/********************************************************** * Copyright 2008-2009 VMware, Inc. All rights reserved. * * 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 AUTHORS OR COPYRIGHT HOLDERS * 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 "util/u_format.h" #include "util/u_inlines.h" #include "util/u_memory.h" #include "pipe/p_defines.h" #include "util/u_upload_mgr.h" #include "svga_screen.h" #include "svga_context.h" #include "svga_state.h" #include "svga_cmd.h" #include "svga_tgsi.h" #include "svga_debug.h" #include "svga_resource_buffer.h" #include "svga_shader.h" #include "svga_hw_reg.h" /* * Don't try to send more than 4kb of successive constants. */ #define MAX_CONST_REG_COUNT 256 /**< number of float[4] constants */ /** * Extra space for svga-specific VS/PS constants (such as texcoord * scale factors, vertex transformation scale/translation). */ #define MAX_EXTRA_CONSTS 32 /** Guest-backed surface constant buffers must be this size */ #define GB_CONSTBUF_SIZE (SVGA3D_CONSTREG_MAX) /** * Emit any extra shader-type-independent shader constants into the buffer * pointed to by 'dest'. * \return number of float[4] constants put into the 'dest' buffer */ static unsigned svga_get_extra_constants_common(struct svga_context *svga, const struct svga_shader_variant *variant, enum pipe_shader_type shader, float *dest) { uint32_t *dest_u = (uint32_t *) dest; // uint version of dest unsigned i; unsigned count = 0; for (i = 0; i < variant->key.num_textures; i++) { struct pipe_sampler_view *sv = svga->curr.sampler_views[shader][i]; if (sv) { struct pipe_resource *tex = sv->texture; /* Scaling factors needed for handling unnormalized texture coordinates * for texture rectangles. */ if (variant->key.tex[i].unnormalized) { /* debug/sanity check */ assert(variant->key.tex[i].width_height_idx == count); *dest++ = 1.0 / (float)tex->width0; *dest++ = 1.0 / (float)tex->height0; *dest++ = 1.0; *dest++ = 1.0; count++; } /* Store the sizes for texture buffers. */ if (tex->target == PIPE_BUFFER) { unsigned bytes_per_element = util_format_get_blocksize(sv->format); *dest_u++ = tex->width0 / bytes_per_element; *dest_u++ = 1; *dest_u++ = 1; *dest_u++ = 1; count++; } } } return count; } /** * Emit any extra fragment shader constants into the buffer pointed * to by 'dest'. * \return number of float[4] constants put into the dest buffer */ static unsigned svga_get_extra_fs_constants(struct svga_context *svga, float *dest) { const struct svga_shader_variant *variant = svga->state.hw_draw.fs; unsigned count = 0; count += svga_get_extra_constants_common(svga, variant, PIPE_SHADER_FRAGMENT, dest); assert(count <= MAX_EXTRA_CONSTS); return count; } /** * Emit extra constants needed for prescale computation into the * the buffer pointed to by '*dest'. The updated buffer pointer * will be returned in 'dest'. */ static unsigned svga_get_prescale_constants(struct svga_context *svga, float **dest) { memcpy(*dest, svga->state.hw_clear.prescale.scale, 4 * sizeof(float)); *dest += 4; memcpy(*dest, svga->state.hw_clear.prescale.translate, 4 * sizeof(float)); *dest += 4; return 2; } /** * Emit extra constants needed for point sprite emulation. */ static unsigned svga_get_pt_sprite_constants(struct svga_context *svga, float **dest) { struct svga_screen *screen = svga_screen(svga->pipe.screen); float *dst = *dest; dst[0] = 1.0 / (svga->curr.viewport.scale[0] * 2); dst[1] = 1.0 / (svga->curr.viewport.scale[1] * 2); dst[2] = svga->curr.rast->pointsize; dst[3] = screen->maxPointSize; *dest = *dest + 4; return 1; } /** * Emit user-defined clip plane coefficients into the buffer pointed to * by '*dest'. The updated buffer pointer will be returned in 'dest'. */ static unsigned svga_get_clip_plane_constants(struct svga_context *svga, const struct svga_shader_variant *variant, float **dest) { unsigned count = 0; /* SVGA_NEW_CLIP */ if (svga_have_vgpu10(svga)) { /* append user-defined clip plane coefficients onto constant buffer */ unsigned clip_planes = variant->key.clip_plane_enable; while (clip_planes) { int i = u_bit_scan(&clip_planes); COPY_4V(*dest, svga->curr.clip.ucp[i]); *dest += 4; count += 1; } } return count; } /** * Emit any extra vertex shader constants into the buffer pointed * to by 'dest'. * In particular, these would be the scale and bias factors computed * from the framebuffer size which are used to copy with differences in * GL vs D3D coordinate spaces. See svga_tgsi_insn.c for more info. * \return number of float[4] constants put into the dest buffer */ static unsigned svga_get_extra_vs_constants(struct svga_context *svga, float *dest) { const struct svga_shader_variant *variant = svga->state.hw_draw.vs; unsigned count = 0; /* SVGA_NEW_VS_VARIANT */ if (variant->key.vs.need_prescale) { count += svga_get_prescale_constants(svga, &dest); } if (variant->key.vs.undo_viewport) { /* Used to convert window coords back to NDC coords */ dest[0] = 1.0f / svga->curr.viewport.scale[0]; dest[1] = 1.0f / svga->curr.viewport.scale[1]; dest[2] = -svga->curr.viewport.translate[0]; dest[3] = -svga->curr.viewport.translate[1]; dest += 4; count += 1; } /* SVGA_NEW_CLIP */ count += svga_get_clip_plane_constants(svga, variant, &dest); /* common constants */ count += svga_get_extra_constants_common(svga, variant, PIPE_SHADER_VERTEX, dest); assert(count <= MAX_EXTRA_CONSTS); return count; } /** * Emit any extra geometry shader constants into the buffer pointed * to by 'dest'. */ static unsigned svga_get_extra_gs_constants(struct svga_context *svga, float *dest) { const struct svga_shader_variant *variant = svga->state.hw_draw.gs; unsigned count = 0; /* SVGA_NEW_GS_VARIANT */ /* Constants for point sprite * These are used in the transformed gs that supports point sprite. * They need to be added before the prescale constants. */ if (variant->key.gs.wide_point) { count += svga_get_pt_sprite_constants(svga, &dest); } if (variant->key.gs.need_prescale) { count += svga_get_prescale_constants(svga, &dest); } /* SVGA_NEW_CLIP */ count += svga_get_clip_plane_constants(svga, variant, &dest); /* common constants */ count += svga_get_extra_constants_common(svga, variant, PIPE_SHADER_GEOMETRY, dest); assert(count <= MAX_EXTRA_CONSTS); return count; } /* * Check and emit a range of shader constant registers, trying to coalesce * successive shader constant updates in a single command in order to save * space on the command buffer. This is a HWv8 feature. */ static enum pipe_error emit_const_range(struct svga_context *svga, enum pipe_shader_type shader, unsigned offset, unsigned count, const float (*values)[4]) { unsigned i, j; enum pipe_error ret; assert(shader == PIPE_SHADER_VERTEX || shader == PIPE_SHADER_FRAGMENT); assert(!svga_have_vgpu10(svga)); #ifdef DEBUG if (offset + count > SVGA3D_CONSTREG_MAX) { debug_printf("svga: too many constants (offset %u + count %u = %u (max = %u))\n", offset, count, offset + count, SVGA3D_CONSTREG_MAX); } #endif if (offset > SVGA3D_CONSTREG_MAX) { /* This isn't OK, but if we propagate an error all the way up we'll * just get into more trouble. * XXX note that offset is always zero at this time so this is moot. */ return PIPE_OK; } if (offset + count > SVGA3D_CONSTREG_MAX) { /* Just drop the extra constants for now. * Ideally we should not have allowed the app to create a shader * that exceeds our constant buffer size but there's no way to * express that in gallium at this time. */ count = SVGA3D_CONSTREG_MAX - offset; } i = 0; while (i < count) { if (memcmp(svga->state.hw_draw.cb[shader][offset + i], values[i], 4 * sizeof(float)) != 0) { /* Found one dirty constant */ if (SVGA_DEBUG & DEBUG_CONSTS) debug_printf("%s %s %d: %f %f %f %f\n", __FUNCTION__, shader == PIPE_SHADER_VERTEX ? "VERT" : "FRAG", offset + i, values[i][0], values[i][1], values[i][2], values[i][3]); /* Look for more consecutive dirty constants. */ j = i + 1; while (j < count && j < i + MAX_CONST_REG_COUNT && memcmp(svga->state.hw_draw.cb[shader][offset + j], values[j], 4 * sizeof(float)) != 0) { if (SVGA_DEBUG & DEBUG_CONSTS) debug_printf("%s %s %d: %f %f %f %f\n", __FUNCTION__, shader == PIPE_SHADER_VERTEX ? "VERT" : "FRAG", offset + j, values[j][0], values[j][1], values[j][2], values[j][3]); ++j; } assert(j >= i + 1); /* Send them all together. */ if (svga_have_gb_objects(svga)) { ret = SVGA3D_SetGBShaderConstsInline(svga->swc, offset + i, /* start */ j - i, /* count */ svga_shader_type(shader), SVGA3D_CONST_TYPE_FLOAT, values + i); } else { ret = SVGA3D_SetShaderConsts(svga->swc, offset + i, j - i, svga_shader_type(shader), SVGA3D_CONST_TYPE_FLOAT, values + i); } if (ret != PIPE_OK) { return ret; } /* * Local copy of the hardware state. */ memcpy(svga->state.hw_draw.cb[shader][offset + i], values[i], (j - i) * 4 * sizeof(float)); i = j + 1; svga->hud.num_const_updates++; } else { ++i; } } return PIPE_OK; } /** * Emit all the constants in a constant buffer for a shader stage. * On VGPU10, emit_consts_vgpu10 is used instead. */ static enum pipe_error emit_consts_vgpu9(struct svga_context *svga, enum pipe_shader_type shader) { const struct pipe_constant_buffer *cbuf; struct pipe_transfer *transfer = NULL; unsigned count; const float (*data)[4] = NULL; enum pipe_error ret = PIPE_OK; const unsigned offset = 0; assert(shader < PIPE_SHADER_TYPES); assert(!svga_have_vgpu10(svga)); /* Only one constant buffer per shader is supported before VGPU10. * This is only an approximate check against that. */ assert(svga->curr.constbufs[shader][1].buffer == NULL); cbuf = &svga->curr.constbufs[shader][0]; if (svga->curr.constbufs[shader][0].buffer) { /* emit user-provided constants */ data = (const float (*)[4]) pipe_buffer_map(&svga->pipe, svga->curr.constbufs[shader][0].buffer, PIPE_TRANSFER_READ, &transfer); if (!data) { return PIPE_ERROR_OUT_OF_MEMORY; } /* sanity check */ assert(cbuf->buffer->width0 >= cbuf->buffer_size); /* Use/apply the constant buffer size and offsets here */ count = cbuf->buffer_size / (4 * sizeof(float)); data += cbuf->buffer_offset / (4 * sizeof(float)); ret = emit_const_range( svga, shader, offset, count, data ); pipe_buffer_unmap(&svga->pipe, transfer); if (ret != PIPE_OK) { return ret; } } /* emit extra shader constants */ { const struct svga_shader_variant *variant = NULL; unsigned offset; float extras[MAX_EXTRA_CONSTS][4]; unsigned count; switch (shader) { case PIPE_SHADER_VERTEX: variant = svga->state.hw_draw.vs; count = svga_get_extra_vs_constants(svga, (float *) extras); break; case PIPE_SHADER_FRAGMENT: variant = svga->state.hw_draw.fs; count = svga_get_extra_fs_constants(svga, (float *) extras); break; default: assert(!"Unexpected shader type"); count = 0; } assert(variant); offset = variant->shader->info.file_max[TGSI_FILE_CONSTANT] + 1; assert(count <= ARRAY_SIZE(extras)); if (count > 0) { ret = emit_const_range(svga, shader, offset, count, (const float (*) [4])extras); } } return ret; } static enum pipe_error emit_constbuf_vgpu10(struct svga_context *svga, enum pipe_shader_type shader) { const struct pipe_constant_buffer *cbuf; struct pipe_resource *dst_buffer = NULL; enum pipe_error ret = PIPE_OK; struct pipe_transfer *src_transfer; struct svga_winsys_surface *dst_handle; float extras[MAX_EXTRA_CONSTS][4]; unsigned extra_count, extra_size, extra_offset; unsigned new_buf_size; void *src_map = NULL, *dst_map; unsigned offset; const struct svga_shader_variant *variant; unsigned alloc_buf_size; assert(shader == PIPE_SHADER_VERTEX || shader == PIPE_SHADER_GEOMETRY || shader == PIPE_SHADER_FRAGMENT); cbuf = &svga->curr.constbufs[shader][0]; switch (shader) { case PIPE_SHADER_VERTEX: variant = svga->state.hw_draw.vs; extra_count = svga_get_extra_vs_constants(svga, (float *) extras); break; case PIPE_SHADER_FRAGMENT: variant = svga->state.hw_draw.fs; extra_count = svga_get_extra_fs_constants(svga, (float *) extras); break; case PIPE_SHADER_GEOMETRY: variant = svga->state.hw_draw.gs; extra_count = svga_get_extra_gs_constants(svga, (float *) extras); break; default: assert(!"Unexpected shader type"); /* Don't return an error code since we don't want to keep re-trying * this function and getting stuck in an infinite loop. */ return PIPE_OK; } assert(variant); /* Compute extra constants size and offset in bytes */ extra_size = extra_count * 4 * sizeof(float); extra_offset = 4 * sizeof(float) * variant->extra_const_start; if (cbuf->buffer_size + extra_size == 0) return PIPE_OK; /* nothing to do */ /* Typically, the cbuf->buffer here is a user-space buffer so mapping * it is really cheap. If we ever get real HW buffers for constants * we should void mapping and instead use a ResourceCopy command. */ if (cbuf->buffer_size > 0) { src_map = pipe_buffer_map_range(&svga->pipe, cbuf->buffer, cbuf->buffer_offset, cbuf->buffer_size, PIPE_TRANSFER_READ, &src_transfer); assert(src_map); if (!src_map) { return PIPE_ERROR_OUT_OF_MEMORY; } } /* The new/dest buffer's size must be large enough to hold the original, * user-specified constants, plus the extra constants. * The size of the original constant buffer _should_ agree with what the * shader is expecting, but it might not (it's not enforced anywhere by * gallium). */ new_buf_size = MAX2(cbuf->buffer_size, extra_offset) + extra_size; /* According to the DX10 spec, the constant buffer size must be * in multiples of 16. */ new_buf_size = align(new_buf_size, 16); /* Constant buffer size in the upload buffer must be in multiples of 256. * In order to maximize the chance of merging the upload buffer chunks * when svga_buffer_add_range() is called, * the allocate buffer size needs to be in multiples of 256 as well. * Otherwise, since there is gap between each dirty range of the upload buffer, * each dirty range will end up in its own UPDATE_GB_IMAGE command. */ alloc_buf_size = align(new_buf_size, CONST0_UPLOAD_ALIGNMENT); u_upload_alloc(svga->const0_upload, 0, alloc_buf_size, CONST0_UPLOAD_ALIGNMENT, &offset, &dst_buffer, &dst_map); if (!dst_map) { if (src_map) pipe_buffer_unmap(&svga->pipe, src_transfer); return PIPE_ERROR_OUT_OF_MEMORY; } if (src_map) { memcpy(dst_map, src_map, cbuf->buffer_size); pipe_buffer_unmap(&svga->pipe, src_transfer); } if (extra_size) { assert(extra_offset + extra_size <= new_buf_size); memcpy((char *) dst_map + extra_offset, extras, extra_size); } /* Get winsys handle for the constant buffer */ if (svga->state.hw_draw.const0_buffer == dst_buffer && svga->state.hw_draw.const0_handle) { /* re-reference already mapped buffer */ dst_handle = svga->state.hw_draw.const0_handle; } else { /* we must unmap the buffer before getting the winsys handle */ u_upload_unmap(svga->const0_upload); dst_handle = svga_buffer_handle(svga, dst_buffer, PIPE_BIND_CONSTANT_BUFFER); if (!dst_handle) { pipe_resource_reference(&dst_buffer, NULL); return PIPE_ERROR_OUT_OF_MEMORY; } /* save the buffer / handle for next time */ pipe_resource_reference(&svga->state.hw_draw.const0_buffer, dst_buffer); svga->state.hw_draw.const0_handle = dst_handle; } /* Issue the SetSingleConstantBuffer command */ assert(new_buf_size % 16 == 0); ret = SVGA3D_vgpu10_SetSingleConstantBuffer(svga->swc, 0, /* index */ svga_shader_type(shader), dst_handle, offset, new_buf_size); if (ret != PIPE_OK) { pipe_resource_reference(&dst_buffer, NULL); return ret; } /* Save this const buffer until it's replaced in the future. * Otherwise, all references to the buffer will go away after the * command buffer is submitted, it'll get recycled and we will have * incorrect constant buffer bindings. */ pipe_resource_reference(&svga->state.hw_draw.constbuf[shader], dst_buffer); svga->state.hw_draw.default_constbuf_size[shader] = new_buf_size; pipe_resource_reference(&dst_buffer, NULL); svga->hud.num_const_buf_updates++; return ret; } static enum pipe_error emit_consts_vgpu10(struct svga_context *svga, enum pipe_shader_type shader) { enum pipe_error ret; unsigned dirty_constbufs; unsigned enabled_constbufs; /* Emit 0th constant buffer (with extra constants) */ ret = emit_constbuf_vgpu10(svga, shader); if (ret != PIPE_OK) { return ret; } enabled_constbufs = svga->state.hw_draw.enabled_constbufs[shader] | 1u; /* Emit other constant buffers (UBOs) */ dirty_constbufs = svga->state.dirty_constbufs[shader] & ~1u; while (dirty_constbufs) { unsigned index = u_bit_scan(&dirty_constbufs); unsigned offset = svga->curr.constbufs[shader][index].buffer_offset; unsigned size = svga->curr.constbufs[shader][index].buffer_size; struct svga_buffer *buffer = svga_buffer(svga->curr.constbufs[shader][index].buffer); struct svga_winsys_surface *handle; if (buffer) { handle = svga_buffer_handle(svga, &buffer->b.b, PIPE_BIND_CONSTANT_BUFFER); enabled_constbufs |= 1 << index; } else { handle = NULL; enabled_constbufs &= ~(1 << index); assert(offset == 0); assert(size == 0); } if (size % 16 != 0) { /* GL's buffer range sizes can be any number of bytes but the * SVGA3D device requires a multiple of 16 bytes. */ const unsigned total_size = buffer->b.b.width0; if (offset + align(size, 16) <= total_size) { /* round up size to multiple of 16 */ size = align(size, 16); } else { /* round down to mulitple of 16 (this may cause rendering problems * but should avoid a device error). */ size &= ~15; } } assert(size % 16 == 0); ret = SVGA3D_vgpu10_SetSingleConstantBuffer(svga->swc, index, svga_shader_type(shader), handle, offset, size); if (ret != PIPE_OK) return ret; svga->hud.num_const_buf_updates++; } svga->state.hw_draw.enabled_constbufs[shader] = enabled_constbufs; svga->state.dirty_constbufs[shader] = 0; return ret; } static enum pipe_error emit_fs_consts(struct svga_context *svga, unsigned dirty) { const struct svga_shader_variant *variant = svga->state.hw_draw.fs; enum pipe_error ret = PIPE_OK; /* SVGA_NEW_FS_VARIANT */ if (!variant) return PIPE_OK; /* SVGA_NEW_FS_CONST_BUFFER */ if (svga_have_vgpu10(svga)) { ret = emit_consts_vgpu10(svga, PIPE_SHADER_FRAGMENT); } else { ret = emit_consts_vgpu9(svga, PIPE_SHADER_FRAGMENT); } return ret; } struct svga_tracked_state svga_hw_fs_constants = { "hw fs params", (SVGA_NEW_FS_CONST_BUFFER | SVGA_NEW_FS_VARIANT | SVGA_NEW_TEXTURE_CONSTS), emit_fs_consts }; static enum pipe_error emit_vs_consts(struct svga_context *svga, unsigned dirty) { const struct svga_shader_variant *variant = svga->state.hw_draw.vs; enum pipe_error ret = PIPE_OK; /* SVGA_NEW_VS_VARIANT */ if (!variant) return PIPE_OK; /* SVGA_NEW_VS_CONST_BUFFER */ if (svga_have_vgpu10(svga)) { ret = emit_consts_vgpu10(svga, PIPE_SHADER_VERTEX); } else { ret = emit_consts_vgpu9(svga, PIPE_SHADER_VERTEX); } return ret; } struct svga_tracked_state svga_hw_vs_constants = { "hw vs params", (SVGA_NEW_PRESCALE | SVGA_NEW_VS_CONST_BUFFER | SVGA_NEW_VS_VARIANT), emit_vs_consts }; static enum pipe_error emit_gs_consts(struct svga_context *svga, unsigned dirty) { const struct svga_shader_variant *variant = svga->state.hw_draw.gs; enum pipe_error ret = PIPE_OK; /* SVGA_NEW_GS_VARIANT */ if (!variant) return PIPE_OK; /* SVGA_NEW_GS_CONST_BUFFER */ if (svga_have_vgpu10(svga)) { /** * If only the rasterizer state has changed and the current geometry * shader does not emit wide points, then there is no reason to * re-emit the GS constants, so skip it. */ if (dirty == SVGA_NEW_RAST && !variant->key.gs.wide_point) return PIPE_OK; ret = emit_consts_vgpu10(svga, PIPE_SHADER_GEOMETRY); } return ret; } struct svga_tracked_state svga_hw_gs_constants = { "hw gs params", (SVGA_NEW_GS_CONST_BUFFER | SVGA_NEW_RAST | SVGA_NEW_GS_VARIANT), emit_gs_consts };