X-Git-Url: http://de.git.xonotic.org/?a=blobdiff_plain;f=gl_rmain.c;h=81287ccb794e7826f03cb40b4cb003b6cba3eee9;hb=adf6ed953670cfef7cdc05928435c159f3af228a;hp=ef397326df69437841a17eba60140ee11e9f5b37;hpb=c9f91aea3b2cb789ece619699e7f41db9b88443a;p=xonotic%2Fdarkplaces.git diff --git a/gl_rmain.c b/gl_rmain.c index ef397326..81287ccb 100644 --- a/gl_rmain.c +++ b/gl_rmain.c @@ -44,6 +44,7 @@ static qboolean r_loadgloss; qboolean r_loadfog; static qboolean r_loaddds; static qboolean r_savedds; +static qboolean r_gpuskeletal; // // screen size info @@ -229,6 +230,7 @@ cvar_t r_test = {0, "r_test", "0", "internal development use only, leave it alon cvar_t r_batch_multidraw = {CVAR_SAVE, "r_batch_multidraw", "1", "issue multiple glDrawElements calls when rendering a batch of surfaces with the same texture (otherwise the index data is copied to make it one draw)"}; cvar_t r_batch_multidraw_mintriangles = {CVAR_SAVE, "r_batch_multidraw_mintriangles", "0", "minimum number of triangles to activate multidraw path (copying small groups of triangles may be faster)"}; cvar_t r_batch_debugdynamicvertexpath = {CVAR_SAVE, "r_batch_debugdynamicvertexpath", "0", "force the dynamic batching code path for debugging purposes"}; +cvar_t r_batch_dynamicbuffer = {CVAR_SAVE, "r_batch_dynamicbuffer", "0", "use vertex/index buffers for drawing dynamic and copytriangles batches"}; cvar_t r_glsl_saturation = {CVAR_SAVE, "r_glsl_saturation", "1", "saturation multiplier (only working in glsl!)"}; cvar_t r_glsl_saturation_redcompensate = {CVAR_SAVE, "r_glsl_saturation_redcompensate", "0", "a 'vampire sight' addition to desaturation effect, does compensation for red color, r_glsl_restart is required"}; @@ -236,6 +238,13 @@ cvar_t r_glsl_saturation_redcompensate = {CVAR_SAVE, "r_glsl_saturation_redcompe cvar_t r_glsl_vertextextureblend_usebothalphas = {CVAR_SAVE, "r_glsl_vertextextureblend_usebothalphas", "0", "use both alpha layers on vertex blended surfaces, each alpha layer sets amount of 'blend leak' on another layer, requires mod_q3shader_force_terrain_alphaflag on."}; cvar_t r_framedatasize = {CVAR_SAVE, "r_framedatasize", "0.5", "size of renderer data cache used during one frame (for skeletal animation caching, light processing, etc)"}; +cvar_t r_buffermegs[R_BUFFERDATA_COUNT] = +{ + {CVAR_SAVE, "r_buffermegs_vertex", "4", "vertex buffer size for one frame"}, + {CVAR_SAVE, "r_buffermegs_index16", "1", "index buffer size for one frame (16bit indices)"}, + {CVAR_SAVE, "r_buffermegs_index32", "1", "index buffer size for one frame (32bit indices)"}, + {CVAR_SAVE, "r_buffermegs_uniform", "0.25", "uniform buffer size for one frame"}, +}; extern cvar_t v_glslgamma; extern cvar_t v_glslgamma_2d; @@ -247,6 +256,8 @@ r_framebufferstate_t r_fb; /// shadow volume bsp struct with automatically growing nodes buffer svbsp_t r_svbsp; +int r_uniformbufferalignment = 32; // dynamically updated to match GL_UNIFORM_BUFFER_OFFSET_ALIGNMENT + rtexture_t *r_texture_blanknormalmap; rtexture_t *r_texture_white; rtexture_t *r_texture_grey128; @@ -602,13 +613,17 @@ static void R_BuildFogHeightTexture(void) //======================================================================================================================================================= -static const char *builtinshaderstring = +static const char *builtinshaderstrings[] = +{ #include "shader_glsl.h" -; +0 +}; -const char *builtinhlslshaderstring = +const char *builtinhlslshaderstrings[] = +{ #include "shader_hlsl.h" -; +0 +}; char *glslshaderstring = NULL; char *hlslshaderstring = NULL; @@ -624,9 +639,7 @@ shaderpermutationinfo_t; typedef struct shadermodeinfo_s { - const char *vertexfilename; - const char *geometryfilename; - const char *fragmentfilename; + const char *filename; const char *pretext; const char *name; } @@ -671,46 +684,44 @@ shaderpermutationinfo_t shaderpermutationinfo[SHADERPERMUTATION_COUNT] = // NOTE: MUST MATCH ORDER OF SHADERMODE_* ENUMS! shadermodeinfo_t glslshadermodeinfo[SHADERMODE_COUNT] = { - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_GENERIC\n", " generic"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_POSTPROCESS\n", " postprocess"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_DEPTH_OR_SHADOW\n", " depth/shadow"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_FLATCOLOR\n", " flatcolor"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_VERTEXCOLOR\n", " vertexcolor"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_LIGHTMAP\n", " lightmap"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_FAKELIGHT\n", " fakelight"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_MODELSPACE\n", " lightdirectionmap_modelspace"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_TANGENTSPACE\n", " lightdirectionmap_tangentspace"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_LIGHTMAP\n", " lightdirectionmap_forced_lightmap"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_VERTEXCOLOR\n", " lightdirectionmap_forced_vertexcolor"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_LIGHTDIRECTION\n", " lightdirection"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_LIGHTSOURCE\n", " lightsource"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_REFRACTION\n", " refraction"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_WATER\n", " water"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_SHOWDEPTH\n", " showdepth"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_DEFERREDGEOMETRY\n", " deferredgeometry"}, - {"glsl/default.glsl", NULL, "glsl/default.glsl", "#define MODE_DEFERREDLIGHTSOURCE\n", " deferredlightsource"}, + {"glsl/default.glsl", "#define MODE_GENERIC\n", " generic"}, + {"glsl/default.glsl", "#define MODE_POSTPROCESS\n", " postprocess"}, + {"glsl/default.glsl", "#define MODE_DEPTH_OR_SHADOW\n", " depth/shadow"}, + {"glsl/default.glsl", "#define MODE_FLATCOLOR\n", " flatcolor"}, + {"glsl/default.glsl", "#define MODE_VERTEXCOLOR\n", " vertexcolor"}, + {"glsl/default.glsl", "#define MODE_LIGHTMAP\n", " lightmap"}, + {"glsl/default.glsl", "#define MODE_FAKELIGHT\n", " fakelight"}, + {"glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_MODELSPACE\n", " lightdirectionmap_modelspace"}, + {"glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_TANGENTSPACE\n", " lightdirectionmap_tangentspace"}, + {"glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_LIGHTMAP\n", " lightdirectionmap_forced_lightmap"}, + {"glsl/default.glsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_VERTEXCOLOR\n", " lightdirectionmap_forced_vertexcolor"}, + {"glsl/default.glsl", "#define MODE_LIGHTDIRECTION\n", " lightdirection"}, + {"glsl/default.glsl", "#define MODE_LIGHTSOURCE\n", " lightsource"}, + {"glsl/default.glsl", "#define MODE_REFRACTION\n", " refraction"}, + {"glsl/default.glsl", "#define MODE_WATER\n", " water"}, + {"glsl/default.glsl", "#define MODE_DEFERREDGEOMETRY\n", " deferredgeometry"}, + {"glsl/default.glsl", "#define MODE_DEFERREDLIGHTSOURCE\n", " deferredlightsource"}, }; shadermodeinfo_t hlslshadermodeinfo[SHADERMODE_COUNT] = { - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_GENERIC\n", " generic"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_POSTPROCESS\n", " postprocess"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_DEPTH_OR_SHADOW\n", " depth/shadow"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_FLATCOLOR\n", " flatcolor"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_VERTEXCOLOR\n", " vertexcolor"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_LIGHTMAP\n", " lightmap"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_FAKELIGHT\n", " fakelight"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_MODELSPACE\n", " lightdirectionmap_modelspace"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_TANGENTSPACE\n", " lightdirectionmap_tangentspace"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_LIGHTMAP\n", " lightdirectionmap_forced_lightmap"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_VERTEXCOLOR\n", " lightdirectionmap_forced_vertexcolor"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_LIGHTDIRECTION\n", " lightdirection"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_LIGHTSOURCE\n", " lightsource"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_REFRACTION\n", " refraction"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_WATER\n", " water"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_SHOWDEPTH\n", " showdepth"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_DEFERREDGEOMETRY\n", " deferredgeometry"}, - {"hlsl/default.hlsl", NULL, "hlsl/default.hlsl", "#define MODE_DEFERREDLIGHTSOURCE\n", " deferredlightsource"}, + {"hlsl/default.hlsl", "#define MODE_GENERIC\n", " generic"}, + {"hlsl/default.hlsl", "#define MODE_POSTPROCESS\n", " postprocess"}, + {"hlsl/default.hlsl", "#define MODE_DEPTH_OR_SHADOW\n", " depth/shadow"}, + {"hlsl/default.hlsl", "#define MODE_FLATCOLOR\n", " flatcolor"}, + {"hlsl/default.hlsl", "#define MODE_VERTEXCOLOR\n", " vertexcolor"}, + {"hlsl/default.hlsl", "#define MODE_LIGHTMAP\n", " lightmap"}, + {"hlsl/default.hlsl", "#define MODE_FAKELIGHT\n", " fakelight"}, + {"hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_MODELSPACE\n", " lightdirectionmap_modelspace"}, + {"hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_TANGENTSPACE\n", " lightdirectionmap_tangentspace"}, + {"hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_LIGHTMAP\n", " lightdirectionmap_forced_lightmap"}, + {"hlsl/default.hlsl", "#define MODE_LIGHTDIRECTIONMAP_FORCED_VERTEXCOLOR\n", " lightdirectionmap_forced_vertexcolor"}, + {"hlsl/default.hlsl", "#define MODE_LIGHTDIRECTION\n", " lightdirection"}, + {"hlsl/default.hlsl", "#define MODE_LIGHTSOURCE\n", " lightsource"}, + {"hlsl/default.hlsl", "#define MODE_REFRACTION\n", " refraction"}, + {"hlsl/default.hlsl", "#define MODE_WATER\n", " water"}, + {"hlsl/default.hlsl", "#define MODE_DEFERREDGEOMETRY\n", " deferredgeometry"}, + {"hlsl/default.hlsl", "#define MODE_DEFERREDLIGHTSOURCE\n", " deferredlightsource"}, }; struct r_glsl_permutation_s; @@ -843,6 +854,10 @@ typedef struct r_glsl_permutation_s int loc_NormalmapScrollBlend; int loc_BounceGridMatrix; int loc_BounceGridIntensity; + /// uniform block bindings + int ubibind_Skeletal_Transform12_UniformBlock; + /// uniform block indices + int ubiloc_Skeletal_Transform12_UniformBlock; } r_glsl_permutation_t; @@ -975,25 +990,68 @@ static r_glsl_permutation_t *R_GLSL_FindPermutation(unsigned int mode, unsigned return p; } -static char *R_GLSL_GetText(const char *filename, qboolean printfromdisknotice) +static char *R_ShaderStrCat(const char **strings) +{ + char *string, *s; + const char **p = strings; + const char *t; + size_t len = 0; + for (p = strings;(t = *p);p++) + len += strlen(t); + len++; + s = string = (char *)Mem_Alloc(r_main_mempool, len); + len = 0; + for (p = strings;(t = *p);p++) + { + len = strlen(t); + memcpy(s, t, len); + s += len; + } + *s = 0; + return string; +} + +static char *R_GetShaderText(const char *filename, qboolean printfromdisknotice, qboolean builtinonly) { char *shaderstring; if (!filename || !filename[0]) return NULL; + // LordHavoc: note that FS_LoadFile appends a 0 byte to make it a valid string, so does R_ShaderStrCat if (!strcmp(filename, "glsl/default.glsl")) { + if (builtinonly) + return R_ShaderStrCat(builtinshaderstrings); if (!glslshaderstring) { glslshaderstring = (char *)FS_LoadFile(filename, r_main_mempool, false, NULL); if (glslshaderstring) Con_DPrintf("Loading shaders from file %s...\n", filename); else - glslshaderstring = (char *)builtinshaderstring; + glslshaderstring = R_ShaderStrCat(builtinshaderstrings); } shaderstring = (char *) Mem_Alloc(r_main_mempool, strlen(glslshaderstring) + 1); memcpy(shaderstring, glslshaderstring, strlen(glslshaderstring) + 1); return shaderstring; } + if (!strcmp(filename, "hlsl/default.hlsl")) + { + if (builtinonly) + return R_ShaderStrCat(builtinhlslshaderstrings); + if (!hlslshaderstring) + { + hlslshaderstring = (char *)FS_LoadFile(filename, r_main_mempool, false, NULL); + if (hlslshaderstring) + Con_DPrintf("Loading shaders from file %s...\n", filename); + else + hlslshaderstring = R_ShaderStrCat(builtinhlslshaderstrings); + } + shaderstring = (char *) Mem_Alloc(r_main_mempool, strlen(hlslshaderstring) + 1); + memcpy(shaderstring, hlslshaderstring, strlen(hlslshaderstring) + 1); + return shaderstring; + } + // we don't have builtin strings for any other files + if (builtinonly) + return NULL; shaderstring = (char *)FS_LoadFile(filename, r_main_mempool, false, NULL); if (shaderstring) { @@ -1007,9 +1065,10 @@ static char *R_GLSL_GetText(const char *filename, qboolean printfromdisknotice) static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode, unsigned int permutation) { int i; + int ubibind; int sampler; shadermodeinfo_t *modeinfo = glslshadermodeinfo + mode; - char *vertexstring, *geometrystring, *fragmentstring; + char *sourcestring; char permutationname[256]; int vertstrings_count = 0; int geomstrings_count = 0; @@ -1024,14 +1083,22 @@ static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode p->program = 0; permutationname[0] = 0; - vertexstring = R_GLSL_GetText(modeinfo->vertexfilename, true); - geometrystring = R_GLSL_GetText(modeinfo->geometryfilename, false); - fragmentstring = R_GLSL_GetText(modeinfo->fragmentfilename, false); + sourcestring = R_GetShaderText(modeinfo->filename, true, false); - strlcat(permutationname, modeinfo->vertexfilename, sizeof(permutationname)); + strlcat(permutationname, modeinfo->filename, sizeof(permutationname)); + // we need 140 for r_glsl_skeletal (GL_ARB_uniform_buffer_object) + if(vid.support.glshaderversion >= 140) + { + vertstrings_list[vertstrings_count++] = "#version 140\n"; + geomstrings_list[geomstrings_count++] = "#version 140\n"; + fragstrings_list[fragstrings_count++] = "#version 140\n"; + vertstrings_list[vertstrings_count++] = "#define GLSL140\n"; + geomstrings_list[geomstrings_count++] = "#define GLSL140\n"; + fragstrings_list[fragstrings_count++] = "#define GLSL140\n"; + } // if we can do #version 130, we should (this improves quality of offset/reliefmapping thanks to textureGrad) - if(vid.support.gl20shaders130) + else if(vid.support.glshaderversion >= 130) { vertstrings_list[vertstrings_count++] = "#version 130\n"; geomstrings_list[geomstrings_count++] = "#version 130\n"; @@ -1082,17 +1149,9 @@ static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode fragstrings_count += shaderstaticparms_count; // now append the shader text itself - vertstrings_list[vertstrings_count++] = vertexstring; - geomstrings_list[geomstrings_count++] = geometrystring; - fragstrings_list[fragstrings_count++] = fragmentstring; - - // if any sources were NULL, clear the respective list - if (!vertexstring) - vertstrings_count = 0; - if (!geometrystring) - geomstrings_count = 0; - if (!fragmentstring) - fragstrings_count = 0; + vertstrings_list[vertstrings_count++] = sourcestring; + geomstrings_list[geomstrings_count++] = sourcestring; + fragstrings_list[fragstrings_count++] = sourcestring; // compile the shader program if (vertstrings_count + geomstrings_count + fragstrings_count) @@ -1104,6 +1163,26 @@ static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode // look up all the uniform variable names we care about, so we don't // have to look them up every time we set them +#if 0 + // debugging aid + { + GLint activeuniformindex = 0; + GLint numactiveuniforms = 0; + char uniformname[128]; + GLsizei uniformnamelength = 0; + GLint uniformsize = 0; + GLenum uniformtype = 0; + memset(uniformname, 0, sizeof(uniformname)); + qglGetProgramiv(p->program, GL_ACTIVE_UNIFORMS, &numactiveuniforms); + Con_Printf("Shader has %i uniforms\n", numactiveuniforms); + for (activeuniformindex = 0;activeuniformindex < numactiveuniforms;activeuniformindex++) + { + qglGetActiveUniform(p->program, activeuniformindex, sizeof(uniformname) - 1, &uniformnamelength, &uniformsize, &uniformtype, uniformname); + Con_Printf("Uniform %i name \"%s\" size %i type %i\n", (int)activeuniformindex, uniformname, (int)uniformsize, (int)uniformtype); + } + } +#endif + p->loc_Texture_First = qglGetUniformLocation(p->program, "Texture_First"); p->loc_Texture_Second = qglGetUniformLocation(p->program, "Texture_Second"); p->loc_Texture_GammaRamps = qglGetUniformLocation(p->program, "Texture_GammaRamps"); @@ -1172,7 +1251,6 @@ static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode p->loc_ShadowMap_Parameters = qglGetUniformLocation(p->program, "ShadowMap_Parameters"); p->loc_ShadowMap_TextureScale = qglGetUniformLocation(p->program, "ShadowMap_TextureScale"); p->loc_SpecularPower = qglGetUniformLocation(p->program, "SpecularPower"); - p->loc_Skeletal_Transform12 = qglGetUniformLocation(p->program, "Skeletal_Transform12"); p->loc_UserVec1 = qglGetUniformLocation(p->program, "UserVec1"); p->loc_UserVec2 = qglGetUniformLocation(p->program, "UserVec2"); p->loc_UserVec3 = qglGetUniformLocation(p->program, "UserVec3"); @@ -1221,6 +1299,7 @@ static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode p->tex_Texture_ReflectMask = -1; p->tex_Texture_ReflectCube = -1; p->tex_Texture_BounceGrid = -1; + // bind the texture samplers in use sampler = 0; if (p->loc_Texture_First >= 0) {p->tex_Texture_First = sampler;qglUniform1i(p->loc_Texture_First , sampler);sampler++;} if (p->loc_Texture_Second >= 0) {p->tex_Texture_Second = sampler;qglUniform1i(p->loc_Texture_Second , sampler);sampler++;} @@ -1251,6 +1330,21 @@ static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode if (p->loc_Texture_ReflectMask >= 0) {p->tex_Texture_ReflectMask = sampler;qglUniform1i(p->loc_Texture_ReflectMask , sampler);sampler++;} if (p->loc_Texture_ReflectCube >= 0) {p->tex_Texture_ReflectCube = sampler;qglUniform1i(p->loc_Texture_ReflectCube , sampler);sampler++;} if (p->loc_Texture_BounceGrid >= 0) {p->tex_Texture_BounceGrid = sampler;qglUniform1i(p->loc_Texture_BounceGrid , sampler);sampler++;} + // get the uniform block indices so we can bind them +#ifndef USE_GLES2 /* FIXME: GLES3 only */ + if (vid.support.arb_uniform_buffer_object) + p->ubiloc_Skeletal_Transform12_UniformBlock = qglGetUniformBlockIndex(p->program, "Skeletal_Transform12_UniformBlock"); + else +#endif + p->ubiloc_Skeletal_Transform12_UniformBlock = -1; + // clear the uniform block bindings + p->ubibind_Skeletal_Transform12_UniformBlock = -1; + // bind the uniform blocks in use + ubibind = 0; +#ifndef USE_GLES2 /* FIXME: GLES3 only */ + if (p->ubiloc_Skeletal_Transform12_UniformBlock >= 0) {p->ubibind_Skeletal_Transform12_UniformBlock = ubibind;qglUniformBlockBinding(p->program, p->ubiloc_Skeletal_Transform12_UniformBlock, ubibind);ubibind++;} +#endif + // we're done compiling and setting up the shader, at least until it is used CHECKGLERROR Con_DPrintf("^5GLSL shader %s compiled (%i textures).\n", permutationname, sampler); } @@ -1258,12 +1352,8 @@ static void R_GLSL_CompilePermutation(r_glsl_permutation_t *p, unsigned int mode Con_Printf("^1GLSL shader %s failed! some features may not work properly.\n", permutationname); // free the strings - if (vertexstring) - Mem_Free(vertexstring); - if (geometrystring) - Mem_Free(geometrystring); - if (fragmentstring) - Mem_Free(fragmentstring); + if (sourcestring) + Mem_Free(sourcestring); } static void R_SetupShader_SetPermutationGLSL(unsigned int mode, unsigned int permutation) @@ -1275,7 +1365,10 @@ static void R_SetupShader_SetPermutationGLSL(unsigned int mode, unsigned int per if (!r_glsl_permutation->program) { if (!r_glsl_permutation->compiled) + { + Con_DPrintf("Compiling shader mode %u permutation %u\n", mode, permutation); R_GLSL_CompilePermutation(perm, mode, permutation); + } if (!r_glsl_permutation->program) { // remove features until we find a valid permutation @@ -1295,7 +1388,7 @@ static void R_SetupShader_SetPermutationGLSL(unsigned int mode, unsigned int per } if (i >= SHADERPERMUTATION_COUNT) { - //Con_Printf("Could not find a working OpenGL 2.0 shader for permutation %s %s\n", shadermodeinfo[mode].vertexfilename, shadermodeinfo[mode].pretext); + //Con_Printf("Could not find a working OpenGL 2.0 shader for permutation %s %s\n", shadermodeinfo[mode].filename, shadermodeinfo[mode].pretext); r_glsl_permutation = R_GLSL_FindPermutation(mode, permutation); qglUseProgram(0);CHECKGLERROR return; // no bit left to clear, entire mode is broken @@ -1308,6 +1401,7 @@ static void R_SetupShader_SetPermutationGLSL(unsigned int mode, unsigned int per if (r_glsl_permutation->loc_ModelViewProjectionMatrix >= 0) qglUniformMatrix4fv(r_glsl_permutation->loc_ModelViewProjectionMatrix, 1, false, gl_modelviewprojection16f); if (r_glsl_permutation->loc_ModelViewMatrix >= 0) qglUniformMatrix4fv(r_glsl_permutation->loc_ModelViewMatrix, 1, false, gl_modelview16f); if (r_glsl_permutation->loc_ClientTime >= 0) qglUniform1f(r_glsl_permutation->loc_ClientTime, cl.time); + CHECKGLERROR } #ifdef SUPPORTD3D @@ -1436,35 +1530,6 @@ static r_hlsl_permutation_t *R_HLSL_FindPermutation(unsigned int mode, unsigned return p; } -static char *R_HLSL_GetText(const char *filename, qboolean printfromdisknotice) -{ - char *shaderstring; - if (!filename || !filename[0]) - return NULL; - if (!strcmp(filename, "hlsl/default.hlsl")) - { - if (!hlslshaderstring) - { - hlslshaderstring = (char *)FS_LoadFile(filename, r_main_mempool, false, NULL); - if (hlslshaderstring) - Con_DPrintf("Loading shaders from file %s...\n", filename); - else - hlslshaderstring = (char *)builtinhlslshaderstring; - } - shaderstring = (char *) Mem_Alloc(r_main_mempool, strlen(hlslshaderstring) + 1); - memcpy(shaderstring, hlslshaderstring, strlen(hlslshaderstring) + 1); - return shaderstring; - } - shaderstring = (char *)FS_LoadFile(filename, r_main_mempool, false, NULL); - if (shaderstring) - { - if (printfromdisknotice) - Con_DPrintf("from disk %s... ", filename); - return shaderstring; - } - return shaderstring; -} - #include //#include //#include @@ -1533,6 +1598,18 @@ static void R_HLSL_CacheShader(r_hlsl_permutation_t *p, const char *cachename, c {"D3DXCompileShader", (void **) &qD3DXCompileShader}, {NULL, NULL} }; + // LordHavoc: the June 2010 SDK lacks these macros to make ID3DXBuffer usable in C, and to make it work in both C and C++ the macros are needed... +#ifndef ID3DXBuffer_GetBufferPointer +#if !defined(__cplusplus) || defined(CINTERFACE) +#define ID3DXBuffer_GetBufferPointer(p) (p)->lpVtbl->GetBufferPointer(p) +#define ID3DXBuffer_GetBufferSize(p) (p)->lpVtbl->GetBufferSize(p) +#define ID3DXBuffer_Release(p) (p)->lpVtbl->Release(p) +#else +#define ID3DXBuffer_GetBufferPointer(p) (p)->GetBufferPointer() +#define ID3DXBuffer_GetBufferSize(p) (p)->GetBufferSize() +#define ID3DXBuffer_Release(p) (p)->Release() +#endif +#endif if (Sys_LoadLibrary(dllnames_d3dx9, &d3dx9_dll, d3dx9_dllfuncs)) { DWORD shaderflags = 0; @@ -1613,7 +1690,7 @@ static void R_HLSL_CompilePermutation(r_hlsl_permutation_t *p, unsigned int mode int geomstring_length = 0; int fragstring_length = 0; char *t; - char *vertexstring, *geometrystring, *fragmentstring; + char *sourcestring; char *vertstring, *geomstring, *fragstring; char permutationname[256]; char cachename[256]; @@ -1632,11 +1709,9 @@ static void R_HLSL_CompilePermutation(r_hlsl_permutation_t *p, unsigned int mode permutationname[0] = 0; cachename[0] = 0; - vertexstring = R_HLSL_GetText(modeinfo->vertexfilename, true); - geometrystring = R_HLSL_GetText(modeinfo->geometryfilename, false); - fragmentstring = R_HLSL_GetText(modeinfo->fragmentfilename, false); + sourcestring = R_GetShaderText(modeinfo->filename, true, false); - strlcat(permutationname, modeinfo->vertexfilename, sizeof(permutationname)); + strlcat(permutationname, modeinfo->filename, sizeof(permutationname)); strlcat(cachename, "hlsl/", sizeof(cachename)); // define HLSL so that the shader can tell apart the HLSL compiler and the Cg compiler @@ -1695,17 +1770,9 @@ static void R_HLSL_CompilePermutation(r_hlsl_permutation_t *p, unsigned int mode cachename[i] = '_'; // now append the shader text itself - vertstrings_list[vertstrings_count++] = vertexstring; - geomstrings_list[geomstrings_count++] = geometrystring; - fragstrings_list[fragstrings_count++] = fragmentstring; - - // if any sources were NULL, clear the respective list - if (!vertexstring) - vertstrings_count = 0; - if (!geometrystring) - geomstrings_count = 0; - if (!fragmentstring) - fragstrings_count = 0; + vertstrings_list[vertstrings_count++] = sourcestring; + geomstrings_list[geomstrings_count++] = sourcestring; + fragstrings_list[fragstrings_count++] = sourcestring; vertstring_length = 0; for (i = 0;i < vertstrings_count;i++) @@ -1743,12 +1810,8 @@ static void R_HLSL_CompilePermutation(r_hlsl_permutation_t *p, unsigned int mode Mem_Free(geomstring); if (fragstring) Mem_Free(fragstring); - if (vertexstring) - Mem_Free(vertexstring); - if (geometrystring) - Mem_Free(geometrystring); - if (fragmentstring) - Mem_Free(fragmentstring); + if (sourcestring) + Mem_Free(sourcestring); } static inline void hlslVSSetParameter16f(D3DVSREGISTER_t r, const float *a) {IDirect3DDevice9_SetVertexShaderConstantF(vid_d3d9dev, r, a, 4);} @@ -1794,7 +1857,7 @@ void R_SetupShader_SetPermutationHLSL(unsigned int mode, unsigned int permutatio } if (i >= SHADERPERMUTATION_COUNT) { - //Con_Printf("Could not find a working HLSL shader for permutation %s %s\n", shadermodeinfo[mode].vertexfilename, shadermodeinfo[mode].pretext); + //Con_Printf("Could not find a working HLSL shader for permutation %s %s\n", shadermodeinfo[mode].filename, shadermodeinfo[mode].pretext); r_hlsl_permutation = R_HLSL_FindPermutation(mode, permutation); return; // no bit left to clear, entire mode is broken } @@ -1820,10 +1883,10 @@ static void R_SetupShader_SetPermutationSoft(unsigned int mode, unsigned int per void R_GLSL_Restart_f(void) { unsigned int i, limit; - if (glslshaderstring && glslshaderstring != builtinshaderstring) + if (glslshaderstring) Mem_Free(glslshaderstring); glslshaderstring = NULL; - if (hlslshaderstring && hlslshaderstring != builtinhlslshaderstring) + if (hlslshaderstring) Mem_Free(hlslshaderstring); hlslshaderstring = NULL; switch(vid.renderpath) @@ -1883,42 +1946,44 @@ void R_GLSL_Restart_f(void) static void R_GLSL_DumpShader_f(void) { - int i; + int i, language, mode, dupe; + char *text; + shadermodeinfo_t *modeinfo; qfile_t *file; - file = FS_OpenRealFile("glsl/default.glsl", "w", false); - if (file) + for (language = 0;language < 2;language++) { - FS_Print(file, "/* The engine may define the following macros:\n"); - FS_Print(file, "#define VERTEX_SHADER\n#define GEOMETRY_SHADER\n#define FRAGMENT_SHADER\n"); - for (i = 0;i < SHADERMODE_COUNT;i++) - FS_Print(file, glslshadermodeinfo[i].pretext); - for (i = 0;i < SHADERPERMUTATION_COUNT;i++) - FS_Print(file, shaderpermutationinfo[i].pretext); - FS_Print(file, "*/\n"); - FS_Print(file, builtinshaderstring); - FS_Close(file); - Con_Printf("glsl/default.glsl written\n"); - } - else - Con_Printf("failed to write to glsl/default.glsl\n"); - - file = FS_OpenRealFile("hlsl/default.hlsl", "w", false); - if (file) - { - FS_Print(file, "/* The engine may define the following macros:\n"); - FS_Print(file, "#define VERTEX_SHADER\n#define GEOMETRY_SHADER\n#define FRAGMENT_SHADER\n"); - for (i = 0;i < SHADERMODE_COUNT;i++) - FS_Print(file, hlslshadermodeinfo[i].pretext); - for (i = 0;i < SHADERPERMUTATION_COUNT;i++) - FS_Print(file, shaderpermutationinfo[i].pretext); - FS_Print(file, "*/\n"); - FS_Print(file, builtinhlslshaderstring); - FS_Close(file); - Con_Printf("hlsl/default.hlsl written\n"); + modeinfo = (language == 0 ? glslshadermodeinfo : hlslshadermodeinfo); + for (mode = 0;mode < SHADERMODE_COUNT;mode++) + { + // don't dump the same file multiple times (most or all shaders come from the same file) + for (dupe = mode - 1;dupe >= 0;dupe--) + if (!strcmp(modeinfo[mode].filename, modeinfo[dupe].filename)) + break; + if (dupe >= 0) + continue; + text = R_GetShaderText(modeinfo[mode].filename, false, true); + if (!text) + continue; + file = FS_OpenRealFile(modeinfo[mode].filename, "w", false); + if (file) + { + FS_Print(file, "/* The engine may define the following macros:\n"); + FS_Print(file, "#define VERTEX_SHADER\n#define GEOMETRY_SHADER\n#define FRAGMENT_SHADER\n"); + for (i = 0;i < SHADERMODE_COUNT;i++) + FS_Print(file, modeinfo[i].pretext); + for (i = 0;i < SHADERPERMUTATION_COUNT;i++) + FS_Print(file, shaderpermutationinfo[i].pretext); + FS_Print(file, "*/\n"); + FS_Print(file, text); + FS_Close(file); + Con_Printf("%s written\n", modeinfo[mode].filename); + } + else + Con_Printf("failed to write to %s\n", modeinfo[mode].filename); + Mem_Free(text); + } } - else - Con_Printf("failed to write to hlsl/default.hlsl\n"); } void R_SetupShader_Generic(rtexture_t *first, rtexture_t *second, int texturemode, int rgbscale, qboolean usegamma, qboolean notrippy, qboolean suppresstexalpha) @@ -1965,8 +2030,10 @@ void R_SetupShader_Generic(rtexture_t *first, rtexture_t *second, int texturemod case RENDERPATH_GL20: case RENDERPATH_GLES2: R_SetupShader_SetPermutationGLSL(SHADERMODE_GENERIC, permutation); - R_Mesh_TexBind(r_glsl_permutation->tex_Texture_First , first ); - R_Mesh_TexBind(r_glsl_permutation->tex_Texture_Second, second); + if (r_glsl_permutation->tex_Texture_First >= 0) + R_Mesh_TexBind(r_glsl_permutation->tex_Texture_First , first ); + if (r_glsl_permutation->tex_Texture_Second >= 0) + R_Mesh_TexBind(r_glsl_permutation->tex_Texture_Second, second); if (r_glsl_permutation->tex_Texture_GammaRamps >= 0) R_Mesh_TexBind(r_glsl_permutation->tex_Texture_GammaRamps, r_texture_gammaramps); break; @@ -1974,12 +2041,18 @@ void R_SetupShader_Generic(rtexture_t *first, rtexture_t *second, int texturemod case RENDERPATH_GLES1: R_Mesh_TexBind(0, first ); R_Mesh_TexCombine(0, GL_MODULATE, GL_MODULATE, 1, 1); + R_Mesh_TexMatrix(0, NULL); R_Mesh_TexBind(1, second); if (second) + { R_Mesh_TexCombine(1, texturemode, texturemode, rgbscale, 1); + R_Mesh_TexMatrix(1, NULL); + } break; case RENDERPATH_GL11: R_Mesh_TexBind(0, first ); + R_Mesh_TexCombine(0, GL_MODULATE, GL_MODULATE, 1, 1); + R_Mesh_TexMatrix(0, NULL); break; case RENDERPATH_SOFT: R_SetupShader_SetPermutationSoft(SHADERMODE_GENERIC, permutation); @@ -1994,13 +2067,16 @@ void R_SetupShader_Generic_NoTexture(qboolean usegamma, qboolean notrippy) R_SetupShader_Generic(NULL, NULL, GL_MODULATE, 1, usegamma, notrippy, false); } -void R_SetupShader_DepthOrShadow(qboolean notrippy, qboolean depthrgb) +void R_SetupShader_DepthOrShadow(qboolean notrippy, qboolean depthrgb, qboolean skeletal) { unsigned int permutation = 0; if (r_trippy.integer && !notrippy) permutation |= SHADERPERMUTATION_TRIPPY; if (depthrgb) permutation |= SHADERPERMUTATION_DEPTHRGB; + if (skeletal) + permutation |= SHADERPERMUTATION_SKELETAL; + if (vid.allowalphatocoverage) GL_AlphaToCoverage(false); switch (vid.renderpath) @@ -2019,6 +2095,9 @@ void R_SetupShader_DepthOrShadow(qboolean notrippy, qboolean depthrgb) case RENDERPATH_GL20: case RENDERPATH_GLES2: R_SetupShader_SetPermutationGLSL(SHADERMODE_DEPTH_OR_SHADOW, permutation); +#ifndef USE_GLES2 /* FIXME: GLES3 only */ + if (r_glsl_permutation->ubiloc_Skeletal_Transform12_UniformBlock >= 0 && rsurface.batchskeletaltransform3x4buffer) qglBindBufferRange(GL_UNIFORM_BUFFER, r_glsl_permutation->ubibind_Skeletal_Transform12_UniformBlock, rsurface.batchskeletaltransform3x4buffer->bufferobject, rsurface.batchskeletaltransform3x4offset, rsurface.batchskeletaltransform3x4size); +#endif break; case RENDERPATH_GL13: case RENDERPATH_GLES1: @@ -2034,41 +2113,6 @@ void R_SetupShader_DepthOrShadow(qboolean notrippy, qboolean depthrgb) } } -void R_SetupShader_ShowDepth(qboolean notrippy) -{ - int permutation = 0; - if (r_trippy.integer && !notrippy) - permutation |= SHADERPERMUTATION_TRIPPY; - if (vid.allowalphatocoverage) - GL_AlphaToCoverage(false); - switch (vid.renderpath) - { - case RENDERPATH_D3D9: -#ifdef SUPPORTHLSL - R_SetupShader_SetPermutationHLSL(SHADERMODE_SHOWDEPTH, permutation); -#endif - break; - case RENDERPATH_D3D10: - Con_DPrintf("FIXME D3D10 %s:%i %s\n", __FILE__, __LINE__, __FUNCTION__); - break; - case RENDERPATH_D3D11: - Con_DPrintf("FIXME D3D11 %s:%i %s\n", __FILE__, __LINE__, __FUNCTION__); - break; - case RENDERPATH_GL20: - case RENDERPATH_GLES2: - R_SetupShader_SetPermutationGLSL(SHADERMODE_SHOWDEPTH, permutation); - break; - case RENDERPATH_GL13: - case RENDERPATH_GLES1: - break; - case RENDERPATH_GL11: - break; - case RENDERPATH_SOFT: - R_SetupShader_SetPermutationSoft(SHADERMODE_SHOWDEPTH, permutation); - break; - } -} - extern qboolean r_shadow_usingdeferredprepass; extern rtexture_t *r_shadow_attenuationgradienttexture; extern rtexture_t *r_shadow_attenuation2dtexture; @@ -2538,7 +2582,7 @@ void R_SetupShader_Surface(const vec3_t lightcolorbase, qboolean modellighting, case RENDERPATH_D3D9: #ifdef SUPPORTD3D RSurf_PrepareVerticesForBatch(BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR | (rsurface.modellightmapcolor4f ? BATCHNEED_VERTEXMESH_VERTEXCOLOR : 0) | BATCHNEED_VERTEXMESH_TEXCOORD | (rsurface.uselightmaptexture ? BATCHNEED_VERTEXMESH_LIGHTMAP : 0) | BATCHNEED_ALLOWMULTIDRAW, texturenumsurfaces, texturesurfacelist); - R_Mesh_PrepareVertices_Mesh(rsurface.batchnumvertices, rsurface.batchvertexmesh, rsurface.batchvertexmeshbuffer); + R_Mesh_PrepareVertices_Mesh(rsurface.batchnumvertices, rsurface.batchvertexmesh, rsurface.batchvertexmesh_vertexbuffer, rsurface.batchvertexmesh_bufferoffset); R_SetupShader_SetPermutationHLSL(mode, permutation); Matrix4x4_ToArrayFloatGL(&rsurface.matrix, m16f);hlslPSSetParameter16f(D3DPSREGISTER_ModelToReflectCube, m16f); if (mode == SHADERMODE_LIGHTSOURCE) @@ -2705,12 +2749,15 @@ void R_SetupShader_Surface(const vec3_t lightcolorbase, qboolean modellighting, else { RSurf_PrepareVerticesForBatch(BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR | (rsurface.modellightmapcolor4f ? BATCHNEED_VERTEXMESH_VERTEXCOLOR : 0) | BATCHNEED_VERTEXMESH_TEXCOORD | (rsurface.uselightmaptexture ? BATCHNEED_VERTEXMESH_LIGHTMAP : 0) | (rsurface.entityskeletaltransform3x4 ? BATCHNEED_VERTEXMESH_SKELETAL : 0) | BATCHNEED_ALLOWMULTIDRAW, texturenumsurfaces, texturesurfacelist); - R_Mesh_PrepareVertices_Mesh(rsurface.batchnumvertices, rsurface.batchvertexmesh, rsurface.batchvertexmeshbuffer); + R_Mesh_PrepareVertices_Mesh(rsurface.batchnumvertices, rsurface.batchvertexmesh, rsurface.batchvertexmesh_vertexbuffer, rsurface.batchvertexmesh_bufferoffset); } // this has to be after RSurf_PrepareVerticesForBatch - if (rsurface.batchskeletaltransform3x4) + if (rsurface.batchskeletaltransform3x4buffer) permutation |= SHADERPERMUTATION_SKELETAL; R_SetupShader_SetPermutationGLSL(mode, permutation); +#ifndef USE_GLES2 /* FIXME: GLES3 only */ + if (r_glsl_permutation->ubiloc_Skeletal_Transform12_UniformBlock >= 0 && rsurface.batchskeletaltransform3x4buffer) qglBindBufferRange(GL_UNIFORM_BUFFER, r_glsl_permutation->ubibind_Skeletal_Transform12_UniformBlock, rsurface.batchskeletaltransform3x4buffer->bufferobject, rsurface.batchskeletaltransform3x4offset, rsurface.batchskeletaltransform3x4size); +#endif if (r_glsl_permutation->loc_ModelToReflectCube >= 0) {Matrix4x4_ToArrayFloatGL(&rsurface.matrix, m16f);qglUniformMatrix4fv(r_glsl_permutation->loc_ModelToReflectCube, 1, false, m16f);} if (mode == SHADERMODE_LIGHTSOURCE) { @@ -2851,8 +2898,6 @@ void R_SetupShader_Surface(const vec3_t lightcolorbase, qboolean modellighting, } } if (r_glsl_permutation->tex_Texture_BounceGrid >= 0) R_Mesh_TexBind(r_glsl_permutation->tex_Texture_BounceGrid, r_shadow_bouncegridtexture); - if (r_glsl_permutation->loc_Skeletal_Transform12 >= 0 && rsurface.batchskeletalnumtransforms > 0) - qglUniform4fv(r_glsl_permutation->loc_Skeletal_Transform12, rsurface.batchskeletalnumtransforms*3, rsurface.batchskeletaltransform3x4); CHECKGLERROR break; case RENDERPATH_GL11: @@ -3975,6 +4020,7 @@ static void gl_main_start(void) r_texture_fogheighttexture = NULL; r_texture_gammaramps = NULL; r_texture_numcubemaps = 0; + r_uniformbufferalignment = 32; r_loaddds = r_texture_dds_load.integer != 0; r_savedds = vid.support.arb_texture_compression && vid.support.ext_texture_compression_s3tc && r_texture_dds_save.integer; @@ -3993,7 +4039,11 @@ static void gl_main_start(void) r_loadnormalmap = true; r_loadgloss = true; r_loadfog = false; - break; +#ifdef GL_UNIFORM_BUFFER_OFFSET_ALIGNMENT + if (vid.support.arb_uniform_buffer_object) + qglGetIntegerv(GL_UNIFORM_BUFFER_OFFSET_ALIGNMENT, &r_uniformbufferalignment); +#endif + break; case RENDERPATH_GL13: case RENDERPATH_GLES1: Cvar_SetValueQuick(&r_textureunits, vid.texunits); @@ -4015,6 +4065,7 @@ static void gl_main_start(void) R_AnimCache_Free(); R_FrameData_Reset(); + R_BufferData_Reset(); r_numqueries = 0; r_maxqueries = 0; @@ -4060,12 +4111,34 @@ static void gl_main_start(void) r_texture_numcubemaps = 0; r_refdef.fogmasktable_density = 0; + +#ifdef __ANDROID__ + // For Steelstorm Android + // FIXME CACHE the program and reload + // FIXME see possible combinations for SS:BR android + Con_DPrintf("Compiling most used shaders for SS:BR android... START\n"); + R_SetupShader_SetPermutationGLSL(0, 12); + R_SetupShader_SetPermutationGLSL(0, 13); + R_SetupShader_SetPermutationGLSL(0, 8388621); + R_SetupShader_SetPermutationGLSL(3, 0); + R_SetupShader_SetPermutationGLSL(3, 2048); + R_SetupShader_SetPermutationGLSL(5, 0); + R_SetupShader_SetPermutationGLSL(5, 2); + R_SetupShader_SetPermutationGLSL(5, 2048); + R_SetupShader_SetPermutationGLSL(5, 8388608); + R_SetupShader_SetPermutationGLSL(11, 1); + R_SetupShader_SetPermutationGLSL(11, 2049); + R_SetupShader_SetPermutationGLSL(11, 8193); + R_SetupShader_SetPermutationGLSL(11, 10241); + Con_DPrintf("Compiling most used shaders for SS:BR android... END\n"); +#endif } static void gl_main_shutdown(void) { R_AnimCache_Free(); R_FrameData_Reset(); + R_BufferData_Reset(); R_Main_FreeViewCache(); @@ -4076,7 +4149,7 @@ static void gl_main_shutdown(void) case RENDERPATH_GL20: case RENDERPATH_GLES1: case RENDERPATH_GLES2: -#ifdef GL_SAMPLES_PASSED_ARB +#if defined(GL_SAMPLES_PASSED_ARB) && !defined(USE_GLES2) if (r_maxqueries) qglDeleteQueriesARB(r_maxqueries, r_queries); #endif @@ -4159,10 +4232,12 @@ static void gl_main_newmap(void) R_Main_FreeViewCache(); R_FrameData_Reset(); + R_BufferData_Reset(); } void GL_Main_Init(void) { + int i; r_main_mempool = Mem_AllocPool("Renderer", 0, NULL); Cmd_AddCommand("r_glsl_restart", R_GLSL_Restart_f, "unloads GLSL shaders, they will then be reloaded as needed"); @@ -4335,8 +4410,18 @@ void GL_Main_Init(void) Cvar_RegisterVariable(&r_glsl_saturation_redcompensate); Cvar_RegisterVariable(&r_glsl_vertextextureblend_usebothalphas); Cvar_RegisterVariable(&r_framedatasize); + for (i = 0;i < R_BUFFERDATA_COUNT;i++) + Cvar_RegisterVariable(&r_buffermegs[i]); + Cvar_RegisterVariable(&r_batch_dynamicbuffer); if (gamemode == GAME_NEHAHRA || gamemode == GAME_TENEBRAE) Cvar_SetValue("r_fullbrights", 0); +#ifdef DP_MOBILETOUCH + // GLES devices have terrible depth precision in general, so... + Cvar_SetValueQuick(&r_nearclip, 4); + Cvar_SetValueQuick(&r_farclip_base, 4096); + Cvar_SetValueQuick(&r_farclip_world, 0); + Cvar_SetValueQuick(&r_useinfinitefarclip, 0); +#endif R_RegisterModule("GL_Main", gl_main_start, gl_main_shutdown, gl_main_newmap, NULL, NULL); } @@ -4385,7 +4470,11 @@ void GL_Init (void) VID_CheckExtensions(); // LordHavoc: report supported extensions +#ifdef CONFIG_MENU Con_DPrintf("\nQuakeC extensions for server and client: %s\nQuakeC extensions for menu: %s\n", vm_sv_extensions, vm_m_extensions ); +#else + Con_DPrintf("\nQuakeC extensions for server and client: %s\n", vm_sv_extensions ); +#endif // clear to black (loading plaque will be seen over this) GL_Clear(GL_COLOR_BUFFER_BIT, NULL, 1.0f, 128); @@ -4520,12 +4609,12 @@ void R_FrameData_Reset(void) } } -static void R_FrameData_Resize(void) +static void R_FrameData_Resize(qboolean mustgrow) { size_t wantedsize; wantedsize = (size_t)(r_framedatasize.value * 1024*1024); wantedsize = bound(65536, wantedsize, 1000*1024*1024); - if (!r_framedata_mem || r_framedata_mem->wantedsize != wantedsize) + if (!r_framedata_mem || r_framedata_mem->wantedsize != wantedsize || mustgrow) { r_framedata_mem_t *newmem = (r_framedata_mem_t *)Mem_Alloc(r_main_mempool, wantedsize); newmem->wantedsize = wantedsize; @@ -4540,7 +4629,7 @@ static void R_FrameData_Resize(void) void R_FrameData_NewFrame(void) { - R_FrameData_Resize(); + R_FrameData_Resize(false); if (!r_framedata_mem) return; // if we ran out of space on the last frame, free the old memory now @@ -4559,6 +4648,7 @@ void R_FrameData_NewFrame(void) void *R_FrameData_Alloc(size_t size) { void *data; + float newvalue; // align to 16 byte boundary - the data pointer is already aligned, so we // only need to ensure the size of every allocation is also aligned @@ -4567,16 +4657,18 @@ void *R_FrameData_Alloc(size_t size) while (!r_framedata_mem || r_framedata_mem->current + size > r_framedata_mem->size) { // emergency - we ran out of space, allocate more memory - Cvar_SetValueQuick(&r_framedatasize, bound(0.25f, r_framedatasize.value * 2.0f, 128.0f)); - R_FrameData_Resize(); + newvalue = bound(0.25f, r_framedatasize.value * 2.0f, 256.0f); + // this might not be a growing it, but we'll allocate another buffer every time + Cvar_SetValueQuick(&r_framedatasize, newvalue); + R_FrameData_Resize(true); } data = r_framedata_mem->data + r_framedata_mem->current; r_framedata_mem->current += size; // count the usage for stats - r_refdef.stats.framedatacurrent = max(r_refdef.stats.framedatacurrent, (int)r_framedata_mem->current); - r_refdef.stats.framedatasize = max(r_refdef.stats.framedatasize, (int)r_framedata_mem->size); + r_refdef.stats[r_stat_framedatacurrent] = max(r_refdef.stats[r_stat_framedatacurrent], (int)r_framedata_mem->current); + r_refdef.stats[r_stat_framedatasize] = max(r_refdef.stats[r_stat_framedatasize], (int)r_framedata_mem->size); return (void *)data; } @@ -4605,6 +4697,160 @@ void R_FrameData_ReturnToMark(void) //================================================================================== +// avoid reusing the same buffer objects on consecutive frames +#define R_BUFFERDATA_CYCLE 3 + +typedef struct r_bufferdata_buffer_s +{ + struct r_bufferdata_buffer_s *purge; // older buffer to free on next frame + size_t size; // how much usable space + size_t current; // how much space in use + r_meshbuffer_t *buffer; // the buffer itself +} +r_bufferdata_buffer_t; + +static int r_bufferdata_cycle = 0; // incremented and wrapped each frame +static r_bufferdata_buffer_t *r_bufferdata_buffer[R_BUFFERDATA_CYCLE][R_BUFFERDATA_COUNT]; + +/// frees all dynamic buffers +void R_BufferData_Reset(void) +{ + int cycle, type; + r_bufferdata_buffer_t **p, *mem; + for (cycle = 0;cycle < R_BUFFERDATA_CYCLE;cycle++) + { + for (type = 0;type < R_BUFFERDATA_COUNT;type++) + { + // free all buffers + p = &r_bufferdata_buffer[cycle][type]; + while (*p) + { + mem = *p; + *p = (*p)->purge; + if (mem->buffer) + R_Mesh_DestroyMeshBuffer(mem->buffer); + Mem_Free(mem); + } + } + } +} + +// resize buffer as needed (this actually makes a new one, the old one will be recycled next frame) +static void R_BufferData_Resize(r_bufferdata_type_t type, qboolean mustgrow, size_t minsize) +{ + r_bufferdata_buffer_t *mem = r_bufferdata_buffer[r_bufferdata_cycle][type]; + size_t size; + float newvalue = r_buffermegs[type].value; + + // increase the cvar if we have to (but only if we already have a mem) + if (mustgrow && mem) + newvalue *= 2.0f; + newvalue = bound(0.25f, newvalue, 256.0f); + while (newvalue * 1024*1024 < minsize) + newvalue *= 2.0f; + + // clamp the cvar to valid range + newvalue = bound(0.25f, newvalue, 256.0f); + if (r_buffermegs[type].value != newvalue) + Cvar_SetValueQuick(&r_buffermegs[type], newvalue); + + // calculate size in bytes + size = (size_t)(newvalue * 1024*1024); + size = bound(131072, size, 256*1024*1024); + + // allocate a new buffer if the size is different (purge old one later) + // or if we were told we must grow the buffer + if (!mem || mem->size != size || mustgrow) + { + mem = (r_bufferdata_buffer_t *)Mem_Alloc(r_main_mempool, sizeof(*mem)); + mem->size = size; + mem->current = 0; + if (type == R_BUFFERDATA_VERTEX) + mem->buffer = R_Mesh_CreateMeshBuffer(NULL, mem->size, "dynamicbuffervertex", false, false, true, false); + else if (type == R_BUFFERDATA_INDEX16) + mem->buffer = R_Mesh_CreateMeshBuffer(NULL, mem->size, "dynamicbufferindex16", true, false, true, true); + else if (type == R_BUFFERDATA_INDEX32) + mem->buffer = R_Mesh_CreateMeshBuffer(NULL, mem->size, "dynamicbufferindex32", true, false, true, false); + else if (type == R_BUFFERDATA_UNIFORM) + mem->buffer = R_Mesh_CreateMeshBuffer(NULL, mem->size, "dynamicbufferuniform", false, true, true, false); + mem->purge = r_bufferdata_buffer[r_bufferdata_cycle][type]; + r_bufferdata_buffer[r_bufferdata_cycle][type] = mem; + } +} + +void R_BufferData_NewFrame(void) +{ + int type; + r_bufferdata_buffer_t **p, *mem; + // cycle to the next frame's buffers + r_bufferdata_cycle = (r_bufferdata_cycle + 1) % R_BUFFERDATA_CYCLE; + // if we ran out of space on the last time we used these buffers, free the old memory now + for (type = 0;type < R_BUFFERDATA_COUNT;type++) + { + if (r_bufferdata_buffer[r_bufferdata_cycle][type]) + { + R_BufferData_Resize((r_bufferdata_type_t)type, false, 131072); + // free all but the head buffer, this is how we recycle obsolete + // buffers after they are no longer in use + p = &r_bufferdata_buffer[r_bufferdata_cycle][type]->purge; + while (*p) + { + mem = *p; + *p = (*p)->purge; + if (mem->buffer) + R_Mesh_DestroyMeshBuffer(mem->buffer); + Mem_Free(mem); + } + // reset the current offset + r_bufferdata_buffer[r_bufferdata_cycle][type]->current = 0; + } + } +} + +r_meshbuffer_t *R_BufferData_Store(size_t datasize, const void *data, r_bufferdata_type_t type, int *returnbufferoffset) +{ + r_bufferdata_buffer_t *mem; + int offset = 0; + int padsize; + + *returnbufferoffset = 0; + + // align size to a byte boundary appropriate for the buffer type, this + // makes all allocations have aligned start offsets + if (type == R_BUFFERDATA_UNIFORM) + padsize = (datasize + r_uniformbufferalignment - 1) & ~(r_uniformbufferalignment - 1); + else + padsize = (datasize + 15) & ~15; + + // if we ran out of space in this buffer we must allocate a new one + if (!r_bufferdata_buffer[r_bufferdata_cycle][type] || r_bufferdata_buffer[r_bufferdata_cycle][type]->current + padsize > r_bufferdata_buffer[r_bufferdata_cycle][type]->size) + R_BufferData_Resize(type, true, padsize); + + // if the resize did not give us enough memory, fail + if (!r_bufferdata_buffer[r_bufferdata_cycle][type] || r_bufferdata_buffer[r_bufferdata_cycle][type]->current + padsize > r_bufferdata_buffer[r_bufferdata_cycle][type]->size) + Sys_Error("R_BufferData_Store: failed to create a new buffer of sufficient size\n"); + + mem = r_bufferdata_buffer[r_bufferdata_cycle][type]; + offset = mem->current; + mem->current += padsize; + + // upload the data to the buffer at the chosen offset + if (offset == 0) + R_Mesh_UpdateMeshBuffer(mem->buffer, NULL, mem->size, false, 0); + R_Mesh_UpdateMeshBuffer(mem->buffer, data, datasize, true, offset); + + // count the usage for stats + r_refdef.stats[r_stat_bufferdatacurrent_vertex + type] = max(r_refdef.stats[r_stat_bufferdatacurrent_vertex + type], (int)mem->current); + r_refdef.stats[r_stat_bufferdatasize_vertex + type] = max(r_refdef.stats[r_stat_bufferdatasize_vertex + type], (int)mem->size); + + // return the buffer offset + *returnbufferoffset = offset; + + return mem->buffer; +} + +//================================================================================== + // LordHavoc: animcache originally written by Echon, rewritten since then /** @@ -4625,13 +4871,24 @@ void R_AnimCache_ClearCache(void) { ent = r_refdef.scene.entities[i]; ent->animcache_vertex3f = NULL; + ent->animcache_vertex3f_vertexbuffer = NULL; + ent->animcache_vertex3f_bufferoffset = 0; ent->animcache_normal3f = NULL; + ent->animcache_normal3f_vertexbuffer = NULL; + ent->animcache_normal3f_bufferoffset = 0; ent->animcache_svector3f = NULL; + ent->animcache_svector3f_vertexbuffer = NULL; + ent->animcache_svector3f_bufferoffset = 0; ent->animcache_tvector3f = NULL; + ent->animcache_tvector3f_vertexbuffer = NULL; + ent->animcache_tvector3f_bufferoffset = 0; ent->animcache_vertexmesh = NULL; - ent->animcache_vertex3fbuffer = NULL; - ent->animcache_vertexmeshbuffer = NULL; + ent->animcache_vertexmesh_vertexbuffer = NULL; + ent->animcache_vertexmesh_bufferoffset = 0; ent->animcache_skeletaltransform3x4 = NULL; + ent->animcache_skeletaltransform3x4buffer = NULL; + ent->animcache_skeletaltransform3x4offset = 0; + ent->animcache_skeletaltransform3x4size = 0; } } @@ -4645,10 +4902,13 @@ static void R_AnimCache_UpdateEntityMeshBuffers(entity_render_t *ent, int numver if (!ent->animcache_vertexmesh && ent->animcache_normal3f) ent->animcache_vertexmesh = (r_vertexmesh_t *)R_FrameData_Alloc(sizeof(r_vertexmesh_t)*numvertices); - // TODO: upload vertex3f buffer? + // TODO: upload vertexbuffer? if (ent->animcache_vertexmesh) { - memcpy(ent->animcache_vertexmesh, ent->model->surfmesh.vertexmesh, sizeof(r_vertexmesh_t)*numvertices); + r_refdef.stats[r_stat_animcache_vertexmesh_count] += 1; + r_refdef.stats[r_stat_animcache_vertexmesh_vertices] += numvertices; + r_refdef.stats[r_stat_animcache_vertexmesh_maxvertices] = max(r_refdef.stats[r_stat_animcache_vertexmesh_maxvertices], numvertices); + memcpy(ent->animcache_vertexmesh, ent->model->surfmesh.data_vertexmesh, sizeof(r_vertexmesh_t)*numvertices); for (i = 0;i < numvertices;i++) memcpy(ent->animcache_vertexmesh[i].vertex3f, ent->animcache_vertex3f + 3*i, sizeof(float[3])); if (ent->animcache_svector3f) @@ -4660,7 +4920,6 @@ static void R_AnimCache_UpdateEntityMeshBuffers(entity_render_t *ent, int numver if (ent->animcache_normal3f) for (i = 0;i < numvertices;i++) memcpy(ent->animcache_vertexmesh[i].normal3f, ent->animcache_normal3f + 3*i, sizeof(float[3])); - // TODO: upload vertexmeshbuffer? } } @@ -4669,154 +4928,64 @@ qboolean R_AnimCache_GetEntity(entity_render_t *ent, qboolean wantnormals, qbool dp_model_t *model = ent->model; int numvertices; - // cache skeletal animation data first (primarily for gpu-skinning) - if (!ent->animcache_skeletaltransform3x4 && model->num_bones > 0 && model->surfmesh.data_skeletalindex4ub) + // see if this ent is worth caching + if (!model || !model->Draw || !model->AnimateVertices) + return false; + // nothing to cache if it contains no animations and has no skeleton + if (!model->surfmesh.isanimated && !(model->num_bones && ent->skeleton && ent->skeleton->relativetransforms)) + return false; + // see if it is already cached for gpuskeletal + if (ent->animcache_skeletaltransform3x4) + return false; + // see if it is already cached as a mesh + if (ent->animcache_vertex3f) { - int i; - int blends; - const skeleton_t *skeleton = ent->skeleton; - const frameblend_t *frameblend = ent->frameblend; - float *boneposerelative; - float m[12]; - static float bonepose[256][12]; - ent->animcache_skeletaltransform3x4 = R_FrameData_Alloc(sizeof(float[3][4]) * model->num_bones); - boneposerelative = ent->animcache_skeletaltransform3x4; - if (skeleton && !skeleton->relativetransforms) - skeleton = NULL; - // resolve hierarchy and make relative transforms (deforms) which the shader wants - if (skeleton) - { - for (i = 0;i < model->num_bones;i++) - { - Matrix4x4_ToArray12FloatD3D(&skeleton->relativetransforms[i], m); - if (model->data_bones[i].parent >= 0) - R_ConcatTransforms(bonepose[model->data_bones[i].parent], m, bonepose[i]); - else - memcpy(bonepose[i], m, sizeof(m)); - - // create a relative deformation matrix to describe displacement - // from the base mesh, which is used by the actual weighting - R_ConcatTransforms(bonepose[i], model->data_baseboneposeinverse + i * 12, boneposerelative + i * 12); - } - } - else - { - for (i = 0;i < model->num_bones;i++) - { - const short * RESTRICT pose7s = model->data_poses7s + 7 * (frameblend[0].subframe * model->num_bones + i); - float lerp = frameblend[0].lerp, - tx = pose7s[0], ty = pose7s[1], tz = pose7s[2], - rx = pose7s[3] * lerp, - ry = pose7s[4] * lerp, - rz = pose7s[5] * lerp, - rw = pose7s[6] * lerp, - dx = tx*rw + ty*rz - tz*ry, - dy = -tx*rz + ty*rw + tz*rx, - dz = tx*ry - ty*rx + tz*rw, - dw = -tx*rx - ty*ry - tz*rz, - scale, sx, sy, sz, sw; - for (blends = 1;blends < MAX_FRAMEBLENDS && frameblend[blends].lerp > 0;blends++) - { - const short * RESTRICT pose7s = model->data_poses7s + 7 * (frameblend[blends].subframe * model->num_bones + i); - float lerp = frameblend[blends].lerp, - tx = pose7s[0], ty = pose7s[1], tz = pose7s[2], - qx = pose7s[3], qy = pose7s[4], qz = pose7s[5], qw = pose7s[6]; - if(rx*qx + ry*qy + rz*qz + rw*qw < 0) lerp = -lerp; - qx *= lerp; - qy *= lerp; - qz *= lerp; - qw *= lerp; - rx += qx; - ry += qy; - rz += qz; - rw += qw; - dx += tx*qw + ty*qz - tz*qy; - dy += -tx*qz + ty*qw + tz*qx; - dz += tx*qy - ty*qx + tz*qw; - dw += -tx*qx - ty*qy - tz*qz; - } - scale = 1.0f / (rx*rx + ry*ry + rz*rz + rw*rw); - sx = rx * scale; - sy = ry * scale; - sz = rz * scale; - sw = rw * scale; - m[0] = sw*rw + sx*rx - sy*ry - sz*rz; - m[1] = 2*(sx*ry - sw*rz); - m[2] = 2*(sx*rz + sw*ry); - m[3] = model->num_posescale*(dx*sw - dy*sz + dz*sy - dw*sx); - m[4] = 2*(sx*ry + sw*rz); - m[5] = sw*rw + sy*ry - sx*rx - sz*rz; - m[6] = 2*(sy*rz - sw*rx); - m[7] = model->num_posescale*(dx*sz + dy*sw - dz*sx - dw*sy); - m[8] = 2*(sx*rz - sw*ry); - m[9] = 2*(sy*rz + sw*rx); - m[10] = sw*rw + sz*rz - sx*rx - sy*ry; - m[11] = model->num_posescale*(dy*sx + dz*sw - dx*sy - dw*sz); - if (i == r_skeletal_debugbone.integer) - m[r_skeletal_debugbonecomponent.integer % 12] += r_skeletal_debugbonevalue.value; - m[3] *= r_skeletal_debugtranslatex.value; - m[7] *= r_skeletal_debugtranslatey.value; - m[11] *= r_skeletal_debugtranslatez.value; - if (model->data_bones[i].parent >= 0) - R_ConcatTransforms(bonepose[model->data_bones[i].parent], m, bonepose[i]); - else - memcpy(bonepose[i], m, sizeof(m)); - // create a relative deformation matrix to describe displacement - // from the base mesh, which is used by the actual weighting - R_ConcatTransforms(bonepose[i], model->data_baseboneposeinverse + i * 12, boneposerelative + i * 12); - } - } + // check if we need to add normals or tangents + if (ent->animcache_normal3f) + wantnormals = false; + if (ent->animcache_svector3f) + wanttangents = false; + if (!wantnormals && !wanttangents) + return false; } - // see if it's already cached this frame - if (ent->animcache_vertex3f) + // check which kind of cache we need to generate + if (r_gpuskeletal && model->num_bones > 0 && model->surfmesh.data_skeletalindex4ub) + { + // cache the skeleton so the vertex shader can use it + r_refdef.stats[r_stat_animcache_skeletal_count] += 1; + r_refdef.stats[r_stat_animcache_skeletal_bones] += model->num_bones; + r_refdef.stats[r_stat_animcache_skeletal_maxbones] = max(r_refdef.stats[r_stat_animcache_skeletal_maxbones], model->num_bones); + ent->animcache_skeletaltransform3x4 = (float *)R_FrameData_Alloc(sizeof(float[3][4]) * model->num_bones); + Mod_Skeletal_BuildTransforms(model, ent->frameblend, ent->skeleton, NULL, ent->animcache_skeletaltransform3x4); + // note: this can fail if the buffer is at the grow limit + ent->animcache_skeletaltransform3x4size = sizeof(float[3][4]) * model->num_bones; + ent->animcache_skeletaltransform3x4buffer = R_BufferData_Store(ent->animcache_skeletaltransform3x4size, ent->animcache_skeletaltransform3x4, R_BUFFERDATA_UNIFORM, &ent->animcache_skeletaltransform3x4offset); + } + else if (ent->animcache_vertex3f) { - // add normals/tangents if needed (this only happens with multiple views, reflections, cameras, etc) + // mesh was already cached but we may need to add normals/tangents + // (this only happens with multiple views, reflections, cameras, etc) if (wantnormals || wanttangents) { - if (ent->animcache_normal3f) - wantnormals = false; - if (ent->animcache_svector3f) - wanttangents = false; - if (wantnormals || wanttangents) + numvertices = model->surfmesh.num_vertices; + if (wantnormals) + ent->animcache_normal3f = (float *)R_FrameData_Alloc(sizeof(float[3])*numvertices); + if (wanttangents) { - numvertices = model->surfmesh.num_vertices; - if (wantnormals) - ent->animcache_normal3f = (float *)R_FrameData_Alloc(sizeof(float[3])*numvertices); - if (wanttangents) - { - ent->animcache_svector3f = (float *)R_FrameData_Alloc(sizeof(float[3])*numvertices); - ent->animcache_tvector3f = (float *)R_FrameData_Alloc(sizeof(float[3])*numvertices); - } - model->AnimateVertices(model, ent->frameblend, ent->skeleton, NULL, wantnormals ? ent->animcache_normal3f : NULL, wanttangents ? ent->animcache_svector3f : NULL, wanttangents ? ent->animcache_tvector3f : NULL); - R_AnimCache_UpdateEntityMeshBuffers(ent, model->surfmesh.num_vertices); + ent->animcache_svector3f = (float *)R_FrameData_Alloc(sizeof(float[3])*numvertices); + ent->animcache_tvector3f = (float *)R_FrameData_Alloc(sizeof(float[3])*numvertices); } + model->AnimateVertices(model, ent->frameblend, ent->skeleton, NULL, wantnormals ? ent->animcache_normal3f : NULL, wanttangents ? ent->animcache_svector3f : NULL, wanttangents ? ent->animcache_tvector3f : NULL); + R_AnimCache_UpdateEntityMeshBuffers(ent, model->surfmesh.num_vertices); + r_refdef.stats[r_stat_animcache_shade_count] += 1; + r_refdef.stats[r_stat_animcache_shade_vertices] += numvertices; + r_refdef.stats[r_stat_animcache_shade_maxvertices] = max(r_refdef.stats[r_stat_animcache_shade_maxvertices], numvertices); } } else { - // see if this ent is worth caching - if (!model || !model->Draw || !model->surfmesh.isanimated || !model->AnimateVertices) - return false; - // skip entity if the shader backend has a cheaper way - if (model->surfmesh.data_skeletalindex4ub && r_glsl_skeletal.integer) - { - switch (vid.renderpath) - { - case RENDERPATH_GL20: - return false; - case RENDERPATH_GL11: - case RENDERPATH_GL13: - case RENDERPATH_GLES1: - case RENDERPATH_GLES2: - case RENDERPATH_D3D9: - case RENDERPATH_D3D10: - case RENDERPATH_D3D11: - case RENDERPATH_SOFT: - break; - } - } - // get some memory for this entity and generate mesh data + // generate mesh cache numvertices = model->surfmesh.num_vertices; ent->animcache_vertex3f = (float *)R_FrameData_Alloc(sizeof(float[3])*numvertices); if (wantnormals) @@ -4828,6 +4997,15 @@ qboolean R_AnimCache_GetEntity(entity_render_t *ent, qboolean wantnormals, qbool } model->AnimateVertices(model, ent->frameblend, ent->skeleton, ent->animcache_vertex3f, ent->animcache_normal3f, ent->animcache_svector3f, ent->animcache_tvector3f); R_AnimCache_UpdateEntityMeshBuffers(ent, model->surfmesh.num_vertices); + if (wantnormals || wanttangents) + { + r_refdef.stats[r_stat_animcache_shade_count] += 1; + r_refdef.stats[r_stat_animcache_shade_vertices] += numvertices; + r_refdef.stats[r_stat_animcache_shade_maxvertices] = max(r_refdef.stats[r_stat_animcache_shade_maxvertices], numvertices); + } + r_refdef.stats[r_stat_animcache_shape_count] += 1; + r_refdef.stats[r_stat_animcache_shape_vertices] += numvertices; + r_refdef.stats[r_stat_animcache_shape_maxvertices] = max(r_refdef.stats[r_stat_animcache_shape_maxvertices], numvertices); } return true; } @@ -5114,7 +5292,7 @@ static void R_DrawModels(void) if (!r_refdef.viewcache.entityvisible[i]) continue; ent = r_refdef.scene.entities[i]; - r_refdef.stats.entities++; + r_refdef.stats[r_stat_entities]++; /* if (ent->model && !strncmp(ent->model->name, "models/proto_", 13)) { @@ -5524,7 +5702,9 @@ void R_EntityMatrix(const matrix4x4_t *matrix) case RENDERPATH_GL11: case RENDERPATH_GL13: case RENDERPATH_GLES1: +#ifndef USE_GLES2 qglLoadMatrixf(gl_modelview16f);CHECKGLERROR +#endif break; case RENDERPATH_SOFT: DPSOFTRAST_UniformMatrix4fv(DPSOFTRAST_UNIFORM_ModelViewProjectionMatrixM1, 1, false, gl_modelviewprojection16f); @@ -5686,10 +5866,10 @@ static void R_Water_StartFrame(void) } else { - for (texturewidth = 1;texturewidth < waterwidth ;texturewidth *= 2); - for (textureheight = 1;textureheight < waterheight;textureheight *= 2); - for (camerawidth = 1;camerawidth <= waterwidth; camerawidth *= 2); camerawidth /= 2; - for (cameraheight = 1;cameraheight <= waterheight;cameraheight *= 2); cameraheight /= 2; + for (texturewidth = 1;texturewidth < waterwidth ;texturewidth *= 2); + for (textureheight = 1;textureheight < waterheight;textureheight *= 2); + for (camerawidth = 1;camerawidth * 2 <= waterwidth ;camerawidth *= 2); + for (cameraheight = 1;cameraheight * 2 <= waterheight;cameraheight *= 2); } // allocate textures as needed @@ -6375,14 +6555,14 @@ static void R_Bloom_MakeTexture(void) rtexture_t *intex; float colorscale = r_bloom_colorscale.value; - r_refdef.stats.bloom++; + r_refdef.stats[r_stat_bloom]++; #if 0 // this copy is unnecessary since it happens in R_BlendView already if (!r_fb.fbo) { R_Mesh_CopyToTexture(r_fb.colortexture, 0, 0, r_refdef.view.viewport.x, r_refdef.view.viewport.y, r_refdef.view.viewport.width, r_refdef.view.viewport.height); - r_refdef.stats.bloom_copypixels += r_refdef.view.viewport.width * r_refdef.view.viewport.height; + r_refdef.stats[r_stat_bloom_copypixels] += r_refdef.view.viewport.width * r_refdef.view.viewport.height; } #endif @@ -6414,14 +6594,14 @@ static void R_Bloom_MakeTexture(void) // TODO: do boxfilter scale-down in shader? R_SetupShader_Generic(r_fb.colortexture, NULL, GL_MODULATE, 1, false, true, true); R_Mesh_Draw(0, 4, 0, 2, polygonelement3i, NULL, 0, polygonelement3s, NULL, 0); - r_refdef.stats.bloom_drawpixels += r_fb.bloomwidth * r_fb.bloomheight; + r_refdef.stats[r_stat_bloom_drawpixels] += r_fb.bloomwidth * r_fb.bloomheight; // we now have a properly scaled bloom image if (!r_fb.bloomfbo[r_fb.bloomindex]) { // copy it into the bloom texture R_Mesh_CopyToTexture(r_fb.bloomtexture[r_fb.bloomindex], 0, 0, r_fb.bloomviewport.x, r_fb.bloomviewport.y, r_fb.bloomviewport.width, r_fb.bloomviewport.height); - r_refdef.stats.bloom_copypixels += r_fb.bloomviewport.width * r_fb.bloomviewport.height; + r_refdef.stats[r_stat_bloom_copypixels] += r_fb.bloomviewport.width * r_fb.bloomviewport.height; } // multiply bloom image by itself as many times as desired @@ -6447,13 +6627,13 @@ static void R_Bloom_MakeTexture(void) R_Mesh_PrepareVertices_Generic_Arrays(4, r_screenvertex3f, NULL, r_fb.bloomtexcoord2f); R_SetupShader_Generic(intex, NULL, GL_MODULATE, 1, false, true, false); R_Mesh_Draw(0, 4, 0, 2, polygonelement3i, NULL, 0, polygonelement3s, NULL, 0); - r_refdef.stats.bloom_drawpixels += r_fb.bloomwidth * r_fb.bloomheight; + r_refdef.stats[r_stat_bloom_drawpixels] += r_fb.bloomwidth * r_fb.bloomheight; if (!r_fb.bloomfbo[r_fb.bloomindex]) { // copy the darkened image to a texture R_Mesh_CopyToTexture(r_fb.bloomtexture[r_fb.bloomindex], 0, 0, r_fb.bloomviewport.x, r_fb.bloomviewport.y, r_fb.bloomviewport.width, r_fb.bloomviewport.height); - r_refdef.stats.bloom_copypixels += r_fb.bloomviewport.width * r_fb.bloomviewport.height; + r_refdef.stats[r_stat_bloom_copypixels] += r_fb.bloomviewport.width * r_fb.bloomviewport.height; } } @@ -6499,7 +6679,7 @@ static void R_Bloom_MakeTexture(void) GL_Color(r, r, r, 1); R_Mesh_PrepareVertices_Generic_Arrays(4, r_screenvertex3f, NULL, r_fb.offsettexcoord2f); R_Mesh_Draw(0, 4, 0, 2, polygonelement3i, NULL, 0, polygonelement3s, NULL, 0); - r_refdef.stats.bloom_drawpixels += r_fb.bloomwidth * r_fb.bloomheight; + r_refdef.stats[r_stat_bloom_drawpixels] += r_fb.bloomwidth * r_fb.bloomheight; GL_BlendFunc(GL_ONE, GL_ONE); } @@ -6507,7 +6687,7 @@ static void R_Bloom_MakeTexture(void) { // copy the vertically or horizontally blurred bloom view to a texture R_Mesh_CopyToTexture(r_fb.bloomtexture[r_fb.bloomindex], 0, 0, r_fb.bloomviewport.x, r_fb.bloomviewport.y, r_fb.bloomviewport.width, r_fb.bloomviewport.height); - r_refdef.stats.bloom_copypixels += r_fb.bloomviewport.width * r_fb.bloomviewport.height; + r_refdef.stats[r_stat_bloom_copypixels] += r_fb.bloomviewport.width * r_fb.bloomviewport.height; } } } @@ -6539,7 +6719,7 @@ static void R_BlendView(int fbo, rtexture_t *depthtexture, rtexture_t *colortext if (!r_fb.fbo) { R_Mesh_CopyToTexture(r_fb.colortexture, 0, 0, r_refdef.view.viewport.x, r_refdef.view.viewport.y, r_refdef.view.viewport.width, r_refdef.view.viewport.height); - r_refdef.stats.bloom_copypixels += r_refdef.view.viewport.width * r_refdef.view.viewport.height; + r_refdef.stats[r_stat_bloom_copypixels] += r_refdef.view.viewport.width * r_refdef.view.viewport.height; } if(!R_Stereo_Active() && (r_motionblur.value > 0 || r_damageblur.value > 0) && r_fb.ghosttexture) @@ -6605,7 +6785,7 @@ static void R_BlendView(int fbo, rtexture_t *depthtexture, rtexture_t *colortext } R_SetupShader_Generic(r_fb.ghosttexture, NULL, GL_MODULATE, 1, false, true, true); R_Mesh_Draw(0, 4, 0, 2, polygonelement3i, NULL, 0, polygonelement3s, NULL, 0); - r_refdef.stats.bloom_drawpixels += r_refdef.view.viewport.width * r_refdef.view.viewport.height; + r_refdef.stats[r_stat_bloom_drawpixels] += r_refdef.view.viewport.width * r_refdef.view.viewport.height; } // updates old view angles for next pass @@ -6613,7 +6793,7 @@ static void R_BlendView(int fbo, rtexture_t *depthtexture, rtexture_t *colortext // copy view into the ghost texture R_Mesh_CopyToTexture(r_fb.ghosttexture, 0, 0, r_refdef.view.viewport.x, r_refdef.view.viewport.y, r_refdef.view.viewport.width, r_refdef.view.viewport.height); - r_refdef.stats.bloom_copypixels += r_refdef.view.viewport.width * r_refdef.view.viewport.height; + r_refdef.stats[r_stat_bloom_copypixels] += r_refdef.view.viewport.width * r_refdef.view.viewport.height; r_fb.ghosttexture_valid = true; } } @@ -6721,7 +6901,7 @@ static void R_BlendView(int fbo, rtexture_t *depthtexture, rtexture_t *colortext break; } R_Mesh_Draw(0, 4, 0, 2, polygonelement3i, NULL, 0, polygonelement3s, NULL, 0); - r_refdef.stats.bloom_drawpixels += r_refdef.view.width * r_refdef.view.height; + r_refdef.stats[r_stat_bloom_drawpixels] += r_refdef.view.width * r_refdef.view.height; break; case RENDERPATH_GL11: case RENDERPATH_GL13: @@ -6869,9 +7049,11 @@ void R_UpdateVariables(void) r_refdef.lightmapintensity = 0; } + r_gpuskeletal = false; switch(vid.renderpath) { case RENDERPATH_GL20: + r_gpuskeletal = vid.support.arb_uniform_buffer_object && r_glsl_skeletal.integer && !r_showsurfaces.integer; // FIXME add r_showsurfaces support to GLSL skeletal! case RENDERPATH_D3D9: case RENDERPATH_D3D10: case RENDERPATH_D3D11: @@ -7012,7 +7194,6 @@ void R_RenderView(void) R_SortEntities(); R_AnimCache_ClearCache(); - R_FrameData_NewFrame(); /* adjust for stereo display */ if(R_Stereo_Active()) @@ -7156,7 +7337,7 @@ void R_RenderScene(int fbo, rtexture_t *depthtexture, rtexture_t *colortexture) if (r_timereport_active) R_TimeReport("beginscene"); - r_refdef.stats.renders++; + r_refdef.stats[r_stat_renders]++; R_UpdateFog(); @@ -7820,10 +8001,22 @@ static float R_EvaluateQ3WaveFunc(q3wavefunc_t func, const float *parms) static void R_tcMod_ApplyToMatrix(matrix4x4_t *texmatrix, q3shaderinfo_layer_tcmod_t *tcmod, int currentmaterialflags) { int w, h, idx; - double f; - double offsetd[2]; + float shadertime; + float f; + float offsetd[2]; float tcmat[12]; matrix4x4_t matrix, temp; + // if shadertime exceeds about 9 hours (32768 seconds), just wrap it, + // it's better to have one huge fixup every 9 hours than gradual + // degradation over time which looks consistently bad after many hours. + // + // tcmod scroll in particular suffers from this degradation which can't be + // effectively worked around even with floor() tricks because we don't + // know if tcmod scroll is the last tcmod being applied, and for clampmap + // a workaround involving floor() would be incorrect anyway... + shadertime = rsurface.shadertime; + if (shadertime >= 32768.0f) + shadertime -= floor(rsurface.shadertime * (1.0f / 32768.0f)) * 32768.0f; switch(tcmod->tcmod) { case Q3TCMOD_COUNT: @@ -7839,16 +8032,17 @@ static void R_tcMod_ApplyToMatrix(matrix4x4_t *texmatrix, q3shaderinfo_layer_tcm Matrix4x4_CreateTranslate(&matrix, 0, 0, 0); break; case Q3TCMOD_ROTATE: - f = tcmod->parms[0] * rsurface.shadertime; Matrix4x4_CreateTranslate(&matrix, 0.5, 0.5, 0); - Matrix4x4_ConcatRotate(&matrix, (f / 360 - floor(f / 360)) * 360, 0, 0, 1); + Matrix4x4_ConcatRotate(&matrix, tcmod->parms[0] * rsurface.shadertime, 0, 0, 1); Matrix4x4_ConcatTranslate(&matrix, -0.5, -0.5, 0); break; case Q3TCMOD_SCALE: Matrix4x4_CreateScale3(&matrix, tcmod->parms[0], tcmod->parms[1], 1); break; case Q3TCMOD_SCROLL: - // extra care is needed because of precision breakdown with large values of time + // this particular tcmod is a "bug for bug" compatible one with regards to + // Quake3, the wrapping is unnecessary with our shadetime fix but quake3 + // specifically did the wrapping and so we must mimic that... offsetd[0] = tcmod->parms[0] * rsurface.shadertime; offsetd[1] = tcmod->parms[1] * rsurface.shadertime; Matrix4x4_CreateTranslate(&matrix, offsetd[0] - floor(offsetd[0]), offsetd[1] - floor(offsetd[1]), 0); @@ -8260,6 +8454,9 @@ void RSurf_ActiveWorldEntity(void) rsurface.basepolygonfactor = r_refdef.polygonfactor; rsurface.basepolygonoffset = r_refdef.polygonoffset; rsurface.entityskeletaltransform3x4 = NULL; + rsurface.entityskeletaltransform3x4buffer = NULL; + rsurface.entityskeletaltransform3x4offset = 0; + rsurface.entityskeletaltransform3x4size = 0;; rsurface.entityskeletalnumtransforms = 0; rsurface.modelvertex3f = model->surfmesh.data_vertex3f; rsurface.modelvertex3f_vertexbuffer = model->surfmesh.vbo_vertexbuffer; @@ -8298,9 +8495,9 @@ void RSurf_ActiveWorldEntity(void) rsurface.modelnumvertices = model->surfmesh.num_vertices; rsurface.modelnumtriangles = model->surfmesh.num_triangles; rsurface.modelsurfaces = model->data_surfaces; - rsurface.modelvertexmesh = model->surfmesh.vertexmesh; - rsurface.modelvertexmeshbuffer = model->surfmesh.vertexmeshbuffer; - rsurface.modelvertex3fbuffer = model->surfmesh.vertex3fbuffer; + rsurface.modelvertexmesh = model->surfmesh.data_vertexmesh; + rsurface.modelvertexmesh_vertexbuffer = model->surfmesh.vbo_vertexbuffer; + rsurface.modelvertexmesh_bufferoffset = model->surfmesh.vbooffset_vertex3f; rsurface.modelgeneratedvertex = false; rsurface.batchgeneratedvertex = false; rsurface.batchfirstvertex = 0; @@ -8335,8 +8532,8 @@ void RSurf_ActiveWorldEntity(void) rsurface.batchskeletalweight4ub_vertexbuffer = NULL; rsurface.batchskeletalweight4ub_bufferoffset = 0; rsurface.batchvertexmesh = NULL; - rsurface.batchvertexmeshbuffer = NULL; - rsurface.batchvertex3fbuffer = NULL; + rsurface.batchvertexmesh_vertexbuffer = NULL; + rsurface.batchvertexmesh_bufferoffset = 0; rsurface.batchelement3i = NULL; rsurface.batchelement3i_indexbuffer = NULL; rsurface.batchelement3i_bufferoffset = 0; @@ -8390,65 +8587,128 @@ void RSurf_ActiveModelEntity(const entity_render_t *ent, qboolean wantnormals, q rsurface.basepolygonoffset += r_polygonoffset_submodel_offset.value; } // if the animcache code decided it should use the shader path, skip the deform step - rsurface.entityskeletaltransform3x4 = ent->animcache_vertex3f ? NULL : ent->animcache_skeletaltransform3x4; + rsurface.entityskeletaltransform3x4 = ent->animcache_skeletaltransform3x4; + rsurface.entityskeletaltransform3x4buffer = ent->animcache_skeletaltransform3x4buffer; + rsurface.entityskeletaltransform3x4offset = ent->animcache_skeletaltransform3x4offset; + rsurface.entityskeletaltransform3x4size = ent->animcache_skeletaltransform3x4size; rsurface.entityskeletalnumtransforms = rsurface.entityskeletaltransform3x4 ? model->num_bones : 0; if (model->surfmesh.isanimated && model->AnimateVertices && !rsurface.entityskeletaltransform3x4) { if (ent->animcache_vertex3f) { + r_refdef.stats[r_stat_batch_entitycache_count]++; + r_refdef.stats[r_stat_batch_entitycache_surfaces] += model->num_surfaces; + r_refdef.stats[r_stat_batch_entitycache_vertices] += model->surfmesh.num_vertices; + r_refdef.stats[r_stat_batch_entitycache_triangles] += model->surfmesh.num_triangles; rsurface.modelvertex3f = ent->animcache_vertex3f; + rsurface.modelvertex3f_vertexbuffer = ent->animcache_vertex3f_vertexbuffer; + rsurface.modelvertex3f_bufferoffset = ent->animcache_vertex3f_bufferoffset; rsurface.modelsvector3f = wanttangents ? ent->animcache_svector3f : NULL; + rsurface.modelsvector3f_vertexbuffer = wanttangents ? ent->animcache_svector3f_vertexbuffer : NULL; + rsurface.modelsvector3f_bufferoffset = wanttangents ? ent->animcache_svector3f_bufferoffset : 0; rsurface.modeltvector3f = wanttangents ? ent->animcache_tvector3f : NULL; + rsurface.modeltvector3f_vertexbuffer = wanttangents ? ent->animcache_tvector3f_vertexbuffer : NULL; + rsurface.modeltvector3f_bufferoffset = wanttangents ? ent->animcache_tvector3f_bufferoffset : 0; rsurface.modelnormal3f = wantnormals ? ent->animcache_normal3f : NULL; + rsurface.modelnormal3f_vertexbuffer = wantnormals ? ent->animcache_normal3f_vertexbuffer : NULL; + rsurface.modelnormal3f_bufferoffset = wantnormals ? ent->animcache_normal3f_bufferoffset : 0; rsurface.modelvertexmesh = ent->animcache_vertexmesh; - rsurface.modelvertexmeshbuffer = ent->animcache_vertexmeshbuffer; - rsurface.modelvertex3fbuffer = ent->animcache_vertex3fbuffer; + rsurface.modelvertexmesh_vertexbuffer = ent->animcache_vertexmesh_vertexbuffer; + rsurface.modelvertexmesh_bufferoffset = ent->animcache_vertexmesh_bufferoffset; } else if (wanttangents) { + r_refdef.stats[r_stat_batch_entityanimate_count]++; + r_refdef.stats[r_stat_batch_entityanimate_surfaces] += model->num_surfaces; + r_refdef.stats[r_stat_batch_entityanimate_vertices] += model->surfmesh.num_vertices; + r_refdef.stats[r_stat_batch_entityanimate_triangles] += model->surfmesh.num_triangles; rsurface.modelvertex3f = (float *)R_FrameData_Alloc(model->surfmesh.num_vertices * sizeof(float[3])); rsurface.modelsvector3f = (float *)R_FrameData_Alloc(model->surfmesh.num_vertices * sizeof(float[3])); rsurface.modeltvector3f = (float *)R_FrameData_Alloc(model->surfmesh.num_vertices * sizeof(float[3])); rsurface.modelnormal3f = (float *)R_FrameData_Alloc(model->surfmesh.num_vertices * sizeof(float[3])); model->AnimateVertices(model, rsurface.frameblend, rsurface.skeleton, rsurface.modelvertex3f, rsurface.modelnormal3f, rsurface.modelsvector3f, rsurface.modeltvector3f); rsurface.modelvertexmesh = NULL; - rsurface.modelvertexmeshbuffer = NULL; - rsurface.modelvertex3fbuffer = NULL; + rsurface.modelvertexmesh_vertexbuffer = NULL; + rsurface.modelvertexmesh_bufferoffset = 0; + rsurface.modelvertex3f_vertexbuffer = NULL; + rsurface.modelvertex3f_bufferoffset = 0; + rsurface.modelvertex3f_vertexbuffer = 0; + rsurface.modelvertex3f_bufferoffset = 0; + rsurface.modelsvector3f_vertexbuffer = 0; + rsurface.modelsvector3f_bufferoffset = 0; + rsurface.modeltvector3f_vertexbuffer = 0; + rsurface.modeltvector3f_bufferoffset = 0; + rsurface.modelnormal3f_vertexbuffer = 0; + rsurface.modelnormal3f_bufferoffset = 0; } else if (wantnormals) { + r_refdef.stats[r_stat_batch_entityanimate_count]++; + r_refdef.stats[r_stat_batch_entityanimate_surfaces] += model->num_surfaces; + r_refdef.stats[r_stat_batch_entityanimate_vertices] += model->surfmesh.num_vertices; + r_refdef.stats[r_stat_batch_entityanimate_triangles] += model->surfmesh.num_triangles; rsurface.modelvertex3f = (float *)R_FrameData_Alloc(model->surfmesh.num_vertices * sizeof(float[3])); rsurface.modelsvector3f = NULL; rsurface.modeltvector3f = NULL; rsurface.modelnormal3f = (float *)R_FrameData_Alloc(model->surfmesh.num_vertices * sizeof(float[3])); model->AnimateVertices(model, rsurface.frameblend, rsurface.skeleton, rsurface.modelvertex3f, rsurface.modelnormal3f, NULL, NULL); rsurface.modelvertexmesh = NULL; - rsurface.modelvertexmeshbuffer = NULL; - rsurface.modelvertex3fbuffer = NULL; + rsurface.modelvertexmesh_vertexbuffer = NULL; + rsurface.modelvertexmesh_bufferoffset = 0; + rsurface.modelvertex3f_vertexbuffer = NULL; + rsurface.modelvertex3f_bufferoffset = 0; + rsurface.modelvertex3f_vertexbuffer = 0; + rsurface.modelvertex3f_bufferoffset = 0; + rsurface.modelsvector3f_vertexbuffer = 0; + rsurface.modelsvector3f_bufferoffset = 0; + rsurface.modeltvector3f_vertexbuffer = 0; + rsurface.modeltvector3f_bufferoffset = 0; + rsurface.modelnormal3f_vertexbuffer = 0; + rsurface.modelnormal3f_bufferoffset = 0; } else { + r_refdef.stats[r_stat_batch_entityanimate_count]++; + r_refdef.stats[r_stat_batch_entityanimate_surfaces] += model->num_surfaces; + r_refdef.stats[r_stat_batch_entityanimate_vertices] += model->surfmesh.num_vertices; + r_refdef.stats[r_stat_batch_entityanimate_triangles] += model->surfmesh.num_triangles; rsurface.modelvertex3f = (float *)R_FrameData_Alloc(model->surfmesh.num_vertices * sizeof(float[3])); rsurface.modelsvector3f = NULL; rsurface.modeltvector3f = NULL; rsurface.modelnormal3f = NULL; model->AnimateVertices(model, rsurface.frameblend, rsurface.skeleton, rsurface.modelvertex3f, NULL, NULL, NULL); rsurface.modelvertexmesh = NULL; - rsurface.modelvertexmeshbuffer = NULL; - rsurface.modelvertex3fbuffer = NULL; - } - rsurface.modelvertex3f_vertexbuffer = 0; - rsurface.modelvertex3f_bufferoffset = 0; - rsurface.modelsvector3f_vertexbuffer = 0; - rsurface.modelsvector3f_bufferoffset = 0; - rsurface.modeltvector3f_vertexbuffer = 0; - rsurface.modeltvector3f_bufferoffset = 0; - rsurface.modelnormal3f_vertexbuffer = 0; - rsurface.modelnormal3f_bufferoffset = 0; + rsurface.modelvertexmesh_vertexbuffer = NULL; + rsurface.modelvertexmesh_bufferoffset = 0; + rsurface.modelvertex3f_vertexbuffer = NULL; + rsurface.modelvertex3f_bufferoffset = 0; + rsurface.modelvertex3f_vertexbuffer = 0; + rsurface.modelvertex3f_bufferoffset = 0; + rsurface.modelsvector3f_vertexbuffer = 0; + rsurface.modelsvector3f_bufferoffset = 0; + rsurface.modeltvector3f_vertexbuffer = 0; + rsurface.modeltvector3f_bufferoffset = 0; + rsurface.modelnormal3f_vertexbuffer = 0; + rsurface.modelnormal3f_bufferoffset = 0; + } rsurface.modelgeneratedvertex = true; } else { + if (rsurface.entityskeletaltransform3x4) + { + r_refdef.stats[r_stat_batch_entityskeletal_count]++; + r_refdef.stats[r_stat_batch_entityskeletal_surfaces] += model->num_surfaces; + r_refdef.stats[r_stat_batch_entityskeletal_vertices] += model->surfmesh.num_vertices; + r_refdef.stats[r_stat_batch_entityskeletal_triangles] += model->surfmesh.num_triangles; + } + else + { + r_refdef.stats[r_stat_batch_entitystatic_count]++; + r_refdef.stats[r_stat_batch_entitystatic_surfaces] += model->num_surfaces; + r_refdef.stats[r_stat_batch_entitystatic_vertices] += model->surfmesh.num_vertices; + r_refdef.stats[r_stat_batch_entitystatic_triangles] += model->surfmesh.num_triangles; + } rsurface.modelvertex3f = model->surfmesh.data_vertex3f; rsurface.modelvertex3f_vertexbuffer = model->surfmesh.vbo_vertexbuffer; rsurface.modelvertex3f_bufferoffset = model->surfmesh.vbooffset_vertex3f; @@ -8461,9 +8721,9 @@ void RSurf_ActiveModelEntity(const entity_render_t *ent, qboolean wantnormals, q rsurface.modelnormal3f = model->surfmesh.data_normal3f; rsurface.modelnormal3f_vertexbuffer = model->surfmesh.vbo_vertexbuffer; rsurface.modelnormal3f_bufferoffset = model->surfmesh.vbooffset_normal3f; - rsurface.modelvertexmesh = model->surfmesh.vertexmesh; - rsurface.modelvertexmeshbuffer = model->surfmesh.vertexmeshbuffer; - rsurface.modelvertex3fbuffer = model->surfmesh.vertex3fbuffer; + rsurface.modelvertexmesh = model->surfmesh.data_vertexmesh; + rsurface.modelvertexmesh_vertexbuffer = model->surfmesh.vbo_vertexbuffer; + rsurface.modelvertexmesh_bufferoffset = model->surfmesh.vbooffset_vertex3f; rsurface.modelgeneratedvertex = false; } rsurface.modellightmapcolor4f = model->surfmesh.data_lightmapcolor4f; @@ -8524,8 +8784,8 @@ void RSurf_ActiveModelEntity(const entity_render_t *ent, qboolean wantnormals, q rsurface.batchskeletalweight4ub_vertexbuffer = NULL; rsurface.batchskeletalweight4ub_bufferoffset = 0; rsurface.batchvertexmesh = NULL; - rsurface.batchvertexmeshbuffer = NULL; - rsurface.batchvertex3fbuffer = NULL; + rsurface.batchvertexmesh_vertexbuffer = NULL; + rsurface.batchvertexmesh_bufferoffset = 0; rsurface.batchelement3i = NULL; rsurface.batchelement3i_indexbuffer = NULL; rsurface.batchelement3i_bufferoffset = 0; @@ -8572,7 +8832,14 @@ void RSurf_ActiveCustomEntity(const matrix4x4_t *matrix, const matrix4x4_t *inve rsurface.basepolygonfactor = r_refdef.polygonfactor; rsurface.basepolygonoffset = r_refdef.polygonoffset; rsurface.entityskeletaltransform3x4 = NULL; + rsurface.entityskeletaltransform3x4buffer = NULL; + rsurface.entityskeletaltransform3x4offset = 0; + rsurface.entityskeletaltransform3x4size = 0; rsurface.entityskeletalnumtransforms = 0; + r_refdef.stats[r_stat_batch_entitycustom_count]++; + r_refdef.stats[r_stat_batch_entitycustom_surfaces] += 1; + r_refdef.stats[r_stat_batch_entitycustom_vertices] += rsurface.modelnumvertices; + r_refdef.stats[r_stat_batch_entitycustom_triangles] += rsurface.modelnumtriangles; if (wanttangents) { rsurface.modelvertex3f = (float *)vertex3f; @@ -8595,8 +8862,8 @@ void RSurf_ActiveCustomEntity(const matrix4x4_t *matrix, const matrix4x4_t *inve rsurface.modelnormal3f = NULL; } rsurface.modelvertexmesh = NULL; - rsurface.modelvertexmeshbuffer = NULL; - rsurface.modelvertex3fbuffer = NULL; + rsurface.modelvertexmesh_vertexbuffer = NULL; + rsurface.modelvertexmesh_bufferoffset = 0; rsurface.modelvertex3f_vertexbuffer = 0; rsurface.modelvertex3f_bufferoffset = 0; rsurface.modelsvector3f_vertexbuffer = 0; @@ -8662,8 +8929,8 @@ void RSurf_ActiveCustomEntity(const matrix4x4_t *matrix, const matrix4x4_t *inve rsurface.batchskeletalweight4ub_vertexbuffer = NULL; rsurface.batchskeletalweight4ub_bufferoffset = 0; rsurface.batchvertexmesh = NULL; - rsurface.batchvertexmeshbuffer = NULL; - rsurface.batchvertex3fbuffer = NULL; + rsurface.batchvertexmesh_vertexbuffer = NULL; + rsurface.batchvertexmesh_bufferoffset = 0; rsurface.batchelement3i = NULL; rsurface.batchelement3i_indexbuffer = NULL; rsurface.batchelement3i_bufferoffset = 0; @@ -8745,6 +9012,7 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const int surfacefirstvertex; int surfaceendvertex; int surfacenumvertices; + int batchnumsurfaces = texturenumsurfaces; int batchnumvertices; int batchnumtriangles; int needsupdate; @@ -8786,6 +9054,13 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const batchnumtriangles += surfacenumtriangles; } + r_refdef.stats[r_stat_batch_batches]++; + if (gaps) + r_refdef.stats[r_stat_batch_withgaps]++; + r_refdef.stats[r_stat_batch_surfaces] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_vertices] += batchnumvertices; + r_refdef.stats[r_stat_batch_triangles] += batchnumtriangles; + // we now know the vertex range used, and if there are any gaps in it rsurface.batchfirstvertex = firstvertex; rsurface.batchnumvertices = endvertex - firstvertex; @@ -8801,11 +9076,27 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const // a cvar to force the dynamic vertex path to be taken, for debugging if (r_batch_debugdynamicvertexpath.integer) + { + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_cvar] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_cvar] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_cvar] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_cvar] += batchnumtriangles; + } dynamicvertex = true; + } // if there is a chance of animated vertex colors, it's a dynamic batch if ((batchneed & (BATCHNEED_VERTEXMESH_VERTEXCOLOR | BATCHNEED_ARRAY_VERTEXCOLOR)) && texturesurfacelist[0]->lightmapinfo) { + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_lightmapvertex] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_lightmapvertex] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_lightmapvertex] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_lightmapvertex] += batchnumtriangles; + } dynamicvertex = true; needsupdate |= BATCHNEED_VERTEXMESH_VERTEXCOLOR; } @@ -8827,16 +9118,37 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const case Q3DEFORM_NONE: break; case Q3DEFORM_AUTOSPRITE: + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_deformvertexes_autosprite] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_deformvertexes_autosprite] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_deformvertexes_autosprite] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_deformvertexes_autosprite] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX | BATCHNEED_ARRAY_NORMAL | BATCHNEED_ARRAY_VECTOR | BATCHNEED_ARRAY_TEXCOORD; needsupdate |= BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR; break; case Q3DEFORM_AUTOSPRITE2: + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_deformvertexes_autosprite2] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_deformvertexes_autosprite2] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_deformvertexes_autosprite2] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_deformvertexes_autosprite2] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX | BATCHNEED_ARRAY_TEXCOORD; needsupdate |= BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR; break; case Q3DEFORM_NORMAL: + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_deformvertexes_normal] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_deformvertexes_normal] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_deformvertexes_normal] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_deformvertexes_normal] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX | BATCHNEED_ARRAY_NORMAL | BATCHNEED_ARRAY_TEXCOORD; needsupdate |= BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR; @@ -8844,11 +9156,25 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const case Q3DEFORM_WAVE: if(!R_TestQ3WaveFunc(deform->wavefunc, deform->waveparms)) break; // if wavefunc is a nop, ignore this transform + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_deformvertexes_wave] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_deformvertexes_wave] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_deformvertexes_wave] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_deformvertexes_wave] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX | BATCHNEED_ARRAY_NORMAL | BATCHNEED_ARRAY_TEXCOORD; needsupdate |= BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR; break; case Q3DEFORM_BULGE: + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_deformvertexes_bulge] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_deformvertexes_bulge] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_deformvertexes_bulge] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_deformvertexes_bulge] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX | BATCHNEED_ARRAY_NORMAL | BATCHNEED_ARRAY_TEXCOORD; needsupdate |= BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR; @@ -8856,6 +9182,13 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const case Q3DEFORM_MOVE: if(!R_TestQ3WaveFunc(deform->wavefunc, deform->waveparms)) break; // if wavefunc is a nop, ignore this transform + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_deformvertexes_move] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_deformvertexes_move] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_deformvertexes_move] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_deformvertexes_move] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX; needsupdate |= BATCHNEED_VERTEXMESH_VERTEX; @@ -8868,16 +9201,37 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const case Q3TCGEN_TEXTURE: break; case Q3TCGEN_LIGHTMAP: + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_tcgen_lightmap] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_tcgen_lightmap] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_tcgen_lightmap] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_tcgen_lightmap] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_LIGHTMAP; needsupdate |= BATCHNEED_VERTEXMESH_LIGHTMAP; break; case Q3TCGEN_VECTOR: + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_tcgen_vector] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_tcgen_vector] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_tcgen_vector] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_tcgen_vector] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX; needsupdate |= BATCHNEED_VERTEXMESH_TEXCOORD; break; case Q3TCGEN_ENVIRONMENT: + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_tcgen_environment] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_tcgen_environment] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_tcgen_environment] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_tcgen_environment] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX | BATCHNEED_ARRAY_NORMAL; needsupdate |= BATCHNEED_VERTEXMESH_TEXCOORD; @@ -8885,6 +9239,13 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const } if (rsurface.texture->tcmods[0].tcmod == Q3TCMOD_TURBULENT) { + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_tcmod_turbulent] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_tcmod_turbulent] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_tcmod_turbulent] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_tcmod_turbulent] += batchnumtriangles; + } dynamicvertex = true; batchneed |= BATCHNEED_ARRAY_VERTEX | BATCHNEED_ARRAY_TEXCOORD; needsupdate |= BATCHNEED_VERTEXMESH_TEXCOORD; @@ -8892,20 +9253,36 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const if (!rsurface.modelvertexmesh && (batchneed & (BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR | BATCHNEED_VERTEXMESH_VERTEXCOLOR | BATCHNEED_VERTEXMESH_TEXCOORD | BATCHNEED_VERTEXMESH_LIGHTMAP))) { + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_interleavedarrays] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_interleavedarrays] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_interleavedarrays] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_interleavedarrays] += batchnumtriangles; + } dynamicvertex = true; needsupdate |= (batchneed & (BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR | BATCHNEED_VERTEXMESH_VERTEXCOLOR | BATCHNEED_VERTEXMESH_TEXCOORD | BATCHNEED_VERTEXMESH_LIGHTMAP)); } // when the model data has no vertex buffer (dynamic mesh), we need to // eliminate gaps - if (vid.useinterleavedarrays ? !rsurface.modelvertexmeshbuffer : !rsurface.modelvertex3f_vertexbuffer) + if (vid.useinterleavedarrays && !rsurface.modelvertexmesh_vertexbuffer) batchneed |= BATCHNEED_NOGAPS; // the caller can specify BATCHNEED_NOGAPS to force a batch with // firstvertex = 0 and endvertex = numvertices (no gaps, no firstvertex), // we ensure this by treating the vertex batch as dynamic... if ((batchneed & BATCHNEED_NOGAPS) && (gaps || firstvertex > 0)) + { + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_nogaps] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_nogaps] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_nogaps] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_nogaps] += batchnumtriangles; + } dynamicvertex = true; + } if (dynamicvertex) { @@ -8921,11 +9298,29 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const // if needsupdate, we have to do a dynamic vertex batch for sure if (needsupdate & batchneed) + { + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_derived] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_derived] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_derived] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_derived] += batchnumtriangles; + } dynamicvertex = true; + } // see if we need to build vertexmesh from arrays if (!rsurface.modelvertexmesh && (batchneed & (BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR | BATCHNEED_VERTEXMESH_VERTEXCOLOR | BATCHNEED_VERTEXMESH_TEXCOORD | BATCHNEED_VERTEXMESH_LIGHTMAP))) + { + if (!dynamicvertex) + { + r_refdef.stats[r_stat_batch_dynamic_batches_because_interleavedarrays] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces_because_interleavedarrays] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices_because_interleavedarrays] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles_because_interleavedarrays] += batchnumtriangles; + } dynamicvertex = true; + } // if we're going to have to apply the skeletal transform manually, we need to batch the skeletal data if (dynamicvertex && rsurface.entityskeletaltransform3x4) @@ -8958,9 +9353,9 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const rsurface.batchskeletalweight4ub = rsurface.modelskeletalweight4ub; rsurface.batchskeletalweight4ub_vertexbuffer = rsurface.modelskeletalweight4ub_vertexbuffer; rsurface.batchskeletalweight4ub_bufferoffset = rsurface.modelskeletalweight4ub_bufferoffset; - rsurface.batchvertex3fbuffer = rsurface.modelvertex3fbuffer; rsurface.batchvertexmesh = rsurface.modelvertexmesh; - rsurface.batchvertexmeshbuffer = rsurface.modelvertexmeshbuffer; + rsurface.batchvertexmesh_vertexbuffer = rsurface.modelvertexmesh_vertexbuffer; + rsurface.batchvertexmesh_bufferoffset = rsurface.modelvertexmesh_bufferoffset; rsurface.batchelement3i = rsurface.modelelement3i; rsurface.batchelement3i_indexbuffer = rsurface.modelelement3i_indexbuffer; rsurface.batchelement3i_bufferoffset = rsurface.modelelement3i_bufferoffset; @@ -8968,6 +9363,9 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const rsurface.batchelement3s_indexbuffer = rsurface.modelelement3s_indexbuffer; rsurface.batchelement3s_bufferoffset = rsurface.modelelement3s_bufferoffset; rsurface.batchskeletaltransform3x4 = rsurface.entityskeletaltransform3x4; + rsurface.batchskeletaltransform3x4buffer = rsurface.entityskeletaltransform3x4buffer; + rsurface.batchskeletaltransform3x4offset = rsurface.entityskeletaltransform3x4offset; + rsurface.batchskeletaltransform3x4size = rsurface.entityskeletaltransform3x4size; rsurface.batchskeletalnumtransforms = rsurface.entityskeletalnumtransforms; // if any dynamic vertex processing has to occur in software, we copy the @@ -8996,6 +9394,10 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const // otherwise use the original static buffer with an appropriate offset if (gaps) { + r_refdef.stats[r_stat_batch_copytriangles_batches] += 1; + r_refdef.stats[r_stat_batch_copytriangles_surfaces] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_copytriangles_vertices] += batchnumvertices; + r_refdef.stats[r_stat_batch_copytriangles_triangles] += batchnumtriangles; if ((batchneed & BATCHNEED_ALLOWMULTIDRAW) && r_batch_multidraw.integer && batchnumtriangles >= r_batch_multidraw_mintriangles.integer) { rsurface.batchmultidraw = true; @@ -9026,6 +9428,21 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const for (i = 0;i < numtriangles*3;i++) rsurface.batchelement3s[i] = rsurface.batchelement3i[i]; } + // upload buffer data for the copytriangles batch + if (((r_batch_dynamicbuffer.integer || gl_vbo_dynamicindex.integer) && vid.support.arb_vertex_buffer_object && gl_vbo.integer) || vid.forcevbo) + { + if (rsurface.batchelement3s) + rsurface.batchelement3s_indexbuffer = R_BufferData_Store(rsurface.batchnumtriangles * sizeof(short[3]), rsurface.batchelement3s, R_BUFFERDATA_INDEX16, &rsurface.batchelement3s_bufferoffset); + else if (rsurface.batchelement3i) + rsurface.batchelement3i_indexbuffer = R_BufferData_Store(rsurface.batchnumtriangles * sizeof(int[3]), rsurface.batchelement3i, R_BUFFERDATA_INDEX32, &rsurface.batchelement3i_bufferoffset); + } + } + else + { + r_refdef.stats[r_stat_batch_fast_batches] += 1; + r_refdef.stats[r_stat_batch_fast_surfaces] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_fast_vertices] += batchnumvertices; + r_refdef.stats[r_stat_batch_fast_triangles] += batchnumtriangles; } return; } @@ -9034,6 +9451,10 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const // we only directly handle separate array data in this case and then // generate interleaved data if needed... rsurface.batchgeneratedvertex = true; + r_refdef.stats[r_stat_batch_dynamic_batches] += 1; + r_refdef.stats[r_stat_batch_dynamic_surfaces] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamic_vertices] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamic_triangles] += batchnumtriangles; // now copy the vertex data into a combined array and make an index array // (this is what Quake3 does all the time) @@ -9042,9 +9463,9 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const // need actual vertex positions and normals //if (dynamicvertex) { - rsurface.batchvertex3fbuffer = NULL; rsurface.batchvertexmesh = NULL; - rsurface.batchvertexmeshbuffer = NULL; + rsurface.batchvertexmesh_vertexbuffer = NULL; + rsurface.batchvertexmesh_bufferoffset = 0; rsurface.batchvertex3f = NULL; rsurface.batchvertex3f_vertexbuffer = NULL; rsurface.batchvertex3f_bufferoffset = 0; @@ -9078,6 +9499,9 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const rsurface.batchelement3s = NULL; rsurface.batchelement3s_indexbuffer = NULL; rsurface.batchelement3s_bufferoffset = 0; + rsurface.batchskeletaltransform3x4buffer = NULL; + rsurface.batchskeletaltransform3x4offset = 0; + rsurface.batchskeletaltransform3x4size = 0; // we'll only be setting up certain arrays as needed if (batchneed & (BATCHNEED_VERTEXMESH_VERTEX | BATCHNEED_VERTEXMESH_NORMAL | BATCHNEED_VERTEXMESH_VECTOR | BATCHNEED_VERTEXMESH_VERTEXCOLOR | BATCHNEED_VERTEXMESH_TEXCOORD | BATCHNEED_VERTEXMESH_LIGHTMAP)) rsurface.batchvertexmesh = (r_vertexmesh_t *)R_FrameData_Alloc(batchnumvertices * sizeof(r_vertexmesh_t)); @@ -9211,6 +9635,10 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const float w[4]; float m[3][4], n[3][4]; float tp[3], ts[3], tt[3], tn[3]; + r_refdef.stats[r_stat_batch_dynamicskeletal_batches] += 1; + r_refdef.stats[r_stat_batch_dynamicskeletal_surfaces] += batchnumsurfaces; + r_refdef.stats[r_stat_batch_dynamicskeletal_vertices] += batchnumvertices; + r_refdef.stats[r_stat_batch_dynamicskeletal_triangles] += batchnumtriangles; si = rsurface.batchskeletalindex4ub; sw = rsurface.batchskeletalweight4ub; vp = rsurface.batchvertex3f; @@ -9662,77 +10090,80 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const } } - // generate texcoords based on the chosen texcoord source - switch(rsurface.texture->tcgen.tcgen) + if (rsurface.batchtexcoordtexture2f) { - default: - case Q3TCGEN_TEXTURE: - break; - case Q3TCGEN_LIGHTMAP: -// rsurface.batchtexcoordtexture2f = R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); -// rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; -// rsurface.batchtexcoordtexture2f_bufferoffset = 0; - if (rsurface.batchtexcoordlightmap2f) - memcpy(rsurface.batchtexcoordlightmap2f, rsurface.batchtexcoordtexture2f, batchnumvertices * sizeof(float[2])); - break; - case Q3TCGEN_VECTOR: -// rsurface.batchtexcoordtexture2f = R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); -// rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; -// rsurface.batchtexcoordtexture2f_bufferoffset = 0; - for (j = 0;j < batchnumvertices;j++) - { - rsurface.batchtexcoordtexture2f[j*2+0] = DotProduct(rsurface.batchvertex3f + 3*j, rsurface.texture->tcgen.parms); - rsurface.batchtexcoordtexture2f[j*2+1] = DotProduct(rsurface.batchvertex3f + 3*j, rsurface.texture->tcgen.parms + 3); - } - break; - case Q3TCGEN_ENVIRONMENT: - // make environment reflections using a spheremap - rsurface.batchtexcoordtexture2f = (float *)R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); - rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; - rsurface.batchtexcoordtexture2f_bufferoffset = 0; - for (j = 0;j < batchnumvertices;j++) + // generate texcoords based on the chosen texcoord source + switch(rsurface.texture->tcgen.tcgen) { - // identical to Q3A's method, but executed in worldspace so - // carried models can be shiny too + default: + case Q3TCGEN_TEXTURE: + break; + case Q3TCGEN_LIGHTMAP: + // rsurface.batchtexcoordtexture2f = R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); + // rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; + // rsurface.batchtexcoordtexture2f_bufferoffset = 0; + if (rsurface.batchtexcoordlightmap2f) + memcpy(rsurface.batchtexcoordtexture2f, rsurface.batchtexcoordlightmap2f, batchnumvertices * sizeof(float[2])); + break; + case Q3TCGEN_VECTOR: + // rsurface.batchtexcoordtexture2f = R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); + // rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; + // rsurface.batchtexcoordtexture2f_bufferoffset = 0; + for (j = 0;j < batchnumvertices;j++) + { + rsurface.batchtexcoordtexture2f[j*2+0] = DotProduct(rsurface.batchvertex3f + 3*j, rsurface.texture->tcgen.parms); + rsurface.batchtexcoordtexture2f[j*2+1] = DotProduct(rsurface.batchvertex3f + 3*j, rsurface.texture->tcgen.parms + 3); + } + break; + case Q3TCGEN_ENVIRONMENT: + // make environment reflections using a spheremap + rsurface.batchtexcoordtexture2f = (float *)R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); + rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; + rsurface.batchtexcoordtexture2f_bufferoffset = 0; + for (j = 0;j < batchnumvertices;j++) + { + // identical to Q3A's method, but executed in worldspace so + // carried models can be shiny too - float viewer[3], d, reflected[3], worldreflected[3]; + float viewer[3], d, reflected[3], worldreflected[3]; - VectorSubtract(rsurface.localvieworigin, rsurface.batchvertex3f + 3*j, viewer); - // VectorNormalize(viewer); + VectorSubtract(rsurface.localvieworigin, rsurface.batchvertex3f + 3*j, viewer); + // VectorNormalize(viewer); - d = DotProduct(rsurface.batchnormal3f + 3*j, viewer); + d = DotProduct(rsurface.batchnormal3f + 3*j, viewer); - reflected[0] = rsurface.batchnormal3f[j*3+0]*2*d - viewer[0]; - reflected[1] = rsurface.batchnormal3f[j*3+1]*2*d - viewer[1]; - reflected[2] = rsurface.batchnormal3f[j*3+2]*2*d - viewer[2]; - // note: this is proportinal to viewer, so we can normalize later + reflected[0] = rsurface.batchnormal3f[j*3+0]*2*d - viewer[0]; + reflected[1] = rsurface.batchnormal3f[j*3+1]*2*d - viewer[1]; + reflected[2] = rsurface.batchnormal3f[j*3+2]*2*d - viewer[2]; + // note: this is proportinal to viewer, so we can normalize later - Matrix4x4_Transform3x3(&rsurface.matrix, reflected, worldreflected); - VectorNormalize(worldreflected); + Matrix4x4_Transform3x3(&rsurface.matrix, reflected, worldreflected); + VectorNormalize(worldreflected); - // note: this sphere map only uses world x and z! - // so positive and negative y will LOOK THE SAME. - rsurface.batchtexcoordtexture2f[j*2+0] = 0.5 + 0.5 * worldreflected[1]; - rsurface.batchtexcoordtexture2f[j*2+1] = 0.5 - 0.5 * worldreflected[2]; + // note: this sphere map only uses world x and z! + // so positive and negative y will LOOK THE SAME. + rsurface.batchtexcoordtexture2f[j*2+0] = 0.5 + 0.5 * worldreflected[1]; + rsurface.batchtexcoordtexture2f[j*2+1] = 0.5 - 0.5 * worldreflected[2]; + } + break; } - break; - } - // the only tcmod that needs software vertex processing is turbulent, so - // check for it here and apply the changes if needed - // and we only support that as the first one - // (handling a mixture of turbulent and other tcmods would be problematic - // without punting it entirely to a software path) - if (rsurface.texture->tcmods[0].tcmod == Q3TCMOD_TURBULENT) - { - amplitude = rsurface.texture->tcmods[0].parms[1]; - animpos = rsurface.texture->tcmods[0].parms[2] + rsurface.shadertime * rsurface.texture->tcmods[0].parms[3]; -// rsurface.batchtexcoordtexture2f = R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); -// rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; -// rsurface.batchtexcoordtexture2f_bufferoffset = 0; - for (j = 0;j < batchnumvertices;j++) - { - rsurface.batchtexcoordtexture2f[j*2+0] += amplitude * sin(((rsurface.batchvertex3f[j*3+0] + rsurface.batchvertex3f[j*3+2]) * 1.0 / 1024.0f + animpos) * M_PI * 2); - rsurface.batchtexcoordtexture2f[j*2+1] += amplitude * sin(((rsurface.batchvertex3f[j*3+1] ) * 1.0 / 1024.0f + animpos) * M_PI * 2); + // the only tcmod that needs software vertex processing is turbulent, so + // check for it here and apply the changes if needed + // and we only support that as the first one + // (handling a mixture of turbulent and other tcmods would be problematic + // without punting it entirely to a software path) + if (rsurface.texture->tcmods[0].tcmod == Q3TCMOD_TURBULENT) + { + amplitude = rsurface.texture->tcmods[0].parms[1]; + animpos = rsurface.texture->tcmods[0].parms[2] + rsurface.shadertime * rsurface.texture->tcmods[0].parms[3]; + // rsurface.batchtexcoordtexture2f = R_FrameData_Alloc(batchnumvertices * sizeof(float[2])); + // rsurface.batchtexcoordtexture2f_vertexbuffer = NULL; + // rsurface.batchtexcoordtexture2f_bufferoffset = 0; + for (j = 0;j < batchnumvertices;j++) + { + rsurface.batchtexcoordtexture2f[j*2+0] += amplitude * sin(((rsurface.batchvertex3f[j*3+0] + rsurface.batchvertex3f[j*3+2]) * 1.0 / 1024.0f + animpos) * M_PI * 2); + rsurface.batchtexcoordtexture2f[j*2+1] += amplitude * sin(((rsurface.batchvertex3f[j*3+1] ) * 1.0 / 1024.0f + animpos) * M_PI * 2); + } } } @@ -9740,7 +10171,8 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const { // convert the modified arrays to vertex structs // rsurface.batchvertexmesh = R_FrameData_Alloc(batchnumvertices * sizeof(r_vertexmesh_t)); -// rsurface.batchvertexmeshbuffer = NULL; +// rsurface.batchvertexmesh_vertexbuffer = NULL; +// rsurface.batchvertexmesh_bufferoffset = 0; if (batchneed & BATCHNEED_VERTEXMESH_VERTEX) for (j = 0, vertexmesh = rsurface.batchvertexmesh;j < batchnumvertices;j++, vertexmesh++) VectorCopy(rsurface.batchvertex3f + 3*j, vertexmesh->vertex3f); @@ -9773,6 +10205,38 @@ void RSurf_PrepareVerticesForBatch(int batchneed, int texturenumsurfaces, const } } } + + // upload buffer data for the dynamic batch + if (((r_batch_dynamicbuffer.integer || gl_vbo_dynamicvertex.integer || gl_vbo_dynamicindex.integer) && vid.support.arb_vertex_buffer_object && gl_vbo.integer) || vid.forcevbo) + { + if (rsurface.batchvertexmesh) + rsurface.batchvertexmesh_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(r_vertexmesh_t), rsurface.batchvertexmesh, R_BUFFERDATA_VERTEX, &rsurface.batchvertexmesh_bufferoffset); + else + { + if (rsurface.batchvertex3f) + rsurface.batchvertex3f_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(float[3]), rsurface.batchvertex3f, R_BUFFERDATA_VERTEX, &rsurface.batchvertex3f_bufferoffset); + if (rsurface.batchsvector3f) + rsurface.batchsvector3f_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(float[3]), rsurface.batchsvector3f, R_BUFFERDATA_VERTEX, &rsurface.batchsvector3f_bufferoffset); + if (rsurface.batchtvector3f) + rsurface.batchtvector3f_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(float[3]), rsurface.batchtvector3f, R_BUFFERDATA_VERTEX, &rsurface.batchtvector3f_bufferoffset); + if (rsurface.batchnormal3f) + rsurface.batchnormal3f_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(float[3]), rsurface.batchnormal3f, R_BUFFERDATA_VERTEX, &rsurface.batchnormal3f_bufferoffset); + if (rsurface.batchlightmapcolor4f) + rsurface.batchlightmapcolor4f_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(float[4]), rsurface.batchlightmapcolor4f, R_BUFFERDATA_VERTEX, &rsurface.batchlightmapcolor4f_bufferoffset); + if (rsurface.batchtexcoordtexture2f) + rsurface.batchtexcoordtexture2f_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(float[2]), rsurface.batchtexcoordtexture2f, R_BUFFERDATA_VERTEX, &rsurface.batchtexcoordtexture2f_bufferoffset); + if (rsurface.batchtexcoordlightmap2f) + rsurface.batchtexcoordlightmap2f_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(float[2]), rsurface.batchtexcoordlightmap2f, R_BUFFERDATA_VERTEX, &rsurface.batchtexcoordlightmap2f_bufferoffset); + if (rsurface.batchskeletalindex4ub) + rsurface.batchskeletalindex4ub_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(unsigned char[4]), rsurface.batchskeletalindex4ub, R_BUFFERDATA_VERTEX, &rsurface.batchskeletalindex4ub_bufferoffset); + if (rsurface.batchskeletalweight4ub) + rsurface.batchskeletalweight4ub_vertexbuffer = R_BufferData_Store(rsurface.batchnumvertices * sizeof(unsigned char[4]), rsurface.batchskeletalweight4ub, R_BUFFERDATA_VERTEX, &rsurface.batchskeletalweight4ub_bufferoffset); + } + if (rsurface.batchelement3s) + rsurface.batchelement3s_indexbuffer = R_BufferData_Store(rsurface.batchnumtriangles * sizeof(short[3]), rsurface.batchelement3s, R_BUFFERDATA_INDEX16, &rsurface.batchelement3s_bufferoffset); + else if (rsurface.batchelement3i) + rsurface.batchelement3i_indexbuffer = R_BufferData_Store(rsurface.batchnumtriangles * sizeof(int[3]), rsurface.batchelement3i, R_BUFFERDATA_INDEX32, &rsurface.batchelement3i_bufferoffset); + } } void RSurf_DrawBatch(void) @@ -9997,6 +10461,8 @@ static void RSurf_DrawBatch_GL11_Lightmap(float r, float g, float b, float a, qb R_Mesh_ColorPointer(4, GL_FLOAT, sizeof(float[4]), rsurface.passcolor4f, rsurface.passcolor4f_vertexbuffer, rsurface.passcolor4f_bufferoffset); GL_Color(r, g, b, a); R_Mesh_TexBind(0, rsurface.lightmaptexture); + R_Mesh_TexCombine(0, GL_MODULATE, GL_MODULATE, 1, 1); + R_Mesh_TexMatrix(0, NULL); RSurf_DrawBatch(); } @@ -10192,17 +10658,14 @@ static void R_DrawTextureSurfaceList_Sky(int texturenumsurfaces, const msurface_ R_Mesh_ResetTextureState(); if (skyrendermasked) { - R_SetupShader_DepthOrShadow(false, false); + R_SetupShader_DepthOrShadow(false, false, false); // depth-only (masking) GL_ColorMask(0,0,0,0); // just to make sure that braindead drivers don't draw // anything despite that colormask... GL_BlendFunc(GL_ZERO, GL_ONE); RSurf_PrepareVerticesForBatch(BATCHNEED_ARRAY_VERTEX | BATCHNEED_ALLOWMULTIDRAW, texturenumsurfaces, texturesurfacelist); - if (rsurface.batchvertex3fbuffer) - R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3fbuffer); - else - R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3f_vertexbuffer); + R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3f_vertexbuffer, rsurface.batchvertex3f_bufferoffset); } else { @@ -10812,14 +11275,11 @@ static void R_DrawSurface_TransparentCallback(const entity_render_t *ent, const GL_BlendFunc(GL_ONE, GL_ZERO); GL_DepthMask(true); // R_Mesh_ResetTextureState(); - R_SetupShader_DepthOrShadow(false, false); } RSurf_SetupDepthAndCulling(); RSurf_PrepareVerticesForBatch(BATCHNEED_ARRAY_VERTEX | BATCHNEED_ALLOWMULTIDRAW, texturenumsurfaces, texturesurfacelist); - if (rsurface.batchvertex3fbuffer) - R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3fbuffer); - else - R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3f_vertexbuffer); + R_SetupShader_DepthOrShadow(false, false, !!rsurface.batchskeletaltransform3x4); + R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3f_vertexbuffer, rsurface.batchvertex3f_bufferoffset); RSurf_DrawBatch(); } if (setup) @@ -10911,10 +11371,8 @@ static void R_DrawTextureSurfaceList_DepthOnly(int texturenumsurfaces, const msu return; RSurf_SetupDepthAndCulling(); RSurf_PrepareVerticesForBatch(BATCHNEED_ARRAY_VERTEX | BATCHNEED_ALLOWMULTIDRAW, texturenumsurfaces, texturesurfacelist); - if (rsurface.batchvertex3fbuffer) - R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3fbuffer); - else - R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3f_vertexbuffer); + R_Mesh_PrepareVertices_Vertex3f(rsurface.batchnumvertices, rsurface.batchvertex3f, rsurface.batchvertex3f_vertexbuffer, rsurface.batchvertex3f_bufferoffset); + R_SetupShader_DepthOrShadow(false, false, !!rsurface.batchskeletaltransform3x4); RSurf_DrawBatch(); } @@ -11745,7 +12203,7 @@ static void R_DrawModelDecals_Entity(entity_render_t *ent) if (numtris > 0) { - r_refdef.stats.drawndecals += numtris; + r_refdef.stats[r_stat_drawndecals] += numtris; // now render the decals all at once // (this assumes they all use one particle font texture!) @@ -11783,7 +12241,7 @@ static void R_DrawModelDecals(void) for (i = 0;i < r_refdef.scene.numentities;i++) numdecals += r_refdef.scene.entities[i]->decalsystem.numdecals; - r_refdef.stats.totaldecals += numdecals; + r_refdef.stats[r_stat_totaldecals] += numdecals; if (r_showsurfaces.integer) return; @@ -11803,10 +12261,9 @@ extern cvar_t mod_collision_bih; static void R_DrawDebugModel(void) { entity_render_t *ent = rsurface.entity; - int i, j, k, l, flagsmask; + int i, j, flagsmask; const msurface_t *surface; dp_model_t *model = ent->model; - vec3_t v; if (!sv.active && !cls.demoplayback && ent != r_refdef.scene.worldentity) return; @@ -11938,6 +12395,8 @@ static void R_DrawDebugModel(void) if (r_shownormals.value != 0 && qglBegin) { + int l, k; + vec3_t v; if (r_showdisabledepthtest.integer) { GL_BlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA); @@ -12103,9 +12562,9 @@ void R_DrawWorldSurfaces(qboolean skysurfaces, qboolean writedepth, qboolean dep // add to stats if desired if (r_speeds.integer && !skysurfaces && !depthonly) { - r_refdef.stats.world_surfaces += numsurfacelist; + r_refdef.stats[r_stat_world_surfaces] += numsurfacelist; for (j = 0;j < numsurfacelist;j++) - r_refdef.stats.world_triangles += r_surfacelist[j]->num_triangles; + r_refdef.stats[r_stat_world_triangles] += r_surfacelist[j]->num_triangles; } rsurface.entity = NULL; // used only by R_GetCurrentTexture and RSurf_ActiveWorldEntity/RSurf_ActiveModelEntity @@ -12239,9 +12698,9 @@ void R_DrawModelSurfaces(entity_render_t *ent, qboolean skysurfaces, qboolean wr // add to stats if desired if (r_speeds.integer && !skysurfaces && !depthonly) { - r_refdef.stats.entities_surfaces += numsurfacelist; + r_refdef.stats[r_stat_entities_surfaces] += numsurfacelist; for (j = 0;j < numsurfacelist;j++) - r_refdef.stats.entities_triangles += r_surfacelist[j]->num_triangles; + r_refdef.stats[r_stat_entities_triangles] += r_surfacelist[j]->num_triangles; } rsurface.entity = NULL; // used only by R_GetCurrentTexture and RSurf_ActiveWorldEntity/RSurf_ActiveModelEntity