/********************************************************** * 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_inlines.h" #include "pipe/p_defines.h" #include "util/format/u_format.h" #include "util/u_math.h" #include "util/u_memory.h" #include "util/u_bitmask.h" #include "tgsi/tgsi_ureg.h" #include "svga_context.h" #include "svga_state.h" #include "svga_cmd.h" #include "svga_shader.h" #include "svga_resource_texture.h" #include "svga_tgsi.h" #include "svga_format.h" #include "svga_hw_reg.h" /** * If we fail to compile a fragment shader (because it uses too many * registers, for example) we'll use a dummy/fallback shader that * simply emits a constant color (red for debug, black for release). * We hit this with the Unigine/Heaven demo when Shaders = High. * With black, the demo still looks good. */ static const struct tgsi_token * get_dummy_fragment_shader(void) { #ifdef DEBUG static const float color[4] = { 1.0, 0.0, 0.0, 0.0 }; /* red */ #else static const float color[4] = { 0.0, 0.0, 0.0, 0.0 }; /* black */ #endif struct ureg_program *ureg; const struct tgsi_token *tokens; struct ureg_src src; struct ureg_dst dst; ureg = ureg_create(PIPE_SHADER_FRAGMENT); if (!ureg) return NULL; dst = ureg_DECL_output(ureg, TGSI_SEMANTIC_COLOR, 0); src = ureg_DECL_immediate(ureg, color, 4); ureg_MOV(ureg, dst, src); ureg_END(ureg); tokens = ureg_get_tokens(ureg, NULL); ureg_destroy(ureg); return tokens; } static struct svga_shader_variant * translate_fragment_program(struct svga_context *svga, const struct svga_fragment_shader *fs, const struct svga_compile_key *key) { if (svga_have_vgpu10(svga)) { return svga_tgsi_vgpu10_translate(svga, &fs->base, key, PIPE_SHADER_FRAGMENT); } else { return svga_tgsi_vgpu9_translate(svga, &fs->base, key, PIPE_SHADER_FRAGMENT); } } /** * Replace the given shader's instruction with a simple constant-color * shader. We use this when normal shader translation fails. */ static struct svga_shader_variant * get_compiled_dummy_shader(struct svga_context *svga, struct svga_fragment_shader *fs, const struct svga_compile_key *key) { const struct tgsi_token *dummy = get_dummy_fragment_shader(); struct svga_shader_variant *variant; if (!dummy) { return NULL; } FREE((void *) fs->base.tokens); fs->base.tokens = dummy; tgsi_scan_shader(fs->base.tokens, &fs->base.info); fs->generic_inputs = svga_get_generic_inputs_mask(&fs->base.info); svga_remap_generics(fs->generic_inputs, fs->generic_remap_table); variant = translate_fragment_program(svga, fs, key); return variant; } /** * Translate TGSI shader into an svga shader variant. */ static enum pipe_error compile_fs(struct svga_context *svga, struct svga_fragment_shader *fs, const struct svga_compile_key *key, struct svga_shader_variant **out_variant) { struct svga_shader_variant *variant; enum pipe_error ret = PIPE_ERROR; variant = translate_fragment_program(svga, fs, key); if (variant == NULL) { debug_printf("Failed to compile fragment shader," " using dummy shader instead.\n"); variant = get_compiled_dummy_shader(svga, fs, key); } else if (svga_shader_too_large(svga, variant)) { /* too big, use dummy shader */ debug_printf("Shader too large (%u bytes)," " using dummy shader instead.\n", (unsigned) (variant->nr_tokens * sizeof(variant->tokens[0]))); /* Free the too-large variant */ svga_destroy_shader_variant(svga, variant); /* Use simple pass-through shader instead */ variant = get_compiled_dummy_shader(svga, fs, key); } if (!variant) { return PIPE_ERROR; } ret = svga_define_shader(svga, variant); if (ret != PIPE_OK) { svga_destroy_shader_variant(svga, variant); return ret; } *out_variant = variant; /* insert variant at head of linked list */ variant->next = fs->base.variants; fs->base.variants = variant; return PIPE_OK; } /* SVGA_NEW_TEXTURE_BINDING * SVGA_NEW_RAST * SVGA_NEW_NEED_SWTNL * SVGA_NEW_SAMPLER */ static enum pipe_error make_fs_key(const struct svga_context *svga, struct svga_fragment_shader *fs, struct svga_compile_key *key) { const enum pipe_shader_type shader = PIPE_SHADER_FRAGMENT; unsigned i; memset(key, 0, sizeof *key); memcpy(key->generic_remap_table, fs->generic_remap_table, sizeof(fs->generic_remap_table)); /* SVGA_NEW_GS, SVGA_NEW_VS */ if (svga->curr.gs) { key->fs.gs_generic_outputs = svga->curr.gs->generic_outputs; } else { key->fs.vs_generic_outputs = svga->curr.vs->generic_outputs; } /* Only need fragment shader fixup for twoside lighting if doing * hwtnl. Otherwise the draw module does the whole job for us. * * SVGA_NEW_SWTNL */ if (!svga->state.sw.need_swtnl) { /* SVGA_NEW_RAST, SVGA_NEW_REDUCED_PRIMITIVE */ key->fs.light_twoside = svga->curr.rast->templ.light_twoside; key->fs.front_ccw = svga->curr.rast->templ.front_ccw; key->fs.pstipple = (svga->curr.rast->templ.poly_stipple_enable && svga->curr.reduced_prim == PIPE_PRIM_TRIANGLES); key->fs.aa_point = (svga->curr.rast->templ.point_smooth && svga->curr.reduced_prim == PIPE_PRIM_POINTS && (svga->curr.rast->pointsize > 1.0 || svga->curr.vs->base.info.writes_psize)); if (key->fs.aa_point) { assert(svga->curr.gs != NULL); assert(svga->curr.gs->aa_point_coord_index != -1); key->fs.aa_point_coord_index = svga->curr.gs->aa_point_coord_index; } } /* The blend workaround for simulating logicop xor behaviour * requires that the incoming fragment color be white. This change * achieves that by creating a variant of the current fragment * shader that overrides all output colors with 1,1,1,1 * * This will work for most shaders, including those containing * TEXKIL and/or depth-write. However, it will break on the * combination of xor-logicop plus alphatest. * * Ultimately, we could implement alphatest in the shader using * texkil prior to overriding the outgoing fragment color. * * SVGA_NEW_BLEND */ key->fs.white_fragments = svga->curr.blend->need_white_fragments; key->fs.alpha_to_one = svga->curr.blend->alpha_to_one; #ifdef DEBUG /* * We expect a consistent set of samplers and sampler views. * Do some debug checks/warnings here. */ { static boolean warned = FALSE; unsigned i, n = MAX2(svga->curr.num_sampler_views[shader], svga->curr.num_samplers[shader]); /* Only warn once to prevent too much debug output */ if (!warned) { if (svga->curr.num_sampler_views[shader] != svga->curr.num_samplers[shader]) { debug_printf("svga: mismatched number of sampler views (%u) " "vs. samplers (%u)\n", svga->curr.num_sampler_views[shader], svga->curr.num_samplers[shader]); } for (i = 0; i < n; i++) { if ((svga->curr.sampler_views[shader][i] == NULL) != (svga->curr.sampler[shader][i] == NULL)) debug_printf("sampler_view[%u] = %p but sampler[%u] = %p\n", i, svga->curr.sampler_views[shader][i], i, svga->curr.sampler[shader][i]); } warned = TRUE; } } #endif /* XXX: want to limit this to the textures that the shader actually * refers to. * * SVGA_NEW_TEXTURE_BINDING | SVGA_NEW_SAMPLER */ svga_init_shader_key_common(svga, shader, key); for (i = 0; i < svga->curr.num_samplers[shader]; ++i) { struct pipe_sampler_view *view = svga->curr.sampler_views[shader][i]; const struct svga_sampler_state *sampler = svga->curr.sampler[shader][i]; if (view) { struct pipe_resource *tex = view->texture; if (tex->target != PIPE_BUFFER) { struct svga_texture *stex = svga_texture(tex); SVGA3dSurfaceFormat format = stex->key.format; if (!svga_have_vgpu10(svga) && (format == SVGA3D_Z_D16 || format == SVGA3D_Z_D24X8 || format == SVGA3D_Z_D24S8)) { /* If we're sampling from a SVGA3D_Z_D16, SVGA3D_Z_D24X8, * or SVGA3D_Z_D24S8 surface, we'll automatically get * shadow comparison. But we only get LEQUAL mode. * Set TEX_COMPARE_NONE here so we don't emit the extra FS * code for shadow comparison. */ key->tex[i].compare_mode = PIPE_TEX_COMPARE_NONE; key->tex[i].compare_func = PIPE_FUNC_NEVER; /* These depth formats _only_ support comparison mode and * not ordinary sampling so warn if the later is expected. */ if (sampler->compare_mode != PIPE_TEX_COMPARE_R_TO_TEXTURE) { debug_warn_once("Unsupported shadow compare mode"); } /* The shader translation code can emit code to * handle ALWAYS and NEVER compare functions */ else if (sampler->compare_func == PIPE_FUNC_ALWAYS || sampler->compare_func == PIPE_FUNC_NEVER) { key->tex[i].compare_mode = sampler->compare_mode; key->tex[i].compare_func = sampler->compare_func; } else if (sampler->compare_func != PIPE_FUNC_LEQUAL) { debug_warn_once("Unsupported shadow compare function"); } } else { /* For other texture formats, just use the compare func/mode * as-is. Should be no-ops for color textures. For depth * textures, we do not get automatic depth compare. We have * to do it ourselves in the shader. And we don't get PCF. */ key->tex[i].compare_mode = sampler->compare_mode; key->tex[i].compare_func = sampler->compare_func; } } } } /* sprite coord gen state */ key->sprite_coord_enable = svga->curr.rast->templ.sprite_coord_enable; key->sprite_origin_lower_left = (svga->curr.rast->templ.sprite_coord_mode == PIPE_SPRITE_COORD_LOWER_LEFT); key->fs.flatshade = svga->curr.rast->templ.flatshade; /* SVGA_NEW_DEPTH_STENCIL_ALPHA */ if (svga_have_vgpu10(svga)) { /* Alpha testing is not supported in integer-valued render targets. */ if (svga_has_any_integer_cbufs(svga)) { key->fs.alpha_func = SVGA3D_CMP_ALWAYS; key->fs.alpha_ref = 0; } else { key->fs.alpha_func = svga->curr.depth->alphafunc; key->fs.alpha_ref = svga->curr.depth->alpharef; } } /* SVGA_NEW_FRAME_BUFFER | SVGA_NEW_BLEND */ if (fs->base.info.properties[TGSI_PROPERTY_FS_COLOR0_WRITES_ALL_CBUFS] || svga->curr.blend->need_white_fragments) { /* Replicate color0 output (or white) to N colorbuffers */ key->fs.write_color0_to_n_cbufs = svga->curr.framebuffer.nr_cbufs; } return PIPE_OK; } /** * svga_reemit_fs_bindings - Reemit the fragment shader bindings */ enum pipe_error svga_reemit_fs_bindings(struct svga_context *svga) { enum pipe_error ret; assert(svga->rebind.flags.fs); assert(svga_have_gb_objects(svga)); if (!svga->state.hw_draw.fs) return PIPE_OK; if (!svga_need_to_rebind_resources(svga)) { ret = svga->swc->resource_rebind(svga->swc, NULL, svga->state.hw_draw.fs->gb_shader, SVGA_RELOC_READ); } else { if (svga_have_vgpu10(svga)) ret = SVGA3D_vgpu10_SetShader(svga->swc, SVGA3D_SHADERTYPE_PS, svga->state.hw_draw.fs->gb_shader, svga->state.hw_draw.fs->id); else ret = SVGA3D_SetGBShader(svga->swc, SVGA3D_SHADERTYPE_PS, svga->state.hw_draw.fs->gb_shader); } if (ret != PIPE_OK) return ret; svga->rebind.flags.fs = FALSE; return PIPE_OK; } static enum pipe_error emit_hw_fs(struct svga_context *svga, unsigned dirty) { struct svga_shader_variant *variant = NULL; enum pipe_error ret = PIPE_OK; struct svga_fragment_shader *fs = svga->curr.fs; struct svga_compile_key key; SVGA_STATS_TIME_PUSH(svga_sws(svga), SVGA_STATS_TIME_EMITFS); /* Disable rasterization if rasterizer_discard flag is set or * vs/gs does not output position. */ svga->disable_rasterizer = svga->curr.rast->templ.rasterizer_discard || (svga->curr.gs && !svga->curr.gs->base.info.writes_position) || (!svga->curr.gs && !svga->curr.vs->base.info.writes_position); /* Set FS to NULL when rasterization is to be disabled */ if (svga->disable_rasterizer) { /* Set FS to NULL if it has not been done */ if (svga->state.hw_draw.fs) { ret = svga_set_shader(svga, SVGA3D_SHADERTYPE_PS, NULL); if (ret != PIPE_OK) goto done; } svga->rebind.flags.fs = FALSE; svga->state.hw_draw.fs = NULL; goto done; } /* SVGA_NEW_BLEND * SVGA_NEW_TEXTURE_BINDING * SVGA_NEW_RAST * SVGA_NEW_NEED_SWTNL * SVGA_NEW_SAMPLER * SVGA_NEW_FRAME_BUFFER * SVGA_NEW_DEPTH_STENCIL_ALPHA * SVGA_NEW_VS */ ret = make_fs_key(svga, fs, &key); if (ret != PIPE_OK) goto done; variant = svga_search_shader_key(&fs->base, &key); if (!variant) { ret = compile_fs(svga, fs, &key, &variant); if (ret != PIPE_OK) goto done; } assert(variant); if (variant != svga->state.hw_draw.fs) { ret = svga_set_shader(svga, SVGA3D_SHADERTYPE_PS, variant); if (ret != PIPE_OK) goto done; svga->rebind.flags.fs = FALSE; svga->dirty |= SVGA_NEW_FS_VARIANT; svga->state.hw_draw.fs = variant; } done: SVGA_STATS_TIME_POP(svga_sws(svga)); return ret; } struct svga_tracked_state svga_hw_fs = { "fragment shader (hwtnl)", (SVGA_NEW_FS | SVGA_NEW_GS | SVGA_NEW_VS | SVGA_NEW_TEXTURE_BINDING | SVGA_NEW_NEED_SWTNL | SVGA_NEW_RAST | SVGA_NEW_STIPPLE | SVGA_NEW_REDUCED_PRIMITIVE | SVGA_NEW_SAMPLER | SVGA_NEW_FRAME_BUFFER | SVGA_NEW_DEPTH_STENCIL_ALPHA | SVGA_NEW_BLEND), emit_hw_fs };