct psp_context *psp, + struct amdgpu_firmware_info *ucode) +{ + struct amdgpu_device *adev = psp->adev; + const struct sdma_firmware_header_v1_0 *sdma_hdr = + (const struct sdma_firmware_header_v1_0 *) + adev->sdma.instance[ucode->ucode_id - AMDGPU_UCODE_ID_SDMA0].fw->data; + const struct gfx_firmware_header_v1_0 *ce_hdr = + (const struct gfx_firmware_header_v1_0 *)adev->gfx.ce_fw->data; + const struct gfx_firmware_header_v1_0 *pfp_hdr = + (const struct gfx_firmware_header_v1_0 *)adev->gfx.pfp_fw->data; + const struct gfx_firmware_header_v1_0 *me_hdr = + (const struct gfx_firmware_header_v1_0 *)adev->gfx.me_fw->data; + const struct gfx_firmware_header_v1_0 *mec_hdr = + (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data; + const struct rlc_firmware_header_v2_0 *rlc_hdr = + (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; + const struct smc_firmware_header_v1_0 *smc_hdr = + (const struct smc_firmware_header_v1_0 *)adev->pm.fw->data; + + switch (ucode->ucode_id) { + case AMDGPU_UCODE_ID_SDMA0: + case AMDGPU_UCODE_ID_SDMA1: + case AMDGPU_UCODE_ID_SDMA2: + case AMDGPU_UCODE_ID_SDMA3: + case AMDGPU_UCODE_ID_SDMA4: + case AMDGPU_UCODE_ID_SDMA5: + case AMDGPU_UCODE_ID_SDMA6: + case AMDGPU_UCODE_ID_SDMA7: + amdgpu_ucode_print_sdma_hdr(&sdma_hdr->header); + break; + case AMDGPU_UCODE_ID_CP_CE: + amdgpu_ucode_print_gfx_hdr(&ce_hdr->header); + break; + case AMDGPU_UCODE_ID_CP_PFP: + amdgpu_ucode_print_gfx_hdr(&pfp_hdr->header); + break; + case AMDGPU_UCODE_ID_CP_ME: + amdgpu_ucode_print_gfx_hdr(&me_hdr->header); + break; + case AMDGPU_UCODE_ID_CP_MEC1: + amdgpu_ucode_print_gfx_hdr(&mec_hdr->header); + break; + case AMDGPU_UCODE_ID_RLC_G: + amdgpu_ucode_print_rlc_hdr(&rlc_hdr->header); + break; + case AMDGPU_UCODE_ID_SMC: + amdgpu_ucode_print_smc_hdr(&smc_hdr->header); + break; + default: + break; + } +} + static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode, struct psp_gfx_cmd_resp *cmd) { @@ -1019,14 +1076,19 @@ out: ucode->ucode_id == AMDGPU_UCODE_ID_CP_MEC2_JT)) /* skip mec JT when autoload is enabled */ continue; + /* Renoir only needs to load mec jump table one time */ + if (adev->asic_type == CHIP_RENOIR && + ucode->ucode_id == AMDGPU_UCODE_ID_CP_MEC2_JT) + continue; + + psp_print_fw_hdr(psp, ucode); ret = psp_execute_np_fw_load(psp, ucode); if (ret) return ret; /* Start rlc autoload after psp recieved all the gfx firmware */ - if (ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM || - (adev->asic_type == CHIP_NAVI12 && ucode->ucode_id == AMDGPU_UCODE_ID_RLC_G)) { + if (ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM) { ret = psp_rlc_autoload(psp); if (ret) { DRM_ERROR("Failed to start rlc autoload\n"); @@ -1154,7 +1216,7 @@ static int psp_hw_fini(void *handle) psp_ring_destroy(psp, PSP_RING_TYPE__KM); - amdgpu_bo_free_kernel(&psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf); + amdgpu_bo_free_kernel(&psp->tmr_bo, &psp->tmr_mc_addr, NULL); amdgpu_bo_free_kernel(&psp->fw_pri_bo, &psp->fw_pri_mc_addr, &psp->fw_pri_buf); amdgpu_bo_free_kernel(&psp->fence_buf_bo, @@ -1358,3 +1420,12 @@ const struct amdgpu_ip_block_version psp_v11_0_ip_block = .rev = 0, .funcs = &psp_ip_funcs, }; + +const struct amdgpu_ip_block_version psp_v12_0_ip_block = +{ + .type = AMD_IP_BLOCK_TYPE_PSP, + .major = 12, + .minor = 0, + .rev = 0, + .funcs = &psp_ip_funcs, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index e0fc2a790e53..bc0947f6bc8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -90,7 +90,6 @@ struct psp_funcs int (*ring_destroy)(struct psp_context *psp, enum psp_ring_type ring_type); int (*cmd_submit)(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, uint64_t cmd_buf_mc_addr, uint64_t fence_mc_addr, int index); bool (*compare_sram_data)(struct psp_context *psp, @@ -172,7 +171,6 @@ struct psp_context /* tmr buffer */ struct amdgpu_bo *tmr_bo; uint64_t tmr_mc_addr; - void *tmr_buf; /* asd firmware and buffer */ const struct firmware *asd_fw; @@ -223,8 +221,8 @@ struct amdgpu_psp_funcs { #define psp_ring_create(psp, type) (psp)->funcs->ring_create((psp), (type)) #define psp_ring_stop(psp, type) (psp)->funcs->ring_stop((psp), (type)) #define psp_ring_destroy(psp, type) ((psp)->funcs->ring_destroy((psp), (type))) -#define psp_cmd_submit(psp, ucode, cmd_mc, fence_mc, index) \ - (psp)->funcs->cmd_submit((psp), (ucode), (cmd_mc), (fence_mc), (index)) +#define psp_cmd_submit(psp, cmd_mc, fence_mc, index) \ + (psp)->funcs->cmd_submit((psp), (cmd_mc), (fence_mc), (index)) #define psp_compare_sram_data(psp, ucode, type) \ (psp)->funcs->compare_sram_data((psp), (ucode), (type)) #define psp_init_microcode(psp) \ @@ -270,6 +268,7 @@ extern int psp_wait_for(struct psp_context *psp, uint32_t reg_index, uint32_t field_val, uint32_t mask, bool check_changed); extern const struct amdgpu_ip_block_version psp_v10_0_ip_block; +extern const struct amdgpu_ip_block_version psp_v12_0_ip_block; int psp_gpu_reset(struct amdgpu_device *adev); int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 523f43732dee..016ea274b955 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -131,6 +131,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, char err[9] = "ue"; int op = -1; int block_id; + uint32_t sub_block; u64 address, value; if (*pos) @@ -169,11 +170,12 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, data->op = op; if (op == 2) { - if (sscanf(str, "%*s %*s %*s %llu %llu", - &address, &value) != 2) - if (sscanf(str, "%*s %*s %*s 0x%llx 0x%llx", - &address, &value) != 2) + if (sscanf(str, "%*s %*s %*s %u %llu %llu", + &sub_block, &address, &value) != 3) + if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx", + &sub_block, &address, &value) != 3) return -EINVAL; + data->head.sub_block_index = sub_block; data->inject.address = address; data->inject.value = value; } @@ -218,7 +220,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, * write the struct to the control node. * * bash: - * echo op block [error [address value]] > .../ras/ras_ctrl + * echo op block [error [sub_blcok address value]] > .../ras/ras_ctrl * op: disable, enable, inject * disable: only block is needed * enable: block and error are needed @@ -228,10 +230,11 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, * error: ue, ce * ue: multi_uncorrectable * ce: single_correctable + * sub_block: sub block index, pass 0 if there is no sub block * * here are some examples for bash commands, - * echo inject umc ue 0x0 0x0 > /sys/kernel/debug/dri/0/ras/ras_ctrl - * echo inject umc ce 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl + * echo inject umc ue 0x0 0x0 0x0 > /sys/kernel/debug/dri/0/ras/ras_ctrl + * echo inject umc ce 0 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl * echo disable umc > /sys/kernel/debug/dri/0/ras/ras_ctrl * * How to check the result? @@ -611,6 +614,10 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, if (adev->gfx.funcs->query_ras_error_count) adev->gfx.funcs->query_ras_error_count(adev, &err_data); break; + case AMDGPU_RAS_BLOCK__MMHUB: + if (adev->mmhub_funcs->query_ras_error_count) + adev->mmhub_funcs->query_ras_error_count(adev, &err_data); + break; default: break; } @@ -656,6 +663,7 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, ret = -EINVAL; break; case AMDGPU_RAS_BLOCK__UMC: + case AMDGPU_RAS_BLOCK__MMHUB: ret = psp_ras_trigger_error(&adev->psp, &block_info); break; default: @@ -680,7 +688,7 @@ int amdgpu_ras_error_cure(struct amdgpu_device *adev, } /* get the total error counts on all IPs */ -int amdgpu_ras_query_error_count(struct amdgpu_device *adev, +unsigned long amdgpu_ras_query_error_count(struct amdgpu_device *adev, bool is_ce) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); @@ -688,7 +696,7 @@ int amdgpu_ras_query_error_count(struct amdgpu_device *adev, struct ras_err_data data = {0, 0}; if (!con) - return -EINVAL; + return 0; list_for_each_entry(obj, &con->head, node) { struct ras_query_if info = { @@ -696,7 +704,7 @@ int amdgpu_ras_query_error_count(struct amdgpu_device *adev, }; if (amdgpu_ras_error_query(adev, &info)) - return -EINVAL; + return 0; data.ce_count += info.ce_count; data.ue_count += info.ue_count; @@ -785,25 +793,8 @@ static ssize_t amdgpu_ras_sysfs_features_read(struct device *dev, { struct amdgpu_ras *con = container_of(attr, struct amdgpu_ras, features_attr); - struct drm_device *ddev = dev_get_drvdata(dev); - struct amdgpu_device *adev = ddev->dev_private; - struct ras_common_if head; - int ras_block_count = AMDGPU_RAS_BLOCK_COUNT; - int i, enabled; - ssize_t s; - s = scnprintf(buf, PAGE_SIZE, "feature mask: 0x%x\n", con->features); - - for (i = 0; i < ras_block_count; i++) { - head.block = i; - enabled = amdgpu_ras_is_feature_enabled(adev, &head); - - s += scnprintf(&buf[s], PAGE_SIZE - s, - "%s ras feature mask: %s\n", - ras_block_str(i), enabled?"on":"off"); - } - - return s; + return scnprintf(buf, PAGE_SIZE, "feature mask: 0x%x\n", con->features); } static int amdgpu_ras_sysfs_create_feature_node(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 2765f2dbb1e6..02a51e3dfa14 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -484,7 +484,7 @@ int amdgpu_ras_request_reset_on_boot(struct amdgpu_device *adev, void amdgpu_ras_resume(struct amdgpu_device *adev); void amdgpu_ras_suspend(struct amdgpu_device *adev); -int amdgpu_ras_query_error_count(struct amdgpu_device *adev, +unsigned long amdgpu_ras_query_error_count(struct amdgpu_device *adev, bool is_ce); /* error handling functions */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 3e8f9072561e..b44b07100564 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -1745,7 +1745,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) r = ttm_bo_device_init(&adev->mman.bdev, &amdgpu_bo_driver, adev->ddev->anon_inode->i_mapping, - adev->need_dma32); + dma_addressing_limited(adev->dev)); if (r) { DRM_ERROR("failed initializing buffer object driver(%d).\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index dd18ebc2eb01..3a6115ad0196 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -83,8 +83,8 @@ void amdgpu_ucode_print_smc_hdr(const struct common_firmware_header *hdr) const struct smc_firmware_header_v2_0 *v2_hdr = container_of(v1_hdr, struct smc_firmware_header_v2_0, v1_0); - DRM_INFO("ppt_offset_bytes: %u\n", le32_to_cpu(v2_hdr->ppt_offset_bytes)); - DRM_INFO("ppt_size_bytes: %u\n", le32_to_cpu(v2_hdr->ppt_size_bytes)); + DRM_DEBUG("ppt_offset_bytes: %u\n", le32_to_cpu(v2_hdr->ppt_offset_bytes)); + DRM_DEBUG("ppt_size_bytes: %u\n", le32_to_cpu(v2_hdr->ppt_size_bytes)); } else { DRM_ERROR("Unknown SMC ucode version: %u.%u\n", version_major, version_minor); } @@ -360,6 +360,7 @@ amdgpu_ucode_get_load_type(struct amdgpu_device *adev, int load_type) case CHIP_RAVEN: case CHIP_VEGA12: case CHIP_VEGA20: + case CHIP_RENOIR: case CHIP_NAVI10: case CHIP_NAVI14: case CHIP_NAVI12: @@ -369,6 +370,7 @@ amdgpu_ucode_get_load_type(struct amdgpu_device *adev, int load_type) return AMDGPU_FW_LOAD_PSP; case CHIP_ARCTURUS: return AMDGPU_FW_LOAD_DIRECT; + default: DRM_ERROR("Unknown firmware load type\n"); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index 47086cdbb413..7a6beb2e7c4e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -47,6 +47,7 @@ #define FIRMWARE_PICASSO "amdgpu/picasso_vcn.bin" #define FIRMWARE_RAVEN2 "amdgpu/raven2_vcn.bin" #define FIRMWARE_ARCTURUS "amdgpu/arcturus_vcn.bin" +#define FIRMWARE_RENOIR "amdgpu/renoir_vcn.bin" #define FIRMWARE_NAVI10 "amdgpu/navi10_vcn.bin" #define FIRMWARE_NAVI14 "amdgpu/navi14_vcn.bin" #define FIRMWARE_NAVI12 "amdgpu/navi12_vcn.bin" @@ -55,6 +56,7 @@ MODULE_FIRMWARE(FIRMWARE_RAVEN); MODULE_FIRMWARE(FIRMWARE_PICASSO); MODULE_FIRMWARE(FIRMWARE_RAVEN2); MODULE_FIRMWARE(FIRMWARE_ARCTURUS); +MODULE_FIRMWARE(FIRMWARE_RENOIR); MODULE_FIRMWARE(FIRMWARE_NAVI10); MODULE_FIRMWARE(FIRMWARE_NAVI14); MODULE_FIRMWARE(FIRMWARE_NAVI12); @@ -83,6 +85,12 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev) case CHIP_ARCTURUS: fw_name = FIRMWARE_ARCTURUS; break; + case CHIP_RENOIR: + fw_name = FIRMWARE_RENOIR; + if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) && + (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)) + adev->vcn.indirect_sram = true; + break; case CHIP_NAVI10: fw_name = FIRMWARE_NAVI10; if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) && diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index b7665b31a2ca..e2fb141ff2e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -2863,6 +2863,13 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, uns WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)), "CPU update of VM recommended only for large BAR system\n"); + if (vm->use_cpu_for_update) + vm->update_funcs = &amdgpu_vm_cpu_funcs; + else + vm->update_funcs = &amdgpu_vm_sdma_funcs; + dma_fence_put(vm->last_update); + vm->last_update = NULL; + if (vm->pasid) { unsigned long flags; diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c new file mode 100644 index 000000000000..d9cc746af5e6 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c @@ -0,0 +1,103 @@ +/* + * Copyright 2016 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. + * + */ +#include "amdgpu.h" +#include "athub_v1_0.h" + +#include "athub/athub_1_0_offset.h" +#include "athub/athub_1_0_sh_mask.h" +#include "vega10_enum.h" + +#include "soc15_common.h" + +static void athub_update_medium_grain_clock_gating(struct amdgpu_device *adev, + bool enable) +{ + uint32_t def, data; + + def = data = RREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL); + + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG)) + data |= ATHUB_MISC_CNTL__CG_ENABLE_MASK; + else + data &= ~ATHUB_MISC_CNTL__CG_ENABLE_MASK; + + if (def != data) + WREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL, data); +} + +static void athub_update_medium_grain_light_sleep(struct amdgpu_device *adev, + bool enable) +{ + uint32_t def, data; + + def = data = RREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL); + + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_LS) && + (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS)) + data |= ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK; + else + data &= ~ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK; + + if(def != data) + WREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL, data); +} + +int athub_v1_0_set_clockgating(struct amdgpu_device *adev, + enum amd_clockgating_state state) +{ + if (amdgpu_sriov_vf(adev)) + return 0; + + switch (adev->asic_type) { + case CHIP_VEGA10: + case CHIP_VEGA12: + case CHIP_VEGA20: + case CHIP_RAVEN: + athub_update_medium_grain_clock_gating(adev, + state == AMD_CG_STATE_GATE ? true : false); + athub_update_medium_grain_light_sleep(adev, + state == AMD_CG_STATE_GATE ? true : false); + break; + default: + break; + } + + return 0; +} + +void athub_v1_0_get_clockgating(struct amdgpu_device *adev, u32 *flags) +{ + int data; + + if (amdgpu_sriov_vf(adev)) + *flags = 0; + + /* AMD_CG_SUPPORT_ATHUB_MGCG */ + data = RREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL); + if (data & ATHUB_MISC_CNTL__CG_ENABLE_MASK) + *flags |= AMD_CG_SUPPORT_ATHUB_MGCG; + + /* AMD_CG_SUPPORT_ATHUB_LS */ + if (data & ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK) + *flags |= AMD_CG_SUPPORT_ATHUB_LS; +} diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.h new file mode 100644 index 000000000000..b279af59e34f --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.h @@ -0,0 +1,30 @@ +/* + * Copyright 2016 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. + * + */ +#ifndef __ATHUB_V1_0_H__ +#define __ATHUB_V1_0_H__ + +int athub_v1_0_set_clockgating(struct amdgpu_device *adev, + enum amd_clockgating_state state); +void athub_v1_0_get_clockgating(struct amdgpu_device *adev, u32 *flags); + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c index 7e6c0bc3e8dd..ceb9aa4df0e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c @@ -75,6 +75,7 @@ int athub_v2_0_set_clockgating(struct amdgpu_device *adev, switch (adev->asic_type) { case CHIP_NAVI10: case CHIP_NAVI14: + case CHIP_NAVI12: athub_v2_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); athub_v2_0_update_medium_grain_light_sleep(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 1ffd1963e765..645550e7caf5 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -236,6 +236,7 @@ static void dce_v10_0_page_flip(struct amdgpu_device *adev, int crtc_id, u64 crtc_base, bool async) { struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id]; + struct drm_framebuffer *fb = amdgpu_crtc->base.primary->fb; u32 tmp; /* flip at hsync for async, default is vsync */ @@ -243,6 +244,9 @@ static void dce_v10_0_page_flip(struct amdgpu_device *adev, tmp = REG_SET_FIELD(tmp, GRPH_FLIP_CONTROL, GRPH_SURFACE_UPDATE_H_RETRACE_EN, async ? 1 : 0); WREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset, tmp); + /* update pitch */ + WREG32(mmGRPH_PITCH + amdgpu_crtc->crtc_offset, + fb->pitches[0] / fb->format->cpp[0]); /* update the primary scanout address */ WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset, upper_32_bits(crtc_base)); diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 9e0782b54066..d9f470632b2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -254,6 +254,7 @@ static void dce_v11_0_page_flip(struct amdgpu_device *adev, int crtc_id, u64 crtc_base, bool async) { struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id]; + struct drm_framebuffer *fb = amdgpu_crtc->base.primary->fb; u32 tmp; /* flip immediate for async, default is vsync */ @@ -261,6 +262,9 @@ static void dce_v11_0_page_flip(struct amdgpu_device *adev, tmp = REG_SET_FIELD(tmp, GRPH_FLIP_CONTROL, GRPH_SURFACE_UPDATE_IMMEDIATE_EN, async ? 1 : 0); WREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset, tmp); + /* update pitch */ + WREG32(mmGRPH_PITCH + amdgpu_crtc->crtc_offset, + fb->pitches[0] / fb->format->cpp[0]); /* update the scanout addresses */ WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset, upper_32_bits(crtc_base)); diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index 4bf453e07dca..3eb2e7429269 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -191,10 +191,14 @@ static void dce_v6_0_page_flip(struct amdgpu_device *adev, int crtc_id, u64 crtc_base, bool async) { struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id]; + struct drm_framebuffer *fb = amdgpu_crtc->base.primary->fb; /* flip at hsync for async, default is vsync */ WREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset, async ? GRPH_FLIP_CONTROL__GRPH_SURFACE_UPDATE_H_RETRACE_EN_MASK : 0); + /* update pitch */ + WREG32(mmGRPH_PITCH + amdgpu_crtc->crtc_offset, + fb->pitches[0] / fb->format->cpp[0]); /* update the scanout addresses */ WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset, upper_32_bits(crtc_base)); diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index b23418ca8f6a..a16c5e9e610e 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -184,10 +184,14 @@ static void dce_v8_0_page_flip(struct amdgpu_device *adev, int crtc_id, u64 crtc_base, bool async) { struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id]; + struct drm_framebuffer *fb = amdgpu_crtc->base.primary->fb; /* flip at hsync for async, default is vsync */ WREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset, async ? GRPH_FLIP_CONTROL__GRPH_SURFACE_UPDATE_H_RETRACE_EN_MASK : 0); + /* update pitch */ + WREG32(mmGRPH_PITCH + amdgpu_crtc->crtc_offset, + fb->pitches[0] / fb->format->cpp[0]); /* update the primary scanout addresses */ WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset, upper_32_bits(crtc_base)); diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c index 4c6d792d51a5..f1fadb5ba191 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c @@ -458,6 +458,7 @@ static int dce_virtual_hw_init(void *handle) case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_ARCTURUS: + case CHIP_RENOIR: case CHIP_NAVI10: case CHIP_NAVI14: case CHIP_NAVI12: diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 43427a3148b7..79d3fbd3ba63 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -1748,9 +1748,12 @@ static void gfx_v10_0_init_csb(struct amdgpu_device *adev) static void gfx_v10_0_init_pg(struct amdgpu_device *adev) { + int i; + gfx_v10_0_init_csb(adev); - amdgpu_gmc_flush_gpu_tlb(adev, 0, 0); + for (i = 0; i < adev->num_vmhubs; i++) + amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); /* TODO: init power gating */ return; @@ -4373,15 +4376,6 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, unsigned vmid = AMDGPU_JOB_GET_VMID(job); u32 header, control = 0; - /* Prevent a hw deadlock due to a wave ID mismatch between ME and GDS. - * This resets the wave ID counters. (needed by transform feedback) - * TODO: This might only be needed on a VMID switch when we change - * the GDS OA mapping, not sure. - */ - amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); - amdgpu_ring_write(ring, mmVGT_GS_MAX_WAVE_ID); - amdgpu_ring_write(ring, ring->adev->gds.vgt_gs_max_wave_id); - if (ib->flags & AMDGPU_IB_FLAG_CE) header = PACKET3(PACKET3_INDIRECT_BUFFER_CNST, 2); else @@ -5128,7 +5122,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = { 5 + /* HDP_INVL */ 8 + 8 + /* FENCE x2 */ 2, /* SWITCH_BUFFER */ - .emit_ib_size = 7, /* gfx_v10_0_ring_emit_ib_gfx */ + .emit_ib_size = 4, /* gfx_v10_0_ring_emit_ib_gfx */ .emit_ib = gfx_v10_0_ring_emit_ib_gfx, .emit_fence = gfx_v10_0_ring_emit_fence, .emit_pipeline_sync = gfx_v10_0_ring_emit_pipeline_sync, @@ -5281,7 +5275,6 @@ static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev) default: adev->gds.gds_size = 0x10000; adev->gds.gds_compute_max_wave_id = 0x4ff; - adev->gds.vgt_gs_max_wave_id = 0x3ff; break; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 52a6fd12e266..1221ebfe9945 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -36,10 +36,10 @@ #include "gc/gc_9_0_offset.h" #include "gc/gc_9_0_sh_mask.h" + #include "vega10_enum.h" #include "hdp/hdp_4_0_offset.h" -#include "soc15.h" #include "soc15_common.h" #include "clearstate_gfx9.h" #include "v9_structs.h" @@ -60,6 +60,9 @@ #define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN_MASK 0x00000001L #define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS_MASK 0x00000006L +#define mmGCEA_PROBE_MAP 0x070c +#define mmGCEA_PROBE_MAP_BASE_IDX 0 + MODULE_FIRMWARE("amdgpu/vega10_ce.bin"); MODULE_FIRMWARE("amdgpu/vega10_pfp.bin"); MODULE_FIRMWARE("amdgpu/vega10_me.bin"); @@ -108,6 +111,13 @@ MODULE_FIRMWARE("amdgpu/arcturus_mec.bin"); MODULE_FIRMWARE("amdgpu/arcturus_mec2.bin"); MODULE_FIRMWARE("amdgpu/arcturus_rlc.bin"); +MODULE_FIRMWARE("amdgpu/renoir_ce.bin"); +MODULE_FIRMWARE("amdgpu/renoir_pfp.bin"); +MODULE_FIRMWARE("amdgpu/renoir_me.bin"); +MODULE_FIRMWARE("amdgpu/renoir_mec.bin"); +MODULE_FIRMWARE("amdgpu/renoir_mec2.bin"); +MODULE_FIRMWARE("amdgpu/renoir_rlc.bin"); + #define mmTCP_CHAN_STEER_0_ARCT 0x0b03 #define mmTCP_CHAN_STEER_0_ARCT_BASE_IDX 0 #define mmTCP_CHAN_STEER_1_ARCT 0x0b04 @@ -611,6 +621,22 @@ static const struct soc15_reg_golden golden_settings_gc_9_1_rv2[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmWD_UTCL1_CNTL, 0x3f8fffff, 0x08000080), }; +static const struct soc15_reg_golden golden_settings_gc_9_1_rn[] = +{ + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL, 0xfffdf3cf, 0x00014104), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_2, 0xff7fffff, 0x0a000000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG2, 0xf00fffff, 0x00000400), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0xf3e777ff, 0x24000042), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG_READ, 0xf3e777ff, 0x24000042), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE, 0x3fffffff, 0x00000001), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_1, 0xffffffff, 0x04040000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfffffeef, 0x010b0000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_HI, 0xffffffff, 0x00000000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_LO, 0xffffffff, 0x00003120), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCEA_PROBE_MAP, 0xffffffff, 0x0000cccc), +}; + static const struct soc15_reg_golden golden_settings_gc_9_x_common[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCP_SD_CNTL, 0xffffffff, 0x000001ff), @@ -755,6 +781,11 @@ static void gfx_v9_0_init_golden_registers(struct amdgpu_device *adev) golden_settings_gc_9_1_rv1, ARRAY_SIZE(golden_settings_gc_9_1_rv1)); break; + case CHIP_RENOIR: + soc15_program_register_sequence(adev, + golden_settings_gc_9_1_rn, + ARRAY_SIZE(golden_settings_gc_9_1_rn)); + return; /* for renoir, don't need common goldensetting */ default: break; } @@ -1008,6 +1039,10 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev) (adev->gfx.rlc_feature_version < 1) || !adev->gfx.rlc.is_rlc_v2_1) adev->pm.pp_feature &= ~PP_GFXOFF_MASK; + if (adev->pm.pp_feature & PP_GFXOFF_MASK) + adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG | + AMD_PG_SUPPORT_CP | + AMD_PG_SUPPORT_RLC_SMU_HS; break; default: break; @@ -1343,6 +1378,9 @@ static int gfx_v9_0_init_microcode(struct amdgpu_device *adev) case CHIP_ARCTURUS: chip_name = "arcturus"; break; + case CHIP_RENOIR: + chip_name = "renoir"; + break; default: BUG(); } @@ -1602,7 +1640,7 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev) return r; } - if (adev->asic_type == CHIP_RAVEN) { + if (adev->asic_type == CHIP_RAVEN || adev->asic_type == CHIP_RENOIR) { /* TODO: double check the cp_table_size for RV */ adev->gfx.rlc.cp_table_size = ALIGN(96 * 5 * 4, 2048) + (64 * 1024); /* JT + GDS */ r = amdgpu_gfx_rlc_init_cpt(adev); @@ -1612,6 +1650,7 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_RAVEN: + case CHIP_RENOIR: gfx_v9_0_init_lbpw(adev); break; case CHIP_VEGA20: @@ -1863,6 +1902,16 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) gb_addr_config &= ~0xf3e777ff; gb_addr_config |= 0x22014042; break; + case CHIP_RENOIR: + adev->gfx.config.max_hw_contexts = 8; + adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; + adev->gfx.config.sc_prim_fifo_size_backend = 0x100; + adev->gfx.config.sc_hiz_tile_fifo_size = 0x80; + adev->gfx.config.sc_earlyz_tile_fifo_size = 0x4C0; + gb_addr_config = RREG32_SOC15(GC, 0, mmGB_ADDR_CONFIG); + gb_addr_config &= ~0xf3e777ff; + gb_addr_config |= 0x22010042; + break; default: BUG(); break; @@ -2140,6 +2189,7 @@ static int gfx_v9_0_sw_init(void *handle) case CHIP_VEGA20: case CHIP_RAVEN: case CHIP_ARCTURUS: + case CHIP_RENOIR: adev->gfx.mec.num_mec = 2; break; default: @@ -2297,7 +2347,7 @@ static int gfx_v9_0_sw_fini(void *handle) gfx_v9_0_mec_fini(adev); gfx_v9_0_ngg_fini(adev); amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj); - if (adev->asic_type == CHIP_RAVEN) { + if (adev->asic_type == CHIP_RAVEN || adev->asic_type == CHIP_RENOIR) { amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, &adev->gfx.rlc.cp_table_gpu_addr, (void **)&adev->gfx.rlc.cp_table_ptr); @@ -2976,6 +3026,7 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_RAVEN: + case CHIP_RENOIR: if (amdgpu_lbpw == 0) gfx_v9_0_enable_lbpw(adev, false); else @@ -4511,6 +4562,9 @@ static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev, { amdgpu_gfx_rlc_enter_safe_mode(adev); + if (is_support_sw_smu(adev) && !enable) + smu_set_gfx_cgpg(&adev->smu, enable); + if ((adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) && enable) { gfx_v9_0_enable_gfx_cg_power_gating(adev, true); if (adev->pg_flags & AMD_PG_SUPPORT_GFX_PIPELINE) @@ -4622,6 +4676,9 @@ static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev, { uint32_t data, def; + if (adev->asic_type == CHIP_ARCTURUS) + return; + amdgpu_gfx_rlc_enter_safe_mode(adev); /* Enable 3D CGCG/CGLS */ @@ -4687,8 +4744,12 @@ static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev /* enable cgcg FSM(0x0000363F) */ def = RREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL); - data = (0x36 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) | - RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK; + if (adev->asic_type == CHIP_ARCTURUS) + data = (0x2000 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) | + RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK; + else + data = (0x36 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) | + RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK; if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS) data |= (0x000F << RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT) | RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK; @@ -4760,6 +4821,7 @@ static int gfx_v9_0_set_powergating_state(void *handle, switch (adev->asic_type) { case CHIP_RAVEN: + case CHIP_RENOIR: if (!enable) { amdgpu_gfx_off_ctrl(adev, false); cancel_delayed_work_sync(&adev->gfx.gfx_off_delay_work); @@ -4778,6 +4840,8 @@ static int gfx_v9_0_set_powergating_state(void *handle, gfx_v9_0_enable_cp_power_gating(adev, false); /* update gfx cgpg state */ + if (is_support_sw_smu(adev) && enable) + smu_set_gfx_cgpg(&adev->smu, enable); gfx_v9_0_update_gfx_cg_power_gating(adev, enable); /* update mgcg state */ @@ -4814,6 +4878,8 @@ static int gfx_v9_0_set_clockgating_state(void *handle, case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: + case CHIP_ARCTURUS: + case CHIP_RENOIR: gfx_v9_0_update_gfx_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); break; @@ -5396,7 +5462,7 @@ static void gfx_v9_0_ring_soft_recovery(struct amdgpu_ring *ring, unsigned vmid) value = REG_SET_FIELD(value, SQ_CMD, MODE, 0x01); value = REG_SET_FIELD(value, SQ_CMD, CHECK_VMID, 1); value = REG_SET_FIELD(value, SQ_CMD, VM_ID, vmid); - WREG32(mmSQ_CMD, value); + WREG32_SOC15(GC, 0, mmSQ_CMD, value); } static void gfx_v9_0_set_gfx_eop_interrupt_state(struct amdgpu_device *adev, @@ -5978,6 +6044,9 @@ static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, if (adev->asic_type != CHIP_VEGA20) return -EINVAL; + if (info->head.sub_block_index >= ARRAY_SIZE(ras_gfx_subblocks)) + return -EINVAL; + if (!ras_gfx_subblocks[info->head.sub_block_index].name) return -EPERM; @@ -6285,6 +6354,7 @@ static void gfx_v9_0_set_rlc_funcs(struct amdgpu_device *adev) case CHIP_VEGA20: case CHIP_RAVEN: case CHIP_ARCTURUS: + case CHIP_RENOIR: adev->gfx.rlc.funcs = &gfx_v9_0_rlc_funcs; break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index 8ce5bf5feb45..8b789f750b72 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -140,7 +140,7 @@ static void gfxhub_v2_0_init_cache_regs(struct amdgpu_device *adev) /* XXX for emulation, Refer to closed source code.*/ tmp = REG_SET_FIELD(tmp, GCVM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE, 0); - tmp = REG_SET_FIELD(tmp, GCVM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 1); + tmp = REG_SET_FIELD(tmp, GCVM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0); tmp = REG_SET_FIELD(tmp, GCVM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1); tmp = REG_SET_FIELD(tmp, GCVM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0); WREG32_SOC15(GC, 0, mmGCVM_L2_CNTL, tmp); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index f585fc92871b..8e8d712f5a6e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -141,17 +141,40 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, } if (printk_ratelimit()) { + struct amdgpu_task_info task_info; + + memset(&task_info, 0, sizeof(struct amdgpu_task_info)); + amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); + dev_err(adev->dev, - "[%s] VMC page fault (src_id:%u ring:%u vmid:%u pasid:%u)\n", + "[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, " + "for process %s pid %d thread %s pid %d)\n", entry->vmid_src ? "mmhub" : "gfxhub", entry->src_id, entry->ring_id, entry->vmid, - entry->pasid); - dev_err(adev->dev, " at page 0x%016llx from %d\n", + entry->pasid, task_info.process_name, task_info.tgid, + task_info.task_name, task_info.pid); + dev_err(adev->dev, " in page starting at address 0x%016llx from client %d\n", addr, entry->client_id); - if (!amdgpu_sriov_vf(adev)) + if (!amdgpu_sriov_vf(adev)) { dev_err(adev->dev, - "VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", + "GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); + dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + REG_GET_FIELD(status, + GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); + dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + REG_GET_FIELD(status, + GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); + dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + REG_GET_FIELD(status, + GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); + dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + REG_GET_FIELD(status, + GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); + dev_err(adev->dev, "\t RW: 0x%lx\n", + REG_GET_FIELD(status, + GCVM_L2_PROTECTION_FAULT_STATUS, RW)); + } } return 0; @@ -230,8 +253,8 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, * * Flush the TLB for the requested page table. */ -static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, - uint32_t vmid, uint32_t flush_type) +static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, + uint32_t vmhub, uint32_t flush_type) { struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; struct dma_fence *fence; @@ -244,7 +267,14 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, mutex_lock(&adev->mman.gtt_window_lock); - gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB_0, 0); + if (vmhub == AMDGPU_MMHUB_0) { + gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB_0, 0); + mutex_unlock(&adev->mman.gtt_window_lock); + return; + } + + BUG_ON(vmhub != AMDGPU_GFXHUB_0); + if (!adev->mman.buffer_funcs_enabled || !adev->ib_pool_ready || adev->in_gpu_reset) { @@ -592,7 +622,6 @@ static unsigned gmc_v10_0_get_vbios_fb_size(struct amdgpu_device *adev) static int gmc_v10_0_sw_init(void *handle) { int r; - int dma_bits; struct amdgpu_device *adev = (struct amdgpu_device *)handle; gfxhub_v2_0_init(adev); @@ -642,26 +671,10 @@ static int gmc_v10_0_sw_init(void *handle) else adev->gmc.stolen_size = 9 * 1024 *1024; - /* - * Set DMA mask + need_dma32 flags. - * PCIE - can handle 44-bits. - * IGP - can handle 44-bits - * PCI - dma32 for legacy pci gart, 44 bits on navi10 - */ - adev->need_dma32 = false; - dma_bits = adev->need_dma32 ? 32 : 44; - - r = pci_set_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); + r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { - adev->need_dma32 = true; - dma_bits = 32; printk(KERN_WARNING "amdgpu: No suitable DMA available.\n"); - } - - r = pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); - if (r) { - pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32)); - printk(KERN_WARNING "amdgpu: No coherent DMA available.\n"); + return r; } r = gmc_v10_0_mc_init(adev); @@ -773,7 +786,8 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev) gfxhub_v2_0_set_fault_enable_default(adev, value); mmhub_v2_0_set_fault_enable_default(adev, value); - gmc_v10_0_flush_gpu_tlb(adev, 0, 0); + gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB_0, 0); + gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB_0, 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index ca8dbe91cc8b..9fb1765e92d1 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -362,8 +362,8 @@ static int gmc_v6_0_mc_init(struct amdgpu_device *adev) return 0; } -static void gmc_v6_0_flush_gpu_tlb(struct amdgpu_device *adev, - uint32_t vmid, uint32_t flush_type) +static void gmc_v6_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, + uint32_t vmhub, uint32_t flush_type) { WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid); } @@ -571,7 +571,7 @@ static int gmc_v6_0_gart_enable(struct amdgpu_device *adev) else gmc_v6_0_set_fault_enable_default(adev, true); - gmc_v6_0_flush_gpu_tlb(adev, 0, 0); + gmc_v6_0_flush_gpu_tlb(adev, 0, 0, 0); dev_info(adev->dev, "PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); @@ -839,9 +839,10 @@ static unsigned gmc_v6_0_get_vbios_fb_size(struct amdgpu_device *adev) static int gmc_v6_0_sw_init(void *handle) { int r; - int dma_bits; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->num_vmhubs = 1; + if (adev->flags & AMD_IS_APU) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; } else { @@ -862,20 +863,12 @@ static int gmc_v6_0_sw_init(void *handle) adev->gmc.mc_mask = 0xffffffffffULL; - adev->need_dma32 = false; - dma_bits = adev->need_dma32 ? 32 : 40; - r = pci_set_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); + r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { - adev->need_dma32 = true; - dma_bits = 32; dev_warn(adev->dev, "amdgpu: No suitable DMA available.\n"); + return r; } - r = pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); - if (r) { - pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32)); - dev_warn(adev->dev, "amdgpu: No coherent DMA available.\n"); - } - adev->need_swiotlb = drm_need_swiotlb(dma_bits); + adev->need_swiotlb = drm_need_swiotlb(44); r = gmc_v6_0_init_microcode(adev); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 57f80065d57a..0c3d9bc3a641 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -433,8 +433,8 @@ static int gmc_v7_0_mc_init(struct amdgpu_device *adev) * * Flush the TLB for the requested page table (CIK). */ -static void gmc_v7_0_flush_gpu_tlb(struct amdgpu_device *adev, - uint32_t vmid, uint32_t flush_type) +static void gmc_v7_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, + uint32_t vmhub, uint32_t flush_type) { /* bits 0-15 are the VM contexts0-15 */ WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid); @@ -677,7 +677,7 @@ static int gmc_v7_0_gart_enable(struct amdgpu_device *adev) WREG32(mmCHUB_CONTROL, tmp); } - gmc_v7_0_flush_gpu_tlb(adev, 0, 0); + gmc_v7_0_flush_gpu_tlb(adev, 0, 0, 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); @@ -959,9 +959,10 @@ static unsigned gmc_v7_0_get_vbios_fb_size(struct amdgpu_device *adev) static int gmc_v7_0_sw_init(void *handle) { int r; - int dma_bits; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->num_vmhubs = 1; + if (adev->flags & AMD_IS_APU) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; } else { @@ -990,25 +991,12 @@ static int gmc_v7_0_sw_init(void *handle) */ adev->gmc.mc_mask = 0xffffffffffULL; /* 40 bit MC */ - /* set DMA mask + need_dma32 flags. - * PCIE - can handle 40-bits. - * IGP - can handle 40-bits - * PCI - dma32 for legacy pci gart, 40 bits on newer asics - */ - adev->need_dma32 = false; - dma_bits = adev->need_dma32 ? 32 : 40; - r = pci_set_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); + r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(40)); if (r) { - adev->need_dma32 = true; - dma_bits = 32; pr_warn("amdgpu: No suitable DMA available\n"); + return r; } - r = pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); - if (r) { - pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32)); - pr_warn("amdgpu: No coherent DMA available\n"); - } - adev->need_swiotlb = drm_need_swiotlb(dma_bits); + adev->need_swiotlb = drm_need_swiotlb(40); r = gmc_v7_0_init_microcode(adev); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 9238280d1ff7..ea764dd9245d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -635,8 +635,8 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev) * * Flush the TLB for the requested page table (VI). */ -static void gmc_v8_0_flush_gpu_tlb(struct amdgpu_device *adev, - uint32_t vmid, uint32_t flush_type) +static void gmc_v8_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, + uint32_t vmhub, uint32_t flush_type) { /* bits 0-15 are the VM contexts0-15 */ WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid); @@ -921,7 +921,7 @@ static int gmc_v8_0_gart_enable(struct amdgpu_device *adev) else gmc_v8_0_set_fault_enable_default(adev, true); - gmc_v8_0_flush_gpu_tlb(adev, 0, 0); + gmc_v8_0_flush_gpu_tlb(adev, 0, 0, 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); @@ -1079,9 +1079,10 @@ static unsigned gmc_v8_0_get_vbios_fb_size(struct amdgpu_device *adev) static int gmc_v8_0_sw_init(void *handle) { int r; - int dma_bits; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->num_vmhubs = 1; + if (adev->flags & AMD_IS_APU) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; } else { @@ -1116,25 +1117,12 @@ static int gmc_v8_0_sw_init(void *handle) */ adev->gmc.mc_mask = 0xffffffffffULL; /* 40 bit MC */ - /* set DMA mask + need_dma32 flags. - * PCIE - can handle 40-bits. - * IGP - can handle 40-bits - * PCI - dma32 for legacy pci gart, 40 bits on newer asics - */ - adev->need_dma32 = false; - dma_bits = adev->need_dma32 ? 32 : 40; - r = pci_set_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); + r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(40)); if (r) { - adev->need_dma32 = true; - dma_bits = 32; pr_warn("amdgpu: No suitable DMA available\n"); + return r; } - r = pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); - if (r) { - pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32)); - pr_warn("amdgpu: No coherent DMA available\n"); - } - adev->need_swiotlb = drm_need_swiotlb(dma_bits); + adev->need_swiotlb = drm_need_swiotlb(40); r = gmc_v8_0_init_microcode(adev); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 0c77b9f244bb..7ae5f5860f31 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -47,6 +47,7 @@ #include "gfxhub_v1_0.h" #include "mmhub_v1_0.h" +#include "athub_v1_0.h" #include "gfxhub_v1_1.h" #include "mmhub_v9_4.h" #include "umc_v6_1.h" @@ -266,7 +267,7 @@ static int gmc_v9_0_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - struct ras_common_if *ras_if = adev->gmc.ras_if; + struct ras_common_if *ras_if = adev->gmc.umc_ras_if; struct ras_dispatch_if ih_data = { .entry = entry, }; @@ -390,6 +391,9 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); + dev_err(adev->dev, "\t RW: 0x%lx\n", + REG_GET_FIELD(status, + VM_L2_PROTECTION_FAULT_STATUS, RW)); } } @@ -452,44 +456,45 @@ static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid, * * Flush the TLB for the requested page table using certain type. */ -static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, - uint32_t vmid, uint32_t flush_type) +static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, + uint32_t vmhub, uint32_t flush_type) { const unsigned eng = 17; - unsigned i, j; + u32 j, tmp; + struct amdgpu_vmhub *hub; - for (i = 0; i < adev->num_vmhubs; ++i) { - struct amdgpu_vmhub *hub = &adev->vmhub[i]; - u32 tmp = gmc_v9_0_get_invalidate_req(vmid, flush_type); + BUG_ON(vmhub >= adev->num_vmhubs); - /* This is necessary for a HW workaround under SRIOV as well - * as GFXOFF under bare metal - */ - if (adev->gfx.kiq.ring.sched.ready && - (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) && - !adev->in_gpu_reset) { - uint32_t req = hub->vm_inv_eng0_req + eng; - uint32_t ack = hub->vm_inv_eng0_ack + eng; - - amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, tmp, - 1 << vmid); - continue; - } + hub = &adev->vmhub[vmhub]; + tmp = gmc_v9_0_get_invalidate_req(vmid, flush_type); - spin_lock(&adev->gmc.invalidate_lock); - WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp); - for (j = 0; j < adev->usec_timeout; j++) { - tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng); - if (tmp & (1 << vmid)) - break; - udelay(1); - } - spin_unlock(&adev->gmc.invalidate_lock); - if (j < adev->usec_timeout) - continue; + /* This is necessary for a HW workaround under SRIOV as well + * as GFXOFF under bare metal + */ + if (adev->gfx.kiq.ring.sched.ready && + (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) && + !adev->in_gpu_reset) { + uint32_t req = hub->vm_inv_eng0_req + eng; + uint32_t ack = hub->vm_inv_eng0_ack + eng; + + amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, tmp, + 1 << vmid); + return; + } - DRM_ERROR("Timeout waiting for VM flush ACK!\n"); + spin_lock(&adev->gmc.invalidate_lock); + WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp); + for (j = 0; j < adev->usec_timeout; j++) { + tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng); + if (tmp & (1 << vmid)) + break; + udelay(1); } + spin_unlock(&adev->gmc.invalidate_lock); + if (j < adev->usec_timeout) + return; + + DRM_ERROR("Timeout waiting for VM flush ACK!\n"); } static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, @@ -656,6 +661,17 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) } } +static void gmc_v9_0_set_mmhub_funcs(struct amdgpu_device *adev) +{ + switch (adev->asic_type) { + case CHIP_VEGA20: + adev->mmhub_funcs = &mmhub_v1_0_funcs; + break; + default: + break; + } +} + static int gmc_v9_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -663,6 +679,7 @@ static int gmc_v9_0_early_init(void *handle) gmc_v9_0_set_gmc_funcs(adev); gmc_v9_0_set_irq_funcs(adev); gmc_v9_0_set_umc_funcs(adev); + gmc_v9_0_set_mmhub_funcs(adev); adev->gmc.shared_aperture_start = 0x2000000000000000ULL; adev->gmc.shared_aperture_end = @@ -690,6 +707,7 @@ static bool gmc_v9_0_keep_stolen_memory(struct amdgpu_device *adev) case CHIP_VEGA10: case CHIP_RAVEN: case CHIP_ARCTURUS: + case CHIP_RENOIR: return true; case CHIP_VEGA12: case CHIP_VEGA20: @@ -728,27 +746,25 @@ static int gmc_v9_0_allocate_vm_inv_eng(struct amdgpu_device *adev) return 0; } -static int gmc_v9_0_ecc_late_init(void *handle) +static int gmc_v9_0_ecc_ras_block_late_init(void *handle, + struct ras_fs_if *fs_info, struct ras_common_if *ras_block) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct ras_common_if **ras_if = &adev->gmc.ras_if; + struct ras_common_if **ras_if = NULL; struct ras_ih_if ih_info = { .cb = gmc_v9_0_process_ras_data_cb, }; - struct ras_fs_if fs_info = { - .sysfs_name = "umc_err_count", - .debugfs_name = "umc_err_inject", - }; - struct ras_common_if ras_block = { - .block = AMDGPU_RAS_BLOCK__UMC, - .type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, - .sub_block_index = 0, - .name = "umc", - }; int r; - if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) { - amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0); + if (ras_block->block == AMDGPU_RAS_BLOCK__UMC) + ras_if = &adev->gmc.umc_ras_if; + else if (ras_block->block == AMDGPU_RAS_BLOCK__MMHUB) + ras_if = &adev->gmc.mmhub_ras_if; + else + BUG(); + + if (!amdgpu_ras_is_supported(adev, ras_block->block)) { + amdgpu_ras_feature_enable_on_boot(adev, ras_block, 0); return 0; } @@ -763,7 +779,7 @@ static int gmc_v9_0_ecc_late_init(void *handle) if (r == -EAGAIN) { /* request a gpu reset. will run again. */ amdgpu_ras_request_reset_on_boot(adev, - AMDGPU_RAS_BLOCK__UMC); + ras_block->block); return 0; } /* fail to enable ras, cleanup all. */ @@ -777,41 +793,46 @@ static int gmc_v9_0_ecc_late_init(void *handle) if (!*ras_if) return -ENOMEM; - **ras_if = ras_block; + **ras_if = *ras_block; r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1); if (r) { if (r == -EAGAIN) { amdgpu_ras_request_reset_on_boot(adev, - AMDGPU_RAS_BLOCK__UMC); + ras_block->block); r = 0; } goto feature; } ih_info.head = **ras_if; - fs_info.head = **ras_if; + fs_info->head = **ras_if; - r = amdgpu_ras_interrupt_add_handler(adev, &ih_info); - if (r) - goto interrupt; + if (ras_block->block == AMDGPU_RAS_BLOCK__UMC) { + r = amdgpu_ras_interrupt_add_handler(adev, &ih_info); + if (r) + goto interrupt; + } - amdgpu_ras_debugfs_create(adev, &fs_info); + amdgpu_ras_debugfs_create(adev, fs_info); - r = amdgpu_ras_sysfs_create(adev, &fs_info); + r = amdgpu_ras_sysfs_create(adev, fs_info); if (r) goto sysfs; resume: - r = amdgpu_irq_get(adev, &adev->gmc.ecc_irq, 0); - if (r) - goto irq; + if (ras_block->block == AMDGPU_RAS_BLOCK__UMC) { + r = amdgpu_irq_get(adev, &adev->gmc.ecc_irq, 0); + if (r) + goto irq; + } return 0; irq: amdgpu_ras_sysfs_remove(adev, *ras_if); sysfs: amdgpu_ras_debugfs_remove(adev, *ras_if); - amdgpu_ras_interrupt_remove_handler(adev, &ih_info); + if (ras_block->block == AMDGPU_RAS_BLOCK__UMC) + amdgpu_ras_interrupt_remove_handler(adev, &ih_info); interrupt: amdgpu_ras_feature_enable(adev, *ras_if, 0); feature: @@ -820,6 +841,40 @@ feature: return r; } +static int gmc_v9_0_ecc_late_init(void *handle) +{ + int r; + + struct ras_fs_if umc_fs_info = { + .sysfs_name = "umc_err_count", + .debugfs_name = "umc_err_inject", + }; + struct ras_common_if umc_ras_block = { + .block = AMDGPU_RAS_BLOCK__UMC, + .type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + .sub_block_index = 0, + .name = "umc", + }; + struct ras_fs_if mmhub_fs_info = { + .sysfs_name = "mmhub_err_count", + .debugfs_name = "mmhub_err_inject", + }; + struct ras_common_if mmhub_ras_block = { + .block = AMDGPU_RAS_BLOCK__MMHUB, + .type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + .sub_block_index = 0, + .name = "mmhub", + }; + + r = gmc_v9_0_ecc_ras_block_late_init(handle, + &umc_fs_info, &umc_ras_block); + if (r) + return r; + + r = gmc_v9_0_ecc_ras_block_late_init(handle, + &mmhub_fs_info, &mmhub_ras_block); + return r; +} static int gmc_v9_0_late_init(void *handle) { @@ -869,18 +924,17 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc) { u64 base = 0; - if (!amdgpu_sriov_vf(adev)) { - if (adev->asic_type == CHIP_ARCTURUS) - base = mmhub_v9_4_get_fb_location(adev); - else - base = mmhub_v1_0_get_fb_location(adev); - } + + if (adev->asic_type == CHIP_ARCTURUS) + base = mmhub_v9_4_get_fb_location(adev); + else if (!amdgpu_sriov_vf(adev)) + base = mmhub_v1_0_get_fb_location(adev); + /* add the xgmi offset of the physical node */ base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; amdgpu_gmc_vram_location(adev, mc, base); amdgpu_gmc_gart_location(adev, mc); - if (!amdgpu_sriov_vf(adev)) - amdgpu_gmc_agp_location(adev, mc); + amdgpu_gmc_agp_location(adev, mc); /* base offset of vram pages */ adev->vm_manager.vram_base_offset = gfxhub_v1_0_get_mc_fb_offset(adev); @@ -959,6 +1013,7 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev) adev->gmc.gart_size = 512ULL << 20; break; case CHIP_RAVEN: /* DCE SG support */ + case CHIP_RENOIR: adev->gmc.gart_size = 1024ULL << 20; break; } @@ -1009,6 +1064,7 @@ static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_RAVEN: + case CHIP_RENOIR: viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION); size = (REG_GET_FIELD(viewport, HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) * @@ -1037,7 +1093,6 @@ static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) static int gmc_v9_0_sw_init(void *handle) { int r; - int dma_bits; struct amdgpu_device *adev = (struct amdgpu_device *)handle; gfxhub_v1_0_init(adev); @@ -1065,8 +1120,10 @@ static int gmc_v9_0_sw_init(void *handle) case CHIP_VEGA10: case CHIP_VEGA12: case CHIP_VEGA20: + case CHIP_RENOIR: adev->num_vmhubs = 2; + /* * To fulfill 4-level page support, * vm size is 256TB (48bit), maximum size of Vega10, @@ -1119,25 +1176,12 @@ static int gmc_v9_0_sw_init(void *handle) */ adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */ - /* set DMA mask + need_dma32 flags. - * PCIE - can handle 44-bits. - * IGP - can handle 44-bits - * PCI - dma32 for legacy pci gart, 44 bits on vega10 - */ - adev->need_dma32 = false; - dma_bits = adev->need_dma32 ? 32 : 44; - r = pci_set_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); + r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { - adev->need_dma32 = true; - dma_bits = 32; printk(KERN_WARNING "amdgpu: No suitable DMA available.\n"); + return r; } - r = pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits)); - if (r) { - pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32)); - printk(KERN_WARNING "amdgpu: No coherent DMA available.\n"); - } - adev->need_swiotlb = drm_need_swiotlb(dma_bits); + adev->need_swiotlb = drm_need_swiotlb(44); if (adev->gmc.xgmi.supported) { r = gfxhub_v1_1_get_xgmi_info(adev); @@ -1180,21 +1224,32 @@ static int gmc_v9_0_sw_fini(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC) && - adev->gmc.ras_if) { - struct ras_common_if *ras_if = adev->gmc.ras_if; + adev->gmc.umc_ras_if) { + struct ras_common_if *ras_if = adev->gmc.umc_ras_if; struct ras_ih_if ih_info = { .head = *ras_if, }; - /*remove fs first*/ + /* remove fs first */ amdgpu_ras_debugfs_remove(adev, ras_if); amdgpu_ras_sysfs_remove(adev, ras_if); - /*remove the IH*/ + /* remove the IH */ amdgpu_ras_interrupt_remove_handler(adev, &ih_info); amdgpu_ras_feature_enable(adev, ras_if, 0); kfree(ras_if); } + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB) && + adev->gmc.mmhub_ras_if) { + struct ras_common_if *ras_if = adev->gmc.mmhub_ras_if; + + /* remove fs and disable ras feature */ + amdgpu_ras_debugfs_remove(adev, ras_if); + amdgpu_ras_sysfs_remove(adev, ras_if); + amdgpu_ras_feature_enable(adev, ras_if, 0); + kfree(ras_if); + } + amdgpu_gem_force_release(adev); amdgpu_vm_manager_fini(adev); @@ -1227,6 +1282,7 @@ static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev) case CHIP_VEGA12: break; case CHIP_RAVEN: + /* TODO for renoir */ soc15_program_register_sequence(adev, golden_settings_athub_1_0_0, ARRAY_SIZE(golden_settings_athub_1_0_0)); @@ -1243,7 +1299,7 @@ static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev) */ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) { - int r; + int r, i; bool value; u32 tmp; @@ -1261,6 +1317,7 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_RAVEN: + /* TODO for renoir */ mmhub_v1_0_update_power_gating(adev, true); break; default: @@ -1299,7 +1356,9 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) mmhub_v9_4_set_fault_enable_default(adev, value); else mmhub_v1_0_set_fault_enable_default(adev, value); - gmc_v9_0_flush_gpu_tlb(adev, 0, 0); + + for (i = 0; i < adev->num_vmhubs; ++i) + gmc_v9_0_flush_gpu_tlb(adev, 0, i, 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), @@ -1408,9 +1467,13 @@ static int gmc_v9_0_set_clockgating_state(void *handle, struct amdgpu_device *adev = (struct amdgpu_device *)handle; if (adev->asic_type == CHIP_ARCTURUS) - return 0; + mmhub_v9_4_set_clockgating(adev, state); + else + mmhub_v1_0_set_clockgating(adev, state); + + athub_v1_0_set_clockgating(adev, state); - return mmhub_v1_0_set_clockgating(adev, state); + return 0; } static void gmc_v9_0_get_clockgating_state(void *handle, u32 *flags) @@ -1418,9 +1481,11 @@ static void gmc_v9_0_get_clockgating_state(void *handle, u32 *flags) struct amdgpu_device *adev = (struct amdgpu_device *)handle; if (adev->asic_type == CHIP_ARCTURUS) - return; + mmhub_v9_4_get_clockgating(adev, flags); + else + mmhub_v1_0_get_clockgating(adev, flags); - mmhub_v1_0_get_clockgating(adev, flags); + athub_v1_0_get_clockgating(adev, flags); } static int gmc_v9_0_set_powergating_state(void *handle, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 292f3b1cddf2..04cd4b6f95d4 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -21,13 +21,13 @@ * */ #include "amdgpu.h" +#include "amdgpu_ras.h" #include "mmhub_v1_0.h" #include "mmhub/mmhub_1_0_offset.h" #include "mmhub/mmhub_1_0_sh_mask.h" #include "mmhub/mmhub_1_0_default.h" -#include "athub/athub_1_0_offset.h" -#include "athub/athub_1_0_sh_mask.h" +#include "mmhub/mmhub_9_4_0_offset.h" #include "vega10_enum.h" #include "soc15_common.h" @@ -35,6 +35,9 @@ #define mmDAGB0_CNTL_MISC2_RV 0x008f #define mmDAGB0_CNTL_MISC2_RV_BASE_IDX 0 +#define EA_EDC_CNT_MASK 0x3 +#define EA_EDC_CNT_SHIFT 0x2 + u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev) { u64 base = RREG32_SOC15(MMHUB, 0, mmMC_VM_FB_LOCATION_BASE); @@ -491,22 +494,6 @@ static void mmhub_v1_0_update_medium_grain_clock_gating(struct amdgpu_device *ad WREG32_SOC15(MMHUB, 0, mmDAGB1_CNTL_MISC2, data2); } -static void athub_update_medium_grain_clock_gating(struct amdgpu_device *adev, - bool enable) -{ - uint32_t def, data; - - def = data = RREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL); - - if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG)) - data |= ATHUB_MISC_CNTL__CG_ENABLE_MASK; - else - data &= ~ATHUB_MISC_CNTL__CG_ENABLE_MASK; - - if (def != data) - WREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL, data); -} - static void mmhub_v1_0_update_medium_grain_light_sleep(struct amdgpu_device *adev, bool enable) { @@ -523,23 +510,6 @@ static void mmhub_v1_0_update_medium_grain_light_sleep(struct amdgpu_device *ade WREG32_SOC15(MMHUB, 0, mmATC_L2_MISC_CG, data); } -static void athub_update_medium_grain_light_sleep(struct amdgpu_device *adev, - bool enable) -{ - uint32_t def, data; - - def = data = RREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL); - - if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_LS) && - (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS)) - data |= ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK; - else - data &= ~ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK; - - if(def != data) - WREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL, data); -} - int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev, enum amd_clockgating_state state) { @@ -551,14 +521,11 @@ int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev, case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: + case CHIP_RENOIR: mmhub_v1_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); - athub_update_medium_grain_clock_gating(adev, - state == AMD_CG_STATE_GATE ? true : false); mmhub_v1_0_update_medium_grain_light_sleep(adev, state == AMD_CG_STATE_GATE ? true : false); - athub_update_medium_grain_light_sleep(adev, - state == AMD_CG_STATE_GATE ? true : false); break; default: break; @@ -569,18 +536,85 @@ int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev, void mmhub_v1_0_get_clockgating(struct amdgpu_device *adev, u32 *flags) { - int data; + int data, data1; if (amdgpu_sriov_vf(adev)) *flags = 0; + data = RREG32_SOC15(MMHUB, 0, mmATC_L2_MISC_CG); + + data1 = RREG32_SOC15(MMHUB, 0, mmDAGB0_CNTL_MISC2); + /* AMD_CG_SUPPORT_MC_MGCG */ - data = RREG32_SOC15(ATHUB, 0, mmATHUB_MISC_CNTL); - if (data & ATHUB_MISC_CNTL__CG_ENABLE_MASK) + if ((data & ATC_L2_MISC_CG__ENABLE_MASK) && + !(data1 & (DAGB0_CNTL_MISC2__DISABLE_WRREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK))) *flags |= AMD_CG_SUPPORT_MC_MGCG; /* AMD_CG_SUPPORT_MC_LS */ - data = RREG32_SOC15(MMHUB, 0, mmATC_L2_MISC_CG); if (data & ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK) *flags |= AMD_CG_SUPPORT_MC_LS; } + +static void mmhub_v1_0_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + int i; + uint32_t ea0_edc_cnt, ea0_edc_cnt2; + uint32_t ea1_edc_cnt, ea1_edc_cnt2; + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + + /* EDC CNT will be cleared automatically after read */ + ea0_edc_cnt = RREG32_SOC15(MMHUB, 0, mmMMEA0_EDC_CNT_VG20); + ea0_edc_cnt2 = RREG32_SOC15(MMHUB, 0, mmMMEA0_EDC_CNT2_VG20); + ea1_edc_cnt = RREG32_SOC15(MMHUB, 0, mmMMEA1_EDC_CNT_VG20); + ea1_edc_cnt2 = RREG32_SOC15(MMHUB, 0, mmMMEA1_EDC_CNT2_VG20); + + /* error count of each error type is recorded by 2 bits, + * ce and ue count in EDC_CNT + */ + for (i = 0; i < 5; i++) { + err_data->ce_count += (ea0_edc_cnt & EA_EDC_CNT_MASK); + err_data->ce_count += (ea1_edc_cnt & EA_EDC_CNT_MASK); + ea0_edc_cnt >>= EA_EDC_CNT_SHIFT; + ea1_edc_cnt >>= EA_EDC_CNT_SHIFT; + err_data->ue_count += (ea0_edc_cnt & EA_EDC_CNT_MASK); + err_data->ue_count += (ea1_edc_cnt & EA_EDC_CNT_MASK); + ea0_edc_cnt >>= EA_EDC_CNT_SHIFT; + ea1_edc_cnt >>= EA_EDC_CNT_SHIFT; + } + /* successive ue count in EDC_CNT */ + for (i = 0; i < 5; i++) { + err_data->ue_count += (ea0_edc_cnt & EA_EDC_CNT_MASK); + err_data->ue_count += (ea1_edc_cnt & EA_EDC_CNT_MASK); + ea0_edc_cnt >>= EA_EDC_CNT_SHIFT; + ea1_edc_cnt >>= EA_EDC_CNT_SHIFT; + } + + /* ce and ue count in EDC_CNT2 */ + for (i = 0; i < 3; i++) { + err_data->ce_count += (ea0_edc_cnt2 & EA_EDC_CNT_MASK); + err_data->ce_count += (ea1_edc_cnt2 & EA_EDC_CNT_MASK); + ea0_edc_cnt2 >>= EA_EDC_CNT_SHIFT; + ea1_edc_cnt2 >>= EA_EDC_CNT_SHIFT; + err_data->ue_count += (ea0_edc_cnt2 & EA_EDC_CNT_MASK); + err_data->ue_count += (ea1_edc_cnt2 & EA_EDC_CNT_MASK); + ea0_edc_cnt2 >>= EA_EDC_CNT_SHIFT; + ea1_edc_cnt2 >>= EA_EDC_CNT_SHIFT; + } + /* successive ue count in EDC_CNT2 */ + for (i = 0; i < 6; i++) { + err_data->ue_count += (ea0_edc_cnt2 & EA_EDC_CNT_MASK); + err_data->ue_count += (ea1_edc_cnt2 & EA_EDC_CNT_MASK); + ea0_edc_cnt2 >>= EA_EDC_CNT_SHIFT; + ea1_edc_cnt2 >>= EA_EDC_CNT_SHIFT; + } +} + +const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs = { + .query_ras_error_count = mmhub_v1_0_query_ras_error_count, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h index 0de0fdf98c00..c43319e8f945 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h @@ -23,6 +23,8 @@ #ifndef __MMHUB_V1_0_H__ #define __MMHUB_V1_0_H__ +extern const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs; + u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev); int mmhub_v1_0_gart_enable(struct amdgpu_device *adev); void mmhub_v1_0_gart_disable(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index d2f4775299c7..3542c203c3c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -126,7 +126,7 @@ static void mmhub_v2_0_init_cache_regs(struct amdgpu_device *adev) /* XXX for emulation, Refer to closed source code.*/ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE, 0); - tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 1); + tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0); tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1); tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0); WREG32_SOC15(MMHUB, 0, mmMMVM_L2_CNTL, tmp); @@ -407,6 +407,7 @@ int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev, switch (adev->asic_type) { case CHIP_NAVI10: case CHIP_NAVI14: + case CHIP_NAVI12: mmhub_v2_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); mmhub_v2_0_update_medium_grain_light_sleep(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index 33b0de54a5da..0cf7ef44b4b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -515,3 +515,128 @@ void mmhub_v9_4_init(struct amdgpu_device *adev) i * MMHUB_INSTANCE_REGISTER_OFFSET; } } + +static void mmhub_v9_4_update_medium_grain_clock_gating(struct amdgpu_device *adev, + bool enable) +{ + uint32_t def, data, def1, data1; + int i, j; + int dist = mmDAGB1_CNTL_MISC2 - mmDAGB0_CNTL_MISC2; + + for (i = 0; i < MMHUB_NUM_INSTANCES; i++) { + def = data = RREG32_SOC15_OFFSET(MMHUB, 0, + mmATCL2_0_ATC_L2_MISC_CG, + i * MMHUB_INSTANCE_REGISTER_OFFSET); + + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG)) + data |= ATCL2_0_ATC_L2_MISC_CG__ENABLE_MASK; + else + data &= ~ATCL2_0_ATC_L2_MISC_CG__ENABLE_MASK; + + if (def != data) + WREG32_SOC15_OFFSET(MMHUB, 0, mmATCL2_0_ATC_L2_MISC_CG, + i * MMHUB_INSTANCE_REGISTER_OFFSET, data); + + for (j = 0; j < 5; j++) { + def1 = data1 = RREG32_SOC15_OFFSET(MMHUB, 0, + mmDAGB0_CNTL_MISC2, + i * MMHUB_INSTANCE_REGISTER_OFFSET + + j * dist); + if (enable && + (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG)) { + data1 &= + ~(DAGB0_CNTL_MISC2__DISABLE_WRREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK); + } else { + data1 |= + (DAGB0_CNTL_MISC2__DISABLE_WRREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK); + } + + if (def1 != data1) + WREG32_SOC15_OFFSET(MMHUB, 0, + mmDAGB0_CNTL_MISC2, + i * MMHUB_INSTANCE_REGISTER_OFFSET + + j * dist, data1); + + if (i == 1 && j == 3) + break; + } + } +} + +static void mmhub_v9_4_update_medium_grain_light_sleep(struct amdgpu_device *adev, + bool enable) +{ + uint32_t def, data; + int i; + + for (i = 0; i < MMHUB_NUM_INSTANCES; i++) { + def = data = RREG32_SOC15_OFFSET(MMHUB, 0, + mmATCL2_0_ATC_L2_MISC_CG, + i * MMHUB_INSTANCE_REGISTER_OFFSET); + + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_LS)) + data |= ATCL2_0_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK; + else + data &= ~ATCL2_0_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK; + + if (def != data) + WREG32_SOC15_OFFSET(MMHUB, 0, mmATCL2_0_ATC_L2_MISC_CG, + i * MMHUB_INSTANCE_REGISTER_OFFSET, data); + } +} + +int mmhub_v9_4_set_clockgating(struct amdgpu_device *adev, + enum amd_clockgating_state state) +{ + if (amdgpu_sriov_vf(adev)) + return 0; + + switch (adev->asic_type) { + case CHIP_ARCTURUS: + mmhub_v9_4_update_medium_grain_clock_gating(adev, + state == AMD_CG_STATE_GATE ? true : false); + mmhub_v9_4_update_medium_grain_light_sleep(adev, + state == AMD_CG_STATE_GATE ? true : false); + break; + default: + break; + } + + return 0; +} + +void mmhub_v9_4_get_clockgating(struct amdgpu_device *adev, u32 *flags) +{ + int data, data1; + + if (amdgpu_sriov_vf(adev)) + *flags = 0; + + /* AMD_CG_SUPPORT_MC_MGCG */ + data = RREG32_SOC15(MMHUB, 0, mmATCL2_0_ATC_L2_MISC_CG); + + data1 = RREG32_SOC15(MMHUB, 0, mmATCL2_0_ATC_L2_MISC_CG); + + if ((data & ATCL2_0_ATC_L2_MISC_CG__ENABLE_MASK) && + !(data1 & (DAGB0_CNTL_MISC2__DISABLE_WRREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK))) + *flags |= AMD_CG_SUPPORT_MC_MGCG; + + /* AMD_CG_SUPPORT_MC_LS */ + if (data & ATCL2_0_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK) + *flags |= AMD_CG_SUPPORT_MC_LS; +} diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h index 9ba3dd808826..d435cfcec1a8 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h @@ -29,5 +29,8 @@ void mmhub_v9_4_gart_disable(struct amdgpu_device *adev); void mmhub_v9_4_set_fault_enable_default(struct amdgpu_device *adev, bool value); void mmhub_v9_4_init(struct amdgpu_device *adev); +int mmhub_v9_4_set_clockgating(struct amdgpu_device *adev, + enum amd_clockgating_state state); +void mmhub_v9_4_get_clockgating(struct amdgpu_device *adev, u32 *flags); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c index 73419fa38159..74eecb768a82 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c @@ -91,6 +91,26 @@ static void nbio_v7_0_sdma_doorbell_range(struct amdgpu_device *adev, int instan WREG32(reg, doorbell_range); } +static void nbio_v7_0_vcn_doorbell_range(struct amdgpu_device *adev, bool use_doorbell, + int doorbell_index, int instance) +{ + u32 reg = SOC15_REG_OFFSET(NBIO, 0, mmBIF_MMSCH0_DOORBELL_RANGE); + + u32 doorbell_range = RREG32(reg); + + if (use_doorbell) { + doorbell_range = REG_SET_FIELD(doorbell_range, + BIF_MMSCH0_DOORBELL_RANGE, OFFSET, + doorbell_index); + doorbell_range = REG_SET_FIELD(doorbell_range, + BIF_MMSCH0_DOORBELL_RANGE, SIZE, 8); + } else + doorbell_range = REG_SET_FIELD(doorbell_range, + BIF_MMSCH0_DOORBELL_RANGE, SIZE, 0); + + WREG32(reg, doorbell_range); +} + static void nbio_v7_0_enable_doorbell_aperture(struct amdgpu_device *adev, bool enable) { @@ -282,6 +302,7 @@ const struct amdgpu_nbio_funcs nbio_v7_0_funcs = { .hdp_flush = nbio_v7_0_hdp_flush, .get_memsize = nbio_v7_0_get_memsize, .sdma_doorbell_range = nbio_v7_0_sdma_doorbell_range, + .vcn_doorbell_range = nbio_v7_0_vcn_doorbell_range, .enable_doorbell_aperture = nbio_v7_0_enable_doorbell_aperture, .enable_doorbell_selfring_aperture = nbio_v7_0_enable_doorbell_selfring_aperture, .ih_doorbell_range = nbio_v7_0_ih_doorbell_range, diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 3e67536f0dc9..dbd577a58899 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -576,7 +576,6 @@ static const struct amdgpu_asic_funcs nv_asic_funcs = static int nv_common_early_init(void *handle) { - bool psp_enabled = false; struct amdgpu_device *adev = (struct amdgpu_device *)handle; adev->smc_rreg = NULL; @@ -593,10 +592,6 @@ static int nv_common_early_init(void *handle) adev->asic_funcs = &nv_asic_funcs; - if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP) && - (amdgpu_ip_block_mask & (1 << AMD_IP_BLOCK_TYPE_PSP))) - psp_enabled = true; - adev->rev_id = nv_get_rev_id(adev); adev->external_rev_id = 0xff; switch (adev->asic_type) { @@ -617,7 +612,6 @@ static int nv_common_early_init(void *handle) AMD_CG_SUPPORT_BIF_LS; adev->pg_flags = AMD_PG_SUPPORT_VCN | AMD_PG_SUPPORT_VCN_DPG | - AMD_PG_SUPPORT_MMHUB | AMD_PG_SUPPORT_ATHUB; adev->external_rev_id = adev->rev_id + 0x1; break; @@ -641,7 +635,21 @@ static int nv_common_early_init(void *handle) adev->external_rev_id = adev->rev_id + 20; break; case CHIP_NAVI12: - adev->cg_flags = 0; + adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | + AMD_CG_SUPPORT_GFX_MGLS | + AMD_CG_SUPPORT_GFX_CGCG | + AMD_CG_SUPPORT_GFX_CP_LS | + AMD_CG_SUPPORT_GFX_RLC_LS | + AMD_CG_SUPPORT_IH_CG | + AMD_CG_SUPPORT_HDP_MGCG | + AMD_CG_SUPPORT_HDP_LS | + AMD_CG_SUPPORT_SDMA_MGCG | + AMD_CG_SUPPORT_SDMA_LS | + AMD_CG_SUPPORT_MC_MGCG | + AMD_CG_SUPPORT_MC_LS | + AMD_CG_SUPPORT_ATHUB_MGCG | + AMD_CG_SUPPORT_ATHUB_LS | + AMD_CG_SUPPORT_VCN_MGCG; adev->pg_flags = AMD_PG_SUPPORT_VCN_DPG; adev->external_rev_id = adev->rev_id + 0xa; break; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c index ce1ea31feee0..5d95e614369a 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c @@ -190,7 +190,6 @@ static int psp_v10_0_ring_destroy(struct psp_context *psp, } static int psp_v10_0_cmd_submit(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, uint64_t cmd_buf_mc_addr, uint64_t fence_mc_addr, int index) { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index f0a0ecb07818..10166104b8a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -498,7 +498,6 @@ static int psp_v11_0_ring_destroy(struct psp_context *psp, } static int psp_v11_0_cmd_submit(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, uint64_t cmd_buf_mc_addr, uint64_t fence_mc_addr, int index) { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c new file mode 100644 index 000000000000..c72e43f8e0be --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c @@ -0,0 +1,565 @@ +/* + * 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. + */ + +#include <linux/firmware.h> +#include <linux/module.h> +#include "amdgpu.h" +#include "amdgpu_psp.h" +#include "amdgpu_ucode.h" +#include "soc15_common.h" +#include "psp_v12_0.h" + +#include "mp/mp_12_0_0_offset.h" +#include "mp/mp_12_0_0_sh_mask.h" +#include "gc/gc_9_0_offset.h" +#include "sdma0/sdma0_4_0_offset.h" +#include "nbio/nbio_7_4_offset.h" + +#include "oss/osssys_4_0_offset.h" +#include "oss/osssys_4_0_sh_mask.h" + +MODULE_FIRMWARE("amdgpu/renoir_asd.bin"); +/* address block */ +#define smnMP1_FIRMWARE_FLAGS 0x3010024 + +static int psp_v12_0_init_microcode(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + const char *chip_name; + char fw_name[30]; + int err = 0; + const struct psp_firmware_header_v1_0 *asd_hdr; + + DRM_DEBUG("\n"); + + switch (adev->asic_type) { + case CHIP_RENOIR: + chip_name = "renoir"; + break; + default: + BUG(); + } + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); + err = request_firmware(&adev->psp.asd_fw, fw_name, adev->dev); + if (err) + goto out1; + + err = amdgpu_ucode_validate(adev->psp.asd_fw); + if (err) + goto out1; + + asd_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.asd_fw->data; + adev->psp.asd_fw_version = le32_to_cpu(asd_hdr->header.ucode_version); + adev->psp.asd_feature_version = le32_to_cpu(asd_hdr->ucode_feature_version); + adev->psp.asd_ucode_size = le32_to_cpu(asd_hdr->header.ucode_size_bytes); + adev->psp.asd_start_addr = (uint8_t *)asd_hdr + + le32_to_cpu(asd_hdr->header.ucode_array_offset_bytes); + + return 0; + +out1: + release_firmware(adev->psp.asd_fw); + adev->psp.asd_fw = NULL; + + return err; +} + +static int psp_v12_0_bootloader_load_sysdrv(struct psp_context *psp) +{ + int ret; + uint32_t psp_gfxdrv_command_reg = 0; + struct amdgpu_device *adev = psp->adev; + uint32_t sol_reg; + + /* Check sOS sign of life register to confirm sys driver and sOS + * are already been loaded. + */ + sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); + if (sol_reg) { + psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); + printk("sos fw version = 0x%x.\n", psp->sos_fw_version); + return 0; + } + + /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + memset(psp->fw_pri_buf, 0, PSP_1_MEG); + + /* Copy PSP System Driver binary to memory */ + memcpy(psp->fw_pri_buf, psp->sys_start_addr, psp->sys_bin_size); + + /* Provide the sys driver to bootloader */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, + (uint32_t)(psp->fw_pri_mc_addr >> 20)); + psp_gfxdrv_command_reg = 1 << 16; + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, + psp_gfxdrv_command_reg); + + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); + + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + + return ret; +} + +static int psp_v12_0_bootloader_load_sos(struct psp_context *psp) +{ + int ret; + unsigned int psp_gfxdrv_command_reg = 0; + struct amdgpu_device *adev = psp->adev; + uint32_t sol_reg; + + /* Check sOS sign of life register to confirm sys driver and sOS + * are already been loaded. + */ + sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); + if (sol_reg) + return 0; + + /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + memset(psp->fw_pri_buf, 0, PSP_1_MEG); + + /* Copy Secure OS binary to PSP memory */ + memcpy(psp->fw_pri_buf, psp->sos_start_addr, psp->sos_bin_size); + + /* Provide the PSP secure OS to bootloader */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, + (uint32_t)(psp->fw_pri_mc_addr >> 20)); + psp_gfxdrv_command_reg = 2 << 16; + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, + psp_gfxdrv_command_reg); + + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_81), + RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), + 0, true); + + return ret; +} + +static void psp_v12_0_reroute_ih(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + uint32_t tmp; + + /* Change IH ring for VMC */ + tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1244b); + tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, CLIENT_TYPE, 1); + tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1); + + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 3); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET); + + mdelay(20); + psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x8000FFFF, false); + + /* Change IH ring for UMC */ + tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1216b); + tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1); + + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 4); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET); + + mdelay(20); + psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x8000FFFF, false); +} + +static int psp_v12_0_ring_init(struct psp_context *psp, + enum psp_ring_type ring_type) +{ + int ret = 0; + struct psp_ring *ring; + struct amdgpu_device *adev = psp->adev; + + psp_v12_0_reroute_ih(psp); + + ring = &psp->km_ring; + + ring->ring_type = ring_type; + + /* allocate 4k Page of Local Frame Buffer memory for ring */ + ring->ring_size = 0x1000; + ret = amdgpu_bo_create_kernel(adev, ring->ring_size, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM, + &adev->firmware.rbuf, + &ring->ring_mem_mc_addr, + (void **)&ring->ring_mem); + if (ret) { + ring->ring_size = 0; + return ret; + } + + return 0; +} + +static bool psp_v12_0_support_vmr_ring(struct psp_context *psp) +{ + if (amdgpu_sriov_vf(psp->adev) && psp->sos_fw_version > 0x80045) + return true; + return false; +} + +static int psp_v12_0_ring_create(struct psp_context *psp, + enum psp_ring_type ring_type) +{ + int ret = 0; + unsigned int psp_ring_reg = 0; + struct psp_ring *ring = &psp->km_ring; + struct amdgpu_device *adev = psp->adev; + + if (psp_v12_0_support_vmr_ring(psp)) { + /* Write low address of the ring to C2PMSG_102 */ + psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, psp_ring_reg); + /* Write high address of the ring to C2PMSG_103 */ + psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_103, psp_ring_reg); + + /* Write the ring initialization command to C2PMSG_101 */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, + GFX_CTRL_CMD_ID_INIT_GPCOM_RING); + + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); + + /* Wait for response flag (bit 31) in C2PMSG_101 */ + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + 0x80000000, 0x8000FFFF, false); + + } else { + /* Write low address of the ring to C2PMSG_69 */ + psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, psp_ring_reg); + /* Write high address of the ring to C2PMSG_70 */ + psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, psp_ring_reg); + /* Write size of ring to C2PMSG_71 */ + psp_ring_reg = ring->ring_size; + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_71, psp_ring_reg); + /* Write the ring initialization command to C2PMSG_64 */ + psp_ring_reg = ring_type; + psp_ring_reg = psp_ring_reg << 16; + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg); + + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); + + /* Wait for response flag (bit 31) in C2PMSG_64 */ + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x8000FFFF, false); + } + + return ret; +} + +static int psp_v12_0_ring_stop(struct psp_context *psp, + enum psp_ring_type ring_type) +{ + int ret = 0; + struct amdgpu_device *adev = psp->adev; + + /* Write the ring destroy command*/ + if (psp_v12_0_support_vmr_ring(psp)) + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, + GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING); + else + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, + GFX_CTRL_CMD_ID_DESTROY_RINGS); + + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); + + /* Wait for response flag (bit 31) */ + if (psp_v12_0_support_vmr_ring(psp)) + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + 0x80000000, 0x80000000, false); + else + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x80000000, false); + + return ret; +} + +static int psp_v12_0_ring_destroy(struct psp_context *psp, + enum psp_ring_type ring_type) +{ + int ret = 0; + struct psp_ring *ring = &psp->km_ring; + struct amdgpu_device *adev = psp->adev; + + ret = psp_v12_0_ring_stop(psp, ring_type); + if (ret) + DRM_ERROR("Fail to stop psp ring\n"); + + amdgpu_bo_free_kernel(&adev->firmware.rbuf, + &ring->ring_mem_mc_addr, + (void **)&ring->ring_mem); + + return ret; +} + +static int psp_v12_0_cmd_submit(struct psp_context *psp, + uint64_t cmd_buf_mc_addr, uint64_t fence_mc_addr, + int index) +{ + unsigned int psp_write_ptr_reg = 0; + struct psp_gfx_rb_frame *write_frame = psp->km_ring.ring_mem; + struct psp_ring *ring = &psp->km_ring; + struct psp_gfx_rb_frame *ring_buffer_start = ring->ring_mem; + struct psp_gfx_rb_frame *ring_buffer_end = ring_buffer_start + + ring->ring_size / sizeof(struct psp_gfx_rb_frame) - 1; + struct amdgpu_device *adev = psp->adev; + uint32_t ring_size_dw = ring->ring_size / 4; + uint32_t rb_frame_size_dw = sizeof(struct psp_gfx_rb_frame) / 4; + + /* KM (GPCOM) prepare write pointer */ + if (psp_v12_0_support_vmr_ring(psp)) + psp_write_ptr_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102); + else + psp_write_ptr_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67); + + /* Update KM RB frame pointer to new frame */ + /* write_frame ptr increments by size of rb_frame in bytes */ + /* psp_write_ptr_reg increments by size of rb_frame in DWORDs */ + if ((psp_write_ptr_reg % ring_size_dw) == 0) + write_frame = ring_buffer_start; + else + write_frame = ring_buffer_start + (psp_write_ptr_reg / rb_frame_size_dw); + /* Check invalid write_frame ptr address */ + if ((write_frame < ring_buffer_start) || (ring_buffer_end < write_frame)) { + DRM_ERROR("ring_buffer_start = %p; ring_buffer_end = %p; write_frame = %p\n", + ring_buffer_start, ring_buffer_end, write_frame); + DRM_ERROR("write_frame is pointing to address out of bounds\n"); + return -EINVAL; + } + + /* Initialize KM RB frame */ + memset(write_frame, 0, sizeof(struct psp_gfx_rb_frame)); + + /* Update KM RB frame */ + write_frame->cmd_buf_addr_hi = upper_32_bits(cmd_buf_mc_addr); + write_frame->cmd_buf_addr_lo = lower_32_bits(cmd_buf_mc_addr); + write_frame->fence_addr_hi = upper_32_bits(fence_mc_addr); + write_frame->fence_addr_lo = lower_32_bits(fence_mc_addr); + write_frame->fence_value = index; + + /* Update the write Pointer in DWORDs */ + psp_write_ptr_reg = (psp_write_ptr_reg + rb_frame_size_dw) % ring_size_dw; + if (psp_v12_0_support_vmr_ring(psp)) { + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, psp_write_ptr_reg); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_CONSUME_CMD); + } else + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67, psp_write_ptr_reg); + + return 0; +} + +static int +psp_v12_0_sram_map(struct amdgpu_device *adev, + unsigned int *sram_offset, unsigned int *sram_addr_reg_offset, + unsigned int *sram_data_reg_offset, + enum AMDGPU_UCODE_ID ucode_id) +{ + int ret = 0; + + switch (ucode_id) { +/* TODO: needs to confirm */ +#if 0 + case AMDGPU_UCODE_ID_SMC: + *sram_offset = 0; + *sram_addr_reg_offset = 0; + *sram_data_reg_offset = 0; + break; +#endif + + case AMDGPU_UCODE_ID_CP_CE: + *sram_offset = 0x0; + *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_ADDR); + *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_DATA); + break; + + case AMDGPU_UCODE_ID_CP_PFP: + *sram_offset = 0x0; + *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_ADDR); + *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_DATA); + break; + + case AMDGPU_UCODE_ID_CP_ME: + *sram_offset = 0x0; + *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_ADDR); + *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_DATA); + break; + + case AMDGPU_UCODE_ID_CP_MEC1: + *sram_offset = 0x10000; + *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_ADDR); + *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_DATA); + break; + + case AMDGPU_UCODE_ID_CP_MEC2: + *sram_offset = 0x10000; + *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_ADDR); + *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_DATA); + break; + + case AMDGPU_UCODE_ID_RLC_G: + *sram_offset = 0x2000; + *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_ADDR); + *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_DATA); + break; + + case AMDGPU_UCODE_ID_SDMA0: + *sram_offset = 0x0; + *sram_addr_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_ADDR); + *sram_data_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_DATA); + break; + +/* TODO: needs to confirm */ +#if 0 + case AMDGPU_UCODE_ID_SDMA1: + *sram_offset = ; + *sram_addr_reg_offset = ; + break; + + case AMDGPU_UCODE_ID_UVD: + *sram_offset = ; + *sram_addr_reg_offset = ; + break; + + case AMDGPU_UCODE_ID_VCE: + *sram_offset = ; + *sram_addr_reg_offset = ; + break; +#endif + + case AMDGPU_UCODE_ID_MAXIMUM: + default: + ret = -EINVAL; + break; + } + + return ret; +} + +static bool psp_v12_0_compare_sram_data(struct psp_context *psp, + struct amdgpu_firmware_info *ucode, + enum AMDGPU_UCODE_ID ucode_type) +{ + int err = 0; + unsigned int fw_sram_reg_val = 0; + unsigned int fw_sram_addr_reg_offset = 0; + unsigned int fw_sram_data_reg_offset = 0; + unsigned int ucode_size; + uint32_t *ucode_mem = NULL; + struct amdgpu_device *adev = psp->adev; + + err = psp_v12_0_sram_map(adev, &fw_sram_reg_val, &fw_sram_addr_reg_offset, + &fw_sram_data_reg_offset, ucode_type); + if (err) + return false; + + WREG32(fw_sram_addr_reg_offset, fw_sram_reg_val); + + ucode_size = ucode->ucode_size; + ucode_mem = (uint32_t *)ucode->kaddr; + while (ucode_size) { + fw_sram_reg_val = RREG32(fw_sram_data_reg_offset); + + if (*ucode_mem != fw_sram_reg_val) + return false; + + ucode_mem++; + /* 4 bytes */ + ucode_size -= 4; + } + + return true; +} + +static int psp_v12_0_mode1_reset(struct psp_context *psp) +{ + int ret; + uint32_t offset; + struct amdgpu_device *adev = psp->adev; + + offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); + + ret = psp_wait_for(psp, offset, 0x80000000, 0x8000FFFF, false); + + if (ret) { + DRM_INFO("psp is not working correctly before mode1 reset!\n"); + return -EINVAL; + } + + /*send the mode 1 reset command*/ + WREG32(offset, GFX_CTRL_CMD_ID_MODE1_RST); + + msleep(500); + + offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); + + ret = psp_wait_for(psp, offset, 0x80000000, 0x80000000, false); + + if (ret) { + DRM_INFO("psp mode 1 reset failed!\n"); + return -EINVAL; + } + + DRM_INFO("psp mode1 reset succeed \n"); + + return 0; +} + +static const struct psp_funcs psp_v12_0_funcs = { + .init_microcode = psp_v12_0_init_microcode, + .bootloader_load_sysdrv = psp_v12_0_bootloader_load_sysdrv, + .bootloader_load_sos = psp_v12_0_bootloader_load_sos, + .ring_init = psp_v12_0_ring_init, + .ring_create = psp_v12_0_ring_create, + .ring_stop = psp_v12_0_ring_stop, + .ring_destroy = psp_v12_0_ring_destroy, + .cmd_submit = psp_v12_0_cmd_submit, + .compare_sram_data = psp_v12_0_compare_sram_data, + .mode1_reset = psp_v12_0_mode1_reset, +}; + +void psp_v12_0_set_psp_funcs(struct psp_context *psp) +{ + psp->funcs = &psp_v12_0_funcs; +} diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.h b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.h new file mode 100644 index 000000000000..241693ab1990 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.h @@ -0,0 +1,30 @@ +/* + * 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. + * + */ +#ifndef __PSP_V12_0_H__ +#define __PSP_V12_0_H__ + +#include "amdgpu_psp.h" + +void psp_v12_0_set_psp_funcs(struct psp_context *psp); + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index c2ebc0020e5d..d2c727f6a8bd 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -411,7 +411,6 @@ static int psp_v3_1_ring_destroy(struct psp_context *psp, } static int psp_v3_1_cmd_submit(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, uint64_t cmd_buf_mc_addr, uint64_t fence_mc_addr, int index) { diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index c04259182614..9b7e660828c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -68,6 +68,7 @@ MODULE_FIRMWARE("amdgpu/raven_sdma.bin"); MODULE_FIRMWARE("amdgpu/picasso_sdma.bin"); MODULE_FIRMWARE("amdgpu/raven2_sdma.bin"); MODULE_FIRMWARE("amdgpu/arcturus_sdma.bin"); +MODULE_FIRMWARE("amdgpu/renoir_sdma.bin"); #define SDMA0_POWER_CNTL__ON_OFF_CONDITION_HOLD_TIME_MASK 0x000000F8L #define SDMA0_POWER_CNTL__ON_OFF_STATUS_DURATION_TIME_MASK 0xFC000000L @@ -243,6 +244,18 @@ static const struct soc15_reg_golden golden_settings_sdma_arct[] = SOC15_REG_GOLDEN_VALUE(SDMA7, 0, mmSDMA7_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002) }; +static const struct soc15_reg_golden golden_settings_sdma_4_3[] = { + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_CHICKEN_BITS, 0xfe931f07, 0x02831f07), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_CLK_CTRL, 0xffffffff, 0x3f000100), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0018773f, 0x00000002), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00000002), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_POWER_CNTL, 0x003fff07, 0x40000051), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0), +}; + static u32 sdma_v4_0_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 offset) { @@ -367,6 +380,11 @@ static void sdma_v4_0_init_golden_registers(struct amdgpu_device *adev) golden_settings_sdma_rv1, ARRAY_SIZE(golden_settings_sdma_rv1)); break; + case CHIP_RENOIR: + soc15_program_register_sequence(adev, + golden_settings_sdma_4_3, + ARRAY_SIZE(golden_settings_sdma_4_3)); + break; default: break; } @@ -452,6 +470,9 @@ static int sdma_v4_0_init_microcode(struct amdgpu_device *adev) case CHIP_ARCTURUS: chip_name = "arcturus"; break; + case CHIP_RENOIR: + chip_name = "renoir"; + break; default: BUG(); } @@ -1640,7 +1661,7 @@ static int sdma_v4_0_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - if (adev->asic_type == CHIP_RAVEN) + if (adev->asic_type == CHIP_RAVEN || adev->asic_type == CHIP_RENOIR) adev->sdma.num_instances = 1; else if (adev->asic_type == CHIP_ARCTURUS) adev->sdma.num_instances = 8; @@ -2086,61 +2107,35 @@ static void sdma_v4_0_update_medium_grain_clock_gating( bool enable) { uint32_t data, def; + int i; if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) { - /* enable sdma0 clock gating */ - def = data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_CLK_CTRL)); - data &= ~(SDMA0_CLK_CTRL__SOFT_OVERRIDE7_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE6_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE5_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE4_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE3_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE2_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE1_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE0_MASK); - if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_CLK_CTRL), data); - - if (adev->sdma.num_instances > 1) { - def = data = RREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_CLK_CTRL)); - data &= ~(SDMA1_CLK_CTRL__SOFT_OVERRIDE7_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE6_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE5_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE4_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE3_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE2_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE1_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE0_MASK); + for (i = 0; i < adev->sdma.num_instances; i++) { + def = data = RREG32_SDMA(i, mmSDMA0_CLK_CTRL); + data &= ~(SDMA0_CLK_CTRL__SOFT_OVERRIDE7_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE6_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE5_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE4_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE3_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE2_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE1_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE0_MASK); if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_CLK_CTRL), data); + WREG32_SDMA(i, mmSDMA0_CLK_CTRL, data); } } else { - /* disable sdma0 clock gating */ - def = data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_CLK_CTRL)); - data |= (SDMA0_CLK_CTRL__SOFT_OVERRIDE7_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE6_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE5_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE4_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE3_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE2_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE1_MASK | - SDMA0_CLK_CTRL__SOFT_OVERRIDE0_MASK); - - if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_CLK_CTRL), data); - - if (adev->sdma.num_instances > 1) { - def = data = RREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_CLK_CTRL)); - data |= (SDMA1_CLK_CTRL__SOFT_OVERRIDE7_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE6_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE5_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE4_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE3_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE2_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE1_MASK | - SDMA1_CLK_CTRL__SOFT_OVERRIDE0_MASK); + for (i = 0; i < adev->sdma.num_instances; i++) { + def = data = RREG32_SDMA(i, mmSDMA0_CLK_CTRL); + data |= (SDMA0_CLK_CTRL__SOFT_OVERRIDE7_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE6_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE5_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE4_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE3_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE2_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE1_MASK | + SDMA0_CLK_CTRL__SOFT_OVERRIDE0_MASK); if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_CLK_CTRL), data); + WREG32_SDMA(i, mmSDMA0_CLK_CTRL, data); } } } @@ -2151,34 +2146,23 @@ static void sdma_v4_0_update_medium_grain_light_sleep( bool enable) { uint32_t data, def; + int i; if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) { - /* 1-not override: enable sdma0 mem light sleep */ - def = data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_POWER_CNTL)); - data |= SDMA0_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; - if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_POWER_CNTL), data); - - /* 1-not override: enable sdma1 mem light sleep */ - if (adev->sdma.num_instances > 1) { - def = data = RREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_POWER_CNTL)); - data |= SDMA1_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; + for (i = 0; i < adev->sdma.num_instances; i++) { + /* 1-not override: enable sdma mem light sleep */ + def = data = RREG32_SDMA(0, mmSDMA0_POWER_CNTL); + data |= SDMA0_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_POWER_CNTL), data); + WREG32_SDMA(0, mmSDMA0_POWER_CNTL, data); } } else { - /* 0-override:disable sdma0 mem light sleep */ - def = data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_POWER_CNTL)); - data &= ~SDMA0_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; - if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_POWER_CNTL), data); - - /* 0-override:disable sdma1 mem light sleep */ - if (adev->sdma.num_instances > 1) { - def = data = RREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_POWER_CNTL)); - data &= ~SDMA1_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; + for (i = 0; i < adev->sdma.num_instances; i++) { + /* 0-override:disable sdma mem light sleep */ + def = data = RREG32_SDMA(0, mmSDMA0_POWER_CNTL); + data &= ~SDMA0_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; if (def != data) - WREG32(SOC15_REG_OFFSET(SDMA1, 0, mmSDMA1_POWER_CNTL), data); + WREG32_SDMA(0, mmSDMA0_POWER_CNTL, data); } } } @@ -2196,6 +2180,8 @@ static int sdma_v4_0_set_clockgating_state(void *handle, case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: + case CHIP_ARCTURUS: + case CHIP_RENOIR: sdma_v4_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); sdma_v4_0_update_medium_grain_light_sleep(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 3e180152c5ad..fa2f70ce2e2b 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -1516,6 +1516,7 @@ static int sdma_v5_0_set_clockgating_state(void *handle, switch (adev->asic_type) { case CHIP_NAVI10: case CHIP_NAVI14: + case CHIP_NAVI12: sdma_v5_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); sdma_v5_0_update_medium_grain_light_sleep(adev, @@ -1627,7 +1628,8 @@ static const struct amdgpu_irq_src_funcs sdma_v5_0_illegal_inst_irq_funcs = { static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev) { - adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST; + adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_INSTANCE0 + + adev->sdma.num_instances; adev->sdma.trap_irq.funcs = &sdma_v5_0_trap_irq_funcs; adev->sdma.illegal_inst_irq.funcs = &sdma_v5_0_illegal_inst_irq_funcs; } diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 5116d0bf9e4a..53f21c567e97 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -63,6 +63,7 @@ #include "uvd_v7_0.h" #include "vce_v4_0.h" #include "vcn_v1_0.h" +#include "vcn_v2_0.h" #include "vcn_v2_5.h" #include "dce_virtual.h" #include "mxgpu_ai.h" @@ -508,6 +509,15 @@ static int soc15_asic_baco_reset(struct amdgpu_device *adev) return 0; } +static int soc15_mode2_reset(struct amdgpu_device *adev) +{ + if (!adev->powerplay.pp_funcs || + !adev->powerplay.pp_funcs->asic_reset_mode_2) + return -ENOENT; + + return adev->powerplay.pp_funcs->asic_reset_mode_2(adev->powerplay.pp_handle); +} + static enum amd_reset_method soc15_asic_reset_method(struct amdgpu_device *adev) { @@ -546,14 +556,14 @@ soc15_asic_reset_method(struct amdgpu_device *adev) static int soc15_asic_reset(struct amdgpu_device *adev) { - int ret; - - if (soc15_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) - ret = soc15_asic_baco_reset(adev); - else - ret = soc15_asic_mode1_reset(adev); - - return ret; + switch (soc15_asic_reset_method(adev)) { + case AMD_RESET_METHOD_BACO: + return soc15_asic_baco_reset(adev); + case AMD_RESET_METHOD_MODE2: + return soc15_mode2_reset(adev); + default: + return soc15_asic_mode1_reset(adev); + } } /*static int soc15_set_uvd_clock(struct amdgpu_device *adev, u32 clock, @@ -637,6 +647,7 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) case CHIP_VEGA10: case CHIP_VEGA12: case CHIP_RAVEN: + case CHIP_RENOIR: vega10_reg_base_init(adev); break; case CHIP_VEGA20: @@ -743,6 +754,20 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block); amdgpu_device_ip_block_add(adev, &vcn_v2_5_ip_block); break; + case CHIP_RENOIR: + amdgpu_device_ip_block_add(adev, &vega10_common_ip_block); + amdgpu_device_ip_block_add(adev, &gmc_v9_0_ip_block); + amdgpu_device_ip_block_add(adev, &vega10_ih_ip_block); + if (likely(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP)) + amdgpu_device_ip_block_add(adev, &psp_v12_0_ip_block); + if (is_support_sw_smu(adev)) + amdgpu_device_ip_block_add(adev, &smu_v12_0_ip_block); + amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block); + amdgpu_device_ip_block_add(adev, &sdma_v4_0_ip_block); + if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) + amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block); + amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block); + break; default: return -EINVAL; } @@ -1111,18 +1136,54 @@ static int soc15_common_early_init(void *handle) adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN; } - - if (adev->pm.pp_feature & PP_GFXOFF_MASK) - adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG | - AMD_PG_SUPPORT_CP | - AMD_PG_SUPPORT_RLC_SMU_HS; break; case CHIP_ARCTURUS: adev->asic_funcs = &vega20_asic_funcs; - adev->cg_flags = 0; + adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | + AMD_CG_SUPPORT_GFX_MGLS | + AMD_CG_SUPPORT_GFX_CGCG | + AMD_CG_SUPPORT_GFX_CGLS | + AMD_CG_SUPPORT_GFX_CP_LS | + AMD_CG_SUPPORT_HDP_MGCG | + AMD_CG_SUPPORT_HDP_LS | + AMD_CG_SUPPORT_SDMA_MGCG | + AMD_CG_SUPPORT_SDMA_LS | + AMD_CG_SUPPORT_MC_MGCG | + AMD_CG_SUPPORT_MC_LS; adev->pg_flags = 0; adev->external_rev_id = adev->rev_id + 0x32; break; + case CHIP_RENOIR: + adev->asic_funcs = &soc15_asic_funcs; + adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | + AMD_CG_SUPPORT_GFX_MGLS | + AMD_CG_SUPPORT_GFX_3D_CGCG | + AMD_CG_SUPPORT_GFX_3D_CGLS | + AMD_CG_SUPPORT_GFX_CGCG | + AMD_CG_SUPPORT_GFX_CGLS | + AMD_CG_SUPPORT_GFX_CP_LS | + AMD_CG_SUPPORT_MC_MGCG | + AMD_CG_SUPPORT_MC_LS | + AMD_CG_SUPPORT_SDMA_MGCG | + AMD_CG_SUPPORT_SDMA_LS | + AMD_CG_SUPPORT_BIF_LS | + AMD_CG_SUPPORT_HDP_LS | + AMD_CG_SUPPORT_ROM_MGCG | + AMD_CG_SUPPORT_VCN_MGCG | + AMD_CG_SUPPORT_IH_CG | + AMD_CG_SUPPORT_ATHUB_LS | + AMD_CG_SUPPORT_ATHUB_MGCG | + AMD_CG_SUPPORT_DF_MGCG; + adev->pg_flags = AMD_PG_SUPPORT_SDMA | + AMD_PG_SUPPORT_VCN | + AMD_PG_SUPPORT_VCN_DPG; + adev->external_rev_id = adev->rev_id + 0x91; + + if (adev->pm.pp_feature & PP_GFXOFF_MASK) + adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG | + AMD_PG_SUPPORT_CP | + AMD_PG_SUPPORT_RLC_SMU_HS; + break; default: /* FIXME: not supported yet */ return -EINVAL; @@ -1256,7 +1317,8 @@ static void soc15_update_hdp_light_sleep(struct amdgpu_device *adev, bool enable { uint32_t def, data; - if (adev->asic_type == CHIP_VEGA20) { + if (adev->asic_type == CHIP_VEGA20 || + adev->asic_type == CHIP_ARCTURUS) { def = data = RREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_CTRL)); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS)) @@ -1375,6 +1437,7 @@ static int soc15_common_set_clockgating_state(void *handle, state == AMD_CG_STATE_GATE ? true : false); break; case CHIP_RAVEN: + case CHIP_RENOIR: adev->nbio_funcs->update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); adev->nbio_funcs->update_medium_grain_light_sleep(adev, @@ -1388,6 +1451,10 @@ static int soc15_common_set_clockgating_state(void *handle, soc15_update_rom_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE ? true : false); break; + case CHIP_ARCTURUS: + soc15_update_hdp_light_sleep(adev, + state == AMD_CG_STATE_GATE ? true : false); + break; default: break; } diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index c1c0a39ae269..4fac1e91a054 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -247,7 +247,7 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev) WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl); } - if (adev->asic_type == CHIP_ARCTURUS && + if ((adev->asic_type == CHIP_ARCTURUS || adev->asic_type == CHIP_RENOIR) && adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { if (adev->irq.ih.use_bus_addr) { ih_chicken = RREG32_SOC15(OSSSYS, 0, mmIH_CHICKEN); diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c index a8e92638a2e8..bd0580334f83 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c @@ -81,6 +81,10 @@ void vega10_doorbell_index_init(struct amdgpu_device *adev) adev->doorbell_index.uvd_vce.vce_ring2_3 = AMDGPU_DOORBELL64_VCE_RING2_3; adev->doorbell_index.uvd_vce.vce_ring4_5 = AMDGPU_DOORBELL64_VCE_RING4_5; adev->doorbell_index.uvd_vce.vce_ring6_7 = AMDGPU_DOORBELL64_VCE_RING6_7; + adev->doorbell_index.vcn.vcn_ring0_1 = AMDGPU_DOORBELL64_VCN0_1; + adev->doorbell_index.vcn.vcn_ring2_3 = AMDGPU_DOORBELL64_VCN2_3; + adev->doorbell_index.vcn.vcn_ring4_5 = AMDGPU_DOORBELL64_VCN4_5; + adev->doorbell_index.vcn.vcn_ring6_7 = AMDGPU_DOORBELL64_VCN6_7; adev->doorbell_index.first_non_cp = AMDGPU_DOORBELL64_FIRST_NON_CP; adev->doorbell_index.last_non_cp = AMDGPU_DOORBELL64_LAST_NON_CP; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 3b9fe629a126..0dc1084b5e82 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -42,6 +42,7 @@ static atomic_t kfd_locked = ATOMIC_INIT(0); #ifdef KFD_SUPPORT_IOMMU_V2 static const struct kfd_device_info kaveri_device_info = { .asic_family = CHIP_KAVERI, + .asic_name = "kaveri", .max_pasid_bits = 16, /* max num of queues for KV.TODO should be a dynamic value */ .max_no_of_hqd = 24, @@ -60,6 +61,7 @@ static const struct kfd_device_info kaveri_device_info = { static const struct kfd_device_info carrizo_device_info = { .asic_family = CHIP_CARRIZO, + .asic_name = "carrizo", .max_pasid_bits = 16, /* max num of queues for CZ.TODO should be a dynamic value */ .max_no_of_hqd = 24, @@ -78,6 +80,7 @@ static const struct kfd_device_info carrizo_device_info = { static const struct kfd_device_info raven_device_info = { .asic_family = CHIP_RAVEN, + .asic_name = "raven", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 8, @@ -96,6 +99,7 @@ static const struct kfd_device_info raven_device_info = { static const struct kfd_device_info hawaii_device_info = { .asic_family = CHIP_HAWAII, + .asic_name = "hawaii", .max_pasid_bits = 16, /* max num of queues for KV.TODO should be a dynamic value */ .max_no_of_hqd = 24, @@ -114,6 +118,7 @@ static const struct kfd_device_info hawaii_device_info = { static const struct kfd_device_info tonga_device_info = { .asic_family = CHIP_TONGA, + .asic_name = "tonga", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -131,6 +136,7 @@ static const struct kfd_device_info tonga_device_info = { static const struct kfd_device_info fiji_device_info = { .asic_family = CHIP_FIJI, + .asic_name = "fiji", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -148,6 +154,7 @@ static const struct kfd_device_info fiji_device_info = { static const struct kfd_device_info fiji_vf_device_info = { .asic_family = CHIP_FIJI, + .asic_name = "fiji", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -166,6 +173,7 @@ static const struct kfd_device_info fiji_vf_device_info = { static const struct kfd_device_info polaris10_device_info = { .asic_family = CHIP_POLARIS10, + .asic_name = "polaris10", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -183,6 +191,7 @@ static const struct kfd_device_info polaris10_device_info = { static const struct kfd_device_info polaris10_vf_device_info = { .asic_family = CHIP_POLARIS10, + .asic_name = "polaris10", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -200,6 +209,7 @@ static const struct kfd_device_info polaris10_vf_device_info = { static const struct kfd_device_info polaris11_device_info = { .asic_family = CHIP_POLARIS11, + .asic_name = "polaris11", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -217,6 +227,7 @@ static const struct kfd_device_info polaris11_device_info = { static const struct kfd_device_info polaris12_device_info = { .asic_family = CHIP_POLARIS12, + .asic_name = "polaris12", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -234,6 +245,7 @@ static const struct kfd_device_info polaris12_device_info = { static const struct kfd_device_info vegam_device_info = { .asic_family = CHIP_VEGAM, + .asic_name = "vegam", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 4, @@ -251,6 +263,7 @@ static const struct kfd_device_info vegam_device_info = { static const struct kfd_device_info vega10_device_info = { .asic_family = CHIP_VEGA10, + .asic_name = "vega10", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 8, @@ -268,6 +281,7 @@ static const struct kfd_device_info vega10_device_info = { static const struct kfd_device_info vega10_vf_device_info = { .asic_family = CHIP_VEGA10, + .asic_name = "vega10", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 8, @@ -285,6 +299,7 @@ static const struct kfd_device_info vega10_vf_device_info = { static const struct kfd_device_info vega12_device_info = { .asic_family = CHIP_VEGA12, + .asic_name = "vega12", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 8, @@ -302,6 +317,7 @@ static const struct kfd_device_info vega12_device_info = { static const struct kfd_device_info vega20_device_info = { .asic_family = CHIP_VEGA20, + .asic_name = "vega20", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 8, @@ -319,6 +335,7 @@ static const struct kfd_device_info vega20_device_info = { static const struct kfd_device_info arcturus_device_info = { .asic_family = CHIP_ARCTURUS, + .asic_name = "arcturus", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 8, @@ -336,6 +353,7 @@ static const struct kfd_device_info arcturus_device_info = { static const struct kfd_device_info navi10_device_info = { .asic_family = CHIP_NAVI10, + .asic_name = "navi10", .max_pasid_bits = 16, .max_no_of_hqd = 24, .doorbell_size = 8, @@ -472,6 +490,7 @@ static const struct kfd_deviceid supported_devices[] = { { 0x738C, &arcturus_device_info }, /* Arcturus */ { 0x7388, &arcturus_device_info }, /* Arcturus */ { 0x738E, &arcturus_device_info }, /* Arcturus */ + { 0x7390, &arcturus_device_info }, /* Arcturus vf */ { 0x7310, &navi10_device_info }, /* Navi10 */ { 0x7312, &navi10_device_info }, /* Navi10 */ { 0x7318, &navi10_device_info }, /* Navi10 */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index fe1ce348fdcd..d985e31fcc1e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -1814,7 +1814,8 @@ out_free: return NULL; } -void deallocate_hiq_sdma_mqd(struct kfd_dev *dev, struct kfd_mem_obj *mqd) +static void deallocate_hiq_sdma_mqd(struct kfd_dev *dev, + struct kfd_mem_obj *mqd) { WARN(!mqd, "No hiq sdma mqd trunk to free"); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 3933fb6a371e..3bb75d11a662 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -195,6 +195,7 @@ struct kfd_event_interrupt_class { struct kfd_device_info { enum amd_asic_type asic_family; + const char *asic_name; const struct kfd_event_interrupt_class *event_interrupt_class; unsigned int max_pasid_bits; unsigned int max_no_of_hqd; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 8f1076c0c88a..0c6ac043ae3c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -801,6 +801,8 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, return ret; } + amdgpu_vm_set_task_info(pdd->vm); + ret = kfd_process_device_reserve_ib_mem(pdd); if (ret) goto err_reserve_ib_mem; @@ -1042,7 +1044,6 @@ static void restore_process_worker(struct work_struct *work) { struct delayed_work *dwork; struct kfd_process *p; - struct kfd_process_device *pdd; int ret = 0; dwork = to_delayed_work(work); @@ -1051,16 +1052,6 @@ static void restore_process_worker(struct work_struct *work) * lifetime of this thread, kfd_process p will be valid */ p = container_of(dwork, struct kfd_process, restore_work); - - /* Call restore_process_bos on the first KGD device. This function - * takes care of restoring the whole process including other devices. - * Restore can fail if enough memory is not available. If so, - * reschedule again. - */ - pdd = list_first_entry(&p->per_device_data, - struct kfd_process_device, - per_device_list); - pr_debug("Started restoring pasid %d\n", p->pasid); /* Setting last_restore_timestamp before successful restoration. diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 36fa98fe858b..7551761f2aa9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -406,8 +406,6 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, char *buffer) { struct kfd_topology_device *dev; - char public_name[KFD_TOPOLOGY_PUBLIC_NAME_SIZE]; - uint32_t i; uint32_t log_max_watch_addr; /* Making sure that the buffer is an empty string */ @@ -422,14 +420,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, if (strcmp(attr->name, "name") == 0) { dev = container_of(attr, struct kfd_topology_device, attr_name); - for (i = 0; i < KFD_TOPOLOGY_PUBLIC_NAME_SIZE; i++) { - public_name[i] = - (char)dev->node_props.marketing_name[i]; - if (dev->node_props.marketing_name[i] == 0) - break; - } - public_name[KFD_TOPOLOGY_PUBLIC_NAME_SIZE-1] = 0x0; - return sysfs_show_str_val(buffer, public_name); + + return sysfs_show_str_val(buffer, dev->node_props.name); } dev = container_of(attr, struct kfd_topology_device, @@ -1274,6 +1266,10 @@ int kfd_topology_add_device(struct kfd_dev *gpu) */ amdgpu_amdkfd_get_cu_info(dev->gpu->kgd, &cu_info); + + strncpy(dev->node_props.name, gpu->device_info->asic_name, + KFD_TOPOLOGY_PUBLIC_NAME_SIZE); + dev->node_props.simd_arrays_per_engine = cu_info.num_shader_arrays_per_engine; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h index 276354aa0fcc..d4718d58d0f2 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h @@ -27,7 +27,7 @@ #include <linux/list.h> #include "kfd_crat.h" -#define KFD_TOPOLOGY_PUBLIC_NAME_SIZE 128 +#define KFD_TOPOLOGY_PUBLIC_NAME_SIZE 32 #define HSA_CAP_HOT_PLUGGABLE 0x00000001 #define HSA_CAP_ATS_PRESENT 0x00000002 @@ -81,7 +81,7 @@ struct kfd_node_properties { int32_t drm_render_minor; uint32_t num_sdma_engines; uint32_t num_sdma_xgmi_engines; - uint16_t marketing_name[KFD_TOPOLOGY_PUBLIC_NAME_SIZE]; + char name[KFD_TOPOLOGY_PUBLIC_NAME_SIZE]; }; #define HSA_MEM_HEAP_TYPE_SYSTEM 0 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 67f8aee4cd1b..0a71ed1e7762 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -694,6 +694,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) 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 @@ -3006,6 +3009,8 @@ 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) @@ -3071,6 +3076,7 @@ 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 @@ -3142,13 +3148,25 @@ 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); } @@ -3502,6 +3520,10 @@ 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) { @@ -3574,17 +3596,23 @@ 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); @@ -6003,11 +6031,9 @@ static void amdgpu_dm_enable_crtc_interrupts(struct drm_device *dev, /* The stream has changed so CRC capture needs to re-enabled. */ source = dm_new_crtc_state->crc_src; if (amdgpu_dm_is_valid_crc_source(source)) { - dm_new_crtc_state->crc_src = AMDGPU_DM_PIPE_CRC_SOURCE_NONE; - if (source == AMDGPU_DM_PIPE_CRC_SOURCE_CRTC) - amdgpu_dm_crtc_set_crc_source(crtc, "crtc"); - else if (source == AMDGPU_DM_PIPE_CRC_SOURCE_DPRX) - amdgpu_dm_crtc_set_crc_source(crtc, "dprx"); + amdgpu_dm_crtc_configure_crc_source( + crtc, dm_new_crtc_state, + dm_new_crtc_state->crc_src); } #endif } @@ -6058,23 +6084,8 @@ 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 (amdgpu_dm_is_valid_crc_source(dm_new_crtc_state->crc_src)) - 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_src = AMDGPU_DM_PIPE_CRC_SOURCE_NONE; - + 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 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 365aaef3ecaf..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 @@ -97,17 +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; - struct amdgpu_dm_connector *aconn; + 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; - - enum amdgpu_dm_pipe_crc_source source = dm_parse_crc_source(src_name); + int ret = 0; if (source < 0) { DRM_DEBUG_DRIVER("Unknown CRC source %s for CRTC%d\n", @@ -115,14 +150,34 @@ 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; + + 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); + + 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; } enable = amdgpu_dm_is_valid_crc_source(source); + crtc_state = to_dm_crtc_state(crtc->state); - mutex_lock(&adev->dm.dc_lock); /* * USER REQ SRC | CURRENT SRC | BEHAVIOR * ----------------------------- @@ -137,38 +192,41 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) * 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))) { - aconn = stream_state->link->priv; + (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); - mutex_unlock(&adev->dm.dc_lock); - return -EINVAL; + ret = -EINVAL; + goto cleanup; } aux = &aconn->dm_dp_aux.aux; if (!aux) { DRM_DEBUG_DRIVER("No dp aux for amd connector\n"); - mutex_unlock(&adev->dm.dc_lock); - return -EINVAL; - } - } else if (dm_is_crc_source_crtc(source)) { - if (!dc_stream_configure_crc(stream_state->ctx->dc, stream_state, - enable, enable)) { - mutex_unlock(&adev->dm.dc_lock); - return -EINVAL; + ret = -EINVAL; + goto cleanup; } } - /* configure dithering */ - if (!dm_need_crc_dither(source)) - dc_stream_set_dither_option(stream_state, DITHER_OPTION_TRUN8); - else if (!dm_need_crc_dither(crtc_state->crc_src)) - dc_stream_set_dither_option(stream_state, DITHER_OPTION_DEFAULT); - - mutex_unlock(&adev->dm.dc_lock); + 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 @@ -176,11 +234,15 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) */ enabled = amdgpu_dm_is_valid_crc_source(crtc_state->crc_src); if (!enabled && enable) { - drm_crtc_vblank_get(crtc); + 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"); - return -EINVAL; + ret = -EINVAL; + goto cleanup; } } } else if (enabled && !enable) { @@ -188,7 +250,8 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) if (dm_is_crc_source_dprx(source)) { if (drm_dp_stop_crc(aux)) { DRM_DEBUG_DRIVER("dp stop crc failed\n"); - return -EINVAL; + ret = -EINVAL; + goto cleanup; } } } @@ -197,7 +260,14 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) /* 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; } /** 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 index 14de7301c28d..f7d731797d3f 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h @@ -26,6 +26,9 @@ #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, @@ -44,6 +47,9 @@ static inline bool amdgpu_dm_is_valid_crc_source(enum amdgpu_dm_pipe_crc_source /* 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, 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/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/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/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c index 9f12e21f8b9b..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" @@ -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 061c6e3a3088..383f4f8db8f4 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -705,6 +705,13 @@ 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) +{ + /* we are ok with all levels */ + return 4; +} + bool dcn_validate_bandwidth( struct dc *dc, struct dc_state *context, @@ -732,6 +739,7 @@ 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 +1276,7 @@ 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/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/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 7ff0396956b3..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 @@ -104,7 +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; @@ -114,28 +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, @@ -146,12 +192,14 @@ 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) { @@ -177,6 +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); @@ -202,10 +251,12 @@ 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; @@ -213,35 +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 || (force_reset && safe_to_lower)) { - 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 && diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 252b621d93a9..87ca5a290d12 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" @@ -290,7 +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; } @@ -959,7 +962,7 @@ 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)) @@ -971,13 +974,22 @@ 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; - tg = dc->res_pool->timing_generators[inst]; + 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[tg_inst]; if (!tg->funcs->is_matching_timing) return false; @@ -990,10 +1002,11 @@ 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; @@ -1183,8 +1196,8 @@ 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; @@ -1204,11 +1217,11 @@ 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]; @@ -1219,6 +1232,12 @@ 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++) { @@ -1242,7 +1261,7 @@ 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) @@ -1602,6 +1621,9 @@ 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; } @@ -1678,6 +1700,8 @@ 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 && @@ -1844,9 +1868,7 @@ 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) @@ -1872,7 +1894,7 @@ 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); @@ -1880,10 +1902,12 @@ 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 } @@ -1900,13 +1924,21 @@ 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); } @@ -1996,6 +2028,7 @@ 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; @@ -2110,7 +2143,7 @@ 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; @@ -2148,16 +2181,6 @@ 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); @@ -2247,6 +2270,14 @@ 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); 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 193d6f14e684..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) { @@ -1387,57 +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; @@ -1472,6 +1461,16 @@ static enum dc_status enable_link_dp( #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); @@ -1493,7 +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, @@ -1508,7 +1508,7 @@ 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; @@ -2767,21 +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 + /* 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_enable(pipe_ctx, true); - pipe_ctx->stream_res.tg->funcs->wait_for_state( - pipe_ctx->stream_res.tg, - CRTC_STATE_VBLANK); + 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); @@ -2798,7 +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; @@ -2833,7 +2839,7 @@ 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 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 08bd9c96b9b0..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 @@ -965,6 +965,7 @@ static inline enum link_training_result perform_link_training_int( static void initialize_training_settings( struct dc_link *link, const struct dc_link_settings *link_setting, + const struct dc_link_training_overrides *overrides, struct link_training_settings *lt_settings) { uint32_t lane; @@ -997,23 +998,23 @@ static void initialize_training_settings( /* Initialize link spread */ if (link->dp_ss_off) lt_settings->link_settings.link_spread = LINK_SPREAD_DISABLED; - else if (link->preferred_training_settings.downspread != NULL) + else if (overrides->downspread != NULL) lt_settings->link_settings.link_spread - = *link->preferred_training_settings.downspread + = *overrides->downspread ? LINK_SPREAD_05_DOWNSPREAD_30KHZ : LINK_SPREAD_DISABLED; else lt_settings->link_settings.link_spread = LINK_SPREAD_05_DOWNSPREAD_30KHZ; /* Initialize lane settings overrides */ - if (link->preferred_training_settings.voltage_swing != NULL) - lt_settings->voltage_swing = link->preferred_training_settings.voltage_swing; + if (overrides->voltage_swing != NULL) + lt_settings->voltage_swing = overrides->voltage_swing; - if (link->preferred_training_settings.pre_emphasis != NULL) - lt_settings->pre_emphasis = link->preferred_training_settings.pre_emphasis; + if (overrides->pre_emphasis != NULL) + lt_settings->pre_emphasis = overrides->pre_emphasis; - if (link->preferred_training_settings.post_cursor2 != NULL) - lt_settings->post_cursor2 = link->preferred_training_settings.post_cursor2; + 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++) { @@ -1032,23 +1033,23 @@ static void initialize_training_settings( } /* Initialize training timings */ - if (link->preferred_training_settings.cr_pattern_time != NULL) - lt_settings->cr_pattern_time = *link->preferred_training_settings.cr_pattern_time; + if (overrides->cr_pattern_time != NULL) + lt_settings->cr_pattern_time = *overrides->cr_pattern_time; else - lt_settings->cr_pattern_time = 100; + lt_settings->cr_pattern_time = get_training_aux_rd_interval(link, 100); - if (link->preferred_training_settings.eq_pattern_time != NULL) - lt_settings->eq_pattern_time = *link->preferred_training_settings.eq_pattern_time; + 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 (link->preferred_training_settings.pattern_for_eq != NULL) - lt_settings->pattern_for_eq = *link->preferred_training_settings.pattern_for_eq; + 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 (link->preferred_training_settings.enhanced_framing != NULL) - lt_settings->enhanced_framing = *link->preferred_training_settings.enhanced_framing; + if (overrides->enhanced_framing != NULL) + lt_settings->enhanced_framing = *overrides->enhanced_framing; else lt_settings->enhanced_framing = 1; } @@ -1139,7 +1140,11 @@ bool dc_link_dp_perform_link_training_skip_aux( 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, <_settings); + initialize_training_settings( + link, + link_setting, + &link->preferred_training_settings, + <_settings); /* 1. Perform_clock_recovery_sequence. */ @@ -1184,7 +1189,11 @@ enum link_training_result dc_link_dp_perform_link_training( bool fec_enable; #endif - initialize_training_settings(link, link_setting, <_settings); + 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); @@ -1247,6 +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 */ @@ -1401,7 +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; @@ -1425,17 +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. @@ -1505,6 +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) @@ -2307,6 +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; @@ -2514,13 +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; @@ -2925,14 +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; - if (bot_odm_pipe) { - struct output_pixel_processor *bot_opp = bot_odm_pipe->stream_res.opp; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + 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, + 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; + + 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, @@ -2961,14 +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; - if (bot_odm_pipe) { - struct output_pixel_processor *bot_opp = bot_odm_pipe->stream_res.opp; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + 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, + 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; + + 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, @@ -3009,7 +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; } @@ -3158,6 +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) { 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 878f47b59d5a..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 @@ -55,6 +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)); } @@ -245,46 +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, @@ -312,7 +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); @@ -326,7 +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, @@ -379,10 +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) @@ -398,55 +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 */ -void 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 (dc_is_dp_signal(stream->signal) && !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)) { + 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, @@ -458,14 +442,18 @@ void set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) OPTC_DSC_DISABLED, 0, 0); /* disable DSC in stream encoder */ - if (dc_is_dp_signal(stream->signal) && !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); } } @@ -482,18 +470,59 @@ bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable) if (enable) { if (dp_set_dsc_on_rx(pipe_ctx, true)) { - set_dsc_on_stream(pipe_ctx, true); + dp_set_dsc_on_stream(pipe_ctx, true); result = true; } } else { dp_set_dsc_on_rx(pipe_ctx, false); - set_dsc_on_stream(pipe_ctx, false); + dp_set_dsc_on_stream(pipe_ctx, false); result = true; } 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; @@ -503,9 +532,9 @@ bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx) if (!dsc) return false; - set_dsc_on_stream(pipe_ctx, true); + 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 c227b86420a0..5fb0c3d672e4 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -172,13 +172,11 @@ struct resource_pool *dc_create_resource_pool(struct dc *dc, 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 +263,10 @@ 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++; } @@ -1119,25 +1115,21 @@ 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; @@ -1153,31 +1145,20 @@ 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]; } @@ -1191,8 +1172,7 @@ 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) @@ -1206,7 +1186,7 @@ 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) @@ -1267,39 +1247,41 @@ 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; @@ -1308,35 +1290,6 @@ 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, @@ -1363,12 +1316,6 @@ 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; @@ -1383,13 +1330,10 @@ 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)); - } } } @@ -1675,24 +1619,25 @@ 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]; } @@ -1752,50 +1697,45 @@ enum dc_status dc_remove_stream_from_ctx( { int i; struct dc_context *dc_ctx = dc->ctx; - struct pipe_ctx *del_pipe = NULL; - - /* 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]); + struct pipe_ctx *del_pipe = resource_get_head_pipe_for_stream(&new_ctx->res_ctx, stream); + struct pipe_ctx *odm_pipe; - del_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; + } - 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); + odm_pipe = del_pipe->next_odm_pipe; - if (del_pipe->stream_res.audio) - update_audio_usage( - &new_ctx->res_ctx, - dc->res_pool, - del_pipe->stream_res.audio, - 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); - resource_unreference_clock_source(&new_ctx->res_ctx, - dc->res_pool, - del_pipe->clock_source); + if (del_pipe->stream_res.audio) + update_audio_usage( + &new_ctx->res_ctx, + dc->res_pool, + del_pipe->stream_res.audio, + false); - if (dc->res_pool->funcs->remove_stream_from_ctx) - dc->res_pool->funcs->remove_stream_from_ctx(dc, new_ctx, stream); + resource_unreference_clock_source(&new_ctx->res_ctx, + dc->res_pool, + del_pipe->clock_source); - memset(del_pipe, 0, sizeof(*del_pipe)); - if (odm_pipe) - memset(odm_pipe, 0, sizeof(*odm_pipe)); + if (dc->res_pool->funcs->remove_stream_from_ctx) + dc->res_pool->funcs->remove_stream_from_ctx(dc, new_ctx, stream); - break; - } - } + while (odm_pipe) { + struct pipe_ctx *next_odm_pipe = odm_pipe->next_odm_pipe; - 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) @@ -1896,7 +1836,7 @@ 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)) @@ -1908,28 +1848,37 @@ 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]); + + 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]; - 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 (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; @@ -2491,6 +2440,12 @@ 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/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index c585e16bc9f1..42b6a6e41c0b 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 "3.2.42" +#define DC_VER "3.2.48" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -121,7 +121,7 @@ struct dc_caps { struct dc_bug_wa { bool no_connect_phy_config; bool dedcn20_305_wa; - struct display_mode_lib alternate_dml; + bool skip_clock_update; }; #endif @@ -220,7 +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 { @@ -423,6 +423,7 @@ struct dc_phy_addr_space_config { } gart_config; bool valid; + uint64_t page_table_default_page_addr; }; struct dc_virtual_addr_space_config { @@ -614,9 +615,12 @@ 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; @@ -747,6 +751,7 @@ 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 */ @@ -776,6 +781,7 @@ struct dc_plane_info { bool global_alpha; int global_alpha_value; bool input_csc_enabled; + int layer_index; }; struct dc_scaling_info { 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 efa7a47f6b7e..ef79a686e4c2 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -128,7 +128,10 @@ struct dc_link_training_overrides { 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 { 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 929c4eadc1dc..0b8700a8a94a 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -516,7 +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; }; @@ -756,6 +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 d6ff5af70c71..9ea75db3484e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -84,6 +84,7 @@ struct dc_link { 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 @@ -228,6 +229,15 @@ enum link_training_result dc_link_dp_perform_link_training( 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); 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 6147530144eb..5a35495bc11d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c @@ -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: @@ -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_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_hwseq.h b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h index 245b80b92681..7d93babaa2fb 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), \ @@ -401,36 +416,34 @@ 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 +451,15 @@ 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 +522,7 @@ 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)\ @@ -576,8 +583,7 @@ 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), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #endif #define HWSEQ_REG_FIELD_LIST(type) \ @@ -612,9 +618,9 @@ 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; \ 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_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c index 84bbff665be9..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 @@ -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..afc61055eca1 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); 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 9c50f09233bb..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 @@ -731,7 +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) { @@ -799,7 +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) { @@ -881,7 +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) { @@ -981,7 +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; @@ -1004,24 +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); @@ -1034,21 +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, @@ -1338,7 +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) { @@ -1406,7 +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, @@ -1417,6 +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 @@ -1726,7 +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}; @@ -1910,8 +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)) { @@ -2076,7 +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( @@ -2755,9 +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_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c index 764329264c3b..c66fe170e1e8 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); 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..3ac4c7e73050 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) }; 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..7d08154e9662 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) }; 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..c8ae3023fda2 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 +119,8 @@ 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; }; /* set field name */ @@ -196,7 +198,9 @@ 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;\ @@ -247,7 +251,7 @@ 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 03f5aa10c4c4..001db49e4bb2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -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; 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 344e446e337d..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; \ @@ -417,8 +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),\ @@ -593,18 +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;\ @@ -722,7 +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( 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 e720be6be369..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); @@ -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 : @@ -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; @@ -2789,14 +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) @@ -2827,6 +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); } @@ -2838,7 +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) @@ -3121,7 +3307,7 @@ static void dcn10_get_clock(struct dc *dc, } 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, @@ -3154,18 +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, @@ -3175,6 +3361,16 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .setup_vupdate_interrupt = dcn10_setup_vupdate_interrupt, .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_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 4f7a10390c57..8b2f29f6dabd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c @@ -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..5a89e462e7cc 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); 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 89e6a4c34018..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 @@ -1004,6 +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 @@ -1529,6 +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, @@ -1563,6 +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 f585e7b620cc..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 @@ -598,6 +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); @@ -615,4 +618,7 @@ void get_audio_clock_info( 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/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_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c index e870caa8d4fa..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; 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 31d6e79ba2b8..b83c022e2c6f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c @@ -380,6 +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; @@ -398,54 +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)*/ + + /*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_BOT, - SDPIF_AGP_BOT, 0X03FFFF); + /*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); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_TOP, - SDPIF_AGP_TOP, 0); + /*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; @@ -582,7 +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..0d0caa6de935 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h @@ -32,10 +32,8 @@ #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 +42,14 @@ 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 +59,9 @@ 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 487de87b03eb..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) { @@ -472,7 +397,7 @@ void hubp2_program_rotation( } void hubp2_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; @@ -1020,6 +945,8 @@ void hubp2_cursor_set_position( 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; @@ -1033,10 +960,16 @@ void hubp2_cursor_set_position( 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) { - src_x_offset = pos->y - pos->y_hotspot - param->viewport.x; - y_hotspot = pos->x_hotspot; - x_hotspot = pos->y_hotspot; + 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) { @@ -1058,13 +991,13 @@ void hubp2_cursor_set_position( if (src_x_offset >= (int)param->viewport.width) cur_en = 0; /* not visible beyond right edge*/ - if (src_x_offset + (int)hubp->curs_attr.width <= 0) + 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 + (int)hubp->curs_attr.height <= 0) + if (src_y_offset + cursor_height <= 0) cur_en = 0; /* not visible beyond top edge*/ if (cur_en && REG_READ(CURSOR_SURFACE_ADDRESS) == 0) @@ -1321,7 +1254,6 @@ static struct hubp_funcs dcn20_hubp_funcs = { .hubp_set_vm_system_aperture_settings = hubp2_set_vm_system_aperture_settings, .set_blank = hubp2_set_blank, .dcc_control = hubp2_dcc_control, - .hubp_update_dchub = hubp2_update_dchub, .mem_program_viewport = min_set_viewport, .set_cursor_attributes = hubp2_cursor_set_attributes, .set_cursor_position = hubp2_cursor_set_position, 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 1c53af4811e8..c4ed8f1b9424 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), \ @@ -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), \ @@ -222,10 +210,6 @@ 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); @@ -268,7 +252,7 @@ bool hubp2_program_surface_flip_and_addr( bool flip_immediate); void hubp2_dcc_control(struct hubp *hubp, bool enable, - bool independent_64b_blks); + enum hubp_ind_block_size independent_64b_blks); void hubp2_program_size( struct hubp *hubp, 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 38b3c89b2a59..d1c56fe9c301 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) { @@ -158,8 +142,7 @@ void dcn20_display_init(struct dc *dc) /* DCCG */ dcn20_dccg_init(hws); - /* Disable all memory low power mode. All memories are enabled. */ - REG_UPDATE(DC_MEM_GLOBAL_PWR_REQ_CNTL, DC_MEM_GLOBAL_PWR_REQ_DIS, 1); + REG_UPDATE(DC_MEM_GLOBAL_PWR_REQ_CNTL, DC_MEM_GLOBAL_PWR_REQ_DIS, 0); /* DCHUB/MMHUBBUB * set global timer refclk divider @@ -184,7 +167,7 @@ void dcn20_display_init(struct dc *dc) REG_WRITE(AZALIA_CONTROLLER_CLOCK_GATING, 0x1); } -static void disable_vga( +void dcn20_disable_vga( struct dce_hwseq *hws) { REG_WRITE(D1VGA_CONTROL, 0); @@ -487,29 +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 */ @@ -535,7 +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)); @@ -559,205 +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); - } - } - -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - /* Power gate DSCs */ - for (i = 0; i < res_pool->res_cap->num_dsc; i++) - dcn20_dsc_pg_control(hws, res_pool->dscs[i]->inst, false); -#endif - - /* 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, @@ -766,11 +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 @@ -781,12 +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. */ @@ -810,11 +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); @@ -835,6 +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); @@ -865,6 +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) @@ -894,7 +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; @@ -1056,14 +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); @@ -1079,7 +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; @@ -1087,8 +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) @@ -1108,10 +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, @@ -1157,6 +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, @@ -1252,7 +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( @@ -1274,8 +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( @@ -1316,17 +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. */ @@ -1354,13 +1130,15 @@ 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) @@ -1408,7 +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 || @@ -1454,6 +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); + } + } } @@ -1525,8 +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( @@ -1616,12 +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 } @@ -1630,12 +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 } @@ -1660,9 +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( @@ -1700,6 +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); } @@ -1765,18 +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); } @@ -1812,9 +1617,23 @@ 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 @@ -1834,6 +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++) @@ -1863,7 +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 || @@ -2148,14 +1971,126 @@ static void dcn20_program_dmdata_engine(struct pipe_ctx *pipe_ctx) 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; @@ -2183,5 +2118,19 @@ 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; + dc->hwss.dsc_pg_control = dcn20_dsc_pg_control; + 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 689c2765b071..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, 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 67f0128f0b38..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); } @@ -498,6 +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, @@ -508,6 +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 e90b6bcad05b..b4e3ce22ed52 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -82,7 +82,8 @@ #include "amdgpu_socbb.h" -#define SOC_BOUNDING_BOX_VALID true +/* NV12 SOC BB is currently in FW, mark SW bounding box invalid. */ +#define SOC_BOUNDING_BOX_VALID false #define DC_LOGGER_INIT(logger) struct _vcs_dpi_ip_params_st dcn2_0_ip = { @@ -268,6 +269,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { .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 @@ -424,7 +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) }; @@ -1317,7 +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; @@ -1335,7 +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) @@ -1479,22 +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 @@ -1593,17 +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; @@ -1626,57 +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( @@ -1789,6 +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: @@ -1841,7 +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: @@ -1863,10 +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; @@ -1915,14 +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; @@ -2106,20 +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; @@ -2128,6 +2161,82 @@ static bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) } #endif +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) +{ + 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; + } + } + + /* + * 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, @@ -2152,6 +2261,38 @@ bool dcn20_fast_validate_bw( 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; @@ -2159,7 +2300,6 @@ bool dcn20_fast_validate_bw( 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; @@ -2167,10 +2307,7 @@ bool dcn20_fast_validate_bw( 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)); @@ -2281,12 +2418,11 @@ bool dcn20_fast_validate_bw( 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 (!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); @@ -2322,16 +2458,20 @@ bool dcn20_fast_validate_bw( 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) { @@ -2370,100 +2510,100 @@ void dcn20_calculate_wm( 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) - 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][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][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][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][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][pipe_split_from[i]]; - 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; + if (!context->res_ctx.pipe_ctx[i].stream) + continue; - pipe_cnt++; - } + 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][context->bw_ctx.dml.vba.maxMpcComb]; - 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); + if (pipe_split_from[i] < 0) { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][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][pipe_idx]; else - pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, - &context->res_ctx, pipes); + 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][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][pipe_split_from[i]]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; } - *out_pipe_cnt = pipe_cnt; + 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; - 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; + pipe_cnt++; + } - /* only pipe 0 is read for voltage and dcf/soc clocks */ - if (vlevel < 1) { - pipes[0].clks_cfg.voltage = 1; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].socclk_mhz; - } - context->bw_ctx.bw.dcn.watermarks.b.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - if (vlevel < 2) { - pipes[0].clks_cfg.voltage = 2; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; - } - context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - if (vlevel < 3) { - pipes[0].clks_cfg.voltage = 3; - pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; - pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; - } - context->bw_ctx.bw.dcn.watermarks.d.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - - 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; - context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; - 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; + 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; + + 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; + + /* only pipe 0 is read for voltage and dcf/soc clocks */ + if (vlevel < 1) { + pipes[0].clks_cfg.voltage = 1; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[1].socclk_mhz; + } + context->bw_ctx.bw.dcn.watermarks.b.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + if (vlevel < 2) { + pipes[0].clks_cfg.voltage = 2; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; + } + context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + if (vlevel < 3) { + pipes[0].clks_cfg.voltage = 3; + pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[2].socclk_mhz; + } + context->bw_ctx.bw.dcn.watermarks.d.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + + 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; + context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + 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( @@ -2472,7 +2612,8 @@ void dcn20_calculate_dlg_params( int pipe_cnt, int vlevel) { - int i, pipe_idx; + 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); @@ -2482,37 +2623,69 @@ void dcn20_calculate_dlg_params( 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; + /* + * 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; + 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; @@ -2612,7 +2785,7 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, goto restore_dml_state; } - // Fallback #1: Try to only support G6 temperature read latency + // 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); @@ -2623,19 +2796,7 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, goto restore_dml_state; } - // Fallback #2: Retry with "new" DCN20 to support G6 temperature read latency - memcpy (&context->bw_ctx.dml, &dc->work_arounds.alternate_dml, sizeof (struct display_mode_lib)); - 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 #2 is supposed to always work. + // ERROR: fallback is supposed to always work. ASSERT(false); restore_dml_state: @@ -2955,6 +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)) @@ -2962,6 +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) { @@ -2972,101 +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); - dcn2_0_soc.writeback_dram_clock_change_latency_us = + // HACK!! Lower uclock latency switch time so we don't switch + 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); } } @@ -3097,14 +3286,14 @@ static bool init_soc_bounding_box(struct dc *dc, } 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; } @@ -3117,6 +3306,12 @@ 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.funcs = &dcn20_res_pool_funcs; @@ -3240,8 +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->work_arounds.alternate_dml, &dcn2_0_soc, &dcn2_0_ip, DML_PROJECT_NAVI10v2); + 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}; @@ -3249,7 +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; @@ -3257,13 +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_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c index 5c2b7b54b126..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. */ @@ -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/dml/dcn20/display_mode_vba_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c index 22455db54980..0fafd693ffb4 100644 --- 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 @@ -1475,6 +1475,33 @@ static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndP / 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 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_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index ab34fd26702f..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 @@ -109,6 +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 88e63f16f7fc..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] = 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/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 fd1fb1653479..000000000000 --- a/drivers/gpu/drm/amd/display/dc/dsc/drm_dsc_dc.c +++ /dev/null @@ -1,453 +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) -#define MAX(x, y) ((x) > (y) ? (x) : (y)) - -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); - -static int compute_offset(struct drm_dsc_config *vdsc_cfg, int pixels_per_group, - int groups_per_line, int grpcnt) -{ - int offset = 0; - int grpcnt_id = DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay, pixels_per_group); - - if (grpcnt <= grpcnt_id) - offset = DIV_ROUND_UP(grpcnt * pixels_per_group * vdsc_cfg->bits_per_pixel, 16); - else - offset = DIV_ROUND_UP(grpcnt_id * pixels_per_group * vdsc_cfg->bits_per_pixel, 16) - - (((grpcnt - grpcnt_id) * vdsc_cfg->slice_bpg_offset) >> 11); - - if (grpcnt <= groups_per_line) - offset += grpcnt * vdsc_cfg->first_line_bpg_offset; - else - offset += groups_per_line * vdsc_cfg->first_line_bpg_offset - - (((grpcnt - groups_per_line) * vdsc_cfg->nfl_bpg_offset) >> 11); - - if (vdsc_cfg->native_420) { - if (grpcnt <= groups_per_line) - offset -= (grpcnt * vdsc_cfg->nsl_bpg_offset) >> 11; - else if (grpcnt <= 2 * groups_per_line) - offset += (grpcnt - groups_per_line) * vdsc_cfg->second_line_bpg_offset - - ((groups_per_line * vdsc_cfg->nsl_bpg_offset) >> 11); - else - offset += (grpcnt - groups_per_line) * vdsc_cfg->second_line_bpg_offset - - (((grpcnt - groups_per_line) * vdsc_cfg->nsl_bpg_offset) >> 11); - } - - return offset; -} - -/** - * 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; - unsigned long max_offset = 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; - } - - if (vdsc_cfg->slice_height > 2) - vdsc_cfg->nsl_bpg_offset = DIV_ROUND_UP((vdsc_cfg->second_line_bpg_offset << 11), - (vdsc_cfg->slice_height - 1)); - else - vdsc_cfg->nsl_bpg_offset = 0; - - if (vdsc_cfg->nsl_bpg_offset > 65535) { - DRM_DEBUG_KMS("NslBpgOffset 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->nsl_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 - */ - if (vdsc_cfg->dsc_version_minor == 2 && (vdsc_cfg->native_420 || vdsc_cfg->native_422)) { - - max_offset = compute_offset(vdsc_cfg, DSC_RC_PIXELS_PER_GROUP, groups_per_line, - DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay, - DSC_RC_PIXELS_PER_GROUP)); - - max_offset = MAX(max_offset, - compute_offset(vdsc_cfg, DSC_RC_PIXELS_PER_GROUP, groups_per_line, - DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay, - groups_per_line))); - - max_offset = MAX(max_offset, - compute_offset(vdsc_cfg, DSC_RC_PIXELS_PER_GROUP, groups_per_line, - DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay, - groups_per_line * 2))); - - rbs_min = vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset + max_offset; - } else { - 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/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 5711f30cf848..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 @@ -196,12 +196,12 @@ 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 = dal_hw_generic_create, - .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_generic_registers = define_generic_registers 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 afb7c0f111bf..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 @@ -212,12 +212,12 @@ static void define_generic_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 = dal_hw_generic_create, - .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_generic_registers = define_generic_registers, 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 f15288c3986e..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 @@ -42,12 +42,9 @@ /* 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/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 a7bc3ee5dfec..d03165e71dc6 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c @@ -290,13 +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); @@ -310,51 +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); - service->factory.funcs->define_generic_registers(pin, 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; } @@ -376,11 +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.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h index 7017c9337348..e15b037f3bcd 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h @@ -28,35 +28,35 @@ 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); diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c index 8b7a8ffe3cd7..69b899741f6d 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.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_generic.h" @@ -43,6 +44,8 @@ #define REG(reg)\ (generic->regs->reg) +struct gpio; + static void dal_hw_generic_construct( struct hw_generic *pin, enum gpio_id id, @@ -106,29 +109,30 @@ static void construct( generic->base.base.funcs = &funcs; } -struct hw_gpio_pin *dal_hw_generic_create( +void dal_hw_generic_init( + struct hw_generic **hw_generic, struct dc_context *ctx, enum gpio_id id, uint32_t en) { - struct hw_generic *generic; - - if (id != GPIO_ID_GENERIC) { + if ((en < GPIO_DDC_LINE_MIN) || (en > GPIO_DDC_LINE_MAX)) { ASSERT_CRITICAL(false); - return NULL; + *hw_generic = NULL; } - if ((en < GPIO_GENERIC_MIN) || (en > GPIO_GENERIC_MAX)) { + *hw_generic = kzalloc(sizeof(struct hw_generic), GFP_KERNEL); + if (!*hw_generic) { ASSERT_CRITICAL(false); - return NULL; + return; } - generic = kzalloc(sizeof(struct hw_generic), GFP_KERNEL); - if (!generic) { - ASSERT_CRITICAL(false); - return NULL; - } + 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); - construct(generic, id, en, ctx); - return &generic->base.base; + 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 index 3ea1c13e3ea6..bd6ffeb5e9df 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h @@ -27,6 +27,7 @@ #define __DAL_HW_generic_H__ #include "generic_regs.h" +#include "hw_gpio.h" struct hw_generic { struct hw_gpio base; @@ -38,9 +39,12 @@ struct hw_generic { #define HW_GENERIC_FROM_BASE(hw_gpio) \ container_of((HW_GPIO_FROM_BASE(hw_gpio)), struct hw_generic, base) -struct hw_gpio_pin *dal_hw_generic_create( +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/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index a148ffde8b12..df28fbc4c63c 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*********************/ @@ -228,14 +223,12 @@ 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 +292,8 @@ 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 c5293f9508fa..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); @@ -62,11 +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); -void set_dsc_on_stream(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_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h index 4b5505fa980c..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 @@ -216,6 +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 @@ -279,8 +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 9b69a06ab46f..474c7194a9f8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h @@ -201,7 +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 61cd4f8752c3..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, @@ -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/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 9f00289bda78..58826be81395 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -199,6 +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. @@ -251,6 +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 28645e10f854..732a93df1844 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -78,6 +78,8 @@ 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 +196,7 @@ 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 +205,7 @@ 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, @@ -236,7 +237,8 @@ struct hw_sequencer_funcs { #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); @@ -280,6 +282,36 @@ 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)( 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 3680846674e8..4eff5d38a2f9 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h @@ -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..1cc1c8ce633b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -179,7 +179,4 @@ 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/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/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/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/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 ed894cddeee5..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) @@ -60,12 +59,12 @@ static struct translate_from_linear_space_args scratch_gamma_args; static struct fixed31_32 pow_buffer[NUM_PTS_IN_REGION]; static struct fixed31_32 gamma_of_2; // 2^gamma int pow_buffer_ptr = -1; - -static const int32_t gamma_numerator01[] = { 31308, 180000, 0}; -static const int32_t gamma_numerator02[] = { 12920, 4500, 0}; -static const int32_t gamma_numerator03[] = { 55, 99, 0}; -static const int32_t gamma_numerator04[] = { 55, 99, 0}; -static const int32_t gamma_numerator05[] = { 2400, 2200, 2200}; + /*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; */ @@ -98,6 +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. */ @@ -162,59 +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); } @@ -270,23 +290,28 @@ 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) { - 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( @@ -302,6 +327,8 @@ static void build_coefficients(struct gamma_coefficients *coefficients, enum gam ++i; } while (i != ARRAY_SIZE(coefficients->a0)); +release: + return ret; } static struct fixed31_32 translate_from_linear_space( @@ -738,11 +765,12 @@ 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 pwl_float_data_ex *rgb = rgb_regamma; @@ -750,9 +778,10 @@ static void build_regamma(struct pwl_float_data_ex *rgb_regamma, coeff = kvzalloc(sizeof(*coeff), GFP_KERNEL); if (!coeff) - return; + goto release; - build_coefficients(coeff, type); + if (!build_coefficients(coeff, type)) + goto release; memset(pow_buffer, 0, NUM_PTS_IN_REGION * sizeof(struct fixed31_32)); pow_buffer_ptr = 0; // see variable definition for more info @@ -768,8 +797,10 @@ static void build_regamma(struct pwl_float_data_ex *rgb_regamma, ++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, @@ -944,15 +975,18 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, 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 @@ -981,11 +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; @@ -993,9 +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; @@ -1004,9 +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; @@ -1015,8 +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; @@ -1675,6 +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; @@ -1683,9 +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, @@ -1886,13 +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; @@ -1979,7 +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); @@ -1993,9 +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; @@ -2004,18 +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; @@ -2065,8 +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); @@ -2076,9 +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; @@ -2087,8 +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); @@ -2098,7 +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 69cecd2ec251..44ddea58523a 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h @@ -92,6 +92,7 @@ struct translate_from_linear_space_args { }; 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..5f4b98df3d92 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, @@ -45,6 +114,25 @@ enum ColorimetryYCCDP { ColorimetryYCC_DP_ITU2020YCbCr = 7, }; +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); +} + 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/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( diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index a0a7211438f2..8889aaceec60 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -142,6 +142,7 @@ enum PP_FEATURE_MASK { enum DC_FEATURE_MASK { DC_FBC_MASK = 0x1, + DC_MULTI_MON_PP_MCLK_SWITCH_MASK = 0x2, }; enum amd_dpm_forced_level; diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_offset.h index 8f515875a34d..f2ae3a58949e 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_offset.h @@ -21,6 +21,27 @@ #ifndef _mmhub_9_4_0_OFFSET_HEADER #define _mmhub_9_4_0_OFFSET_HEADER +/* MMEA */ +#define mmMMEA0_SDP_ARB_FINAL_VG20 0x01ee +#define mmMMEA0_SDP_ARB_FINAL_VG20_BASE_IDX 0 +#define mmMMEA0_EDC_CNT_VG20 0x0206 +#define mmMMEA0_EDC_CNT_VG20_BASE_IDX 0 +#define mmMMEA0_EDC_CNT2_VG20 0x0207 +#define mmMMEA0_EDC_CNT2_VG20_BASE_IDX 0 +#define mmMMEA0_EDC_MODE_VG20 0x0210 +#define mmMMEA0_EDC_MODE_VG20_BASE_IDX 0 +#define mmMMEA0_ERR_STATUS_VG20 0x0211 +#define mmMMEA0_ERR_STATUS_VG20_BASE_IDX 0 +#define mmMMEA1_SDP_ARB_FINAL_VG20 0x032e +#define mmMMEA1_SDP_ARB_FINAL_VG20_BASE_IDX 0 +#define mmMMEA1_EDC_CNT_VG20 0x0346 +#define mmMMEA1_EDC_CNT_VG20_BASE_IDX 0 +#define mmMMEA1_EDC_CNT2_VG20 0x0347 +#define mmMMEA1_EDC_CNT2_VG20_BASE_IDX 0 +#define mmMMEA1_EDC_MODE_VG20 0x0350 +#define mmMMEA1_EDC_MODE_VG20_BASE_IDX 0 +#define mmMMEA1_ERR_STATUS_VG20 0x0351 +#define mmMMEA1_ERR_STATUS_VG20_BASE_IDX 0 // addressBlock: mmhub_utcl2_vmsharedpfdec // base address: 0x6a040 diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_sh_mask.h index 0a6b072d191e..c24259ed12a1 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_sh_mask.h @@ -21,6 +21,228 @@ #ifndef _mmhub_9_4_0_SH_MASK_HEADER #define _mmhub_9_4_0_SH_MASK_HEADER +//MMEA0_SDP_ARB_FINAL +#define MMEA0_SDP_ARB_FINAL__DRAM_BURST_LIMIT__SHIFT 0x0 +#define MMEA0_SDP_ARB_FINAL__GMI_BURST_LIMIT__SHIFT 0x5 +#define MMEA0_SDP_ARB_FINAL__IO_BURST_LIMIT__SHIFT 0xa +#define MMEA0_SDP_ARB_FINAL__BURST_LIMIT_MULTIPLIER__SHIFT 0xf +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC0__SHIFT 0x11 +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC1__SHIFT 0x12 +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC2__SHIFT 0x13 +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC3__SHIFT 0x14 +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC4__SHIFT 0x15 +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC5__SHIFT 0x16 +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC6__SHIFT 0x17 +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC7__SHIFT 0x18 +#define MMEA0_SDP_ARB_FINAL__ERREVENT_ON_ERROR__SHIFT 0x19 +#define MMEA0_SDP_ARB_FINAL__HALTREQ_ON_ERROR__SHIFT 0x1a +#define MMEA0_SDP_ARB_FINAL__DRAM_BURST_LIMIT_MASK 0x0000001FL +#define MMEA0_SDP_ARB_FINAL__GMI_BURST_LIMIT_MASK 0x000003E0L +#define MMEA0_SDP_ARB_FINAL__IO_BURST_LIMIT_MASK 0x00007C00L +#define MMEA0_SDP_ARB_FINAL__BURST_LIMIT_MULTIPLIER_MASK 0x00018000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC0_MASK 0x00020000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC1_MASK 0x00040000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC2_MASK 0x00080000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC3_MASK 0x00100000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC4_MASK 0x00200000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC5_MASK 0x00400000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC6_MASK 0x00800000L +#define MMEA0_SDP_ARB_FINAL__RDONLY_VC7_MASK 0x01000000L +#define MMEA0_SDP_ARB_FINAL__ERREVENT_ON_ERROR_MASK 0x02000000L +#define MMEA0_SDP_ARB_FINAL__HALTREQ_ON_ERROR_MASK 0x04000000L +//MMEA0_EDC_CNT +#define MMEA0_EDC_CNT__DRAMRD_CMDMEM_SEC_COUNT__SHIFT 0x0 +#define MMEA0_EDC_CNT__DRAMRD_CMDMEM_DED_COUNT__SHIFT 0x2 +#define MMEA0_EDC_CNT__DRAMWR_CMDMEM_SEC_COUNT__SHIFT 0x4 +#define MMEA0_EDC_CNT__DRAMWR_CMDMEM_DED_COUNT__SHIFT 0x6 +#define MMEA0_EDC_CNT__DRAMWR_DATAMEM_SEC_COUNT__SHIFT 0x8 +#define MMEA0_EDC_CNT__DRAMWR_DATAMEM_DED_COUNT__SHIFT 0xa +#define MMEA0_EDC_CNT__RRET_TAGMEM_SEC_COUNT__SHIFT 0xc +#define MMEA0_EDC_CNT__RRET_TAGMEM_DED_COUNT__SHIFT 0xe +#define MMEA0_EDC_CNT__WRET_TAGMEM_SEC_COUNT__SHIFT 0x10 +#define MMEA0_EDC_CNT__WRET_TAGMEM_DED_COUNT__SHIFT 0x12 +#define MMEA0_EDC_CNT__DRAMRD_PAGEMEM_SED_COUNT__SHIFT 0x14 +#define MMEA0_EDC_CNT__DRAMWR_PAGEMEM_SED_COUNT__SHIFT 0x16 +#define MMEA0_EDC_CNT__IORD_CMDMEM_SED_COUNT__SHIFT 0x18 +#define MMEA0_EDC_CNT__IOWR_CMDMEM_SED_COUNT__SHIFT 0x1a +#define MMEA0_EDC_CNT__IOWR_DATAMEM_SED_COUNT__SHIFT 0x1c +#define MMEA0_EDC_CNT__DRAMRD_CMDMEM_SEC_COUNT_MASK 0x00000003L +#define MMEA0_EDC_CNT__DRAMRD_CMDMEM_DED_COUNT_MASK 0x0000000CL +#define MMEA0_EDC_CNT__DRAMWR_CMDMEM_SEC_COUNT_MASK 0x00000030L +#define MMEA0_EDC_CNT__DRAMWR_CMDMEM_DED_COUNT_MASK 0x000000C0L +#define MMEA0_EDC_CNT__DRAMWR_DATAMEM_SEC_COUNT_MASK 0x00000300L +#define MMEA0_EDC_CNT__DRAMWR_DATAMEM_DED_COUNT_MASK 0x00000C00L +#define MMEA0_EDC_CNT__RRET_TAGMEM_SEC_COUNT_MASK 0x00003000L +#define MMEA0_EDC_CNT__RRET_TAGMEM_DED_COUNT_MASK 0x0000C000L +#define MMEA0_EDC_CNT__WRET_TAGMEM_SEC_COUNT_MASK 0x00030000L +#define MMEA0_EDC_CNT__WRET_TAGMEM_DED_COUNT_MASK 0x000C0000L +#define MMEA0_EDC_CNT__DRAMRD_PAGEMEM_SED_COUNT_MASK 0x00300000L +#define MMEA0_EDC_CNT__DRAMWR_PAGEMEM_SED_COUNT_MASK 0x00C00000L +#define MMEA0_EDC_CNT__IORD_CMDMEM_SED_COUNT_MASK 0x03000000L +#define MMEA0_EDC_CNT__IOWR_CMDMEM_SED_COUNT_MASK 0x0C000000L +#define MMEA0_EDC_CNT__IOWR_DATAMEM_SED_COUNT_MASK 0x30000000L +//MMEA0_EDC_CNT2 +#define MMEA0_EDC_CNT2__GMIRD_CMDMEM_SEC_COUNT__SHIFT 0x0 +#define MMEA0_EDC_CNT2__GMIRD_CMDMEM_DED_COUNT__SHIFT 0x2 +#define MMEA0_EDC_CNT2__GMIWR_CMDMEM_SEC_COUNT__SHIFT 0x4 +#define MMEA0_EDC_CNT2__GMIWR_CMDMEM_DED_COUNT__SHIFT 0x6 +#define MMEA0_EDC_CNT2__GMIWR_DATAMEM_SEC_COUNT__SHIFT 0x8 +#define MMEA0_EDC_CNT2__GMIWR_DATAMEM_DED_COUNT__SHIFT 0xa +#define MMEA0_EDC_CNT2__GMIRD_PAGEMEM_SED_COUNT__SHIFT 0xc +#define MMEA0_EDC_CNT2__GMIWR_PAGEMEM_SED_COUNT__SHIFT 0xe +#define MMEA0_EDC_CNT2__MAM_D0MEM_SED_COUNT__SHIFT 0x10 +#define MMEA0_EDC_CNT2__MAM_D1MEM_SED_COUNT__SHIFT 0x12 +#define MMEA0_EDC_CNT2__MAM_D2MEM_SED_COUNT__SHIFT 0x14 +#define MMEA0_EDC_CNT2__MAM_D3MEM_SED_COUNT__SHIFT 0x16 +#define MMEA0_EDC_CNT2__GMIRD_CMDMEM_SEC_COUNT_MASK 0x00000003L +#define MMEA0_EDC_CNT2__GMIRD_CMDMEM_DED_COUNT_MASK 0x0000000CL +#define MMEA0_EDC_CNT2__GMIWR_CMDMEM_SEC_COUNT_MASK 0x00000030L +#define MMEA0_EDC_CNT2__GMIWR_CMDMEM_DED_COUNT_MASK 0x000000C0L +#define MMEA0_EDC_CNT2__GMIWR_DATAMEM_SEC_COUNT_MASK 0x00000300L +#define MMEA0_EDC_CNT2__GMIWR_DATAMEM_DED_COUNT_MASK 0x00000C00L +#define MMEA0_EDC_CNT2__GMIRD_PAGEMEM_SED_COUNT_MASK 0x00003000L +#define MMEA0_EDC_CNT2__GMIWR_PAGEMEM_SED_COUNT_MASK 0x0000C000L +#define MMEA0_EDC_CNT2__MAM_D0MEM_SED_COUNT_MASK 0x00030000L +#define MMEA0_EDC_CNT2__MAM_D1MEM_SED_COUNT_MASK 0x000C0000L +#define MMEA0_EDC_CNT2__MAM_D2MEM_SED_COUNT_MASK 0x00300000L +#define MMEA0_EDC_CNT2__MAM_D3MEM_SED_COUNT_MASK 0x00C00000L +//MMEA0_EDC_MODE +#define MMEA0_EDC_MODE__COUNT_FED_OUT__SHIFT 0x10 +#define MMEA0_EDC_MODE__GATE_FUE__SHIFT 0x11 +#define MMEA0_EDC_MODE__DED_MODE__SHIFT 0x14 +#define MMEA0_EDC_MODE__PROP_FED__SHIFT 0x1d +#define MMEA0_EDC_MODE__BYPASS__SHIFT 0x1f +#define MMEA0_EDC_MODE__COUNT_FED_OUT_MASK 0x00010000L +#define MMEA0_EDC_MODE__GATE_FUE_MASK 0x00020000L +#define MMEA0_EDC_MODE__DED_MODE_MASK 0x00300000L +#define MMEA0_EDC_MODE__PROP_FED_MASK 0x20000000L +#define MMEA0_EDC_MODE__BYPASS_MASK 0x80000000L +//MMEA0_ERR_STATUS +#define MMEA0_ERR_STATUS__SDP_RDRSP_STATUS__SHIFT 0x0 +#define MMEA0_ERR_STATUS__SDP_WRRSP_STATUS__SHIFT 0x4 +#define MMEA0_ERR_STATUS__SDP_RDRSP_DATASTATUS__SHIFT 0x8 +#define MMEA0_ERR_STATUS__SDP_RDRSP_DATAPARITY_ERROR__SHIFT 0xa +#define MMEA0_ERR_STATUS__CLEAR_ERROR_STATUS__SHIFT 0xb +#define MMEA0_ERR_STATUS__BUSY_ON_ERROR__SHIFT 0xc +#define MMEA0_ERR_STATUS__FUE_FLAG__SHIFT 0xd +#define MMEA0_ERR_STATUS__SDP_RDRSP_STATUS_MASK 0x0000000FL +#define MMEA0_ERR_STATUS__SDP_WRRSP_STATUS_MASK 0x000000F0L +#define MMEA0_ERR_STATUS__SDP_RDRSP_DATASTATUS_MASK 0x00000300L +#define MMEA0_ERR_STATUS__SDP_RDRSP_DATAPARITY_ERROR_MASK 0x00000400L +#define MMEA0_ERR_STATUS__CLEAR_ERROR_STATUS_MASK 0x00000800L +#define MMEA0_ERR_STATUS__BUSY_ON_ERROR_MASK 0x00001000L +#define MMEA0_ERR_STATUS__FUE_FLAG_MASK 0x00002000L +//MMEA1_SDP_ARB_FINAL +#define MMEA1_SDP_ARB_FINAL__DRAM_BURST_LIMIT__SHIFT 0x0 +#define MMEA1_SDP_ARB_FINAL__GMI_BURST_LIMIT__SHIFT 0x5 +#define MMEA1_SDP_ARB_FINAL__IO_BURST_LIMIT__SHIFT 0xa +#define MMEA1_SDP_ARB_FINAL__BURST_LIMIT_MULTIPLIER__SHIFT 0xf +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC0__SHIFT 0x11 +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC1__SHIFT 0x12 +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC2__SHIFT 0x13 +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC3__SHIFT 0x14 +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC4__SHIFT 0x15 +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC5__SHIFT 0x16 +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC6__SHIFT 0x17 +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC7__SHIFT 0x18 +#define MMEA1_SDP_ARB_FINAL__ERREVENT_ON_ERROR__SHIFT 0x19 +#define MMEA1_SDP_ARB_FINAL__HALTREQ_ON_ERROR__SHIFT 0x1a +#define MMEA1_SDP_ARB_FINAL__DRAM_BURST_LIMIT_MASK 0x0000001FL +#define MMEA1_SDP_ARB_FINAL__GMI_BURST_LIMIT_MASK 0x000003E0L +#define MMEA1_SDP_ARB_FINAL__IO_BURST_LIMIT_MASK 0x00007C00L +#define MMEA1_SDP_ARB_FINAL__BURST_LIMIT_MULTIPLIER_MASK 0x00018000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC0_MASK 0x00020000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC1_MASK 0x00040000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC2_MASK 0x00080000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC3_MASK 0x00100000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC4_MASK 0x00200000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC5_MASK 0x00400000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC6_MASK 0x00800000L +#define MMEA1_SDP_ARB_FINAL__RDONLY_VC7_MASK 0x01000000L +#define MMEA1_SDP_ARB_FINAL__ERREVENT_ON_ERROR_MASK 0x02000000L +#define MMEA1_SDP_ARB_FINAL__HALTREQ_ON_ERROR_MASK 0x04000000L +//MMEA1_EDC_CNT +#define MMEA1_EDC_CNT__DRAMRD_CMDMEM_SEC_COUNT__SHIFT 0x0 +#define MMEA1_EDC_CNT__DRAMRD_CMDMEM_DED_COUNT__SHIFT 0x2 +#define MMEA1_EDC_CNT__DRAMWR_CMDMEM_SEC_COUNT__SHIFT 0x4 +#define MMEA1_EDC_CNT__DRAMWR_CMDMEM_DED_COUNT__SHIFT 0x6 +#define MMEA1_EDC_CNT__DRAMWR_DATAMEM_SEC_COUNT__SHIFT 0x8 +#define MMEA1_EDC_CNT__DRAMWR_DATAMEM_DED_COUNT__SHIFT 0xa +#define MMEA1_EDC_CNT__RRET_TAGMEM_SEC_COUNT__SHIFT 0xc +#define MMEA1_EDC_CNT__RRET_TAGMEM_DED_COUNT__SHIFT 0xe +#define MMEA1_EDC_CNT__WRET_TAGMEM_SEC_COUNT__SHIFT 0x10 +#define MMEA1_EDC_CNT__WRET_TAGMEM_DED_COUNT__SHIFT 0x12 +#define MMEA1_EDC_CNT__DRAMRD_PAGEMEM_SED_COUNT__SHIFT 0x14 +#define MMEA1_EDC_CNT__DRAMWR_PAGEMEM_SED_COUNT__SHIFT 0x16 +#define MMEA1_EDC_CNT__IORD_CMDMEM_SED_COUNT__SHIFT 0x18 +#define MMEA1_EDC_CNT__IOWR_CMDMEM_SED_COUNT__SHIFT 0x1a +#define MMEA1_EDC_CNT__IOWR_DATAMEM_SED_COUNT__SHIFT 0x1c +#define MMEA1_EDC_CNT__DRAMRD_CMDMEM_SEC_COUNT_MASK 0x00000003L +#define MMEA1_EDC_CNT__DRAMRD_CMDMEM_DED_COUNT_MASK 0x0000000CL +#define MMEA1_EDC_CNT__DRAMWR_CMDMEM_SEC_COUNT_MASK 0x00000030L +#define MMEA1_EDC_CNT__DRAMWR_CMDMEM_DED_COUNT_MASK 0x000000C0L +#define MMEA1_EDC_CNT__DRAMWR_DATAMEM_SEC_COUNT_MASK 0x00000300L +#define MMEA1_EDC_CNT__DRAMWR_DATAMEM_DED_COUNT_MASK 0x00000C00L +#define MMEA1_EDC_CNT__RRET_TAGMEM_SEC_COUNT_MASK 0x00003000L +#define MMEA1_EDC_CNT__RRET_TAGMEM_DED_COUNT_MASK 0x0000C000L +#define MMEA1_EDC_CNT__WRET_TAGMEM_SEC_COUNT_MASK 0x00030000L +#define MMEA1_EDC_CNT__WRET_TAGMEM_DED_COUNT_MASK 0x000C0000L +#define MMEA1_EDC_CNT__DRAMRD_PAGEMEM_SED_COUNT_MASK 0x00300000L +#define MMEA1_EDC_CNT__DRAMWR_PAGEMEM_SED_COUNT_MASK 0x00C00000L +#define MMEA1_EDC_CNT__IORD_CMDMEM_SED_COUNT_MASK 0x03000000L +#define MMEA1_EDC_CNT__IOWR_CMDMEM_SED_COUNT_MASK 0x0C000000L +#define MMEA1_EDC_CNT__IOWR_DATAMEM_SED_COUNT_MASK 0x30000000L +//MMEA1_EDC_CNT2 +#define MMEA1_EDC_CNT2__GMIRD_CMDMEM_SEC_COUNT__SHIFT 0x0 +#define MMEA1_EDC_CNT2__GMIRD_CMDMEM_DED_COUNT__SHIFT 0x2 +#define MMEA1_EDC_CNT2__GMIWR_CMDMEM_SEC_COUNT__SHIFT 0x4 +#define MMEA1_EDC_CNT2__GMIWR_CMDMEM_DED_COUNT__SHIFT 0x6 +#define MMEA1_EDC_CNT2__GMIWR_DATAMEM_SEC_COUNT__SHIFT 0x8 +#define MMEA1_EDC_CNT2__GMIWR_DATAMEM_DED_COUNT__SHIFT 0xa +#define MMEA1_EDC_CNT2__GMIRD_PAGEMEM_SED_COUNT__SHIFT 0xc +#define MMEA1_EDC_CNT2__GMIWR_PAGEMEM_SED_COUNT__SHIFT 0xe +#define MMEA1_EDC_CNT2__MAM_D0MEM_SED_COUNT__SHIFT 0x10 +#define MMEA1_EDC_CNT2__MAM_D1MEM_SED_COUNT__SHIFT 0x12 +#define MMEA1_EDC_CNT2__MAM_D2MEM_SED_COUNT__SHIFT 0x14 +#define MMEA1_EDC_CNT2__MAM_D3MEM_SED_COUNT__SHIFT 0x16 +#define MMEA1_EDC_CNT2__GMIRD_CMDMEM_SEC_COUNT_MASK 0x00000003L +#define MMEA1_EDC_CNT2__GMIRD_CMDMEM_DED_COUNT_MASK 0x0000000CL +#define MMEA1_EDC_CNT2__GMIWR_CMDMEM_SEC_COUNT_MASK 0x00000030L +#define MMEA1_EDC_CNT2__GMIWR_CMDMEM_DED_COUNT_MASK 0x000000C0L +#define MMEA1_EDC_CNT2__GMIWR_DATAMEM_SEC_COUNT_MASK 0x00000300L +#define MMEA1_EDC_CNT2__GMIWR_DATAMEM_DED_COUNT_MASK 0x00000C00L +#define MMEA1_EDC_CNT2__GMIRD_PAGEMEM_SED_COUNT_MASK 0x00003000L +#define MMEA1_EDC_CNT2__GMIWR_PAGEMEM_SED_COUNT_MASK 0x0000C000L +#define MMEA1_EDC_CNT2__MAM_D0MEM_SED_COUNT_MASK 0x00030000L +#define MMEA1_EDC_CNT2__MAM_D1MEM_SED_COUNT_MASK 0x000C0000L +#define MMEA1_EDC_CNT2__MAM_D2MEM_SED_COUNT_MASK 0x00300000L +#define MMEA1_EDC_CNT2__MAM_D3MEM_SED_COUNT_MASK 0x00C00000L +//MMEA1_EDC_MODE +#define MMEA1_EDC_MODE__COUNT_FED_OUT__SHIFT 0x10 +#define MMEA1_EDC_MODE__GATE_FUE__SHIFT 0x11 +#define MMEA1_EDC_MODE__DED_MODE__SHIFT 0x14 +#define MMEA1_EDC_MODE__PROP_FED__SHIFT 0x1d +#define MMEA1_EDC_MODE__BYPASS__SHIFT 0x1f +#define MMEA1_EDC_MODE__COUNT_FED_OUT_MASK 0x00010000L +#define MMEA1_EDC_MODE__GATE_FUE_MASK 0x00020000L +#define MMEA1_EDC_MODE__DED_MODE_MASK 0x00300000L +#define MMEA1_EDC_MODE__PROP_FED_MASK 0x20000000L +#define MMEA1_EDC_MODE__BYPASS_MASK 0x80000000L +//MMEA1_ERR_STATUS +#define MMEA1_ERR_STATUS__SDP_RDRSP_STATUS__SHIFT 0x0 +#define MMEA1_ERR_STATUS__SDP_WRRSP_STATUS__SHIFT 0x4 +#define MMEA1_ERR_STATUS__SDP_RDRSP_DATASTATUS__SHIFT 0x8 +#define MMEA1_ERR_STATUS__SDP_RDRSP_DATAPARITY_ERROR__SHIFT 0xa +#define MMEA1_ERR_STATUS__CLEAR_ERROR_STATUS__SHIFT 0xb +#define MMEA1_ERR_STATUS__BUSY_ON_ERROR__SHIFT 0xc +#define MMEA1_ERR_STATUS__FUE_FLAG__SHIFT 0xd +#define MMEA1_ERR_STATUS__SDP_RDRSP_STATUS_MASK 0x0000000FL +#define MMEA1_ERR_STATUS__SDP_WRRSP_STATUS_MASK 0x000000F0L +#define MMEA1_ERR_STATUS__SDP_RDRSP_DATASTATUS_MASK 0x00000300L +#define MMEA1_ERR_STATUS__SDP_RDRSP_DATAPARITY_ERROR_MASK 0x00000400L +#define MMEA1_ERR_STATUS__CLEAR_ERROR_STATUS_MASK 0x00000800L +#define MMEA1_ERR_STATUS__BUSY_ON_ERROR_MASK 0x00001000L +#define MMEA1_ERR_STATUS__FUE_FLAG_MASK 0x00002000L // addressBlock: mmhub_utcl2_vmsharedpfdec //MC_VM_XGMI_LFB_CNTL diff --git a/drivers/gpu/drm/amd/include/asic_reg/mp/mp_12_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/mp/mp_12_0_0_offset.h new file mode 100644 index 000000000000..1fe51fcb648e --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/mp/mp_12_0_0_offset.h @@ -0,0 +1,336 @@ +/* + * Copyright (C) 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) 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. + */ +#ifndef _mp_12_0_0_OFFSET_HEADER +#define _mp_12_0_0_OFFSET_HEADER + + + +// addressBlock: mp_SmuMp0_SmnDec +// base address: 0x0 +#define mmMP0_SMN_C2PMSG_32 0x0060 +#define mmMP0_SMN_C2PMSG_32_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_33 0x0061 +#define mmMP0_SMN_C2PMSG_33_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_34 0x0062 +#define mmMP0_SMN_C2PMSG_34_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_35 0x0063 +#define mmMP0_SMN_C2PMSG_35_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_36 0x0064 +#define mmMP0_SMN_C2PMSG_36_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_37 0x0065 +#define mmMP0_SMN_C2PMSG_37_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_38 0x0066 +#define mmMP0_SMN_C2PMSG_38_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_39 0x0067 +#define mmMP0_SMN_C2PMSG_39_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_40 0x0068 +#define mmMP0_SMN_C2PMSG_40_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_41 0x0069 +#define mmMP0_SMN_C2PMSG_41_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_42 0x006a +#define mmMP0_SMN_C2PMSG_42_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_43 0x006b +#define mmMP0_SMN_C2PMSG_43_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_44 0x006c +#define mmMP0_SMN_C2PMSG_44_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_45 0x006d +#define mmMP0_SMN_C2PMSG_45_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_46 0x006e +#define mmMP0_SMN_C2PMSG_46_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_47 0x006f +#define mmMP0_SMN_C2PMSG_47_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_48 0x0070 +#define mmMP0_SMN_C2PMSG_48_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_49 0x0071 +#define mmMP0_SMN_C2PMSG_49_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_50 0x0072 +#define mmMP0_SMN_C2PMSG_50_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_51 0x0073 +#define mmMP0_SMN_C2PMSG_51_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_52 0x0074 +#define mmMP0_SMN_C2PMSG_52_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_53 0x0075 +#define mmMP0_SMN_C2PMSG_53_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_54 0x0076 +#define mmMP0_SMN_C2PMSG_54_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_55 0x0077 +#define mmMP0_SMN_C2PMSG_55_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_56 0x0078 +#define mmMP0_SMN_C2PMSG_56_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_57 0x0079 +#define mmMP0_SMN_C2PMSG_57_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_58 0x007a +#define mmMP0_SMN_C2PMSG_58_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_59 0x007b +#define mmMP0_SMN_C2PMSG_59_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_60 0x007c +#define mmMP0_SMN_C2PMSG_60_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_61 0x007d +#define mmMP0_SMN_C2PMSG_61_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_62 0x007e +#define mmMP0_SMN_C2PMSG_62_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_63 0x007f +#define mmMP0_SMN_C2PMSG_63_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_64 0x0080 +#define mmMP0_SMN_C2PMSG_64_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_65 0x0081 +#define mmMP0_SMN_C2PMSG_65_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_66 0x0082 +#define mmMP0_SMN_C2PMSG_66_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_67 0x0083 +#define mmMP0_SMN_C2PMSG_67_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_68 0x0084 +#define mmMP0_SMN_C2PMSG_68_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_69 0x0085 +#define mmMP0_SMN_C2PMSG_69_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_70 0x0086 +#define mmMP0_SMN_C2PMSG_70_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_71 0x0087 +#define mmMP0_SMN_C2PMSG_71_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_72 0x0088 +#define mmMP0_SMN_C2PMSG_72_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_73 0x0089 +#define mmMP0_SMN_C2PMSG_73_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_74 0x008a +#define mmMP0_SMN_C2PMSG_74_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_75 0x008b +#define mmMP0_SMN_C2PMSG_75_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_76 0x008c +#define mmMP0_SMN_C2PMSG_76_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_77 0x008d +#define mmMP0_SMN_C2PMSG_77_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_78 0x008e +#define mmMP0_SMN_C2PMSG_78_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_79 0x008f +#define mmMP0_SMN_C2PMSG_79_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_80 0x0090 +#define mmMP0_SMN_C2PMSG_80_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_81 0x0091 +#define mmMP0_SMN_C2PMSG_81_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_82 0x0092 +#define mmMP0_SMN_C2PMSG_82_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_83 0x0093 +#define mmMP0_SMN_C2PMSG_83_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_84 0x0094 +#define mmMP0_SMN_C2PMSG_84_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_85 0x0095 +#define mmMP0_SMN_C2PMSG_85_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_86 0x0096 +#define mmMP0_SMN_C2PMSG_86_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_87 0x0097 +#define mmMP0_SMN_C2PMSG_87_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_88 0x0098 +#define mmMP0_SMN_C2PMSG_88_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_89 0x0099 +#define mmMP0_SMN_C2PMSG_89_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_90 0x009a +#define mmMP0_SMN_C2PMSG_90_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_91 0x009b +#define mmMP0_SMN_C2PMSG_91_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_92 0x009c +#define mmMP0_SMN_C2PMSG_92_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_93 0x009d +#define mmMP0_SMN_C2PMSG_93_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_94 0x009e +#define mmMP0_SMN_C2PMSG_94_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_95 0x009f +#define mmMP0_SMN_C2PMSG_95_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_96 0x00a0 +#define mmMP0_SMN_C2PMSG_96_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_97 0x00a1 +#define mmMP0_SMN_C2PMSG_97_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_98 0x00a2 +#define mmMP0_SMN_C2PMSG_98_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_99 0x00a3 +#define mmMP0_SMN_C2PMSG_99_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_100 0x00a4 +#define mmMP0_SMN_C2PMSG_100_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_101 0x00a5 +#define mmMP0_SMN_C2PMSG_101_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_102 0x00a6 +#define mmMP0_SMN_C2PMSG_102_BASE_IDX 0 +#define mmMP0_SMN_C2PMSG_103 0x00a7 +#define mmMP0_SMN_C2PMSG_103_BASE_IDX 0 +#define mmMP0_SMN_IH_CREDIT 0x00c1 +#define mmMP0_SMN_IH_CREDIT_BASE_IDX 0 +#define mmMP0_SMN_IH_SW_INT 0x00c2 +#define mmMP0_SMN_IH_SW_INT_BASE_IDX 0 +#define mmMP0_SMN_IH_SW_INT_CTRL 0x00c3 +#define mmMP0_SMN_IH_SW_INT_CTRL_BASE_IDX 0 + + +// addressBlock: mp_SmuMp1_SmnDec +// base address: 0x0 +#define mmMP1_SMN_C2PMSG_32 0x0260 +#define mmMP1_SMN_C2PMSG_32_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_33 0x0261 +#define mmMP1_SMN_C2PMSG_33_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_34 0x0262 +#define mmMP1_SMN_C2PMSG_34_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_35 0x0263 +#define mmMP1_SMN_C2PMSG_35_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_36 0x0264 +#define mmMP1_SMN_C2PMSG_36_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_37 0x0265 +#define mmMP1_SMN_C2PMSG_37_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_38 0x0266 +#define mmMP1_SMN_C2PMSG_38_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_39 0x0267 +#define mmMP1_SMN_C2PMSG_39_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_40 0x0268 +#define mmMP1_SMN_C2PMSG_40_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_41 0x0269 +#define mmMP1_SMN_C2PMSG_41_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_42 0x026a +#define mmMP1_SMN_C2PMSG_42_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_43 0x026b +#define mmMP1_SMN_C2PMSG_43_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_44 0x026c +#define mmMP1_SMN_C2PMSG_44_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_45 0x026d +#define mmMP1_SMN_C2PMSG_45_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_46 0x026e +#define mmMP1_SMN_C2PMSG_46_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_47 0x026f +#define mmMP1_SMN_C2PMSG_47_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_48 0x0270 +#define mmMP1_SMN_C2PMSG_48_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_49 0x0271 +#define mmMP1_SMN_C2PMSG_49_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_50 0x0272 +#define mmMP1_SMN_C2PMSG_50_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_51 0x0273 +#define mmMP1_SMN_C2PMSG_51_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_52 0x0274 +#define mmMP1_SMN_C2PMSG_52_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_53 0x0275 +#define mmMP1_SMN_C2PMSG_53_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_54 0x0276 +#define mmMP1_SMN_C2PMSG_54_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_55 0x0277 +#define mmMP1_SMN_C2PMSG_55_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_56 0x0278 +#define mmMP1_SMN_C2PMSG_56_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_57 0x0279 +#define mmMP1_SMN_C2PMSG_57_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_58 0x027a +#define mmMP1_SMN_C2PMSG_58_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_59 0x027b +#define mmMP1_SMN_C2PMSG_59_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_60 0x027c +#define mmMP1_SMN_C2PMSG_60_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_61 0x027d +#define mmMP1_SMN_C2PMSG_61_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_62 0x027e +#define mmMP1_SMN_C2PMSG_62_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_63 0x027f +#define mmMP1_SMN_C2PMSG_63_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_64 0x0280 +#define mmMP1_SMN_C2PMSG_64_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_65 0x0281 +#define mmMP1_SMN_C2PMSG_65_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_66 0x0282 +#define mmMP1_SMN_C2PMSG_66_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_67 0x0283 +#define mmMP1_SMN_C2PMSG_67_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_68 0x0284 +#define mmMP1_SMN_C2PMSG_68_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_69 0x0285 +#define mmMP1_SMN_C2PMSG_69_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_70 0x0286 +#define mmMP1_SMN_C2PMSG_70_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_71 0x0287 +#define mmMP1_SMN_C2PMSG_71_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_72 0x0288 +#define mmMP1_SMN_C2PMSG_72_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_73 0x0289 +#define mmMP1_SMN_C2PMSG_73_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_74 0x028a +#define mmMP1_SMN_C2PMSG_74_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_75 0x028b +#define mmMP1_SMN_C2PMSG_75_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_76 0x028c +#define mmMP1_SMN_C2PMSG_76_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_77 0x028d +#define mmMP1_SMN_C2PMSG_77_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_78 0x028e +#define mmMP1_SMN_C2PMSG_78_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_79 0x028f +#define mmMP1_SMN_C2PMSG_79_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_80 0x0290 +#define mmMP1_SMN_C2PMSG_80_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_81 0x0291 +#define mmMP1_SMN_C2PMSG_81_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_82 0x0292 +#define mmMP1_SMN_C2PMSG_82_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_83 0x0293 +#define mmMP1_SMN_C2PMSG_83_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_84 0x0294 +#define mmMP1_SMN_C2PMSG_84_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_85 0x0295 +#define mmMP1_SMN_C2PMSG_85_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_86 0x0296 +#define mmMP1_SMN_C2PMSG_86_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_87 0x0297 +#define mmMP1_SMN_C2PMSG_87_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_88 0x0298 +#define mmMP1_SMN_C2PMSG_88_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_89 0x0299 +#define mmMP1_SMN_C2PMSG_89_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_90 0x029a +#define mmMP1_SMN_C2PMSG_90_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_91 0x029b +#define mmMP1_SMN_C2PMSG_91_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_92 0x029c +#define mmMP1_SMN_C2PMSG_92_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_93 0x029d +#define mmMP1_SMN_C2PMSG_93_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_94 0x029e +#define mmMP1_SMN_C2PMSG_94_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_95 0x029f +#define mmMP1_SMN_C2PMSG_95_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_96 0x02a0 +#define mmMP1_SMN_C2PMSG_96_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_97 0x02a1 +#define mmMP1_SMN_C2PMSG_97_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_98 0x02a2 +#define mmMP1_SMN_C2PMSG_98_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_99 0x02a3 +#define mmMP1_SMN_C2PMSG_99_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_100 0x02a4 +#define mmMP1_SMN_C2PMSG_100_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_101 0x02a5 +#define mmMP1_SMN_C2PMSG_101_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_102 0x02a6 +#define mmMP1_SMN_C2PMSG_102_BASE_IDX 0 +#define mmMP1_SMN_C2PMSG_103 0x02a7 +#define mmMP1_SMN_C2PMSG_103_BASE_IDX 0 +#define mmMP1_SMN_IH_CREDIT 0x02c1 +#define mmMP1_SMN_IH_CREDIT_BASE_IDX 0 +#define mmMP1_SMN_IH_SW_INT 0x02c2 +#define mmMP1_SMN_IH_SW_INT_BASE_IDX 0 +#define mmMP1_SMN_IH_SW_INT_CTRL 0x02c3 +#define mmMP1_SMN_IH_SW_INT_CTRL_BASE_IDX 0 +#define mmMP1_SMN_FPS_CNT 0x02c4 +#define mmMP1_SMN_FPS_CNT_BASE_IDX 0 + + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/mp/mp_12_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/mp/mp_12_0_0_sh_mask.h new file mode 100644 index 000000000000..c78151e624b3 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/mp/mp_12_0_0_sh_mask.h @@ -0,0 +1,866 @@ +/* + * Copyright (C) 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) 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. + */ +#ifndef _mp_12_0_0_SH_MASK_HEADER +#define _mp_12_0_0_SH_MASK_HEADER + + +// addressBlock: mp_SmuMp0_SmnDec +//MP0_SMN_C2PMSG_32 +#define MP0_SMN_C2PMSG_32__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_32__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_33 +#define MP0_SMN_C2PMSG_33__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_33__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_34 +#define MP0_SMN_C2PMSG_34__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_34__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_35 +#define MP0_SMN_C2PMSG_35__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_35__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_36 +#define MP0_SMN_C2PMSG_36__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_36__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_37 +#define MP0_SMN_C2PMSG_37__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_37__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_38 +#define MP0_SMN_C2PMSG_38__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_38__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_39 +#define MP0_SMN_C2PMSG_39__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_39__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_40 +#define MP0_SMN_C2PMSG_40__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_40__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_41 +#define MP0_SMN_C2PMSG_41__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_41__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_42 +#define MP0_SMN_C2PMSG_42__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_42__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_43 +#define MP0_SMN_C2PMSG_43__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_43__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_44 +#define MP0_SMN_C2PMSG_44__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_44__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_45 +#define MP0_SMN_C2PMSG_45__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_45__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_46 +#define MP0_SMN_C2PMSG_46__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_46__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_47 +#define MP0_SMN_C2PMSG_47__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_47__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_48 +#define MP0_SMN_C2PMSG_48__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_48__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_49 +#define MP0_SMN_C2PMSG_49__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_49__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_50 +#define MP0_SMN_C2PMSG_50__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_50__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_51 +#define MP0_SMN_C2PMSG_51__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_51__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_52 +#define MP0_SMN_C2PMSG_52__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_52__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_53 +#define MP0_SMN_C2PMSG_53__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_53__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_54 +#define MP0_SMN_C2PMSG_54__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_54__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_55 +#define MP0_SMN_C2PMSG_55__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_55__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_56 +#define MP0_SMN_C2PMSG_56__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_56__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_57 +#define MP0_SMN_C2PMSG_57__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_57__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_58 +#define MP0_SMN_C2PMSG_58__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_58__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_59 +#define MP0_SMN_C2PMSG_59__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_59__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_60 +#define MP0_SMN_C2PMSG_60__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_60__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_61 +#define MP0_SMN_C2PMSG_61__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_61__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_62 +#define MP0_SMN_C2PMSG_62__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_62__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_63 +#define MP0_SMN_C2PMSG_63__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_63__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_64 +#define MP0_SMN_C2PMSG_64__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_64__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_65 +#define MP0_SMN_C2PMSG_65__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_65__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_66 +#define MP0_SMN_C2PMSG_66__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_66__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_67 +#define MP0_SMN_C2PMSG_67__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_67__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_68 +#define MP0_SMN_C2PMSG_68__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_68__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_69 +#define MP0_SMN_C2PMSG_69__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_69__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_70 +#define MP0_SMN_C2PMSG_70__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_70__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_71 +#define MP0_SMN_C2PMSG_71__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_71__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_72 +#define MP0_SMN_C2PMSG_72__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_72__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_73 +#define MP0_SMN_C2PMSG_73__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_73__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_74 +#define MP0_SMN_C2PMSG_74__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_74__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_75 +#define MP0_SMN_C2PMSG_75__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_75__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_76 +#define MP0_SMN_C2PMSG_76__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_76__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_77 +#define MP0_SMN_C2PMSG_77__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_77__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_78 +#define MP0_SMN_C2PMSG_78__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_78__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_79 +#define MP0_SMN_C2PMSG_79__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_79__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_80 +#define MP0_SMN_C2PMSG_80__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_80__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_81 +#define MP0_SMN_C2PMSG_81__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_81__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_82 +#define MP0_SMN_C2PMSG_82__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_82__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_83 +#define MP0_SMN_C2PMSG_83__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_83__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_84 +#define MP0_SMN_C2PMSG_84__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_84__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_85 +#define MP0_SMN_C2PMSG_85__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_85__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_86 +#define MP0_SMN_C2PMSG_86__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_86__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_87 +#define MP0_SMN_C2PMSG_87__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_87__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_88 +#define MP0_SMN_C2PMSG_88__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_88__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_89 +#define MP0_SMN_C2PMSG_89__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_89__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_90 +#define MP0_SMN_C2PMSG_90__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_90__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_91 +#define MP0_SMN_C2PMSG_91__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_91__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_92 +#define MP0_SMN_C2PMSG_92__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_92__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_93 +#define MP0_SMN_C2PMSG_93__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_93__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_94 +#define MP0_SMN_C2PMSG_94__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_94__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_95 +#define MP0_SMN_C2PMSG_95__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_95__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_96 +#define MP0_SMN_C2PMSG_96__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_96__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_97 +#define MP0_SMN_C2PMSG_97__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_97__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_98 +#define MP0_SMN_C2PMSG_98__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_98__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_99 +#define MP0_SMN_C2PMSG_99__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_99__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_100 +#define MP0_SMN_C2PMSG_100__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_100__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_101 +#define MP0_SMN_C2PMSG_101__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_101__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_102 +#define MP0_SMN_C2PMSG_102__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_102__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_C2PMSG_103 +#define MP0_SMN_C2PMSG_103__CONTENT__SHIFT 0x0 +#define MP0_SMN_C2PMSG_103__CONTENT_MASK 0xFFFFFFFFL +//MP0_SMN_IH_CREDIT +#define MP0_SMN_IH_CREDIT__CREDIT_VALUE__SHIFT 0x0 +#define MP0_SMN_IH_CREDIT__CLIENT_ID__SHIFT 0x10 +#define MP0_SMN_IH_CREDIT__CREDIT_VALUE_MASK 0x00000003L +#define MP0_SMN_IH_CREDIT__CLIENT_ID_MASK 0x00FF0000L +//MP0_SMN_IH_SW_INT +#define MP0_SMN_IH_SW_INT__ID__SHIFT 0x0 +#define MP0_SMN_IH_SW_INT__VALID__SHIFT 0x8 +#define MP0_SMN_IH_SW_INT__ID_MASK 0x000000FFL +#define MP0_SMN_IH_SW_INT__VALID_MASK 0x00000100L + + +// addressBlock: mp_SmuMp1_SmnDec +//MP1_SMN_C2PMSG_32 +#define MP1_SMN_C2PMSG_32__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_32__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_33 +#define MP1_SMN_C2PMSG_33__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_33__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_34 +#define MP1_SMN_C2PMSG_34__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_34__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_35 +#define MP1_SMN_C2PMSG_35__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_35__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_36 +#define MP1_SMN_C2PMSG_36__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_36__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_37 +#define MP1_SMN_C2PMSG_37__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_37__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_38 +#define MP1_SMN_C2PMSG_38__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_38__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_39 +#define MP1_SMN_C2PMSG_39__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_39__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_40 +#define MP1_SMN_C2PMSG_40__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_40__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_41 +#define MP1_SMN_C2PMSG_41__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_41__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_42 +#define MP1_SMN_C2PMSG_42__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_42__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_43 +#define MP1_SMN_C2PMSG_43__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_43__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_44 +#define MP1_SMN_C2PMSG_44__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_44__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_45 +#define MP1_SMN_C2PMSG_45__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_45__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_46 +#define MP1_SMN_C2PMSG_46__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_46__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_47 +#define MP1_SMN_C2PMSG_47__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_47__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_48 +#define MP1_SMN_C2PMSG_48__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_48__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_49 +#define MP1_SMN_C2PMSG_49__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_49__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_50 +#define MP1_SMN_C2PMSG_50__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_50__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_51 +#define MP1_SMN_C2PMSG_51__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_51__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_52 +#define MP1_SMN_C2PMSG_52__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_52__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_53 +#define MP1_SMN_C2PMSG_53__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_53__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_54 +#define MP1_SMN_C2PMSG_54__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_54__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_55 +#define MP1_SMN_C2PMSG_55__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_55__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_56 +#define MP1_SMN_C2PMSG_56__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_56__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_57 +#define MP1_SMN_C2PMSG_57__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_57__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_58 +#define MP1_SMN_C2PMSG_58__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_58__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_59 +#define MP1_SMN_C2PMSG_59__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_59__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_60 +#define MP1_SMN_C2PMSG_60__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_60__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_61 +#define MP1_SMN_C2PMSG_61__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_61__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_62 +#define MP1_SMN_C2PMSG_62__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_62__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_63 +#define MP1_SMN_C2PMSG_63__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_63__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_64 +#define MP1_SMN_C2PMSG_64__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_64__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_65 +#define MP1_SMN_C2PMSG_65__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_65__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_66 +#define MP1_SMN_C2PMSG_66__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_66__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_67 +#define MP1_SMN_C2PMSG_67__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_67__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_68 +#define MP1_SMN_C2PMSG_68__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_68__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_69 +#define MP1_SMN_C2PMSG_69__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_69__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_70 +#define MP1_SMN_C2PMSG_70__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_70__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_71 +#define MP1_SMN_C2PMSG_71__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_71__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_72 +#define MP1_SMN_C2PMSG_72__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_72__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_73 +#define MP1_SMN_C2PMSG_73__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_73__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_74 +#define MP1_SMN_C2PMSG_74__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_74__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_75 +#define MP1_SMN_C2PMSG_75__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_75__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_76 +#define MP1_SMN_C2PMSG_76__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_76__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_77 +#define MP1_SMN_C2PMSG_77__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_77__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_78 +#define MP1_SMN_C2PMSG_78__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_78__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_79 +#define MP1_SMN_C2PMSG_79__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_79__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_80 +#define MP1_SMN_C2PMSG_80__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_80__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_81 +#define MP1_SMN_C2PMSG_81__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_81__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_82 +#define MP1_SMN_C2PMSG_82__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_82__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_83 +#define MP1_SMN_C2PMSG_83__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_83__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_84 +#define MP1_SMN_C2PMSG_84__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_84__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_85 +#define MP1_SMN_C2PMSG_85__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_85__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_86 +#define MP1_SMN_C2PMSG_86__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_86__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_87 +#define MP1_SMN_C2PMSG_87__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_87__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_88 +#define MP1_SMN_C2PMSG_88__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_88__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_89 +#define MP1_SMN_C2PMSG_89__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_89__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_90 +#define MP1_SMN_C2PMSG_90__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_90__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_91 +#define MP1_SMN_C2PMSG_91__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_91__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_92 +#define MP1_SMN_C2PMSG_92__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_92__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_93 +#define MP1_SMN_C2PMSG_93__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_93__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_94 +#define MP1_SMN_C2PMSG_94__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_94__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_95 +#define MP1_SMN_C2PMSG_95__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_95__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_96 +#define MP1_SMN_C2PMSG_96__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_96__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_97 +#define MP1_SMN_C2PMSG_97__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_97__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_98 +#define MP1_SMN_C2PMSG_98__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_98__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_99 +#define MP1_SMN_C2PMSG_99__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_99__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_100 +#define MP1_SMN_C2PMSG_100__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_100__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_101 +#define MP1_SMN_C2PMSG_101__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_101__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_102 +#define MP1_SMN_C2PMSG_102__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_102__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_C2PMSG_103 +#define MP1_SMN_C2PMSG_103__CONTENT__SHIFT 0x0 +#define MP1_SMN_C2PMSG_103__CONTENT_MASK 0xFFFFFFFFL +//MP1_SMN_IH_CREDIT +#define MP1_SMN_IH_CREDIT__CREDIT_VALUE__SHIFT 0x0 +#define MP1_SMN_IH_CREDIT__CLIENT_ID__SHIFT 0x10 +#define MP1_SMN_IH_CREDIT__CREDIT_VALUE_MASK 0x00000003L +#define MP1_SMN_IH_CREDIT__CLIENT_ID_MASK 0x00FF0000L +//MP1_SMN_IH_SW_INT +#define MP1_SMN_IH_SW_INT__ID__SHIFT 0x0 +#define MP1_SMN_IH_SW_INT__VALID__SHIFT 0x8 +#define MP1_SMN_IH_SW_INT__ID_MASK 0x000000FFL +#define MP1_SMN_IH_SW_INT__VALID_MASK 0x00000100L +//MP1_SMN_FPS_CNT +#define MP1_SMN_FPS_CNT__COUNT__SHIFT 0x0 +#define MP1_SMN_FPS_CNT__COUNT_MASK 0xFFFFFFFFL + + +// addressBlock: mp_SmuMp0Pub_CruDec +//MP0_IH_CREDIT +#define MP0_IH_CREDIT__CREDIT_VALUE__SHIFT 0x0 +#define MP0_IH_CREDIT__CLIENT_ID__SHIFT 0x10 +#define MP0_IH_CREDIT__CREDIT_VALUE_MASK 0x00000003L +#define MP0_IH_CREDIT__CLIENT_ID_MASK 0x00FF0000L +//MP0_IH_SW_INT +#define MP0_IH_SW_INT__ID__SHIFT 0x0 +#define MP0_IH_SW_INT__VALID__SHIFT 0x8 +#define MP0_IH_SW_INT__ID_MASK 0x000000FFL +#define MP0_IH_SW_INT__VALID_MASK 0x00000100L +//MP0_IH_SW_INT_CTRL +#define MP0_IH_SW_INT_CTRL__INT_MASK__SHIFT 0x0 +#define MP0_IH_SW_INT_CTRL__INT_ACK__SHIFT 0x8 +#define MP0_IH_SW_INT_CTRL__INT_MASK_MASK 0x00000001L +#define MP0_IH_SW_INT_CTRL__INT_ACK_MASK 0x00000100L + + +// addressBlock: mp_SmuMp1Pub_CruDec +//MP1_FIRMWARE_FLAGS +#define MP1_FIRMWARE_FLAGS__INTERRUPTS_ENABLED__SHIFT 0x0 +#define MP1_FIRMWARE_FLAGS__RESERVED__SHIFT 0x1 +#define MP1_FIRMWARE_FLAGS__INTERRUPTS_ENABLED_MASK 0x00000001L +#define MP1_FIRMWARE_FLAGS__RESERVED_MASK 0xFFFFFFFEL +//MP1_C2PMSG_0 +#define MP1_C2PMSG_0__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_0__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_1 +#define MP1_C2PMSG_1__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_1__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_2 +#define MP1_C2PMSG_2__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_2__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_3 +#define MP1_C2PMSG_3__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_3__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_4 +#define MP1_C2PMSG_4__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_4__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_5 +#define MP1_C2PMSG_5__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_5__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_6 +#define MP1_C2PMSG_6__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_6__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_7 +#define MP1_C2PMSG_7__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_7__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_8 +#define MP1_C2PMSG_8__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_8__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_9 +#define MP1_C2PMSG_9__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_9__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_10 +#define MP1_C2PMSG_10__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_10__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_11 +#define MP1_C2PMSG_11__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_11__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_12 +#define MP1_C2PMSG_12__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_12__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_13 +#define MP1_C2PMSG_13__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_13__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_14 +#define MP1_C2PMSG_14__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_14__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_15 +#define MP1_C2PMSG_15__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_15__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_16 +#define MP1_C2PMSG_16__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_16__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_17 +#define MP1_C2PMSG_17__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_17__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_18 +#define MP1_C2PMSG_18__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_18__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_19 +#define MP1_C2PMSG_19__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_19__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_20 +#define MP1_C2PMSG_20__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_20__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_21 +#define MP1_C2PMSG_21__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_21__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_22 +#define MP1_C2PMSG_22__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_22__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_23 +#define MP1_C2PMSG_23__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_23__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_24 +#define MP1_C2PMSG_24__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_24__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_25 +#define MP1_C2PMSG_25__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_25__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_26 +#define MP1_C2PMSG_26__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_26__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_27 +#define MP1_C2PMSG_27__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_27__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_28 +#define MP1_C2PMSG_28__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_28__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_29 +#define MP1_C2PMSG_29__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_29__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_30 +#define MP1_C2PMSG_30__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_30__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_31 +#define MP1_C2PMSG_31__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_31__CONTENT_MASK 0xFFFFFFFFL +//MP1_P2CMSG_0 +#define MP1_P2CMSG_0__CONTENT__SHIFT 0x0 +#define MP1_P2CMSG_0__CONTENT_MASK 0xFFFFFFFFL +//MP1_P2CMSG_1 +#define MP1_P2CMSG_1__CONTENT__SHIFT 0x0 +#define MP1_P2CMSG_1__CONTENT_MASK 0xFFFFFFFFL +//MP1_P2CMSG_2 +#define MP1_P2CMSG_2__CONTENT__SHIFT 0x0 +#define MP1_P2CMSG_2__CONTENT_MASK 0xFFFFFFFFL +//MP1_P2CMSG_3 +#define MP1_P2CMSG_3__CONTENT__SHIFT 0x0 +#define MP1_P2CMSG_3__CONTENT_MASK 0xFFFFFFFFL +//MP1_P2CMSG_INTEN +#define MP1_P2CMSG_INTEN__INTEN__SHIFT 0x0 +#define MP1_P2CMSG_INTEN__INTEN_MASK 0x0000000FL +//MP1_P2CMSG_INTSTS +#define MP1_P2CMSG_INTSTS__INTSTS0__SHIFT 0x0 +#define MP1_P2CMSG_INTSTS__INTSTS1__SHIFT 0x1 +#define MP1_P2CMSG_INTSTS__INTSTS2__SHIFT 0x2 +#define MP1_P2CMSG_INTSTS__INTSTS3__SHIFT 0x3 +#define MP1_P2CMSG_INTSTS__INTSTS0_MASK 0x00000001L +#define MP1_P2CMSG_INTSTS__INTSTS1_MASK 0x00000002L +#define MP1_P2CMSG_INTSTS__INTSTS2_MASK 0x00000004L +#define MP1_P2CMSG_INTSTS__INTSTS3_MASK 0x00000008L +//MP1_C2PMSG_32 +#define MP1_C2PMSG_32__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_32__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_33 +#define MP1_C2PMSG_33__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_33__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_34 +#define MP1_C2PMSG_34__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_34__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_35 +#define MP1_C2PMSG_35__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_35__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_36 +#define MP1_C2PMSG_36__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_36__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_37 +#define MP1_C2PMSG_37__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_37__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_38 +#define MP1_C2PMSG_38__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_38__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_39 +#define MP1_C2PMSG_39__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_39__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_40 +#define MP1_C2PMSG_40__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_40__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_41 +#define MP1_C2PMSG_41__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_41__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_42 +#define MP1_C2PMSG_42__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_42__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_43 +#define MP1_C2PMSG_43__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_43__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_44 +#define MP1_C2PMSG_44__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_44__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_45 +#define MP1_C2PMSG_45__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_45__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_46 +#define MP1_C2PMSG_46__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_46__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_47 +#define MP1_C2PMSG_47__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_47__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_48 +#define MP1_C2PMSG_48__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_48__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_49 +#define MP1_C2PMSG_49__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_49__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_50 +#define MP1_C2PMSG_50__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_50__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_51 +#define MP1_C2PMSG_51__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_51__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_52 +#define MP1_C2PMSG_52__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_52__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_53 +#define MP1_C2PMSG_53__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_53__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_54 +#define MP1_C2PMSG_54__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_54__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_55 +#define MP1_C2PMSG_55__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_55__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_56 +#define MP1_C2PMSG_56__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_56__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_57 +#define MP1_C2PMSG_57__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_57__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_58 +#define MP1_C2PMSG_58__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_58__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_59 +#define MP1_C2PMSG_59__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_59__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_60 +#define MP1_C2PMSG_60__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_60__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_61 +#define MP1_C2PMSG_61__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_61__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_62 +#define MP1_C2PMSG_62__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_62__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_63 +#define MP1_C2PMSG_63__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_63__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_64 +#define MP1_C2PMSG_64__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_64__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_65 +#define MP1_C2PMSG_65__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_65__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_66 +#define MP1_C2PMSG_66__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_66__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_67 +#define MP1_C2PMSG_67__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_67__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_68 +#define MP1_C2PMSG_68__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_68__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_69 +#define MP1_C2PMSG_69__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_69__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_70 +#define MP1_C2PMSG_70__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_70__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_71 +#define MP1_C2PMSG_71__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_71__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_72 +#define MP1_C2PMSG_72__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_72__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_73 +#define MP1_C2PMSG_73__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_73__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_74 +#define MP1_C2PMSG_74__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_74__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_75 +#define MP1_C2PMSG_75__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_75__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_76 +#define MP1_C2PMSG_76__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_76__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_77 +#define MP1_C2PMSG_77__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_77__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_78 +#define MP1_C2PMSG_78__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_78__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_79 +#define MP1_C2PMSG_79__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_79__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_80 +#define MP1_C2PMSG_80__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_80__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_81 +#define MP1_C2PMSG_81__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_81__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_82 +#define MP1_C2PMSG_82__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_82__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_83 +#define MP1_C2PMSG_83__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_83__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_84 +#define MP1_C2PMSG_84__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_84__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_85 +#define MP1_C2PMSG_85__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_85__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_86 +#define MP1_C2PMSG_86__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_86__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_87 +#define MP1_C2PMSG_87__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_87__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_88 +#define MP1_C2PMSG_88__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_88__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_89 +#define MP1_C2PMSG_89__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_89__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_90 +#define MP1_C2PMSG_90__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_90__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_91 +#define MP1_C2PMSG_91__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_91__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_92 +#define MP1_C2PMSG_92__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_92__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_93 +#define MP1_C2PMSG_93__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_93__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_94 +#define MP1_C2PMSG_94__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_94__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_95 +#define MP1_C2PMSG_95__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_95__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_96 +#define MP1_C2PMSG_96__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_96__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_97 +#define MP1_C2PMSG_97__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_97__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_98 +#define MP1_C2PMSG_98__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_98__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_99 +#define MP1_C2PMSG_99__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_99__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_100 +#define MP1_C2PMSG_100__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_100__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_101 +#define MP1_C2PMSG_101__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_101__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_102 +#define MP1_C2PMSG_102__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_102__CONTENT_MASK 0xFFFFFFFFL +//MP1_C2PMSG_103 +#define MP1_C2PMSG_103__CONTENT__SHIFT 0x0 +#define MP1_C2PMSG_103__CONTENT_MASK 0xFFFFFFFFL +//MP1_IH_CREDIT +#define MP1_IH_CREDIT__CREDIT_VALUE__SHIFT 0x0 +#define MP1_IH_CREDIT__CLIENT_ID__SHIFT 0x10 +#define MP1_IH_CREDIT__CREDIT_VALUE_MASK 0x00000003L +#define MP1_IH_CREDIT__CLIENT_ID_MASK 0x00FF0000L +//MP1_IH_SW_INT +#define MP1_IH_SW_INT__ID__SHIFT 0x0 +#define MP1_IH_SW_INT__VALID__SHIFT 0x8 +#define MP1_IH_SW_INT__ID_MASK 0x000000FFL +#define MP1_IH_SW_INT__VALID_MASK 0x00000100L +//MP1_IH_SW_INT_CTRL +#define MP1_IH_SW_INT_CTRL__INT_MASK__SHIFT 0x0 +#define MP1_IH_SW_INT_CTRL__INT_ACK__SHIFT 0x8 +#define MP1_IH_SW_INT_CTRL__INT_MASK_MASK 0x00000001L +#define MP1_IH_SW_INT_CTRL__INT_ACK_MASK 0x00000100L +//MP1_FPS_CNT +#define MP1_FPS_CNT__COUNT__SHIFT 0x0 +#define MP1_FPS_CNT__COUNT_MASK 0xFFFFFFFFL + + +#endif diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index bba1291ae405..0de4e37fe7da 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -310,6 +310,7 @@ struct amd_pm_funcs { int (*set_asic_baco_state)(void *handle, int state); int (*get_ppfeature_status)(void *handle, char *buf); int (*set_ppfeature_status)(void *handle, uint64_t ppfeature_masks); + int (*asic_reset_mode_2)(void *handle); }; #endif diff --git a/drivers/gpu/drm/amd/include/navi12_ip_offset.h b/drivers/gpu/drm/amd/include/navi12_ip_offset.h index 229e8fddfcc1..6c2cc6296c06 100644 --- a/drivers/gpu/drm/amd/include/navi12_ip_offset.h +++ b/drivers/gpu/drm/amd/include/navi12_ip_offset.h @@ -18,8 +18,8 @@ * 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. */ -#ifndef _navi10_ip_offset_HEADER -#define _navi10_ip_offset_HEADER +#ifndef _navi12_ip_offset_HEADER +#define _navi12_ip_offset_HEADER #define MAX_INSTANCE 7 #define MAX_SEGMENT 5 diff --git a/drivers/gpu/drm/amd/powerplay/Makefile b/drivers/gpu/drm/amd/powerplay/Makefile index e05a7e3d6d8d..390345f2d601 100644 --- a/drivers/gpu/drm/amd/powerplay/Makefile +++ b/drivers/gpu/drm/amd/powerplay/Makefile @@ -35,7 +35,7 @@ AMD_POWERPLAY = $(addsuffix /Makefile,$(addprefix $(FULL_AMD_PATH)/powerplay/,$( include $(AMD_POWERPLAY) -POWER_MGR = amd_powerplay.o amdgpu_smu.o smu_v11_0.o vega20_ppt.o arcturus_ppt.o navi10_ppt.o +POWER_MGR = amd_powerplay.o amdgpu_smu.o smu_v11_0.o smu_v12_0.o vega20_ppt.o arcturus_ppt.o navi10_ppt.o renoir_ppt.o AMD_PP_POWER = $(addprefix $(AMD_PP_PATH)/,$(POWER_MGR)) diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c index 2e3d9ef625bf..7ef202761998 100644 --- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c @@ -1508,6 +1508,26 @@ static int pp_set_ppfeature_status(void *handle, uint64_t ppfeature_masks) return ret; } +static int pp_asic_reset_mode_2(void *handle) +{ + struct pp_hwmgr *hwmgr = handle; + int ret = 0; + + if (!hwmgr || !hwmgr->pm_en) + return -EINVAL; + + if (hwmgr->hwmgr_func->asic_reset == NULL) { + pr_info_ratelimited("%s was not implemented.\n", __func__); + return -EINVAL; + } + + mutex_lock(&hwmgr->smu_lock); + ret = hwmgr->hwmgr_func->asic_reset(hwmgr, SMU_ASIC_RESET_MODE_2); + mutex_unlock(&hwmgr->smu_lock); + + return ret; +} + static const struct amd_pm_funcs pp_dpm_funcs = { .load_firmware = pp_dpm_load_fw, .wait_for_fw_loading_complete = pp_dpm_fw_loading_complete, @@ -1564,4 +1584,5 @@ static const struct amd_pm_funcs pp_dpm_funcs = { .set_asic_baco_state = pp_set_asic_baco_state, .get_ppfeature_status = pp_get_ppfeature_status, .set_ppfeature_status = pp_set_ppfeature_status, + .asic_reset_mode_2 = pp_asic_reset_mode_2, }; diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c index dd274922ed8a..99924d3538c2 100644 --- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c @@ -27,6 +27,7 @@ #include "amdgpu_smu.h" #include "soc15_common.h" #include "smu_v11_0.h" +#include "smu_v12_0.h" #include "atom.h" #include "amd_pcie.h" @@ -413,6 +414,9 @@ int smu_common_read_sensor(struct smu_context *smu, enum amd_pp_sensors sensor, struct smu_power_gate *power_gate = &smu_power->power_gate; int ret = 0; + if(!data || !size) + return -EINVAL; + switch (sensor) { case AMDGPU_PP_SENSOR_STABLE_PSTATE_SCLK: *((uint32_t *)data) = smu->pstate_sclk; @@ -626,10 +630,14 @@ int smu_feature_update_enable_state(struct smu_context *smu, uint64_t feature_ma int smu_feature_is_enabled(struct smu_context *smu, enum smu_feature_mask mask) { + struct amdgpu_device *adev = smu->adev; struct smu_feature *feature = &smu->smu_feature; int feature_id; int ret = 0; + if (adev->flags & AMD_IS_APU) + return 0; + feature_id = smu_feature_get_index(smu, mask); if (feature_id < 0) return 0; @@ -732,6 +740,11 @@ static int smu_set_funcs(struct amdgpu_device *adev) smu->od_enabled = true; smu_v11_0_set_smu_funcs(smu); break; + case CHIP_RENOIR: + if (adev->pm.pp_feature & PP_OVERDRIVE_MASK) + smu->od_enabled = true; + smu_v12_0_set_smu_funcs(smu); + break; default: return -EINVAL; } @@ -758,6 +771,7 @@ static int smu_late_init(void *handle) if (!smu->pm_enabled) return 0; + mutex_lock(&smu->mutex); smu_handle_task(&adev->smu, smu->smu_dpm.dpm_level, @@ -1028,6 +1042,9 @@ static int smu_smc_table_hw_init(struct smu_context *smu, struct amdgpu_device *adev = smu->adev; int ret; + if (adev->flags & AMD_IS_APU) + return 0; + if (smu_is_dpm_running(smu) && adev->in_suspend) { pr_info("dpm has been enabled\n"); return 0; @@ -1133,7 +1150,7 @@ static int smu_smc_table_hw_init(struct smu_context *smu, * type of clks. */ if (initialize) { - ret = smu_populate_smc_pptable(smu); + ret = smu_populate_smc_tables(smu); if (ret) return ret; @@ -1250,6 +1267,11 @@ static int smu_hw_init(void *handle) return ret; } + if (adev->flags & AMD_IS_APU) { + smu_powergate_sdma(&adev->smu, false); + smu_powergate_vcn(&adev->smu, false); + } + if (!smu->pm_enabled) return 0; @@ -1297,6 +1319,11 @@ static int smu_hw_fini(void *handle) struct smu_table_context *table_context = &smu->smu_table; int ret = 0; + if (adev->flags & AMD_IS_APU) { + smu_powergate_sdma(&adev->smu, true); + smu_powergate_vcn(&adev->smu, true); + } + kfree(table_context->driver_pptable); table_context->driver_pptable = NULL; @@ -1787,3 +1814,12 @@ const struct amdgpu_ip_block_version smu_v11_0_ip_block = .rev = 0, .funcs = &smu_ip_funcs, }; + +const struct amdgpu_ip_block_version smu_v12_0_ip_block = +{ + .type = AMD_IP_BLOCK_TYPE_SMC, + .major = 12, + .minor = 0, + .rev = 0, + .funcs = &smu_ip_funcs, +}; diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c index cff3777ae5aa..dd6563358e8e 100644 --- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c @@ -51,6 +51,15 @@ #define SMU_FEATURES_HIGH_MASK 0xFFFFFFFF00000000 #define SMU_FEATURES_HIGH_SHIFT 32 +#define SMC_DPM_FEATURE ( \ + FEATURE_DPM_PREFETCHER_MASK | \ + FEATURE_DPM_GFXCLK_MASK | \ + FEATURE_DPM_UCLK_MASK | \ + FEATURE_DPM_SOCCLK_MASK | \ + FEATURE_DPM_MP0CLK_MASK | \ + FEATURE_DPM_FCLK_MASK | \ + FEATURE_DPM_XGMI_MASK) + /* possible frequency drift (1Mhz) */ #define EPSILON 1 @@ -172,7 +181,7 @@ static struct smu_11_0_cmn2aisc_mapping arcturus_workload_map[PP_SMC_POWER_PROFI WORKLOAD_MAP(PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT, WORKLOAD_PPLIB_DEFAULT_BIT), WORKLOAD_MAP(PP_SMC_POWER_PROFILE_POWERSAVING, WORKLOAD_PPLIB_POWER_SAVING_BIT), WORKLOAD_MAP(PP_SMC_POWER_PROFILE_VIDEO, WORKLOAD_PPLIB_VIDEO_BIT), - WORKLOAD_MAP(PP_SMC_POWER_PROFILE_COMPUTE, WORKLOAD_PPLIB_CUSTOM_BIT), + WORKLOAD_MAP(PP_SMC_POWER_PROFILE_COMPUTE, WORKLOAD_PPLIB_COMPUTE_BIT), WORKLOAD_MAP(PP_SMC_POWER_PROFILE_CUSTOM, WORKLOAD_PPLIB_CUSTOM_BIT), }; @@ -215,7 +224,6 @@ static int arcturus_get_smu_feature_index(struct smu_context *smc, uint32_t inde mapping = arcturus_feature_mask_map[index]; if (!(mapping.valid_mapping)) { - pr_warn("Unsupported SMU feature: %d\n", index); return -EINVAL; } @@ -455,7 +463,8 @@ static int arcturus_set_default_dpm_table(struct smu_context *smu) return ret; } } else { - single_dpm_table->count = 0; + single_dpm_table->count = 1; + single_dpm_table->dpm_levels[0].value = smu->smu_table.boot_values.fclk / 100; } arcturus_init_single_dpm_state(&(single_dpm_table->dpm_state)); @@ -869,23 +878,14 @@ static int arcturus_force_clk_levels(struct smu_context *smu, return ret; } -static const struct smu_temperature_range arcturus_thermal_policy[] = -{ - {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, - { 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000}, -}; - static int arcturus_get_thermal_temperature_range(struct smu_context *smu, struct smu_temperature_range *range) { - PPTable_t *pptable = smu->smu_table.driver_pptable; if (!range) return -EINVAL; - memcpy(range, &arcturus_thermal_policy[0], sizeof(struct smu_temperature_range)); - range->max = pptable->TedgeLimit * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; range->edge_emergency_max = (pptable->TedgeLimit + CTF_OFFSET_EDGE) * @@ -899,7 +899,6 @@ static int arcturus_get_thermal_temperature_range(struct smu_context *smu, range->mem_emergency_max = (pptable->TmemLimit + CTF_OFFSET_HBM)* SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - return 0; } @@ -1014,6 +1013,9 @@ static int arcturus_read_sensor(struct smu_context *smu, PPTable_t *pptable = table_context->driver_pptable; int ret = 0; + if (!data || !size) + return -EINVAL; + switch (sensor) { case AMDGPU_PP_SENSOR_MAX_FAN_RPM: *(uint32_t *)data = pptable->FanMaximumRpm; @@ -1038,7 +1040,7 @@ static int arcturus_read_sensor(struct smu_context *smu, *size = 4; break; default: - return -EINVAL; + ret = smu_smc_read_sensor(smu, sensor, data, size); } return ret; @@ -1874,6 +1876,17 @@ static void arcturus_dump_pptable(struct smu_context *smu) } +static bool arcturus_is_dpm_running(struct smu_context *smu) +{ + int ret = 0; + uint32_t feature_mask[2]; + unsigned long feature_enabled; + ret = smu_feature_get_enabled_mask(smu, feature_mask, 2); + feature_enabled = (unsigned long)((uint64_t)feature_mask[0] | + ((uint64_t)feature_mask[1] << 32)); + return !!(feature_enabled & SMC_DPM_FEATURE); +} + static const struct pptable_funcs arcturus_ppt_funcs = { /* translate smu index into arcturus specific index */ .get_smu_msg_index = arcturus_get_smu_msg_index, @@ -1911,6 +1924,7 @@ static const struct pptable_funcs arcturus_ppt_funcs = { /* debug (internal used) */ .dump_pptable = arcturus_dump_pptable, .get_power_limit = arcturus_get_power_limit, + .is_dpm_running = arcturus_is_dpm_running, }; void arcturus_set_ppt_funcs(struct smu_context *smu) @@ -1918,6 +1932,5 @@ void arcturus_set_ppt_funcs(struct smu_context *smu) struct smu_table_context *smu_table = &smu->smu_table; smu->ppt_funcs = &arcturus_ppt_funcs; - smu->smc_if_version = SMU11_DRIVER_IF_VERSION; smu_table->table_count = TABLE_COUNT; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c index 18e780f566fa..1115761982a7 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c @@ -1311,6 +1311,12 @@ static int smu10_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, uin return 0; } +static int smu10_asic_reset(struct pp_hwmgr *hwmgr, enum SMU_ASIC_RESET_MODE mode) +{ + return smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_DeviceDriverReset, + mode); +} static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .backend_init = smu10_hwmgr_backend_init, @@ -1355,6 +1361,7 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .set_hard_min_fclk_by_freq = smu10_set_hard_min_fclk_by_freq, .get_power_profile_mode = smu10_get_power_profile_mode, .set_power_profile_mode = smu10_set_power_profile_mode, + .asic_reset = smu10_asic_reset, }; int smu10_init_function_pointers(struct pp_hwmgr *hwmgr) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index 487aeee1cf8a..34f95e0e3ea4 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -2956,9 +2956,10 @@ static int smu7_apply_state_adjust_rules(struct pp_hwmgr *hwmgr, if (hwmgr->display_config->num_display == 0) disable_mclk_switching = false; else - disable_mclk_switching = ((1 < hwmgr->display_config->num_display) || - disable_mclk_switching_for_frame_lock || - smu7_vblank_too_short(hwmgr, hwmgr->display_config->min_vblank_time)); + disable_mclk_switching = ((1 < hwmgr->display_config->num_display) && + !hwmgr->display_config->multi_monitor_in_sync) || + disable_mclk_switching_for_frame_lock || + smu7_vblank_too_short(hwmgr, hwmgr->display_config->min_vblank_time); sclk = smu7_ps->performance_levels[0].engine_clock; mclk = smu7_ps->performance_levels[0].memory_clock; @@ -4068,6 +4069,11 @@ static int smu7_program_display_gap(struct pp_hwmgr *hwmgr) data->frame_time_x2 = frame_time_in_us * 2 / 100; + if (data->frame_time_x2 < 280) { + pr_debug("%s: enforce minimal VBITimeout: %d -> 280\n", __func__, data->frame_time_x2); + data->frame_time_x2 = 280; + } + display_gap2 = pre_vbi_time_in_us * (ref_clock / 100); cgs_write_ind_register(hwmgr->device, CGS_IND_REG__SMC, ixCG_DISPLAY_GAP_CNTL2, display_gap2); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c index 948c54cb9c5d..d08493b67b67 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c @@ -3220,7 +3220,8 @@ static int vega10_apply_state_adjust_rules(struct pp_hwmgr *hwmgr, if (hwmgr->display_config->num_display == 0) disable_mclk_switching = false; else - disable_mclk_switching = (hwmgr->display_config->num_display > 1) || + disable_mclk_switching = ((1 < hwmgr->display_config->num_display) && + !hwmgr->display_config->multi_monitor_in_sync) || disable_mclk_switching_for_frame_lock || disable_mclk_switching_for_vr || force_mclk_high; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index 0516c294b377..98a6f5305974 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -2101,7 +2101,11 @@ static int vega20_get_gpu_power(struct pp_hwmgr *hwmgr, if (ret) return ret; - *query = metrics_table.CurrSocketPower << 8; + /* For the 40.46 release, they changed the value name */ + if (hwmgr->smu_version == 0x282e00) + *query = metrics_table.AverageSocketPower << 8; + else + *query = metrics_table.CurrSocketPower << 8; return ret; } diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h index f813072ab9e4..c42691a9afd3 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h @@ -222,12 +222,17 @@ struct smu_bios_boot_up_values uint16_t vdd_gfx; uint8_t cooling_id; uint32_t pp_table_id; + uint32_t format_revision; + uint32_t content_revision; + uint32_t fclk; }; enum smu_table_id { SMU_TABLE_PPTABLE = 0, SMU_TABLE_WATERMARKS, + SMU_TABLE_CUSTOM_DPM, + SMU_TABLE_DPMCLOCKS, SMU_TABLE_AVFS, SMU_TABLE_AVFS_PSM_DEBUG, SMU_TABLE_AVFS_FUSE_OVERRIDE, @@ -248,6 +253,7 @@ struct smu_table_context void *hardcode_pptable; unsigned long metrics_time; void *metrics_table; + void *clocks_table; void *max_sustainable_clocks; struct smu_bios_boot_up_values boot_values; @@ -468,8 +474,11 @@ struct smu_funcs int (*get_clk_info_from_vbios)(struct smu_context *smu); int (*check_pptable)(struct smu_context *smu); int (*parse_pptable)(struct smu_context *smu); - int (*populate_smc_pptable)(struct smu_context *smu); + int (*populate_smc_tables)(struct smu_context *smu); int (*check_fw_version)(struct smu_context *smu); + int (*powergate_sdma)(struct smu_context *smu, bool gate); + int (*powergate_vcn)(struct smu_context *smu, bool gate); + int (*set_gfx_cgpg)(struct smu_context *smu, bool enable); int (*write_pptable)(struct smu_context *smu); int (*set_min_dcef_deep_sleep)(struct smu_context *smu); int (*set_tool_table_location)(struct smu_context *smu); @@ -546,6 +555,12 @@ struct smu_funcs ((smu)->funcs->check_fw_status ? (smu)->funcs->check_fw_status((smu)) : 0) #define smu_setup_pptable(smu) \ ((smu)->funcs->setup_pptable ? (smu)->funcs->setup_pptable((smu)) : 0) +#define smu_powergate_sdma(smu, gate) \ + ((smu)->funcs->powergate_sdma ? (smu)->funcs->powergate_sdma((smu), (gate)) : 0) +#define smu_powergate_vcn(smu, gate) \ + ((smu)->funcs->powergate_vcn ? (smu)->funcs->powergate_vcn((smu), (gate)) : 0) +#define smu_set_gfx_cgpg(smu, enabled) \ + ((smu)->funcs->set_gfx_cgpg ? (smu)->funcs->set_gfx_cgpg((smu), (enabled)) : 0) #define smu_get_vbios_bootup_values(smu) \ ((smu)->funcs->get_vbios_bootup_values ? (smu)->funcs->get_vbios_bootup_values((smu)) : 0) #define smu_get_clk_info_from_vbios(smu) \ @@ -554,8 +569,8 @@ struct smu_funcs ((smu)->funcs->check_pptable ? (smu)->funcs->check_pptable((smu)) : 0) #define smu_parse_pptable(smu) \ ((smu)->funcs->parse_pptable ? (smu)->funcs->parse_pptable((smu)) : 0) -#define smu_populate_smc_pptable(smu) \ - ((smu)->funcs->populate_smc_pptable ? (smu)->funcs->populate_smc_pptable((smu)) : 0) +#define smu_populate_smc_tables(smu) \ + ((smu)->funcs->populate_smc_tables ? (smu)->funcs->populate_smc_tables((smu)) : 0) #define smu_check_fw_version(smu) \ ((smu)->funcs->check_fw_version ? (smu)->funcs->check_fw_version((smu)) : 0) #define smu_write_pptable(smu) \ @@ -634,9 +649,9 @@ struct smu_funcs #define smu_start_thermal_control(smu) \ ((smu)->funcs->start_thermal_control? (smu)->funcs->start_thermal_control((smu)) : 0) #define smu_read_sensor(smu, sensor, data, size) \ - ((smu)->funcs->read_sensor? (smu)->funcs->read_sensor((smu), (sensor), (data), (size)) : 0) -#define smu_asic_read_sensor(smu, sensor, data, size) \ ((smu)->ppt_funcs->read_sensor? (smu)->ppt_funcs->read_sensor((smu), (sensor), (data), (size)) : 0) +#define smu_smc_read_sensor(smu, sensor, data, size) \ + ((smu)->funcs->read_sensor? (smu)->funcs->read_sensor((smu), (sensor), (data), (size)) : -EINVAL) #define smu_get_power_profile_mode(smu, buf) \ ((smu)->ppt_funcs->get_power_profile_mode ? (smu)->ppt_funcs->get_power_profile_mode((smu), buf) : 0) #define smu_set_power_profile_mode(smu, param, param_size) \ @@ -730,8 +745,6 @@ struct smu_funcs ((smu)->funcs->register_irq_handler ? (smu)->funcs->register_irq_handler(smu) : 0) #define smu_set_azalia_d3_pme(smu) \ ((smu)->funcs->set_azalia_d3_pme ? (smu)->funcs->set_azalia_d3_pme((smu)) : 0) -#define smu_get_uclk_dpm_states(smu, clocks_in_khz, num_states) \ - ((smu)->ppt_funcs->get_uclk_dpm_states ? (smu)->ppt_funcs->get_uclk_dpm_states((smu), (clocks_in_khz), (num_states)) : 0) #define smu_get_max_sustainable_clocks_by_dc(smu, max_clocks) \ ((smu)->funcs->get_max_sustainable_clocks_by_dc ? (smu)->funcs->get_max_sustainable_clocks_by_dc((smu), (max_clocks)) : 0) #define smu_get_uclk_dpm_states(smu, clocks_in_khz, num_states) \ @@ -754,6 +767,8 @@ extern int smu_get_atom_data_table(struct smu_context *smu, uint32_t table, extern const struct amd_ip_funcs smu_ip_funcs; extern const struct amdgpu_ip_block_version smu_v11_0_ip_block; +extern const struct amdgpu_ip_block_version smu_v12_0_ip_block; + extern int smu_feature_init_dpm(struct smu_context *smu); extern int smu_feature_is_enabled(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h index 07fd64aad2ae..abeff1570277 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h @@ -189,6 +189,14 @@ struct phm_vce_clock_voltage_dependency_table { struct phm_vce_clock_voltage_dependency_record entries[1]; }; + +enum SMU_ASIC_RESET_MODE +{ + SMU_ASIC_RESET_MODE_0, + SMU_ASIC_RESET_MODE_1, + SMU_ASIC_RESET_MODE_2, +}; + struct pp_smumgr_func { char *name; int (*smu_init)(struct pp_hwmgr *hwmgr); @@ -345,6 +353,7 @@ struct pp_hwmgr_func { int (*get_ppfeature_status)(struct pp_hwmgr *hwmgr, char *buf); int (*set_ppfeature_status)(struct pp_hwmgr *hwmgr, uint64_t ppfeature_masks); int (*set_mp1_state)(struct pp_hwmgr *hwmgr, enum pp_mp1_state mp1_state); + int (*asic_reset)(struct pp_hwmgr *hwmgr, enum SMU_ASIC_RESET_MODE mode); }; struct pp_table_func { diff --git a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h index 90879e4092a3..df4677da736c 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h +++ b/drivers/gpu/drm/amd/powerplay/inc/rv_ppsmc.h @@ -59,7 +59,7 @@ #define PPSMC_MSG_SetDriverDramAddrLow 0x1B #define PPSMC_MSG_TransferTableSmu2Dram 0x1C #define PPSMC_MSG_TransferTableDram2Smu 0x1D -#define PPSMC_MSG_ControlGfxRM 0x1E +#define PPSMC_MSG_DeviceDriverReset 0x1E #define PPSMC_MSG_SetGfxclkOverdriveByFreqVid 0x1F #define PPSMC_MSG_SetHardMinDcefclkByFreq 0x20 #define PPSMC_MSG_SetHardMinSocclkByFreq 0x21 diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h index 755d51f9c6a9..fdc6b7a57bc9 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h @@ -27,7 +27,9 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define SMU11_DRIVER_IF_VERSION 0x13 +// Be aware of that the version should be updated in +// smu_v11_0.h, rename is also needed. +// #define SMU11_DRIVER_IF_VERSION 0x13 #define PPTABLE_V20_SMU_VERSION 3 diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h index c7a7953b52b7..e02950b505fa 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h @@ -27,7 +27,7 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define SMU11_DRIVER_IF_VERSION 0x08 +//#define SMU11_DRIVER_IF_VERSION 0x09 #define PPTABLE_ARCTURUS_SMU_VERSION 4 @@ -691,7 +691,12 @@ typedef struct { uint16_t XgmiFclkFreq [NUM_XGMI_PSTATE_LEVELS]; uint16_t XgmiSocVoltage [NUM_XGMI_PSTATE_LEVELS]; - uint32_t BoardReserved[10]; + // GPIO pins for I2C communications with 2nd controller for Input Telemetry Sequence + uint8_t GpioI2cScl; // Serial Clock + uint8_t GpioI2cSda; // Serial Data + uint16_t GpioPadding; + + uint32_t BoardReserved[9]; // Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use @@ -738,7 +743,7 @@ typedef struct { uint32_t Padding[4]; // Padding - ignore - uint32_t MmHubPadding[7]; // SMU internal use + uint32_t MmHubPadding[8]; // SMU internal use } SmuMetrics_t; @@ -748,7 +753,7 @@ typedef struct { float avgPsmVoltage[75]; float minPsmVoltage[75]; - uint32_t MmHubPadding[3]; // SMU internal use + uint32_t MmHubPadding[8]; // SMU internal use } AvfsDebugTable_t; typedef struct { @@ -797,9 +802,9 @@ typedef struct { uint32_t P2VCharzFreq[AVFS_VOLTAGE_COUNT]; // in 10KHz units - uint32_t EnabledAvfsModules; + uint32_t EnabledAvfsModules[2]; - uint32_t MmHubPadding[7]; // SMU internal use + uint32_t MmHubPadding[8]; // SMU internal use } AvfsFuseOverride_t; /* NOT CURRENTLY USED diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h index adbbfebbb1e5..ac0120e384be 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_navi10.h @@ -26,7 +26,9 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define SMU11_DRIVER_IF_VERSION 0x33 +// Be aware of that the version should be updated in +// smu_v11_0.h, maybe rename is also needed. +// #define SMU11_DRIVER_IF_VERSION 0x33 #define PPTABLE_NV10_SMU_VERSION 8 @@ -504,10 +506,11 @@ typedef struct { uint32_t Status; uint16_t DieTemperature; - uint16_t MemoryTemperature; + uint16_t CurrentMemoryTemperature; - uint16_t SelectedCardPower; - uint16_t Reserved4; + uint16_t MemoryTemperature; + uint8_t MemoryHotspotPosition; + uint8_t Reserved4; uint32_t BoardLevelEnergyAccumulator; } OutOfBandMonitor_t; @@ -799,7 +802,12 @@ typedef struct { // Mvdd Svi2 Div Ratio Setting uint32_t MvddRatio; // This is used for MVDD Vid workaround. It has 16 fractional bits (Q16.16) - uint32_t BoardReserved[9]; + uint8_t RenesesLoadLineEnabled; + uint8_t GfxLoadlineResistance; + uint8_t SocLoadlineResistance; + uint8_t Padding8_Loadline; + + uint32_t BoardReserved[8]; // Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use @@ -903,13 +911,22 @@ typedef struct { } Watermarks_t; typedef struct { + uint16_t avgPsmCount[28]; + uint16_t minPsmCount[28]; + float avgPsmVoltage[28]; + float minPsmVoltage[28]; + + uint32_t MmHubPadding[32]; // SMU internal use +} AvfsDebugTable_t_NV14; + +typedef struct { uint16_t avgPsmCount[36]; uint16_t minPsmCount[36]; float avgPsmVoltage[36]; float minPsmVoltage[36]; uint32_t MmHubPadding[8]; // SMU internal use -} AvfsDebugTable_t; +} AvfsDebugTable_t_NV10; typedef struct { uint8_t AvfsVersion; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h b/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h new file mode 100644 index 000000000000..c27c82851468 --- /dev/null +++ b/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h @@ -0,0 +1,217 @@ +/* + * 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. + * + */ + +#ifndef SMU12_DRIVER_IF_H +#define SMU12_DRIVER_IF_H + +// *** IMPORTANT *** +// SMU TEAM: Always increment the interface version if +// any structure is changed in this file +#define SMU12_DRIVER_IF_VERSION 10 + +typedef struct { + int32_t value; + uint32_t numFractionalBits; +} FloatInIntFormat_t; + +typedef enum { + DSPCLK_DCFCLK = 0, + DSPCLK_DISPCLK, + DSPCLK_PIXCLK, + DSPCLK_PHYCLK, + DSPCLK_COUNT, +} DSPCLK_e; + +typedef struct { + uint16_t Freq; // in MHz + uint16_t Vid; // min voltage in SVI2 VID +} DisplayClockTable_t; + +typedef struct { + uint16_t MinClock; // This is either DCFCLK or SOCCLK (in MHz) + uint16_t MaxClock; // This is either DCFCLK or SOCCLK (in MHz) + uint16_t MinMclk; + uint16_t MaxMclk; + + uint8_t WmSetting; + uint8_t WmType; // Used for normal pstate change or memory retraining + uint8_t Padding[2]; +} WatermarkRowGeneric_t; + +#define NUM_WM_RANGES 4 +#define WM_PSTATE_CHG 0 +#define WM_RETRAINING 1 + +typedef enum { + WM_SOCCLK = 0, + WM_DCFCLK, + WM_COUNT, +} WM_CLOCK_e; + +typedef struct { + // Watermarks + WatermarkRowGeneric_t WatermarkRow[WM_COUNT][NUM_WM_RANGES]; + + uint32_t MmHubPadding[7]; // SMU internal use +} Watermarks_t; + +typedef enum { + CUSTOM_DPM_SETTING_GFXCLK, + CUSTOM_DPM_SETTING_CCLK, + CUSTOM_DPM_SETTING_FCLK_CCX, + CUSTOM_DPM_SETTING_FCLK_GFX, + CUSTOM_DPM_SETTING_FCLK_STALLS, + CUSTOM_DPM_SETTING_LCLK, + CUSTOM_DPM_SETTING_COUNT, +} CUSTOM_DPM_SETTING_e; + +typedef struct { + uint8_t ActiveHystLimit; + uint8_t IdleHystLimit; + uint8_t FPS; + uint8_t MinActiveFreqType; + FloatInIntFormat_t MinActiveFreq; + FloatInIntFormat_t PD_Data_limit; + FloatInIntFormat_t PD_Data_time_constant; + FloatInIntFormat_t PD_Data_error_coeff; + FloatInIntFormat_t PD_Data_error_rate_coeff; +} DpmActivityMonitorCoeffExt_t; + +typedef struct { + DpmActivityMonitorCoeffExt_t DpmActivityMonitorCoeff[CUSTOM_DPM_SETTING_COUNT]; +} CustomDpmSettings_t; + + +#define NUM_DCFCLK_DPM_LEVELS 8 +#define NUM_SOCCLK_DPM_LEVELS 8 +#define NUM_FCLK_DPM_LEVELS 4 +#define NUM_MEMCLK_DPM_LEVELS 4 +#define NUM_VCN_DPM_LEVELS 8 + +typedef struct { + uint32_t Freq; // In MHz + uint32_t Vol; // Millivolts with 2 fractional bits +} DpmClock_t; + +typedef struct { + DpmClock_t DcfClocks[NUM_DCFCLK_DPM_LEVELS]; + DpmClock_t SocClocks[NUM_SOCCLK_DPM_LEVELS]; + DpmClock_t FClocks[NUM_FCLK_DPM_LEVELS]; + DpmClock_t MemClocks[NUM_MEMCLK_DPM_LEVELS]; + DpmClock_t VClocks[NUM_VCN_DPM_LEVELS]; + DpmClock_t DClocks[NUM_VCN_DPM_LEVELS]; + + uint8_t NumDcfClkDpmEnabled; + uint8_t NumSocClkDpmEnabled; + uint8_t NumFClkDpmEnabled; + uint8_t NumMemClkDpmEnabled; + uint8_t NumVClkDpmEnabled; + uint8_t NumDClkDpmEnabled; + uint8_t spare[2]; +} DpmClocks_t; + + +typedef enum { + CLOCK_SMNCLK = 0, + CLOCK_SOCCLK, + CLOCK_MP0CLK, + CLOCK_MP1CLK, + CLOCK_MP2CLK, + CLOCK_VCLK, + CLOCK_LCLK, + CLOCK_DCLK, + CLOCK_ACLK, + CLOCK_ISPCLK, + CLOCK_SHUBCLK, + CLOCK_DISPCLK, + CLOCK_DPPCLK, + CLOCK_DPREFCLK, + CLOCK_DCFCLK, + CLOCK_FCLK, + CLOCK_UMCCLK, + CLOCK_GFXCLK, + CLOCK_COUNT, +} CLOCK_IDs_e; + +// Throttler Status Bitmask +#define THROTTLER_STATUS_BIT_SPL 0 +#define THROTTLER_STATUS_BIT_FPPT 1 +#define THROTTLER_STATUS_BIT_SPPT 2 +#define THROTTLER_STATUS_BIT_SPPT_APU 3 +#define THROTTLER_STATUS_BIT_THM_CORE 4 +#define THROTTLER_STATUS_BIT_THM_GFX 5 +#define THROTTLER_STATUS_BIT_THM_SOC 6 +#define THROTTLER_STATUS_BIT_TDC_VDD 7 +#define THROTTLER_STATUS_BIT_TDC_SOC 8 + +typedef struct { + uint16_t ClockFrequency[CLOCK_COUNT]; //[MHz] + + uint16_t AverageGfxclkFrequency; //[MHz] + uint16_t AverageSocclkFrequency; //[MHz] + uint16_t AverageVclkFrequency; //[MHz] + uint16_t AverageFclkFrequency; //[MHz] + + uint16_t AverageGfxActivity; //[centi] + uint16_t AverageUvdActivity; //[centi] + + uint16_t Voltage[2]; //[mV] indices: VDDCR_VDD, VDDCR_SOC + uint16_t Current[2]; //[mA] indices: VDDCR_VDD, VDDCR_SOC + uint16_t Power[2]; //[mW] indices: VDDCR_VDD, VDDCR_SOC + + uint16_t FanPwm; //[milli] + uint16_t CurrentSocketPower; //[mW] + + uint16_t CoreFrequency[8]; //[MHz] + uint16_t CorePower[8]; //[mW] + uint16_t CoreTemperature[8]; //[centi-Celsius] + uint16_t L3Frequency[2]; //[MHz] + uint16_t L3Temperature[2]; //[centi-Celsius] + + uint16_t GfxTemperature; //[centi-Celsius] + uint16_t SocTemperature; //[centi-Celsius] + uint16_t ThrottlerStatus; + uint16_t spare; +} SmuMetrics_t; + + +// Workload bits +#define WORKLOAD_PPLIB_FULL_SCREEN_3D_BIT 0 +#define WORKLOAD_PPLIB_VIDEO_BIT 2 +#define WORKLOAD_PPLIB_VR_BIT 3 +#define WORKLOAD_PPLIB_COMPUTE_BIT 4 +#define WORKLOAD_PPLIB_CUSTOM_BIT 5 +#define WORKLOAD_PPLIB_COUNT 6 + +#define TABLE_BIOS_IF 0 // Called by BIOS +#define TABLE_WATERMARKS 1 // Called by Driver +#define TABLE_CUSTOM_DPM 2 // Called by Driver +#define TABLE_SPARE1 3 +#define TABLE_DPMCLOCKS 4 // Called by Driver +#define TABLE_MOMENTARY_PM 5 // Called by Tools +#define TABLE_MODERN_STDBY 6 // Called by Tools for Modern Standby Log +#define TABLE_SMU_METRICS 7 // Called by Driver +#define TABLE_COUNT 8 + + +#endif diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h b/drivers/gpu/drm/amd/powerplay/inc/smu_types.h index 72962e842d69..052aecc2827a 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_types.h @@ -124,6 +124,50 @@ __SMU_DUMMY_MAP(PowerDownVcn0), \ __SMU_DUMMY_MAP(PowerUpVcn1), \ __SMU_DUMMY_MAP(PowerDownVcn1), \ + __SMU_DUMMY_MAP(PowerUpGfx), \ + __SMU_DUMMY_MAP(PowerDownIspByTile), \ + __SMU_DUMMY_MAP(PowerUpIspByTile), \ + __SMU_DUMMY_MAP(PowerDownSdma), \ + __SMU_DUMMY_MAP(PowerUpSdma), \ + __SMU_DUMMY_MAP(SetHardMinIspclkByFreq), \ + __SMU_DUMMY_MAP(SetHardMinVcn), \ + __SMU_DUMMY_MAP(Spare1), \ + __SMU_DUMMY_MAP(Spare2), \ + __SMU_DUMMY_MAP(SetAllowFclkSwitch), \ + __SMU_DUMMY_MAP(SetMinVideoGfxclkFreq), \ + __SMU_DUMMY_MAP(ActiveProcessNotify), \ + __SMU_DUMMY_MAP(SetCustomPolicy), \ + __SMU_DUMMY_MAP(QueryPowerLimit), \ + __SMU_DUMMY_MAP(SetGfxclkOverdriveByFreqVid), \ + __SMU_DUMMY_MAP(SetHardMinDcfclkByFreq), \ + __SMU_DUMMY_MAP(SetHardMinSocclkByFreq), \ + __SMU_DUMMY_MAP(ControlIgpuATS), \ + __SMU_DUMMY_MAP(SetMinVideoFclkFreq), \ + __SMU_DUMMY_MAP(SetMinDeepSleepDcfclk), \ + __SMU_DUMMY_MAP(ForcePowerDownGfx), \ + __SMU_DUMMY_MAP(SetPhyclkVoltageByFreq), \ + __SMU_DUMMY_MAP(SetDppclkVoltageByFreq), \ + __SMU_DUMMY_MAP(SetSoftMinVcn), \ + __SMU_DUMMY_MAP(EnablePostCode), \ + __SMU_DUMMY_MAP(GetGfxclkFrequency), \ + __SMU_DUMMY_MAP(GetFclkFrequency), \ + __SMU_DUMMY_MAP(GetMinGfxclkFrequency), \ + __SMU_DUMMY_MAP(GetMaxGfxclkFrequency), \ + __SMU_DUMMY_MAP(SetGfxCGPG), \ + __SMU_DUMMY_MAP(SetSoftMaxGfxClk), \ + __SMU_DUMMY_MAP(SetHardMinGfxClk), \ + __SMU_DUMMY_MAP(SetSoftMaxSocclkByFreq), \ + __SMU_DUMMY_MAP(SetSoftMaxFclkByFreq), \ + __SMU_DUMMY_MAP(SetSoftMaxVcn), \ + __SMU_DUMMY_MAP(PowerGateMmHub), \ + __SMU_DUMMY_MAP(UpdatePmeRestore), \ + __SMU_DUMMY_MAP(GpuChangeState), \ + __SMU_DUMMY_MAP(SetPowerLimitPercentage), \ + __SMU_DUMMY_MAP(ForceGfxContentSave), \ + __SMU_DUMMY_MAP(EnableTmdp48MHzRefclkPwrDown), \ + __SMU_DUMMY_MAP(PowerGateAtHub), \ + __SMU_DUMMY_MAP(SetSoftMinJpeg), \ + __SMU_DUMMY_MAP(SetHardMinFclkByFreq), \ #undef __SMU_DUMMY_MAP #define __SMU_DUMMY_MAP(type) SMU_MSG_##type diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h index fcb58012170f..5bda8539447a 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h @@ -25,6 +25,12 @@ #include "amdgpu_smu.h" +#define SMU11_DRIVER_IF_VERSION_INV 0xFFFFFFFF +#define SMU11_DRIVER_IF_VERSION_VG20 0x13 +#define SMU11_DRIVER_IF_VERSION_ARCT 0x09 +#define SMU11_DRIVER_IF_VERSION_NV10 0x33 +#define SMU11_DRIVER_IF_VERSION_NV14 0x34 + /* MP Apertures */ #define MP0_Public 0x03800000 #define MP0_SRAM 0x03900000 @@ -57,6 +63,12 @@ #define WORKLOAD_MAP(profile, workload) \ [profile] = {1, (workload)} +static const struct smu_temperature_range smu11_thermal_policy[] = +{ + {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, + { 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000}, +}; + struct smu_11_0_cmn2aisc_mapping { int valid_mapping; int map_to; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h new file mode 100644 index 000000000000..acf3db12f59f --- /dev/null +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0.h @@ -0,0 +1,42 @@ +/* + * 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. + * + */ +#ifndef __SMU_V12_0_H__ +#define __SMU_V12_0_H__ + +#include "amdgpu_smu.h" + +/* MP Apertures */ +#define MP0_Public 0x03800000 +#define MP0_SRAM 0x03900000 +#define MP1_Public 0x03b00000 +#define MP1_SRAM 0x03c00004 + + +struct smu_12_0_cmn2aisc_mapping { + int valid_mapping; + int map_to; +}; + +void smu_v12_0_set_smu_funcs(struct smu_context *smu); + +#endif diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0_ppsmc.h new file mode 100644 index 000000000000..9ac9f3bd3664 --- /dev/null +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v12_0_ppsmc.h @@ -0,0 +1,106 @@ +/* + * 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. + * + */ + +#ifndef SMU_12_0_PPSMC_H +#define SMU_12_0_PPSMC_H + +// SMU Response Codes: +#define PPSMC_Result_OK 0x1 +#define PPSMC_Result_Failed 0xFF +#define PPSMC_Result_UnknownCmd 0xFE +#define PPSMC_Result_CmdRejectedPrereq 0xFD +#define PPSMC_Result_CmdRejectedBusy 0xFC + + +// Message Definitions: +#define PPSMC_MSG_TestMessage 0x1 +#define PPSMC_MSG_GetSmuVersion 0x2 +#define PPSMC_MSG_GetDriverIfVersion 0x3 +#define PPSMC_MSG_PowerUpGfx 0x6 +#define PPSMC_MSG_EnableGfxOff 0x7 +#define PPSMC_MSG_DisableGfxOff 0x8 +#define PPSMC_MSG_PowerDownIspByTile 0x9 // ISP is power gated by default +#define PPSMC_MSG_PowerUpIspByTile 0xA +#define PPSMC_MSG_PowerDownVcn 0xB // VCN is power gated by default +#define PPSMC_MSG_PowerUpVcn 0xC +#define PPSMC_MSG_PowerDownSdma 0xD // SDMA is power gated by default +#define PPSMC_MSG_PowerUpSdma 0xE +#define PPSMC_MSG_SetHardMinIspclkByFreq 0xF +#define PPSMC_MSG_SetHardMinVcn 0x10 // For wireless display +#define PPSMC_MSG_spare1 0x11 +#define PPSMC_MSG_spare2 0x12 +#define PPSMC_MSG_SetAllowFclkSwitch 0x13 +#define PPSMC_MSG_SetMinVideoGfxclkFreq 0x14 +#define PPSMC_MSG_ActiveProcessNotify 0x15 +#define PPSMC_MSG_SetCustomPolicy 0x16 +#define PPSMC_MSG_SetVideoFps 0x17 +#define PPSMC_MSG_SetDisplayCount 0x18 // Moved to VBIOS +#define PPSMC_MSG_QueryPowerLimit 0x19 //Driver to look up sustainable clocks for VQ +#define PPSMC_MSG_SetDriverDramAddrHigh 0x1A +#define PPSMC_MSG_SetDriverDramAddrLow 0x1B +#define PPSMC_MSG_TransferTableSmu2Dram 0x1C +#define PPSMC_MSG_TransferTableDram2Smu 0x1D +#define PPSMC_MSG_GfxDeviceDriverReset 0x1E +#define PPSMC_MSG_SetGfxclkOverdriveByFreqVid 0x1F +#define PPSMC_MSG_SetHardMinDcfclkByFreq 0x20 // Moved to VBIOS +#define PPSMC_MSG_SetHardMinSocclkByFreq 0x21 +#define PPSMC_MSG_ControlIgpuATS 0x22 +#define PPSMC_MSG_SetMinVideoFclkFreq 0x23 +#define PPSMC_MSG_SetMinDeepSleepDcfclk 0x24 // Moved to VBIOS +#define PPSMC_MSG_ForcePowerDownGfx 0x25 +#define PPSMC_MSG_SetPhyclkVoltageByFreq 0x26 // Moved to VBIOS +#define PPSMC_MSG_SetDppclkVoltageByFreq 0x27 // Moved to VBIOS and is SetDppclkFreq +#define PPSMC_MSG_SetSoftMinVcn 0x28 +#define PPSMC_MSG_EnablePostCode 0x29 +#define PPSMC_MSG_GetGfxclkFrequency 0x2A +#define PPSMC_MSG_GetFclkFrequency 0x2B +#define PPSMC_MSG_GetMinGfxclkFrequency 0x2C +#define PPSMC_MSG_GetMaxGfxclkFrequency 0x2D +#define PPSMC_MSG_SoftReset 0x2E // Not supported +#define PPSMC_MSG_SetGfxCGPG 0x2F +#define PPSMC_MSG_SetSoftMaxGfxClk 0x30 +#define PPSMC_MSG_SetHardMinGfxClk 0x31 +#define PPSMC_MSG_SetSoftMaxSocclkByFreq 0x32 +#define PPSMC_MSG_SetSoftMaxFclkByFreq 0x33 +#define PPSMC_MSG_SetSoftMaxVcn 0x34 +#define PPSMC_MSG_PowerGateMmHub 0x35 +#define PPSMC_MSG_UpdatePmeRestore 0x36 // Moved to VBIOS +#define PPSMC_MSG_GpuChangeState 0x37 +#define PPSMC_MSG_SetPowerLimitPercentage 0x38 +#define PPSMC_MSG_ForceGfxContentSave 0x39 +#define PPSMC_MSG_EnableTmdp48MHzRefclkPwrDown 0x3A // Moved to VBIOS +#define PPSMC_MSG_PowerDownJpeg 0x3B +#define PPSMC_MSG_PowerUpJpeg 0x3C +#define PPSMC_MSG_PowerGateAtHub 0x3D +#define PPSMC_MSG_SetSoftMinJpeg 0x3E +#define PPSMC_MSG_SetHardMinFclkByFreq 0x3F +#define PPSMC_Message_Count 0x40 + + +//Argument for PPSMC_MSG_GpuChangeState +enum { + eGpuChangeState_D0Entry = 1, + eGpuChangeState_D3Entry, +}; + +#endif diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c index b3e66fead779..6f96ed7b9775 100644 --- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c @@ -941,8 +941,6 @@ static int navi10_get_gpu_power(struct smu_context *smu, uint32_t *value) ret = navi10_get_metrics_table(smu, &metrics); if (ret) return ret; - if (ret) - return ret; *value = metrics.AverageSocketPower << 8; @@ -1001,8 +999,6 @@ static int navi10_get_fan_speed_rpm(struct smu_context *smu, ret = navi10_get_metrics_table(smu, &metrics); if (ret) return ret; - if (ret) - return ret; *speed = metrics.CurrFanSpeed; @@ -1386,6 +1382,9 @@ static int navi10_read_sensor(struct smu_context *smu, struct smu_table_context *table_context = &smu->smu_table; PPTable_t *pptable = table_context->driver_pptable; + if(!data || !size) + return -EINVAL; + switch (sensor) { case AMDGPU_PP_SENSOR_MAX_FAN_RPM: *(uint32_t *)data = pptable->FanMaximumRpm; @@ -1407,7 +1406,7 @@ static int navi10_read_sensor(struct smu_context *smu, *size = 4; break; default: - return -EINVAL; + ret = smu_smc_read_sensor(smu, sensor, data, size); } return ret; @@ -1483,6 +1482,10 @@ static int navi10_set_peak_clock_by_device(struct smu_context *smu) static int navi10_set_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level) { int ret = 0; + struct amdgpu_device *adev = smu->adev; + + if (adev->asic_type != CHIP_NAVI10) + return -EINVAL; switch (level) { case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: @@ -1505,9 +1508,8 @@ static int navi10_get_thermal_temperature_range(struct smu_context *smu, if (!range || !powerplay_table) return -EINVAL; - /* The unit is temperature */ - range->min = 0; - range->max = powerplay_table->software_shutdown_temp; + range->max = powerplay_table->software_shutdown_temp * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; return 0; } @@ -1631,6 +1633,5 @@ void navi10_set_ppt_funcs(struct smu_context *smu) struct smu_table_context *smu_table = &smu->smu_table; smu->ppt_funcs = &navi10_ppt_funcs; - smu->smc_if_version = SMU11_DRIVER_IF_VERSION; smu_table->table_count = TABLE_COUNT; } diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c new file mode 100644 index 000000000000..de43159564a5 --- /dev/null +++ b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c @@ -0,0 +1,173 @@ +/* + * 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. + * + */ + +#include "amdgpu.h" +#include "amdgpu_smu.h" +#include "soc15_common.h" +#include "smu_v12_0_ppsmc.h" +#include "smu12_driver_if.h" +#include "smu_v12_0.h" +#include "renoir_ppt.h" + + +#define MSG_MAP(msg, index) \ + [SMU_MSG_##msg] = {1, (index)} + +#define TAB_MAP_VALID(tab) \ + [SMU_TABLE_##tab] = {1, TABLE_##tab} + +#define TAB_MAP_INVALID(tab) \ + [SMU_TABLE_##tab] = {0, TABLE_##tab} + +static struct smu_12_0_cmn2aisc_mapping renoir_message_map[SMU_MSG_MAX_COUNT] = { + MSG_MAP(TestMessage, PPSMC_MSG_TestMessage), + MSG_MAP(GetSmuVersion, PPSMC_MSG_GetSmuVersion), + MSG_MAP(GetDriverIfVersion, PPSMC_MSG_GetDriverIfVersion), + MSG_MAP(PowerUpGfx, PPSMC_MSG_PowerUpGfx), + MSG_MAP(AllowGfxOff, PPSMC_MSG_EnableGfxOff), + MSG_MAP(DisallowGfxOff, PPSMC_MSG_DisableGfxOff), + MSG_MAP(PowerDownIspByTile, PPSMC_MSG_PowerDownIspByTile), + MSG_MAP(PowerUpIspByTile, PPSMC_MSG_PowerUpIspByTile), + MSG_MAP(PowerDownVcn, PPSMC_MSG_PowerDownVcn), + MSG_MAP(PowerUpVcn, PPSMC_MSG_PowerUpVcn), + MSG_MAP(PowerDownSdma, PPSMC_MSG_PowerDownSdma), + MSG_MAP(PowerUpSdma, PPSMC_MSG_PowerUpSdma), + MSG_MAP(SetHardMinIspclkByFreq, PPSMC_MSG_SetHardMinIspclkByFreq), + MSG_MAP(SetHardMinVcn, PPSMC_MSG_SetHardMinVcn), + MSG_MAP(Spare1, PPSMC_MSG_spare1), + MSG_MAP(Spare2, PPSMC_MSG_spare2), + MSG_MAP(SetAllowFclkSwitch, PPSMC_MSG_SetAllowFclkSwitch), + MSG_MAP(SetMinVideoGfxclkFreq, PPSMC_MSG_SetMinVideoGfxclkFreq), + MSG_MAP(ActiveProcessNotify, PPSMC_MSG_ActiveProcessNotify), + MSG_MAP(SetCustomPolicy, PPSMC_MSG_SetCustomPolicy), + MSG_MAP(SetVideoFps, PPSMC_MSG_SetVideoFps), + MSG_MAP(NumOfDisplays, PPSMC_MSG_SetDisplayCount), + MSG_MAP(QueryPowerLimit, PPSMC_MSG_QueryPowerLimit), + MSG_MAP(SetDriverDramAddrHigh, PPSMC_MSG_SetDriverDramAddrHigh), + MSG_MAP(SetDriverDramAddrLow, PPSMC_MSG_SetDriverDramAddrLow), + MSG_MAP(TransferTableSmu2Dram, PPSMC_MSG_TransferTableSmu2Dram), + MSG_MAP(TransferTableDram2Smu, PPSMC_MSG_TransferTableDram2Smu), + MSG_MAP(GfxDeviceDriverReset, PPSMC_MSG_GfxDeviceDriverReset), + MSG_MAP(SetGfxclkOverdriveByFreqVid, PPSMC_MSG_SetGfxclkOverdriveByFreqVid), + MSG_MAP(SetHardMinDcfclkByFreq, PPSMC_MSG_SetHardMinDcfclkByFreq), + MSG_MAP(SetHardMinSocclkByFreq, PPSMC_MSG_SetHardMinSocclkByFreq), + MSG_MAP(ControlIgpuATS, PPSMC_MSG_ControlIgpuATS), + MSG_MAP(SetMinVideoFclkFreq, PPSMC_MSG_SetMinVideoFclkFreq), + MSG_MAP(SetMinDeepSleepDcfclk, PPSMC_MSG_SetMinDeepSleepDcfclk), + MSG_MAP(ForcePowerDownGfx, PPSMC_MSG_ForcePowerDownGfx), + MSG_MAP(SetPhyclkVoltageByFreq, PPSMC_MSG_SetPhyclkVoltageByFreq), + MSG_MAP(SetDppclkVoltageByFreq, PPSMC_MSG_SetDppclkVoltageByFreq), + MSG_MAP(SetSoftMinVcn, PPSMC_MSG_SetSoftMinVcn), + MSG_MAP(EnablePostCode, PPSMC_MSG_EnablePostCode), + MSG_MAP(GetGfxclkFrequency, PPSMC_MSG_GetGfxclkFrequency), + MSG_MAP(GetFclkFrequency, PPSMC_MSG_GetFclkFrequency), + MSG_MAP(GetMinGfxclkFrequency, PPSMC_MSG_GetMinGfxclkFrequency), + MSG_MAP(GetMaxGfxclkFrequency, PPSMC_MSG_GetMaxGfxclkFrequency), + MSG_MAP(SoftReset, PPSMC_MSG_SoftReset), + MSG_MAP(SetGfxCGPG, PPSMC_MSG_SetGfxCGPG), + MSG_MAP(SetSoftMaxGfxClk, PPSMC_MSG_SetSoftMaxGfxClk), + MSG_MAP(SetHardMinGfxClk, PPSMC_MSG_SetHardMinGfxClk), + MSG_MAP(SetSoftMaxSocclkByFreq, PPSMC_MSG_SetSoftMaxSocclkByFreq), + MSG_MAP(SetSoftMaxFclkByFreq, PPSMC_MSG_SetSoftMaxFclkByFreq), + MSG_MAP(SetSoftMaxVcn, PPSMC_MSG_SetSoftMaxVcn), + MSG_MAP(PowerGateMmHub, PPSMC_MSG_PowerGateMmHub), + MSG_MAP(UpdatePmeRestore, PPSMC_MSG_UpdatePmeRestore), + MSG_MAP(GpuChangeState, PPSMC_MSG_GpuChangeState), + MSG_MAP(SetPowerLimitPercentage, PPSMC_MSG_SetPowerLimitPercentage), + MSG_MAP(ForceGfxContentSave, PPSMC_MSG_ForceGfxContentSave), + MSG_MAP(EnableTmdp48MHzRefclkPwrDown, PPSMC_MSG_EnableTmdp48MHzRefclkPwrDown), + MSG_MAP(PowerDownJpeg, PPSMC_MSG_PowerDownJpeg), + MSG_MAP(PowerUpJpeg, PPSMC_MSG_PowerUpJpeg), + MSG_MAP(PowerGateAtHub, PPSMC_MSG_PowerGateAtHub), + MSG_MAP(SetSoftMinJpeg, PPSMC_MSG_SetSoftMinJpeg), + MSG_MAP(SetHardMinFclkByFreq, PPSMC_MSG_SetHardMinFclkByFreq), +}; + +static struct smu_12_0_cmn2aisc_mapping renoir_table_map[SMU_TABLE_COUNT] = { + TAB_MAP_VALID(WATERMARKS), + TAB_MAP_INVALID(CUSTOM_DPM), + TAB_MAP_VALID(DPMCLOCKS), + TAB_MAP_VALID(SMU_METRICS), +}; + +static int renoir_get_smu_msg_index(struct smu_context *smc, uint32_t index) +{ + struct smu_12_0_cmn2aisc_mapping mapping; + + if (index >= SMU_MSG_MAX_COUNT) + return -EINVAL; + + mapping = renoir_message_map[index]; + if (!(mapping.valid_mapping)) + return -EINVAL; + + return mapping.map_to; +} + +static int renoir_get_smu_table_index(struct smu_context *smc, uint32_t index) +{ + struct smu_12_0_cmn2aisc_mapping mapping; + + if (index >= SMU_TABLE_COUNT) + return -EINVAL; + + mapping = renoir_table_map[index]; + if (!(mapping.valid_mapping)) + return -EINVAL; + + return mapping.map_to; +} + +static int renoir_tables_init(struct smu_context *smu, struct smu_table *tables) +{ + struct smu_table_context *smu_table = &smu->smu_table; + + SMU_TABLE_INIT(tables, SMU_TABLE_WATERMARKS, sizeof(Watermarks_t), + PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + SMU_TABLE_INIT(tables, SMU_TABLE_DPMCLOCKS, sizeof(DpmClocks_t), + PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, sizeof(SmuMetrics_t), + PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + + smu_table->clocks_table = kzalloc(sizeof(DpmClocks_t), GFP_KERNEL); + if (!smu_table->clocks_table) + return -ENOMEM; + + return 0; +} + +static const struct pptable_funcs renoir_ppt_funcs = { + .get_smu_msg_index = renoir_get_smu_msg_index, + .get_smu_table_index = renoir_get_smu_table_index, + .tables_init = renoir_tables_init, + .set_power_state = NULL, +}; + +void renoir_set_ppt_funcs(struct smu_context *smu) +{ + struct smu_table_context *smu_table = &smu->smu_table; + + smu->ppt_funcs = &renoir_ppt_funcs; + smu->smc_if_version = SMU12_DRIVER_IF_VERSION; + smu_table->table_count = TABLE_COUNT; +} diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.h b/drivers/gpu/drm/amd/powerplay/renoir_ppt.h new file mode 100644 index 000000000000..e9b7237c0f7f --- /dev/null +++ b/drivers/gpu/drm/amd/powerplay/renoir_ppt.h @@ -0,0 +1,28 @@ +/* + * 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. + * + */ +#ifndef __RENOIR_PPT_H__ +#define __RENOIR_PPT_H__ + +extern void renoir_set_ppt_funcs(struct smu_context *smu); + +#endif diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c index 54618d7d6927..fa0b0256e0bc 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c @@ -274,6 +274,25 @@ static int smu_v11_0_check_fw_version(struct smu_context *smu) smu_minor = (smu_version >> 8) & 0xff; smu_debug = (smu_version >> 0) & 0xff; + switch (smu->adev->asic_type) { + case CHIP_VEGA20: + smu->smc_if_version = SMU11_DRIVER_IF_VERSION_VG20; + break; + case CHIP_ARCTURUS: + smu->smc_if_version = SMU11_DRIVER_IF_VERSION_ARCT; + break; + case CHIP_NAVI10: + smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV10; + break; + case CHIP_NAVI14: + smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV14; + break; + default: + pr_err("smu unsupported asic type:%d.\n", smu->adev->asic_type); + smu->smc_if_version = SMU11_DRIVER_IF_VERSION_INV; + break; + } + /* * 1. if_version mismatch is not critical as our fw is designed * to be backward compatible. @@ -340,7 +359,8 @@ static int smu_v11_0_setup_pptable(struct smu_context *smu) struct amdgpu_device *adev = smu->adev; const struct smc_firmware_header_v1_0 *hdr; int ret, index; - uint32_t size; + uint32_t size = 0; + uint16_t atom_table_size; uint8_t frev, crev; void *table; uint16_t version_major, version_minor; @@ -368,10 +388,11 @@ static int smu_v11_0_setup_pptable(struct smu_context *smu) index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1, powerplayinfo); - ret = smu_get_atom_data_table(smu, index, (uint16_t *)&size, &frev, &crev, + ret = smu_get_atom_data_table(smu, index, &atom_table_size, &frev, &crev, (uint8_t **)&table); if (ret) return ret; + size = atom_table_size; } if (!smu->smu_table.power_play_table) @@ -549,6 +570,9 @@ int smu_v11_0_get_vbios_bootup_values(struct smu_context *smu) smu->smu_table.boot_values.pp_table_id = v_3_3->pplib_pptable_id; } + smu->smu_table.boot_values.format_revision = header->format_revision; + smu->smu_table.boot_values.content_revision = header->content_revision; + return 0; } @@ -628,6 +652,24 @@ static int smu_v11_0_get_clk_info_from_vbios(struct smu_context *smu) output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&input; smu->smu_table.boot_values.dclk = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000; + if ((smu->smu_table.boot_values.format_revision == 3) && + (smu->smu_table.boot_values.content_revision >= 2)) { + memset(&input, 0, sizeof(input)); + input.clk_id = SMU11_SYSPLL1_0_FCLK_ID; + input.syspll_id = SMU11_SYSPLL1_2_ID; + input.command = GET_SMU_CLOCK_INFO_V3_1_GET_CLOCK_FREQ; + index = get_index_into_master_table(atom_master_list_of_command_functions_v2_1, + getsmuclockinfo); + + ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index, + (uint32_t *)&input); + if (ret) + return -EINVAL; + + output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&input; + smu->smu_table.boot_values.fclk = le32_to_cpu(output->atom_smu_outputclkfreq.smu_clock_freq_hz) / 10000; + } + return 0; } @@ -1085,23 +1127,17 @@ static int smu_v11_0_get_current_clk_freq(struct smu_context *smu, } static int smu_v11_0_set_thermal_range(struct smu_context *smu, - struct smu_temperature_range *range) + struct smu_temperature_range range) { struct amdgpu_device *adev = smu->adev; int low = SMU_THERMAL_MINIMUM_ALERT_TEMP; int high = SMU_THERMAL_MAXIMUM_ALERT_TEMP; uint32_t val; - if (!range) - return -EINVAL; - - if (low < range->min) - low = range->min; - if (high > range->max) - high = range->max; - - low = max(SMU_THERMAL_MINIMUM_ALERT_TEMP, range->min); - high = min(SMU_THERMAL_MAXIMUM_ALERT_TEMP, range->max); + low = max(SMU_THERMAL_MINIMUM_ALERT_TEMP, + range.min / SMU_TEMPERATURE_UNITS_PER_CENTIGRADES); + high = min(SMU_THERMAL_MAXIMUM_ALERT_TEMP, + range.max / SMU_TEMPERATURE_UNITS_PER_CENTIGRADES); if (low > high) return -EINVAL; @@ -1137,27 +1173,20 @@ static int smu_v11_0_enable_thermal_alert(struct smu_context *smu) static int smu_v11_0_start_thermal_control(struct smu_context *smu) { int ret = 0; - struct smu_temperature_range range = { - TEMP_RANGE_MIN, - TEMP_RANGE_MAX, - TEMP_RANGE_MAX, - TEMP_RANGE_MIN, - TEMP_RANGE_MAX, - TEMP_RANGE_MAX, - TEMP_RANGE_MIN, - TEMP_RANGE_MAX, - TEMP_RANGE_MAX}; + struct smu_temperature_range range; struct amdgpu_device *adev = smu->adev; if (!smu->pm_enabled) return ret; + memcpy(&range, &smu11_thermal_policy[0], sizeof(struct smu_temperature_range)); + ret = smu_get_thermal_temperature_range(smu, &range); if (ret) return ret; if (smu->smu_table.thermal_controller_type) { - ret = smu_v11_0_set_thermal_range(smu, &range); + ret = smu_v11_0_set_thermal_range(smu, range); if (ret) return ret; @@ -1170,17 +1199,15 @@ static int smu_v11_0_start_thermal_control(struct smu_context *smu) return ret; } - adev->pm.dpm.thermal.min_temp = range.min * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.max_temp = range.max * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.max_edge_emergency_temp = range.edge_emergency_max * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.min_hotspot_temp = range.hotspot_min * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.max_hotspot_crit_temp = range.hotspot_crit_max * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.max_hotspot_emergency_temp = range.hotspot_emergency_max * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.min_mem_temp = range.mem_min * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.max_mem_crit_temp = range.mem_crit_max * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.max_mem_emergency_temp = range.mem_emergency_max * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.min_temp = range.min * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; - adev->pm.dpm.thermal.max_temp = range.max * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + adev->pm.dpm.thermal.min_temp = range.min; + adev->pm.dpm.thermal.max_temp = range.max; + adev->pm.dpm.thermal.max_edge_emergency_temp = range.edge_emergency_max; + adev->pm.dpm.thermal.min_hotspot_temp = range.hotspot_min; + adev->pm.dpm.thermal.max_hotspot_crit_temp = range.hotspot_crit_max; + adev->pm.dpm.thermal.max_hotspot_emergency_temp = range.hotspot_emergency_max; + adev->pm.dpm.thermal.min_mem_temp = range.mem_min; + adev->pm.dpm.thermal.max_mem_crit_temp = range.mem_crit_max; + adev->pm.dpm.thermal.max_mem_emergency_temp = range.mem_emergency_max; return ret; } @@ -1214,6 +1241,10 @@ static int smu_v11_0_read_sensor(struct smu_context *smu, void *data, uint32_t *size) { int ret = 0; + + if(!data || !size) + return -EINVAL; + switch (sensor) { case AMDGPU_PP_SENSOR_GFX_MCLK: ret = smu_get_current_clk_freq(smu, SMU_UCLK, (uint32_t *)data); @@ -1236,10 +1267,6 @@ static int smu_v11_0_read_sensor(struct smu_context *smu, break; } - /* try get sensor data by asic */ - if (ret) - ret = smu_asic_read_sensor(smu, sensor, data, size); - if (ret) *size = 0; @@ -1255,7 +1282,6 @@ smu_v11_0_display_clock_voltage_request(struct smu_context *smu, int ret = 0; enum smu_clk_type clk_select = 0; uint32_t clk_freq = clock_req->clock_freq_in_khz / 1000; - int clk_id; if (!smu->pm_enabled) return -EINVAL; @@ -1290,16 +1316,8 @@ smu_v11_0_display_clock_voltage_request(struct smu_context *smu, if (clk_select == SMU_UCLK && smu->disable_uclk_switch) return 0; - clk_id = smu_clk_get_index(smu, clk_select); - if (clk_id < 0) { - ret = -EINVAL; - goto failed; - } - - mutex_lock(&smu->mutex); - ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq, - (clk_id << 16) | clk_freq); + ret = smu_set_hard_freq_range(smu, clk_select, clk_freq, 0); mutex_unlock(&smu->mutex); if(clk_select == SMU_UCLK) @@ -1646,7 +1664,7 @@ static bool smu_v11_0_baco_is_support(struct smu_context *smu) static enum smu_baco_state smu_v11_0_baco_get_state(struct smu_context *smu) { struct smu_baco_context *smu_baco = &smu->smu_baco; - enum smu_baco_state baco_state = SMU_BACO_STATE_EXIT; + enum smu_baco_state baco_state; mutex_lock(&smu_baco->mutex); baco_state = smu_baco->state; @@ -1718,7 +1736,7 @@ static const struct smu_funcs smu_v11_0_funcs = { .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .check_pptable = smu_v11_0_check_pptable, .parse_pptable = smu_v11_0_parse_pptable, - .populate_smc_pptable = smu_v11_0_populate_smc_pptable, + .populate_smc_tables = smu_v11_0_populate_smc_pptable, .write_pptable = smu_v11_0_write_pptable, .write_watermarks_table = smu_v11_0_write_watermarks_table, .set_min_dcef_deep_sleep = smu_v11_0_set_min_dcef_deep_sleep, diff --git a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c new file mode 100644 index 000000000000..0f5d08ae71ae --- /dev/null +++ b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c @@ -0,0 +1,350 @@ +/* + * 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. + */ + +#include "pp_debug.h" +#include <linux/firmware.h> +#include "amdgpu.h" +#include "amdgpu_smu.h" +#include "atomfirmware.h" +#include "amdgpu_atomfirmware.h" +#include "smu_v12_0.h" +#include "soc15_common.h" +#include "atom.h" +#include "renoir_ppt.h" + +#include "asic_reg/mp/mp_12_0_0_offset.h" +#include "asic_reg/mp/mp_12_0_0_sh_mask.h" + +#define smnMP1_FIRMWARE_FLAGS 0x3010024 + +#define mmSMUIO_GFX_MISC_CNTL 0x00c8 +#define mmSMUIO_GFX_MISC_CNTL_BASE_IDX 0 +#define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS_MASK 0x00000006L +#define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS__SHIFT 0x1 + +static int smu_v12_0_send_msg_without_waiting(struct smu_context *smu, + uint16_t msg) +{ + struct amdgpu_device *adev = smu->adev; + + WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_66, msg); + return 0; +} + +static int smu_v12_0_read_arg(struct smu_context *smu, uint32_t *arg) +{ + struct amdgpu_device *adev = smu->adev; + + *arg = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82); + return 0; +} + +static int smu_v12_0_wait_for_response(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + uint32_t cur_value, i; + + for (i = 0; i < adev->usec_timeout; i++) { + cur_value = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90); + if ((cur_value & MP1_C2PMSG_90__CONTENT_MASK) != 0) + break; + udelay(1); + } + + /* timeout means wrong logic */ + if (i == adev->usec_timeout) + return -ETIME; + + return RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90) == 0x1 ? 0 : -EIO; +} + +static int smu_v12_0_send_msg(struct smu_context *smu, uint16_t msg) +{ + struct amdgpu_device *adev = smu->adev; + int ret = 0, index = 0; + + index = smu_msg_get_index(smu, msg); + if (index < 0) + return index; + + smu_v12_0_wait_for_response(smu); + + WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); + + smu_v12_0_send_msg_without_waiting(smu, (uint16_t)index); + + ret = smu_v12_0_wait_for_response(smu); + + if (ret) + pr_err("Failed to send message 0x%x, response 0x%x\n", index, + ret); + + return ret; + +} + +static int +smu_v12_0_send_msg_with_param(struct smu_context *smu, uint16_t msg, + uint32_t param) +{ + struct amdgpu_device *adev = smu->adev; + int ret = 0, index = 0; + + index = smu_msg_get_index(smu, msg); + if (index < 0) + return index; + + ret = smu_v12_0_wait_for_response(smu); + if (ret) + pr_err("Failed to send message 0x%x, response 0x%x, param 0x%x\n", + index, ret, param); + + WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); + + WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82, param); + + smu_v12_0_send_msg_without_waiting(smu, (uint16_t)index); + + ret = smu_v12_0_wait_for_response(smu); + if (ret) + pr_err("Failed to send message 0x%x, response 0x%x param 0x%x\n", + index, ret, param); + + return ret; +} + +static int smu_v12_0_check_fw_status(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + uint32_t mp1_fw_flags; + + mp1_fw_flags = RREG32_PCIE(MP1_Public | + (smnMP1_FIRMWARE_FLAGS & 0xffffffff)); + + if ((mp1_fw_flags & MP1_FIRMWARE_FLAGS__INTERRUPTS_ENABLED_MASK) >> + MP1_FIRMWARE_FLAGS__INTERRUPTS_ENABLED__SHIFT) + return 0; + + return -EIO; +} + +static int smu_v12_0_check_fw_version(struct smu_context *smu) +{ + uint32_t if_version = 0xff, smu_version = 0xff; + uint16_t smu_major; + uint8_t smu_minor, smu_debug; + int ret = 0; + + ret = smu_get_smc_version(smu, &if_version, &smu_version); + if (ret) + return ret; + + smu_major = (smu_version >> 16) & 0xffff; + smu_minor = (smu_version >> 8) & 0xff; + smu_debug = (smu_version >> 0) & 0xff; + + /* + * 1. if_version mismatch is not critical as our fw is designed + * to be backward compatible. + * 2. New fw usually brings some optimizations. But that's visible + * only on the paired driver. + * Considering above, we just leave user a warning message instead + * of halt driver loading. + */ + if (if_version != smu->smc_if_version) { + pr_info("smu driver if version = 0x%08x, smu fw if version = 0x%08x, " + "smu fw version = 0x%08x (%d.%d.%d)\n", + smu->smc_if_version, if_version, + smu_version, smu_major, smu_minor, smu_debug); + pr_warn("SMU driver if version not matched\n"); + } + + return ret; +} + +static int smu_v12_0_powergate_sdma(struct smu_context *smu, bool gate) +{ + if (!(smu->adev->flags & AMD_IS_APU)) + return 0; + + if (gate) + return smu_send_smc_msg(smu, SMU_MSG_PowerDownSdma); + else + return smu_send_smc_msg(smu, SMU_MSG_PowerUpSdma); +} + +static int smu_v12_0_powergate_vcn(struct smu_context *smu, bool gate) +{ + if (!(smu->adev->flags & AMD_IS_APU)) + return 0; + + if (gate) + return smu_send_smc_msg(smu, SMU_MSG_PowerDownVcn); + else + return smu_send_smc_msg(smu, SMU_MSG_PowerUpVcn); +} + +static int smu_v12_0_set_gfx_cgpg(struct smu_context *smu, bool enable) +{ + if (!(smu->adev->pg_flags & AMD_PG_SUPPORT_GFX_PG)) + return 0; + + return smu_v12_0_send_msg_with_param(smu, + SMU_MSG_SetGfxCGPG, enable ? 1 : 0); +} + +/** + * smu_v12_0_get_gfxoff_status - get gfxoff status + * + * @smu: amdgpu_device pointer + * + * This function will be used to get gfxoff status + * + * Returns 0=GFXOFF(default). + * Returns 1=Transition out of GFX State. + * Returns 2=Not in GFXOFF. + * Returns 3=Transition into GFXOFF. + */ +static uint32_t smu_v12_0_get_gfxoff_status(struct smu_context *smu) +{ + uint32_t reg; + uint32_t gfxOff_Status = 0; + struct amdgpu_device *adev = smu->adev; + + reg = RREG32_SOC15(SMUIO, 0, mmSMUIO_GFX_MISC_CNTL); + gfxOff_Status = (reg & SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS_MASK) + >> SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS__SHIFT; + + return gfxOff_Status; +} + +static int smu_v12_0_gfx_off_control(struct smu_context *smu, bool enable) +{ + int ret = 0, timeout = 500; + + if (enable) { + ret = smu_send_smc_msg(smu, SMU_MSG_AllowGfxOff); + + /* confirm gfx is back to "off" state, timeout is 5 seconds */ + while (!(smu_v12_0_get_gfxoff_status(smu) == 0)) { + msleep(10); + timeout--; + if (timeout == 0) { + DRM_ERROR("enable gfxoff timeout and failed!\n"); + break; + } + } + } else { + ret = smu_send_smc_msg(smu, SMU_MSG_DisallowGfxOff); + + /* confirm gfx is back to "on" state, timeout is 0.5 second */ + while (!(smu_v12_0_get_gfxoff_status(smu) == 2)) { + msleep(1); + timeout--; + if (timeout == 0) { + DRM_ERROR("disable gfxoff timeout and failed!\n"); + break; + } + } + } + + return ret; +} + +static int smu_v12_0_init_smc_tables(struct smu_context *smu) +{ + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *tables = NULL; + + if (smu_table->tables || smu_table->table_count == 0) + return -EINVAL; + + tables = kcalloc(SMU_TABLE_COUNT, sizeof(struct smu_table), + GFP_KERNEL); + if (!tables) + return -ENOMEM; + + smu_table->tables = tables; + + return smu_tables_init(smu, tables); +} + +static int smu_v12_0_fini_smc_tables(struct smu_context *smu) +{ + struct smu_table_context *smu_table = &smu->smu_table; + + if (!smu_table->tables || smu_table->table_count == 0) + return -EINVAL; + + kfree(smu_table->clocks_table); + kfree(smu_table->tables); + + smu_table->clocks_table = NULL; + smu_table->tables = NULL; + + return 0; +} + +static int smu_v12_0_populate_smc_tables(struct smu_context *smu) +{ + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *table = NULL; + + table = &smu_table->tables[SMU_TABLE_DPMCLOCKS]; + if (!table) + return -EINVAL; + + if (!table->cpu_addr) + return -EINVAL; + + return smu_update_table(smu, SMU_TABLE_DPMCLOCKS, 0, smu_table->clocks_table, false); +} + +static const struct smu_funcs smu_v12_0_funcs = { + .check_fw_status = smu_v12_0_check_fw_status, + .check_fw_version = smu_v12_0_check_fw_version, + .powergate_sdma = smu_v12_0_powergate_sdma, + .powergate_vcn = smu_v12_0_powergate_vcn, + .send_smc_msg = smu_v12_0_send_msg, + .send_smc_msg_with_param = smu_v12_0_send_msg_with_param, + .read_smc_arg = smu_v12_0_read_arg, + .set_gfx_cgpg = smu_v12_0_set_gfx_cgpg, + .gfx_off_control = smu_v12_0_gfx_off_control, + .init_smc_tables = smu_v12_0_init_smc_tables, + .fini_smc_tables = smu_v12_0_fini_smc_tables, + .populate_smc_tables = smu_v12_0_populate_smc_tables, +}; + +void smu_v12_0_set_smu_funcs(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + + smu->funcs = &smu_v12_0_funcs; + + switch (adev->asic_type) { + case CHIP_RENOIR: + renoir_set_ppt_funcs(smu); + break; + default: + pr_warn("Unknown asic for smu12\n"); + } +} diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c index 0102e24063d4..64386ee3f878 100644 --- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c @@ -2917,6 +2917,7 @@ static int vega20_get_fan_speed_percent(struct smu_context *smu, static int vega20_get_gpu_power(struct smu_context *smu, uint32_t *value) { + uint32_t smu_version; int ret = 0; SmuMetrics_t metrics; @@ -2927,7 +2928,15 @@ static int vega20_get_gpu_power(struct smu_context *smu, uint32_t *value) if (ret) return ret; - *value = metrics.CurrSocketPower << 8; + ret = smu_get_smc_version(smu, NULL, &smu_version); + if (ret) + return ret; + + /* For the 40.46 release, they changed the value name */ + if (smu_version == 0x282e00) + *value = metrics.AverageSocketPower << 8; + else + *value = metrics.CurrSocketPower << 8; return 0; } @@ -3011,6 +3020,9 @@ static int vega20_read_sensor(struct smu_context *smu, struct smu_table_context *table_context = &smu->smu_table; PPTable_t *pptable = table_context->driver_pptable; + if(!data || !size) + return -EINVAL; + switch (sensor) { case AMDGPU_PP_SENSOR_MAX_FAN_RPM: *(uint32_t *)data = pptable->FanMaximumRpm; @@ -3034,7 +3046,7 @@ static int vega20_read_sensor(struct smu_context *smu, *size = 4; break; default: - return -EINVAL; + ret = smu_smc_read_sensor(smu, sensor, data, size); } return ret; @@ -3110,14 +3122,18 @@ static int vega20_get_thermal_temperature_range(struct smu_context *smu, if (!range || !powerplay_table) return -EINVAL; - /* The unit is temperature */ - range->min = 0; - range->max = powerplay_table->usSoftwareShutdownTemp; - range->edge_emergency_max = (pptable->TedgeLimit + CTF_OFFSET_EDGE); - range->hotspot_crit_max = pptable->ThotspotLimit; - range->hotspot_emergency_max = (pptable->ThotspotLimit + CTF_OFFSET_HOTSPOT); - range->mem_crit_max = pptable->ThbmLimit; - range->mem_emergency_max = (pptable->ThbmLimit + CTF_OFFSET_HBM); + range->max = powerplay_table->usSoftwareShutdownTemp * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + range->edge_emergency_max = (pptable->TedgeLimit + CTF_OFFSET_EDGE) * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + range->hotspot_crit_max = pptable->ThotspotLimit * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + range->hotspot_emergency_max = (pptable->ThotspotLimit + CTF_OFFSET_HOTSPOT) * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + range->mem_crit_max = pptable->ThbmLimit * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + range->mem_emergency_max = (pptable->ThbmLimit + CTF_OFFSET_HBM) * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; return 0; @@ -3173,6 +3189,5 @@ void vega20_set_ppt_funcs(struct smu_context *smu) struct smu_table_context *smu_table = &smu->smu_table; smu->ppt_funcs = &vega20_ppt_funcs; - smu->smc_if_version = SMU11_DRIVER_IF_VERSION; smu_table->table_count = TABLE_COUNT; } |