diff --git a/configs/current.in b/configs/current.in index 1802271..62edfa5 100644 --- a/configs/current.in +++ b/configs/current.in @@ -144,7 +144,7 @@ GLAPI_LIB_DEPS = @GLAPI_LIB_DEPS@ DRI_LIB_DEPS = @DRI_LIB_DEPS@ GALLIUM_DRI_LIB_DEPS = @GALLIUM_DRI_LIB_DEPS@ LIBDRM_CFLAGS = @LIBDRM_CFLAGS@ -LIBDRM_LIB = @LIBDRM_LIBS@ +LIBDRM_LIBS = @LIBDRM_LIBS@ DRI2PROTO_CFLAGS = @DRI2PROTO_CFLAGS@ GLPROTO_CFLAGS = @GLPROTO_CFLAGS@ EXPAT_INCLUDES = @EXPAT_INCLUDES@ diff --git a/docs/GL3.txt b/docs/GL3.txt index 24b70a1..9035779 100644 --- a/docs/GL3.txt +++ b/docs/GL3.txt @@ -34,8 +34,7 @@ sRGB framebuffer format (GL_EXT_framebuffer_sRGB) DONE (i965, r600) glClearBuffer commands DONE glGetStringi command DONE glTexParameterI, glGetTexParameterI commands DONE -glVertexAttribI commands ~50% done (converts int - values to floats) +glVertexAttribI commands DONE Depth format cube textures DONE GLX_ARB_create_context (GLX 1.4 is required) DONE diff --git a/docs/news.html b/docs/news.html index 26e648b..399a537 100644 --- a/docs/news.html +++ b/docs/news.html @@ -9,6 +9,14 @@

News

+

November 16, 2012

+ +

+Mesa 9.0.1 is released. +This is a bug fix release. +

+ +

October 8, 2012

diff --git a/docs/relnotes-9.0.1.html b/docs/relnotes-9.0.1.html index 8a354b0..ed8f1a7 100644 --- a/docs/relnotes-9.0.1.html +++ b/docs/relnotes-9.0.1.html @@ -7,7 +7,7 @@ -

Mesa 9.0.1 Release Notes / November 14th, 2012

+

Mesa 9.0.1 Release Notes / November 16th, 2012

Mesa 9.0.1 is a bug fix release which fixes bugs found since the 9.0 release. @@ -23,7 +23,9 @@ because GL_ARB_compatibility is not supported.

MD5 checksums

-TBD.
+97d6554c05ea7449398afe3a0ede7018  MesaLib-9.0.1.tar.bz2
+fd0fd5a6e56bc3dd210c80e42baef975  MesaLib-9.0.1.tar.gz
+c2683d957acd530a00f747f50317186f  MesaLib-9.0.1.zip
 

New features

diff --git a/src/gallium/auxiliary/draw/draw_vertex.h b/src/gallium/auxiliary/draw/draw_vertex.h index e63cf5f..c87c3d8 100644 --- a/src/gallium/auxiliary/draw/draw_vertex.h +++ b/src/gallium/auxiliary/draw/draw_vertex.h @@ -42,6 +42,7 @@ #include "pipe/p_compiler.h" #include "pipe/p_state.h" #include "util/u_debug.h" +#include "util/u_memory.h" /** @@ -87,7 +88,7 @@ struct vertex_info unsigned interp_mode:4; /**< INTERP_x */ unsigned emit:4; /**< EMIT_x */ unsigned src_index:8; /**< map to post-xform attribs */ - } attrib[PIPE_MAX_SHADER_INPUTS]; + } attrib[PIPE_MAX_SHADER_OUTPUTS]; }; static INLINE size_t @@ -127,7 +128,7 @@ draw_emit_vertex_attr(struct vertex_info *vinfo, uint src_index) { const uint n = vinfo->num_attribs; - assert(n < PIPE_MAX_SHADER_INPUTS); + assert(n < Elements(vinfo->attrib)); vinfo->attrib[n].emit = emit; vinfo->attrib[n].interp_mode = interp; vinfo->attrib[n].src_index = src_index; diff --git a/src/gallium/auxiliary/tgsi/tgsi_ureg.c b/src/gallium/auxiliary/tgsi/tgsi_ureg.c index 3fe78e0..3c2a923 100644 --- a/src/gallium/auxiliary/tgsi/tgsi_ureg.c +++ b/src/gallium/auxiliary/tgsi/tgsi_ureg.c @@ -73,7 +73,7 @@ struct ureg_tokens { #define UREG_MAX_INPUT PIPE_MAX_ATTRIBS #define UREG_MAX_SYSTEM_VALUE PIPE_MAX_ATTRIBS -#define UREG_MAX_OUTPUT PIPE_MAX_ATTRIBS +#define UREG_MAX_OUTPUT PIPE_MAX_SHADER_OUTPUTS #define UREG_MAX_CONSTANT_RANGE 32 #define UREG_MAX_IMMEDIATE 256 #define UREG_MAX_ADDR 2 diff --git a/src/gallium/drivers/nv50/codegen/nv50_ir_from_tgsi.cpp b/src/gallium/drivers/nv50/codegen/nv50_ir_from_tgsi.cpp index 071674b..e73c804 100644 --- a/src/gallium/drivers/nv50/codegen/nv50_ir_from_tgsi.cpp +++ b/src/gallium/drivers/nv50/codegen/nv50_ir_from_tgsi.cpp @@ -257,7 +257,7 @@ unsigned int Instruction::srcMask(unsigned int s) const mask &= 0x9; break; case TGSI_TEXTURE_SHADOW1D: - mask &= 0x5; + mask &= 0xd; break; case TGSI_TEXTURE_1D_ARRAY: case TGSI_TEXTURE_2D: diff --git a/src/gallium/drivers/nv50/nv50_screen.c b/src/gallium/drivers/nv50/nv50_screen.c index 3d510ea..982948e 100644 --- a/src/gallium/drivers/nv50/nv50_screen.c +++ b/src/gallium/drivers/nv50/nv50_screen.c @@ -454,6 +454,18 @@ nv50_screen_init_hwctx(struct nv50_screen *screen) PUSH_DATA (push, (NV50_CB_AUX << 12) | 0xf21); PUSH_DATA (push, (NV50_CB_AUX << 12) | 0xf31); + /* return { 0.0, 0.0, 0.0, 0.0 } on out-of-bounds vtxbuf access */ + BEGIN_NV04(push, NV50_3D(CB_ADDR), 1); + PUSH_DATA (push, ((1 << 9) << 6) | NV50_CB_AUX); + BEGIN_NI04(push, NV50_3D(CB_DATA(0)), 4); + PUSH_DATAf(push, 0.0f); + PUSH_DATAf(push, 0.0f); + PUSH_DATAf(push, 0.0f); + PUSH_DATAf(push, 0.0f); + BEGIN_NV04(push, NV50_3D(VERTEX_RUNOUT_ADDRESS_HIGH), 2); + PUSH_DATAh(push, screen->uniforms->offset + (3 << 16) + (1 << 9)); + PUSH_DATA (push, screen->uniforms->offset + (3 << 16) + (1 << 9)); + /* max TIC (bits 4:8) & TSC bindings, per program type */ for (i = 0; i < 3; ++i) { BEGIN_NV04(push, NV50_3D(TEX_LIMITS(i)), 1); diff --git a/src/gallium/drivers/nvc0/nvc0_context.c b/src/gallium/drivers/nvc0/nvc0_context.c index b021f2e..9c43ae6 100644 --- a/src/gallium/drivers/nvc0/nvc0_context.c +++ b/src/gallium/drivers/nvc0/nvc0_context.c @@ -183,6 +183,7 @@ nvc0_create(struct pipe_screen *pscreen, void *priv) BCTX_REFN_bo(nvc0->bufctx_3d, SCREEN, flags, screen->text); BCTX_REFN_bo(nvc0->bufctx_3d, SCREEN, flags, screen->uniform_bo); BCTX_REFN_bo(nvc0->bufctx_3d, SCREEN, flags, screen->txc); + BCTX_REFN_bo(nvc0->bufctx_3d, SCREEN, flags, screen->poly_cache); flags = NOUVEAU_BO_GART | NOUVEAU_BO_WR; diff --git a/src/gallium/drivers/nvc0/nvc0_screen.c b/src/gallium/drivers/nvc0/nvc0_screen.c index eb74e95..8bb9ff6 100644 --- a/src/gallium/drivers/nvc0/nvc0_screen.c +++ b/src/gallium/drivers/nvc0/nvc0_screen.c @@ -589,7 +589,7 @@ nvc0_screen_create(struct nouveau_device *dev) for (i = 0; i < 5; ++i) { /* TIC and TSC entries for each unit (nve4+ only) */ - /* auxiliary constants (6 user clip planes, base instance id */ + /* auxiliary constants (6 user clip planes, base instance id) */ BEGIN_NVC0(push, NVC0_3D(CB_SIZE), 3); PUSH_DATA (push, 512); PUSH_DATAh(push, screen->uniform_bo->offset + (5 << 16) + (i << 9)); @@ -610,6 +610,21 @@ nvc0_screen_create(struct nouveau_device *dev) BEGIN_NVC0(push, NVC0_3D(LINKED_TSC), 1); PUSH_DATA (push, 0); + /* return { 0.0, 0.0, 0.0, 0.0 } for out-of-bounds vtxbuf access */ + BEGIN_NVC0(push, NVC0_3D(CB_SIZE), 3); + PUSH_DATA (push, 256); + PUSH_DATAh(push, screen->uniform_bo->offset + (5 << 16) + (6 << 9)); + PUSH_DATA (push, screen->uniform_bo->offset + (5 << 16) + (6 << 9)); + BEGIN_1IC0(push, NVC0_3D(CB_POS), 5); + PUSH_DATA (push, 0); + PUSH_DATAf(push, 0.0f); + PUSH_DATAf(push, 0.0f); + PUSH_DATAf(push, 0.0f); + PUSH_DATAf(push, 0.0f); + BEGIN_NVC0(push, NVC0_3D(VERTEX_RUNOUT_ADDRESS_HIGH), 2); + PUSH_DATAh(push, screen->uniform_bo->offset + (5 << 16) + (6 << 9)); + PUSH_DATA (push, screen->uniform_bo->offset + (5 << 16) + (6 << 9)); + /* max MPs * max warps per MP (TODO: ask kernel) */ if (screen->eng3d->oclass >= NVE4_3D_CLASS) screen->tls_size = 8 * 64; @@ -734,10 +749,6 @@ nvc0_screen_create(struct nouveau_device *dev) IMMED_NVC0(push, NVC0_3D(EDGEFLAG), 1); - BEGIN_NVC0(push, NVC0_3D(VERTEX_RUNOUT_ADDRESS_HIGH), 2); - PUSH_DATA (push, 0xab); - PUSH_DATA (push, 0x00000000); - PUSH_KICK (push); screen->tic.entries = CALLOC(4096, sizeof(void *)); diff --git a/src/gallium/drivers/r300/r300_state_derived.c b/src/gallium/drivers/r300/r300_state_derived.c index 940179c..3ccf396 100644 --- a/src/gallium/drivers/r300/r300_state_derived.c +++ b/src/gallium/drivers/r300/r300_state_derived.c @@ -655,6 +655,12 @@ static uint32_t r300_get_border_color(enum pipe_format format, case PIPE_FORMAT_LATC2_UNORM: util_pack_color(border_swizzled, PIPE_FORMAT_R8G8B8A8_UNORM, &uc); return uc.ui; + case PIPE_FORMAT_DXT1_SRGB: + case PIPE_FORMAT_DXT1_SRGBA: + case PIPE_FORMAT_DXT3_SRGBA: + case PIPE_FORMAT_DXT5_SRGBA: + util_pack_color(border_swizzled, PIPE_FORMAT_B8G8R8A8_SRGB, &uc); + return uc.ui; default: util_pack_color(border_swizzled, PIPE_FORMAT_B8G8R8A8_UNORM, &uc); return uc.ui; @@ -685,10 +691,18 @@ static uint32_t r300_get_border_color(enum pipe_format format, default: case 8: - if (desc->channel[0].type == UTIL_FORMAT_TYPE_SIGNED) - util_pack_color(border_swizzled, PIPE_FORMAT_R8G8B8A8_SNORM, &uc); - else - util_pack_color(border_swizzled, PIPE_FORMAT_R8G8B8A8_UNORM, &uc); + if (desc->channel[0].type == UTIL_FORMAT_TYPE_SIGNED) { + util_pack_color(border_swizzled, PIPE_FORMAT_R8G8B8A8_SNORM, &uc); + } else if (desc->colorspace == UTIL_FORMAT_COLORSPACE_SRGB) { + if (desc->nr_channels == 2) { + border_swizzled[3] = border_swizzled[1]; + util_pack_color(border_swizzled, PIPE_FORMAT_L8A8_SRGB, &uc); + } else { + util_pack_color(border_swizzled, PIPE_FORMAT_R8G8B8A8_SRGB, &uc); + } + } else { + util_pack_color(border_swizzled, PIPE_FORMAT_R8G8B8A8_UNORM, &uc); + } break; case 10: diff --git a/src/gallium/drivers/r600/r600_hw_context.c b/src/gallium/drivers/r600/r600_hw_context.c index 7b7b6b1..af27fd9 100644 --- a/src/gallium/drivers/r600/r600_hw_context.c +++ b/src/gallium/drivers/r600/r600_hw_context.c @@ -1080,13 +1080,14 @@ void r600_context_streamout_begin(struct r600_context *ctx) unsigned *stride_in_dw = ctx->vs_shader->so.stride; unsigned buffer_en, i, update_flags = 0; uint64_t va; + unsigned num_cs_dw_streamout_end; buffer_en = (ctx->num_so_targets >= 1 && t[0] ? 1 : 0) | (ctx->num_so_targets >= 2 && t[1] ? 2 : 0) | (ctx->num_so_targets >= 3 && t[2] ? 4 : 0) | (ctx->num_so_targets >= 4 && t[3] ? 8 : 0); - ctx->num_cs_dw_streamout_end = + num_cs_dw_streamout_end = 12 + /* flush_vgt_streamout */ util_bitcount(buffer_en) * 8 + /* STRMOUT_BUFFER_UPDATE */ 3 /* set_streamout_enable(0) */; @@ -1095,11 +1096,15 @@ void r600_context_streamout_begin(struct r600_context *ctx) 12 + /* flush_vgt_streamout */ 6 + /* set_streamout_enable */ util_bitcount(buffer_en) * 7 + /* SET_CONTEXT_REG */ - (ctx->chip_class == R700 ? util_bitcount(buffer_en) * 5 : 0) + /* STRMOUT_BASE_UPDATE */ + (ctx->family >= CHIP_RS780 && + ctx->family <= CHIP_RV740 ? util_bitcount(buffer_en) * 5 : 0) + /* STRMOUT_BASE_UPDATE */ util_bitcount(buffer_en & ctx->streamout_append_bitmask) * 8 + /* STRMOUT_BUFFER_UPDATE */ util_bitcount(buffer_en & ~ctx->streamout_append_bitmask) * 6 + /* STRMOUT_BUFFER_UPDATE */ - (ctx->family > CHIP_R600 && ctx->family < CHIP_RV770 ? 2 : 0) + /* SURFACE_BASE_UPDATE */ - ctx->num_cs_dw_streamout_end, TRUE); + (ctx->family > CHIP_R600 && ctx->family < CHIP_RS780 ? 2 : 0) + /* SURFACE_BASE_UPDATE */ + num_cs_dw_streamout_end, TRUE); + + /* This must be set after r600_need_cs_space. */ + ctx->num_cs_dw_streamout_end = num_cs_dw_streamout_end; if (ctx->chip_class >= EVERGREEN) { evergreen_flush_vgt_streamout(ctx); @@ -1133,7 +1138,7 @@ void r600_context_streamout_begin(struct r600_context *ctx) /* R7xx requires this packet after updating BUFFER_BASE. * Without this, R7xx locks up. */ - if (ctx->chip_class == R700) { + if (ctx->family >= CHIP_RS780 && ctx->family <= CHIP_RV740) { cs->buf[cs->cdw++] = PKT3(PKT3_STRMOUT_BASE_UPDATE, 1, 0); cs->buf[cs->cdw++] = i; cs->buf[cs->cdw++] = va >> 8; @@ -1173,7 +1178,7 @@ void r600_context_streamout_begin(struct r600_context *ctx) } } - if (ctx->family > CHIP_R600 && ctx->family < CHIP_RV770) { + if (ctx->family > CHIP_R600 && ctx->family < CHIP_RS780) { cs->buf[cs->cdw++] = PKT3(PKT3_SURFACE_BASE_UPDATE, 0, 0); cs->buf[cs->cdw++] = update_flags; } diff --git a/src/gallium/drivers/r600/r600_pipe.c b/src/gallium/drivers/r600/r600_pipe.c index 286c676..be30c7e 100644 --- a/src/gallium/drivers/r600/r600_pipe.c +++ b/src/gallium/drivers/r600/r600_pipe.c @@ -443,7 +443,7 @@ static int r600_get_param(struct pipe_screen* pscreen, enum pipe_cap param) return rscreen->has_streamout ? 1 : 0; case PIPE_CAP_MAX_STREAM_OUTPUT_SEPARATE_COMPONENTS: case PIPE_CAP_MAX_STREAM_OUTPUT_INTERLEAVED_COMPONENTS: - return 16*4; + return 32*4; /* Texturing. */ case PIPE_CAP_MAX_TEXTURE_2D_LEVELS: @@ -933,13 +933,19 @@ struct pipe_screen *r600_screen_create(struct radeon_winsys *ws) /* Figure out streamout kernel support. */ switch (rscreen->chip_class) { case R600: - case EVERGREEN: - case CAYMAN: - rscreen->has_streamout = rscreen->info.drm_minor >= 14; + if (rscreen->family < CHIP_RS780) { + rscreen->has_streamout = rscreen->info.drm_minor >= 14; + } else { + rscreen->has_streamout = rscreen->info.drm_minor >= 23; + } break; case R700: rscreen->has_streamout = rscreen->info.drm_minor >= 17; break; + case EVERGREEN: + case CAYMAN: + rscreen->has_streamout = rscreen->info.drm_minor >= 14; + break; } if (r600_init_tiling(rscreen)) { diff --git a/src/gallium/drivers/r600/r600_shader.c b/src/gallium/drivers/r600/r600_shader.c index 21019b7..1f8ab2b 100644 --- a/src/gallium/drivers/r600/r600_shader.c +++ b/src/gallium/drivers/r600/r600_shader.c @@ -4012,6 +4012,23 @@ static int tgsi_tex(struct r600_shader_ctx *ctx) if (r) return r; } + + /* for cube forms of lod and bias we need to route the lod + value into Z */ + if (inst->Instruction.Opcode == TGSI_OPCODE_TXB || + inst->Instruction.Opcode == TGSI_OPCODE_TXL) { + memset(&alu, 0, sizeof(struct r600_bytecode_alu)); + alu.inst = CTX_INST(V_SQ_ALU_WORD1_OP2_SQ_OP2_INST_MOV); + r600_bytecode_src(&alu.src[0], &ctx->src[0], 3); + alu.dst.sel = ctx->temp_reg; + alu.dst.chan = 2; + alu.last = 1; + alu.dst.write = 1; + r = r600_bytecode_add_alu(ctx->bc, &alu); + if (r) + return r; + } + src_loaded = TRUE; src_gpr = ctx->temp_reg; } @@ -4087,17 +4104,12 @@ static int tgsi_tex(struct r600_shader_ctx *ctx) tex.src_rel = ctx->src[0].rel; } - if (inst->Texture.Texture == TGSI_TEXTURE_CUBE) { - tex.src_sel_x = 1; - tex.src_sel_y = 0; - tex.src_sel_z = 3; - tex.src_sel_w = 1; - } - if (inst->Texture.Texture == TGSI_TEXTURE_SHADOWCUBE) { + if (inst->Texture.Texture == TGSI_TEXTURE_CUBE || + inst->Texture.Texture == TGSI_TEXTURE_SHADOWCUBE) { tex.src_sel_x = 1; tex.src_sel_y = 0; tex.src_sel_z = 3; - tex.src_sel_w = 2; /* route Z compare value into W */ + tex.src_sel_w = 2; /* route Z compare or Lod value into W */ } if (inst->Texture.Texture != TGSI_TEXTURE_RECT && diff --git a/src/gallium/include/pipe/p_state.h b/src/gallium/include/pipe/p_state.h index 9ea9c0e..a67c43e 100644 --- a/src/gallium/include/pipe/p_state.h +++ b/src/gallium/include/pipe/p_state.h @@ -59,7 +59,7 @@ extern "C" { #define PIPE_MAX_CONSTANT_BUFFERS 32 #define PIPE_MAX_SAMPLERS 16 #define PIPE_MAX_SHADER_INPUTS 32 -#define PIPE_MAX_SHADER_OUTPUTS 32 +#define PIPE_MAX_SHADER_OUTPUTS 48 /* 32 GENERICs + POS, PSIZE, FOG, etc. */ #define PIPE_MAX_SHADER_SAMPLER_VIEWS 32 #define PIPE_MAX_SHADER_RESOURCES 32 #define PIPE_MAX_TEXTURE_LEVELS 16 @@ -145,7 +145,7 @@ struct pipe_rasterizer_state unsigned line_stipple_factor:8; /**< [1..256] actually */ unsigned line_stipple_pattern:16; - unsigned sprite_coord_enable:PIPE_MAX_SHADER_OUTPUTS; + unsigned sprite_coord_enable; /* bitfield referring to 32 GENERIC inputs */ float line_width; float point_size; /**< used when no per-vertex size */ diff --git a/src/gallium/targets/dri-i915/Makefile b/src/gallium/targets/dri-i915/Makefile index ddebff2..a835ee7 100644 --- a/src/gallium/targets/dri-i915/Makefile +++ b/src/gallium/targets/dri-i915/Makefile @@ -28,6 +28,6 @@ endif include ../Makefile.dri -GALLIUM_DRI_LIB_DEPS += -ldrm_intel +GALLIUM_DRI_LIB_DEPS += $(INTEL_LIBS) symlinks: diff --git a/src/gallium/targets/dri-r300/Makefile b/src/gallium/targets/dri-r300/Makefile index 256aa23..63a2956 100644 --- a/src/gallium/targets/dri-r300/Makefile +++ b/src/gallium/targets/dri-r300/Makefile @@ -21,6 +21,6 @@ DRIVER_DEFINES = \ include ../Makefile.dri -GALLIUM_DRI_LIB_DEPS += -ldrm_radeon +GALLIUM_DRI_LIB_DEPS += $(RADEON_LIBS) symlinks: diff --git a/src/gallium/targets/dri-r600/Makefile b/src/gallium/targets/dri-r600/Makefile index 7402750..2499ede 100644 --- a/src/gallium/targets/dri-r600/Makefile +++ b/src/gallium/targets/dri-r600/Makefile @@ -21,6 +21,6 @@ DRIVER_DEFINES = \ include ../Makefile.dri -GALLIUM_DRI_LIB_DEPS += -ldrm_radeon +GALLIUM_DRI_LIB_DEPS += $(RADEON_LIBS) symlinks: diff --git a/src/gallium/targets/dri-radeonsi/Makefile b/src/gallium/targets/dri-radeonsi/Makefile index b45f243..588946e 100644 --- a/src/gallium/targets/dri-radeonsi/Makefile +++ b/src/gallium/targets/dri-radeonsi/Makefile @@ -21,6 +21,6 @@ DRIVER_DEFINES = \ include ../Makefile.dri -GALLIUM_DRI_LIB_DEPS += -ldrm_radeon +GALLIUM_DRI_LIB_DEPS += $(RADEON_LIBS) symlinks: diff --git a/src/gallium/targets/egl-static/Makefile b/src/gallium/targets/egl-static/Makefile index f05c81d..08626a9 100644 --- a/src/gallium/targets/egl-static/Makefile +++ b/src/gallium/targets/egl-static/Makefile @@ -48,17 +48,17 @@ egl_SYS += $(LIBUDEV_LIBS) $(DLOPEN_LIBS) -lEGL -lm -lpthread ifneq ($(findstring x11, $(EGL_PLATFORMS)),) egl_CPPFLAGS += $(LIBDRM_CFLAGS) egl_LIBS += $(TOP)/src/gallium/winsys/sw/xlib/libws_xlib.a -egl_SYS += -lX11 -lXext -lXfixes $(LIBDRM_LIB) +egl_SYS += -lX11 -lXext -lXfixes $(LIBDRM_LIBS) endif ifneq ($(findstring wayland, $(EGL_PLATFORMS)),) egl_CPPFLAGS += $(LIBDRM_CFLAGS) egl_LIBS += $(TOP)/src/gallium/winsys/sw/wayland/libws_wayland.a egl_LIBS += $(TOP)/src/egl/wayland/wayland-drm/.libs/libwayland-drm.a -egl_SYS += $(LIBDRM_LIB) $(WAYLAND_LIBS) +egl_SYS += $(LIBDRM_LIBS) $(WAYLAND_LIBS) endif ifneq ($(findstring drm, $(EGL_PLATFORMS)),) egl_CPPFLAGS += $(LIBDRM_CFLAGS) -egl_SYS += $(LIBDRM_LIB) -lgbm +egl_SYS += $(LIBDRM_LIBS) -lgbm endif ifneq ($(findstring fbdev, $(EGL_PLATFORMS)),) egl_LIBS += $(TOP)/src/gallium/winsys/sw/fbdev/libfbdev.a @@ -93,7 +93,7 @@ egl_CPPFLAGS += -D_EGL_PIPE_I915=1 egl_LIBS += \ $(TOP)/src/gallium/winsys/i915/drm/libi915drm.a \ $(TOP)/src/gallium/drivers/i915/libi915.a -egl_SYS += -ldrm_intel +egl_SYS += $(INTEL_LIBS) endif # nouveau @@ -105,7 +105,7 @@ egl_LIBS += \ $(TOP)/src/gallium/drivers/nv50/libnv50.a \ $(TOP)/src/gallium/drivers/nvc0/libnvc0.a \ $(TOP)/src/gallium/drivers/nouveau/libnouveau.a -egl_SYS += -ldrm_nouveau +egl_SYS += $(NOUVEAU_LIBS) endif # r300 @@ -115,7 +115,7 @@ egl_CPPFLAGS += -D_EGL_PIPE_R300=1 egl_LIBS += \ $(TOP)/src/gallium/winsys/radeon/drm/libradeonwinsys.a \ $(TOP)/src/gallium/drivers/r300/libr300.a -egl_SYS += -ldrm_radeon +egl_SYS += $(RADEON_LIBS) endif endif @@ -126,7 +126,7 @@ egl_CPPFLAGS += -D_EGL_PIPE_R600=1 egl_LIBS += \ $(TOP)/src/gallium/winsys/radeon/drm/libradeonwinsys.a \ $(TOP)/src/gallium/drivers/r600/libr600.a -egl_SYS += -ldrm_radeon +egl_SYS += $(RADEON_LIBS) endif endif @@ -137,7 +137,7 @@ egl_CPPFLAGS += -D_EGL_PIPE_RADEONSI=1 egl_LIBS += \ $(TOP)/src/gallium/winsys/radeon/drm/libradeonwinsys.a \ $(TOP)/src/gallium/drivers/radeonsi/libradeonsi.a -egl_SYS += -ldrm_radeon +egl_SYS += $(RADEON_LIBS) endif endif diff --git a/src/gallium/targets/gbm/Makefile b/src/gallium/targets/gbm/Makefile index 423debf..17d5648 100644 --- a/src/gallium/targets/gbm/Makefile +++ b/src/gallium/targets/gbm/Makefile @@ -14,7 +14,7 @@ GBM_INCLUDES = \ -I$(TOP)/src/gallium/winsys \ -I$(TOP)/src/gallium/include -GBM_LIBS = $(LIBUDEV_LIBS) $(LIBDRM_LIB) -lm \ +GBM_LIBS = $(LIBUDEV_LIBS) $(LIBDRM_LIBS) -lm \ $(TOP)/src/gallium/state_trackers/gbm/libgbm.a \ $(GALLIUM_PIPE_LOADER_LIBS) $(GALLIUM_AUXILIARIES) diff --git a/src/gallium/targets/pipe-loader/Makefile b/src/gallium/targets/pipe-loader/Makefile index 596539c..f15cd25 100644 --- a/src/gallium/targets/pipe-loader/Makefile +++ b/src/gallium/targets/pipe-loader/Makefile @@ -24,7 +24,7 @@ PIPE_LIBS = \ $(TOP)/src/gallium/drivers/rbug/librbug.a \ $(GALLIUM_AUXILIARIES) -PIPE_SYS = $(LIBDRM_LIB) -lm -lpthread $(DLOPEN_LIBS) +PIPE_SYS = $(LIBDRM_LIBS) -lm -lpthread $(DLOPEN_LIBS) PIPE_CFLAGS = $(LIBDRM_CFLAGS) @@ -34,7 +34,7 @@ PIPE_LDFLAGS = -Wl,--no-undefined i915_LIBS = \ $(TOP)/src/gallium/winsys/i915/drm/libi915drm.a \ $(TOP)/src/gallium/drivers/i915/libi915.a -i915_SYS = -ldrm_intel +i915_SYS = $(INTEL_LIBS) # nouveau pipe driver nouveau_LIBS = \ @@ -43,25 +43,25 @@ nouveau_LIBS = \ $(TOP)/src/gallium/drivers/nv50/libnv50.a \ $(TOP)/src/gallium/drivers/nvc0/libnvc0.a \ $(TOP)/src/gallium/drivers/nouveau/libnouveau.a -nouveau_SYS = -ldrm_nouveau +nouveau_SYS = $(NOUVEAU_LIBS) # r300 pipe driver r300_LIBS = \ $(TOP)/src/gallium/winsys/radeon/drm/libradeonwinsys.a \ $(TOP)/src/gallium/drivers/r300/libr300.a -r300_SYS += -ldrm_radeon +r300_SYS += $(RADEON_LIBS) # r600 pipe driver r600_LIBS = \ $(TOP)/src/gallium/winsys/radeon/drm/libradeonwinsys.a \ $(TOP)/src/gallium/drivers/r600/libr600.a -r600_SYS += -ldrm_radeon +r600_SYS += $(RADEON_LIBS) # radeonsi pipe driver radeonsi_LIBS = \ $(TOP)/src/gallium/winsys/radeon/drm/libradeonwinsys.a \ $(TOP)/src/gallium/drivers/radeonsi/libradeonsi.a -radeonsi_SYS += -ldrm_radeon +radeonsi_SYS += $(RADEON_LIBS) # vmwgfx pipe driver vmwgfx_LIBS = \ diff --git a/src/gallium/targets/vdpau-r300/Makefile b/src/gallium/targets/vdpau-r300/Makefile index 1ada550..e79c920 100644 --- a/src/gallium/targets/vdpau-r300/Makefile +++ b/src/gallium/targets/vdpau-r300/Makefile @@ -21,7 +21,7 @@ C_SOURCES = \ $(COMMON_GALLIUM_SOURCES) \ $(DRIVER_SOURCES) -DRIVER_LIBS = $(shell $(PKG_CONFIG) libdrm --libs) -lXfixes -ldrm_radeon +DRIVER_LIBS = $(LIBDRM_LIBS) $(RADEON_LIBS) -lXfixes include ../Makefile.vdpau diff --git a/src/gallium/targets/vdpau-r600/Makefile b/src/gallium/targets/vdpau-r600/Makefile index 05e0b4f..6ef7f90 100644 --- a/src/gallium/targets/vdpau-r600/Makefile +++ b/src/gallium/targets/vdpau-r600/Makefile @@ -17,7 +17,7 @@ C_SOURCES = \ $(COMMON_GALLIUM_SOURCES) \ $(DRIVER_SOURCES) -DRIVER_LIBS = $(shell $(PKG_CONFIG) libdrm --libs) -lXfixes -ldrm_radeon +DRIVER_LIBS = $(LIBDRM_LIBS) $(RADEON_LIBS) -lXfixes include ../Makefile.vdpau diff --git a/src/gallium/targets/vdpau-radeonsi/Makefile b/src/gallium/targets/vdpau-radeonsi/Makefile index 6dd2be9..4ff6743 100644 --- a/src/gallium/targets/vdpau-radeonsi/Makefile +++ b/src/gallium/targets/vdpau-radeonsi/Makefile @@ -23,7 +23,7 @@ C_SOURCES = \ DRIVER_DEFINES = \ -DGALLIUM_RBUG -DGALLIUM_TRACE -DGALLIUM_NOOP -DRIVER_LIBS = $(shell $(PKG_CONFIG) libdrm --libs) -lXfixes -ldrm_radeon +DRIVER_LIBS = $(LIBDRM_LIBS) $(RADEON_LIBS) -lXfixes include ../Makefile.vdpau diff --git a/src/gallium/targets/xvmc-r300/Makefile b/src/gallium/targets/xvmc-r300/Makefile index 800f8d5..d2ab488 100644 --- a/src/gallium/targets/xvmc-r300/Makefile +++ b/src/gallium/targets/xvmc-r300/Makefile @@ -17,7 +17,7 @@ C_SOURCES = \ $(COMMON_GALLIUM_SOURCES) \ $(DRIVER_SOURCES) -DRIVER_LIBS = $(shell $(PKG_CONFIG) libdrm --libs) -lXfixes -ldrm_radeon +DRIVER_LIBS = $(LIBDRM_LIBS) $(RADEON_LIBS) -lXfixes include ../Makefile.xvmc diff --git a/src/gallium/targets/xvmc-r600/Makefile b/src/gallium/targets/xvmc-r600/Makefile index d9ce72f..3ef5b1a 100644 --- a/src/gallium/targets/xvmc-r600/Makefile +++ b/src/gallium/targets/xvmc-r600/Makefile @@ -17,7 +17,7 @@ C_SOURCES = \ $(COMMON_GALLIUM_SOURCES) \ $(DRIVER_SOURCES) -DRIVER_LIBS = $(shell $(PKG_CONFIG) libdrm --libs) -lXfixes -ldrm_radeon +DRIVER_LIBS = $(LIBDRM_LIBS) $(RADEON_LIBS) -lXfixes include ../Makefile.xvmc diff --git a/src/mesa/drivers/dri/nouveau/nouveau_texture.c b/src/mesa/drivers/dri/nouveau/nouveau_texture.c index 37f7577..288b510 100644 --- a/src/mesa/drivers/dri/nouveau/nouveau_texture.c +++ b/src/mesa/drivers/dri/nouveau/nouveau_texture.c @@ -369,7 +369,7 @@ get_last_level(struct gl_texture_object *t) t->Sampler.MinFilter == GL_LINEAR || !base) return t->BaseLevel; else - return MIN2(t->BaseLevel + base->MaxLog2, t->MaxLevel); + return MIN2(t->BaseLevel + base->MaxNumLevels - 1, t->MaxLevel); } static void diff --git a/src/mesa/drivers/dri/radeon/radeon_mipmap_tree.c b/src/mesa/drivers/dri/radeon/radeon_mipmap_tree.c index 8901090..f4fb3a5 100644 --- a/src/mesa/drivers/dri/radeon/radeon_mipmap_tree.c +++ b/src/mesa/drivers/dri/radeon/radeon_mipmap_tree.c @@ -270,7 +270,7 @@ static void calculate_min_max_lod(struct gl_sampler_object *samp, struct gl_text minLod = MIN2(minLod, tObj->MaxLevel); maxLod = tObj->BaseLevel + (GLint)(samp->MaxLod + 0.5); maxLod = MIN2(maxLod, tObj->MaxLevel); - maxLod = MIN2(maxLod, tObj->Image[0][minLod]->MaxLog2 + minLod); + maxLod = MIN2(maxLod, tObj->Image[0][minLod]->MaxNumLevels - 1 + minLod); maxLod = MAX2(maxLod, minLod); /* need at least one level */ } break; @@ -329,7 +329,7 @@ static GLboolean radeon_miptree_matches_texture(radeon_mipmap_tree *mt, struct g mtBaseLevel = &mt->levels[texObj->BaseLevel - mt->baseLevel]; firstImage = texObj->Image[0][texObj->BaseLevel]; - numLevels = MIN2(texObj->_MaxLevel - texObj->BaseLevel + 1, firstImage->MaxLog2 + 1); + numLevels = MIN2(texObj->_MaxLevel - texObj->BaseLevel + 1, firstImage->MaxNumLevels); if (radeon_is_debug_enabled(RADEON_TEXTURE,RADEON_TRACE)) { fprintf(stderr, "Checking if miptree %p matches texObj %p\n", mt, texObj); @@ -378,7 +378,7 @@ void radeon_try_alloc_miptree(radeonContextPtr rmesa, radeonTexObj *t) } - numLevels = MIN2(texObj->MaxLevel - texObj->BaseLevel + 1, texImg->MaxLog2 + 1); + numLevels = MIN2(texObj->MaxLevel - texObj->BaseLevel + 1, texImg->MaxNumLevels); t->mt = radeon_miptree_create(rmesa, t->base.Target, texImg->TexFormat, texObj->BaseLevel, diff --git a/src/mesa/main/config.h b/src/mesa/main/config.h index ead1d32..3572ae8 100644 --- a/src/mesa/main/config.h +++ b/src/mesa/main/config.h @@ -187,7 +187,7 @@ #define MAX_PROGRAM_CALL_DEPTH 8 #define MAX_PROGRAM_TEMPS 256 #define MAX_PROGRAM_ADDRESS_REGS 2 -#define MAX_VARYING 16 /**< number of float[4] vectors */ +#define MAX_VARYING 32 /**< number of float[4] vectors */ #define MAX_SAMPLERS MAX_TEXTURE_IMAGE_UNITS #define MAX_PROGRAM_INPUTS 32 #define MAX_PROGRAM_OUTPUTS 64 diff --git a/src/mesa/main/context.c b/src/mesa/main/context.c index a21ae10..de385cd 100644 --- a/src/mesa/main/context.c +++ b/src/mesa/main/context.c @@ -624,7 +624,7 @@ _mesa_init_constants(struct gl_context *ctx) #if FEATURE_ARB_vertex_shader ctx->Const.MaxVertexTextureImageUnits = MAX_VERTEX_TEXTURE_IMAGE_UNITS; ctx->Const.MaxCombinedTextureImageUnits = MAX_COMBINED_TEXTURE_IMAGE_UNITS; - ctx->Const.MaxVarying = MAX_VARYING; + ctx->Const.MaxVarying = 16; /* old limit not to break tnl and swrast */ #endif #if FEATURE_ARB_geometry_shader4 ctx->Const.MaxGeometryTextureImageUnits = MAX_GEOMETRY_TEXTURE_IMAGE_UNITS; diff --git a/src/mesa/main/imports.h b/src/mesa/main/imports.h index 551aea7..4b74f70 100644 --- a/src/mesa/main/imports.h +++ b/src/mesa/main/imports.h @@ -81,7 +81,7 @@ extern "C" { * these casts generate warnings. * The following union typedef is used to solve that. */ -typedef union { GLfloat f; GLint i; } fi_type; +typedef union { GLfloat f; GLint i; GLuint u; } fi_type; diff --git a/src/mesa/main/macros.h b/src/mesa/main/macros.h index 5af9487..e7ef877 100644 --- a/src/mesa/main/macros.h +++ b/src/mesa/main/macros.h @@ -171,6 +171,20 @@ extern GLfloat _mesa_ubyte_to_float_color_tab[256]; ub = ((GLubyte) F_TO_I((f) * 255.0F)) #endif +static inline GLfloat INT_AS_FLT(GLint i) +{ + fi_type tmp; + tmp.i = i; + return tmp.f; +} + +static inline GLfloat UINT_AS_FLT(GLuint u) +{ + fi_type tmp; + tmp.u = u; + return tmp.f; +} + /*@}*/ @@ -573,6 +587,31 @@ do { \ /*@}*/ +/** Copy \p sz elements into a homegeneous (4-element) vector, giving + * default values to the remaining components. + * The default values are chosen based on \p type. + */ +static inline void +COPY_CLEAN_4V_TYPE_AS_FLOAT(GLfloat dst[4], int sz, const GLfloat src[4], + GLenum type) +{ + switch (type) { + case GL_FLOAT: + ASSIGN_4V(dst, 0, 0, 0, 1); + break; + case GL_INT: + ASSIGN_4V(dst, INT_AS_FLT(0), INT_AS_FLT(0), + INT_AS_FLT(0), INT_AS_FLT(1)); + break; + case GL_UNSIGNED_INT: + ASSIGN_4V(dst, UINT_AS_FLT(0), UINT_AS_FLT(0), + UINT_AS_FLT(0), UINT_AS_FLT(1)); + break; + default: + ASSERT(0); + } + COPY_SZ_4V(dst, sz, src); +} /** \name Linear interpolation functions */ /*@{*/ diff --git a/src/mesa/main/mtypes.h b/src/mesa/main/mtypes.h index f185ffd..6fce70f 100644 --- a/src/mesa/main/mtypes.h +++ b/src/mesa/main/mtypes.h @@ -1205,7 +1205,8 @@ struct gl_texture_image GLuint WidthLog2; /**< = log2(Width2) */ GLuint HeightLog2; /**< = log2(Height2) */ GLuint DepthLog2; /**< = log2(Depth2) */ - GLuint MaxLog2; /**< = MAX(WidthLog2, HeightLog2) */ + GLuint MaxNumLevels; /**< = maximum possible number of mipmap + levels, computed from the dimensions */ struct gl_texture_object *TexObject; /**< Pointer back to parent object */ GLuint Level; /**< Which mipmap level am I? */ diff --git a/src/mesa/main/teximage.c b/src/mesa/main/teximage.c index 38fa9fa..9edd4d8 100644 --- a/src/mesa/main/teximage.c +++ b/src/mesa/main/teximage.c @@ -1002,6 +1002,43 @@ _mesa_get_texture_dimensions(GLenum target) } +/** + * Return the maximum number of mipmap levels for the given target + * and the dimensions. + * The dimensions are expected not to include the border. + */ +GLsizei +_mesa_get_tex_max_num_levels(GLenum target, GLsizei width, GLsizei height, + GLsizei depth) +{ + GLsizei size; + + switch (target) { + case GL_TEXTURE_1D: + case GL_TEXTURE_1D_ARRAY: + size = width; + break; + case GL_TEXTURE_CUBE_MAP: + case GL_TEXTURE_CUBE_MAP_ARRAY: + ASSERT(width == height); + size = width; + break; + case GL_TEXTURE_2D: + case GL_TEXTURE_2D_ARRAY: + size = MAX2(width, height); + break; + case GL_TEXTURE_3D: + size = MAX3(width, height, depth); + break; + case GL_TEXTURE_RECTANGLE: + return 1; + default: + assert(0); + return 1; + } + + return _mesa_logbase2(size) + 1; +} #if 000 /* not used anymore */ @@ -1187,7 +1224,9 @@ _mesa_init_teximage_fields(struct gl_context *ctx, target); } - img->MaxLog2 = MAX2(img->WidthLog2, img->HeightLog2); + img->MaxNumLevels = + _mesa_get_tex_max_num_levels(target, + img->Width2, img->Height2, img->Depth2); img->TexFormat = format; } diff --git a/src/mesa/main/teximage.h b/src/mesa/main/teximage.h index 36fd1c2..a2565ec 100644 --- a/src/mesa/main/teximage.h +++ b/src/mesa/main/teximage.h @@ -138,6 +138,10 @@ _mesa_tex_target_to_face(GLenum target); extern GLint _mesa_get_texture_dimensions(GLenum target); +extern GLsizei +_mesa_get_tex_max_num_levels(GLenum target, GLsizei width, GLsizei height, + GLsizei depth); + extern GLenum _mesa_es_error_check_format_and_type(GLenum format, GLenum type, unsigned dimensions); diff --git a/src/mesa/main/texstorage.c b/src/mesa/main/texstorage.c index f8a9397..b4dd13e 100644 --- a/src/mesa/main/texstorage.c +++ b/src/mesa/main/texstorage.c @@ -244,7 +244,6 @@ tex_storage_error_check(struct gl_context *ctx, GLuint dims, GLenum target, GLsizei width, GLsizei height, GLsizei depth) { struct gl_texture_object *texObj; - GLuint maxDim; GLboolean legalFormat; /* check internal format - note that only sized formats are allowed */ @@ -324,8 +323,7 @@ tex_storage_error_check(struct gl_context *ctx, GLuint dims, GLenum target, } /* check levels against width/height/depth */ - maxDim = MAX3(width, height, depth); - if (levels > _mesa_logbase2(maxDim) + 1) { + if (levels > _mesa_get_tex_max_num_levels(target, width, height, depth)) { _mesa_error(ctx, GL_INVALID_OPERATION, "glTexStorage%uD(too many levels for max texture dimension)", dims); diff --git a/src/mesa/program/prog_print.c b/src/mesa/program/prog_print.c index bc5dc58..887f020 100644 --- a/src/mesa/program/prog_print.c +++ b/src/mesa/program/prog_print.c @@ -164,7 +164,23 @@ arb_input_attrib_string(GLint index, GLenum progType) "fragment.varying[12]", "fragment.varying[13]", "fragment.varying[14]", - "fragment.varying[15]" /* MAX_VARYING = 16 */ + "fragment.varying[15]", + "fragment.varying[16]", + "fragment.varying[17]", + "fragment.varying[18]", + "fragment.varying[19]", + "fragment.varying[20]", + "fragment.varying[21]", + "fragment.varying[22]", + "fragment.varying[23]", + "fragment.varying[24]", + "fragment.varying[25]", + "fragment.varying[26]", + "fragment.varying[27]", + "fragment.varying[28]", + "fragment.varying[29]", + "fragment.varying[30]", + "fragment.varying[31]", /* MAX_VARYING = 32 */ }; /* sanity checks */ @@ -268,7 +284,23 @@ arb_output_attrib_string(GLint index, GLenum progType) "result.varying[12]", "result.varying[13]", "result.varying[14]", - "result.varying[15]" /* MAX_VARYING = 16 */ + "result.varying[15]", + "result.varying[16]", + "result.varying[17]", + "result.varying[18]", + "result.varying[19]", + "result.varying[20]", + "result.varying[21]", + "result.varying[22]", + "result.varying[23]", + "result.varying[24]", + "result.varying[25]", + "result.varying[26]", + "result.varying[27]", + "result.varying[28]", + "result.varying[29]", + "result.varying[30]", + "result.varying[31]", /* MAX_VARYING = 32 */ }; static const char *const fragResults[] = { "result.depth", /* FRAG_RESULT_DEPTH */ diff --git a/src/mesa/state_tracker/st_atom_sampler.c b/src/mesa/state_tracker/st_atom_sampler.c index adcc7b5..3eba5b1 100644 --- a/src/mesa/state_tracker/st_atom_sampler.c +++ b/src/mesa/state_tracker/st_atom_sampler.c @@ -34,6 +34,7 @@ #include "main/macros.h" #include "main/mtypes.h" +#include "main/glformats.h" #include "main/samplerobj.h" #include "main/texobj.h" @@ -172,12 +173,17 @@ convert_sampler(struct st_context *st, msamp->BorderColor.ui[2] || msamp->BorderColor.ui[3]) { struct gl_texture_image *teximg; + GLboolean is_integer = GL_FALSE; teximg = texobj->Image[0][texobj->BaseLevel]; - st_translate_color(msamp->BorderColor.f, - teximg ? teximg->_BaseFormat : GL_RGBA, - sampler->border_color.f); + if (teximg) { + is_integer = _mesa_is_enum_format_integer(teximg->InternalFormat); + } + + st_translate_color(&msamp->BorderColor, + &sampler->border_color, + teximg ? teximg->_BaseFormat : GL_RGBA, is_integer); } sampler->max_anisotropy = (msamp->MaxAnisotropy == 1.0 ? diff --git a/src/mesa/state_tracker/st_cb_clear.c b/src/mesa/state_tracker/st_cb_clear.c index e731b6b..90eb0af 100644 --- a/src/mesa/state_tracker/st_cb_clear.c +++ b/src/mesa/state_tracker/st_cb_clear.c @@ -37,6 +37,7 @@ #include "main/accum.h" #include "main/formats.h" #include "main/macros.h" +#include "main/glformats.h" #include "program/prog_instruction.h" #include "st_context.h" #include "st_atom.h" @@ -301,9 +302,13 @@ clear_with_quad(struct gl_context *ctx, cso_set_geometry_shader_handle(st->cso_context, NULL); if (ctx->DrawBuffer->_ColorDrawBuffers[0]) { - st_translate_color(ctx->Color.ClearColor.f, - ctx->DrawBuffer->_ColorDrawBuffers[0]->_BaseFormat, - clearColor.f); + struct gl_renderbuffer *rb = ctx->DrawBuffer->_ColorDrawBuffers[0]; + GLboolean is_integer = _mesa_is_enum_format_integer(rb->InternalFormat); + + st_translate_color(&ctx->Color.ClearColor, + &clearColor, + ctx->DrawBuffer->_ColorDrawBuffers[0]->_BaseFormat, + is_integer); } /* draw quad matching scissor rect */ @@ -540,9 +545,13 @@ st_Clear(struct gl_context *ctx, GLbitfield mask) clear_buffers |= PIPE_CLEAR_DEPTHSTENCIL; if (ctx->DrawBuffer->_ColorDrawBuffers[0]) { - st_translate_color(ctx->Color.ClearColor.f, + struct gl_renderbuffer *rb = ctx->DrawBuffer->_ColorDrawBuffers[0]; + GLboolean is_integer = _mesa_is_enum_format_integer(rb->InternalFormat); + + st_translate_color(&ctx->Color.ClearColor, + &clearColor, ctx->DrawBuffer->_ColorDrawBuffers[0]->_BaseFormat, - clearColor.f); + is_integer); } st->pipe->clear(st->pipe, clear_buffers, &clearColor, diff --git a/src/mesa/state_tracker/st_cb_texture.c b/src/mesa/state_tracker/st_cb_texture.c index b8e2538..923a8c9 100644 --- a/src/mesa/state_tracker/st_cb_texture.c +++ b/src/mesa/state_tracker/st_cb_texture.c @@ -237,36 +237,6 @@ default_bindings(struct st_context *st, enum pipe_format format) } -/** Return number of image dimensions (1, 2 or 3) for a texture target. */ -static GLuint -get_texture_dims(GLenum target) -{ - switch (target) { - case GL_TEXTURE_1D: - case GL_TEXTURE_1D_ARRAY_EXT: - case GL_TEXTURE_BUFFER: - return 1; - case GL_TEXTURE_2D: - case GL_TEXTURE_CUBE_MAP_ARB: - case GL_TEXTURE_CUBE_MAP_POSITIVE_X_ARB: - case GL_TEXTURE_CUBE_MAP_NEGATIVE_X_ARB: - case GL_TEXTURE_CUBE_MAP_POSITIVE_Y_ARB: - case GL_TEXTURE_CUBE_MAP_NEGATIVE_Y_ARB: - case GL_TEXTURE_CUBE_MAP_POSITIVE_Z_ARB: - case GL_TEXTURE_CUBE_MAP_NEGATIVE_Z_ARB: - case GL_TEXTURE_RECTANGLE_NV: - case GL_TEXTURE_2D_ARRAY_EXT: - case GL_TEXTURE_EXTERNAL_OES: - return 2; - case GL_TEXTURE_3D: - return 3; - default: - assert(0 && "invalid texture target in get_texture_dims()"); - return 1; - } -} - - /** * Given the size of a mipmap image, try to compute the size of the level=0 * mipmap image. @@ -281,32 +251,57 @@ guess_base_level_size(GLenum target, GLuint width, GLuint height, GLuint depth, GLuint level, GLuint *width0, GLuint *height0, GLuint *depth0) { - const GLuint dims = get_texture_dims(target); - assert(width >= 1); assert(height >= 1); assert(depth >= 1); if (level > 0) { - /* Depending on the image's size, we can't always make a guess here */ - if ((dims >= 1 && width == 1) || - (dims >= 2 && height == 1) || - (dims >= 3 && depth == 1)) { - /* we can't determine the image size at level=0 */ - return GL_FALSE; - } + /* Guess the size of the base level. + * Depending on the image's size, we can't always make a guess here. + */ + switch (target) { + case GL_TEXTURE_1D: + case GL_TEXTURE_1D_ARRAY: + width <<= level; + break; + + case GL_TEXTURE_2D: + case GL_TEXTURE_2D_ARRAY: + /* We can't make a good guess here, because the base level dimensions + * can be non-square. + */ + if (width == 1 || height == 1) { + return GL_FALSE; + } + width <<= level; + height <<= level; + break; + + case GL_TEXTURE_CUBE_MAP: + case GL_TEXTURE_CUBE_MAP_ARRAY: + width <<= level; + height <<= level; + break; + + case GL_TEXTURE_3D: + /* We can't make a good guess here, because the base level dimensions + * can be non-cube. + */ + if (width == 1 || height == 1 || depth == 1) { + return GL_FALSE; + } + width <<= level; + height <<= level; + depth <<= level; + break; + + case GL_TEXTURE_RECTANGLE: + break; - /* grow the image size until we hit level = 0 */ - while (level > 0) { - if (width > 1) - width <<= 1; - if (height > 1) - height <<= 1; - if (depth > 1) - depth <<= 1; - level--; + default: + assert(0); } - } + } *width0 = width; *height0 = height; @@ -371,10 +366,8 @@ guess_and_alloc_texture(struct st_context *st, } else { /* alloc space for a full mipmap */ - GLuint l2width = util_logbase2(width); - GLuint l2height = util_logbase2(height); - GLuint l2depth = util_logbase2(depth); - lastLevel = MAX2(MAX2(l2width, l2height), l2depth); + lastLevel = _mesa_get_tex_max_num_levels(stObj->base.Target, + width, height, depth) - 1; } /* Save the level=0 dimensions */ diff --git a/src/mesa/state_tracker/st_format.c b/src/mesa/state_tracker/st_format.c index 302b58c..75602e5 100644 --- a/src/mesa/state_tracker/st_format.c +++ b/src/mesa/state_tracker/st_format.c @@ -1699,44 +1699,92 @@ st_sampler_compat_formats(enum pipe_format format1, enum pipe_format format2) * Similarly for texture border colors. */ void -st_translate_color(const GLfloat colorIn[4], GLenum baseFormat, - GLfloat colorOut[4]) +st_translate_color(union gl_color_union *colorIn, + union pipe_color_union *colorOut, + GLenum baseFormat, GLboolean is_integer) { - switch (baseFormat) { - case GL_RED: - colorOut[0] = colorIn[0]; - colorOut[1] = 0.0F; - colorOut[2] = 0.0F; - colorOut[3] = 1.0F; - break; - case GL_RG: - colorOut[0] = colorIn[0]; - colorOut[1] = colorIn[1]; - colorOut[2] = 0.0F; - colorOut[3] = 1.0F; - break; - case GL_RGB: - colorOut[0] = colorIn[0]; - colorOut[1] = colorIn[1]; - colorOut[2] = colorIn[2]; - colorOut[3] = 1.0F; - break; - case GL_ALPHA: - colorOut[0] = colorOut[1] = colorOut[2] = 0.0; - colorOut[3] = colorIn[3]; - break; - case GL_LUMINANCE: - colorOut[0] = colorOut[1] = colorOut[2] = colorIn[0]; - colorOut[3] = 1.0; - break; - case GL_LUMINANCE_ALPHA: - colorOut[0] = colorOut[1] = colorOut[2] = colorIn[0]; - colorOut[3] = colorIn[3]; - break; - case GL_INTENSITY: - colorOut[0] = colorOut[1] = colorOut[2] = colorOut[3] = colorIn[0]; - break; - default: - COPY_4V(colorOut, colorIn); + if (is_integer) { + int *in = colorIn->i; + int *out = colorOut->i; + + switch (baseFormat) { + case GL_RED: + out[0] = in[0]; + out[1] = 0; + out[2] = 0; + out[3] = 1; + break; + case GL_RG: + out[0] = in[0]; + out[1] = in[1]; + out[2] = 0; + out[3] = 1; + break; + case GL_RGB: + out[0] = in[0]; + out[1] = in[1]; + out[2] = in[2]; + out[3] = 1; + break; + case GL_ALPHA: + out[0] = out[1] = out[2] = 0; + out[3] = in[3]; + break; + case GL_LUMINANCE: + out[0] = out[1] = out[2] = in[0]; + out[3] = 1; + break; + case GL_LUMINANCE_ALPHA: + out[0] = out[1] = out[2] = in[0]; + out[3] = in[3]; + break; + case GL_INTENSITY: + out[0] = out[1] = out[2] = out[3] = in[0]; + break; + default: + COPY_4V(out, in); + } + } + else { + float *in = colorIn->f; + float *out = colorOut->f; + + switch (baseFormat) { + case GL_RED: + out[0] = in[0]; + out[1] = 0.0F; + out[2] = 0.0F; + out[3] = 1.0F; + break; + case GL_RG: + out[0] = in[0]; + out[1] = in[1]; + out[2] = 0.0F; + out[3] = 1.0F; + break; + case GL_RGB: + out[0] = in[0]; + out[1] = in[1]; + out[2] = in[2]; + out[3] = 1.0F; + break; + case GL_ALPHA: + out[0] = out[1] = out[2] = 0.0F; + out[3] = in[3]; + break; + case GL_LUMINANCE: + out[0] = out[1] = out[2] = in[0]; + out[3] = 1.0F; + break; + case GL_LUMINANCE_ALPHA: + out[0] = out[1] = out[2] = in[0]; + out[3] = in[3]; + break; + case GL_INTENSITY: + out[0] = out[1] = out[2] = out[3] = in[0]; + break; + default: + COPY_4V(out, in); + } } } diff --git a/src/mesa/state_tracker/st_format.h b/src/mesa/state_tracker/st_format.h index 2eef2c0..39397b1 100644 --- a/src/mesa/state_tracker/st_format.h +++ b/src/mesa/state_tracker/st_format.h @@ -75,7 +75,8 @@ st_sampler_compat_formats(enum pipe_format format1, enum pipe_format format2); extern void -st_translate_color(const GLfloat colorIn[4], GLenum baseFormat, - GLfloat colorOut[4]); +st_translate_color(union gl_color_union *colorIn, + union pipe_color_union *colorOut, + GLenum baseFormat, GLboolean is_integer); #endif /* ST_FORMAT_H */ diff --git a/src/mesa/state_tracker/st_gen_mipmap.c b/src/mesa/state_tracker/st_gen_mipmap.c index 8892006..c092613 100644 --- a/src/mesa/state_tracker/st_gen_mipmap.c +++ b/src/mesa/state_tracker/st_gen_mipmap.c @@ -121,30 +121,16 @@ compute_num_levels(struct gl_context *ctx, struct gl_texture_object *texObj, GLenum target) { - if (target == GL_TEXTURE_RECTANGLE_ARB) { - return 1; - } - else { - const struct gl_texture_image *baseImage = - _mesa_get_tex_image(ctx, texObj, target, texObj->BaseLevel); - GLuint size, numLevels; + const struct gl_texture_image *baseImage; + GLuint numLevels; - size = MAX2(baseImage->Width2, baseImage->Height2); - size = MAX2(size, baseImage->Depth2); + baseImage = _mesa_get_tex_image(ctx, texObj, target, texObj->BaseLevel); - numLevels = texObj->BaseLevel; - - while (size > 0) { - numLevels++; - size >>= 1; - } + numLevels = texObj->BaseLevel + baseImage->MaxNumLevels; + numLevels = MIN2(numLevels, texObj->MaxLevel + 1); + assert(numLevels >= 1); - numLevels = MIN2(numLevels, texObj->MaxLevel + 1); - - assert(numLevels >= 1); - - return numLevels; - } + return numLevels; } diff --git a/src/mesa/vbo/vbo_attrib_tmp.h b/src/mesa/vbo/vbo_attrib_tmp.h index d3fc77e..b4a6d73 100644 --- a/src/mesa/vbo/vbo_attrib_tmp.h +++ b/src/mesa/vbo/vbo_attrib_tmp.h @@ -26,38 +26,46 @@ USE OR OTHER DEALINGS IN THE SOFTWARE. **************************************************************************/ /* float */ -#define ATTR1FV( A, V ) ATTR( A, 1, (V)[0], 0, 0, 1 ) -#define ATTR2FV( A, V ) ATTR( A, 2, (V)[0], (V)[1], 0, 1 ) -#define ATTR3FV( A, V ) ATTR( A, 3, (V)[0], (V)[1], (V)[2], 1 ) -#define ATTR4FV( A, V ) ATTR( A, 4, (V)[0], (V)[1], (V)[2], (V)[3] ) +#define ATTR1FV( A, V ) ATTR( A, 1, GL_FLOAT, (V)[0], 0, 0, 1 ) +#define ATTR2FV( A, V ) ATTR( A, 2, GL_FLOAT, (V)[0], (V)[1], 0, 1 ) +#define ATTR3FV( A, V ) ATTR( A, 3, GL_FLOAT, (V)[0], (V)[1], (V)[2], 1 ) +#define ATTR4FV( A, V ) ATTR( A, 4, GL_FLOAT, (V)[0], (V)[1], (V)[2], (V)[3] ) -#define ATTR1F( A, X ) ATTR( A, 1, X, 0, 0, 1 ) -#define ATTR2F( A, X, Y ) ATTR( A, 2, X, Y, 0, 1 ) -#define ATTR3F( A, X, Y, Z ) ATTR( A, 3, X, Y, Z, 1 ) -#define ATTR4F( A, X, Y, Z, W ) ATTR( A, 4, X, Y, Z, W ) +#define ATTR1F( A, X ) ATTR( A, 1, GL_FLOAT, X, 0, 0, 1 ) +#define ATTR2F( A, X, Y ) ATTR( A, 2, GL_FLOAT, X, Y, 0, 1 ) +#define ATTR3F( A, X, Y, Z ) ATTR( A, 3, GL_FLOAT, X, Y, Z, 1 ) +#define ATTR4F( A, X, Y, Z, W ) ATTR( A, 4, GL_FLOAT, X, Y, Z, W ) /* int */ -#define ATTR2IV( A, V ) ATTR( A, 2, (V)[0], (V)[1], 0, 1 ) -#define ATTR3IV( A, V ) ATTR( A, 3, (V)[0], (V)[1], (V)[2], 1 ) -#define ATTR4IV( A, V ) ATTR( A, 4, (V)[0], (V)[1], (V)[2], (V)[3] ) +#define ATTRI( A, N, X, Y, Z, W) ATTR( A, N, GL_INT, \ + INT_AS_FLT(X), INT_AS_FLT(Y), \ + INT_AS_FLT(Z), INT_AS_FLT(W) ) -#define ATTR1I( A, X ) ATTR( A, 1, X, 0, 0, 1 ) -#define ATTR2I( A, X, Y ) ATTR( A, 2, X, Y, 0, 1 ) -#define ATTR3I( A, X, Y, Z ) ATTR( A, 3, X, Y, Z, 1 ) -#define ATTR4I( A, X, Y, Z, W ) ATTR( A, 4, X, Y, Z, W ) +#define ATTR2IV( A, V ) ATTRI( A, 2, (V)[0], (V)[1], 0, 1 ) +#define ATTR3IV( A, V ) ATTRI( A, 3, (V)[0], (V)[1], (V)[2], 1 ) +#define ATTR4IV( A, V ) ATTRI( A, 4, (V)[0], (V)[1], (V)[2], (V)[3] ) + +#define ATTR1I( A, X ) ATTRI( A, 1, X, 0, 0, 1 ) +#define ATTR2I( A, X, Y ) ATTRI( A, 2, X, Y, 0, 1 ) +#define ATTR3I( A, X, Y, Z ) ATTRI( A, 3, X, Y, Z, 1 ) +#define ATTR4I( A, X, Y, Z, W ) ATTRI( A, 4, X, Y, Z, W ) /* uint */ -#define ATTR2UIV( A, V ) ATTR( A, 2, (V)[0], (V)[1], 0, 1 ) -#define ATTR3UIV( A, V ) ATTR( A, 3, (V)[0], (V)[1], (V)[2], 1 ) -#define ATTR4UIV( A, V ) ATTR( A, 4, (V)[0], (V)[1], (V)[2], (V)[3] ) +#define ATTRUI( A, N, X, Y, Z, W) ATTR( A, N, GL_UNSIGNED_INT, \ + UINT_AS_FLT(X), UINT_AS_FLT(Y), \ + UINT_AS_FLT(Z), UINT_AS_FLT(W) ) + +#define ATTR2UIV( A, V ) ATTRUI( A, 2, (V)[0], (V)[1], 0, 1 ) +#define ATTR3UIV( A, V ) ATTRUI( A, 3, (V)[0], (V)[1], (V)[2], 1 ) +#define ATTR4UIV( A, V ) ATTRUI( A, 4, (V)[0], (V)[1], (V)[2], (V)[3] ) -#define ATTR1UI( A, X ) ATTR( A, 1, X, 0, 0, 1 ) -#define ATTR2UI( A, X, Y ) ATTR( A, 2, X, Y, 0, 1 ) -#define ATTR3UI( A, X, Y, Z ) ATTR( A, 3, X, Y, Z, 1 ) -#define ATTR4UI( A, X, Y, Z, W ) ATTR( A, 4, X, Y, Z, W ) +#define ATTR1UI( A, X ) ATTRUI( A, 1, X, 0, 0, 1 ) +#define ATTR2UI( A, X, Y ) ATTRUI( A, 2, X, Y, 0, 1 ) +#define ATTR3UI( A, X, Y, Z ) ATTRUI( A, 3, X, Y, Z, 1 ) +#define ATTR4UI( A, X, Y, Z, W ) ATTRUI( A, 4, X, Y, Z, W ) -#define MAT_ATTR( A, N, V ) ATTR( A, N, (V)[0], (V)[1], (V)[2], (V)[3] ) +#define MAT_ATTR( A, N, V ) ATTR( A, N, GL_FLOAT, (V)[0], (V)[1], (V)[2], (V)[3] ) static inline float conv_ui10_to_norm_float(unsigned ui10) { @@ -69,20 +77,20 @@ static inline float conv_ui2_to_norm_float(unsigned ui2) return (float)(ui2) / 3.0; } -#define ATTRUI10_1( A, UI ) ATTR( A, 1, (UI) & 0x3ff, 0, 0, 1 ) -#define ATTRUI10_2( A, UI ) ATTR( A, 2, (UI) & 0x3ff, ((UI) >> 10) & 0x3ff, 0, 1 ) -#define ATTRUI10_3( A, UI ) ATTR( A, 3, (UI) & 0x3ff, ((UI) >> 10) & 0x3ff, ((UI) >> 20) & 0x3ff, 1 ) -#define ATTRUI10_4( A, UI ) ATTR( A, 4, (UI) & 0x3ff, ((UI) >> 10) & 0x3ff, ((UI) >> 20) & 0x3ff, ((UI) >> 30) & 0x3 ) +#define ATTRUI10_1( A, UI ) ATTR( A, 1, GL_FLOAT, (UI) & 0x3ff, 0, 0, 1 ) +#define ATTRUI10_2( A, UI ) ATTR( A, 2, GL_FLOAT, (UI) & 0x3ff, ((UI) >> 10) & 0x3ff, 0, 1 ) +#define ATTRUI10_3( A, UI ) ATTR( A, 3, GL_FLOAT, (UI) & 0x3ff, ((UI) >> 10) & 0x3ff, ((UI) >> 20) & 0x3ff, 1 ) +#define ATTRUI10_4( A, UI ) ATTR( A, 4, GL_FLOAT, (UI) & 0x3ff, ((UI) >> 10) & 0x3ff, ((UI) >> 20) & 0x3ff, ((UI) >> 30) & 0x3 ) -#define ATTRUI10N_1( A, UI ) ATTR( A, 1, conv_ui10_to_norm_float((UI) & 0x3ff), 0, 0, 1 ) -#define ATTRUI10N_2( A, UI ) ATTR( A, 2, \ +#define ATTRUI10N_1( A, UI ) ATTR( A, 1, GL_FLOAT, conv_ui10_to_norm_float((UI) & 0x3ff), 0, 0, 1 ) +#define ATTRUI10N_2( A, UI ) ATTR( A, 2, GL_FLOAT, \ conv_ui10_to_norm_float((UI) & 0x3ff), \ conv_ui10_to_norm_float(((UI) >> 10) & 0x3ff), 0, 1 ) -#define ATTRUI10N_3( A, UI ) ATTR( A, 3, \ +#define ATTRUI10N_3( A, UI ) ATTR( A, 3, GL_FLOAT, \ conv_ui10_to_norm_float((UI) & 0x3ff), \ conv_ui10_to_norm_float(((UI) >> 10) & 0x3ff), \ conv_ui10_to_norm_float(((UI) >> 20) & 0x3ff), 1 ) -#define ATTRUI10N_4( A, UI ) ATTR( A, 4, \ +#define ATTRUI10N_4( A, UI ) ATTR( A, 4, GL_FLOAT, \ conv_ui10_to_norm_float((UI) & 0x3ff), \ conv_ui10_to_norm_float(((UI) >> 10) & 0x3ff), \ conv_ui10_to_norm_float(((UI) >> 20) & 0x3ff), \ @@ -119,30 +127,30 @@ static inline float conv_i2_to_norm_float(int i2) return (float)val.x; } -#define ATTRI10_1( A, I10 ) ATTR( A, 1, conv_i10_to_i((I10) & 0x3ff), 0, 0, 1 ) -#define ATTRI10_2( A, I10 ) ATTR( A, 2, \ +#define ATTRI10_1( A, I10 ) ATTR( A, 1, GL_FLOAT, conv_i10_to_i((I10) & 0x3ff), 0, 0, 1 ) +#define ATTRI10_2( A, I10 ) ATTR( A, 2, GL_FLOAT, \ conv_i10_to_i((I10) & 0x3ff), \ conv_i10_to_i(((I10) >> 10) & 0x3ff), 0, 1 ) -#define ATTRI10_3( A, I10 ) ATTR( A, 3, \ +#define ATTRI10_3( A, I10 ) ATTR( A, 3, GL_FLOAT, \ conv_i10_to_i((I10) & 0x3ff), \ conv_i10_to_i(((I10) >> 10) & 0x3ff), \ conv_i10_to_i(((I10) >> 20) & 0x3ff), 1 ) -#define ATTRI10_4( A, I10 ) ATTR( A, 4, \ +#define ATTRI10_4( A, I10 ) ATTR( A, 4, GL_FLOAT, \ conv_i10_to_i((I10) & 0x3ff), \ conv_i10_to_i(((I10) >> 10) & 0x3ff), \ conv_i10_to_i(((I10) >> 20) & 0x3ff), \ conv_i2_to_i(((I10) >> 30) & 0x3)) -#define ATTRI10N_1( A, I10 ) ATTR( A, 1, conv_i10_to_norm_float((I10) & 0x3ff), 0, 0, 1 ) -#define ATTRI10N_2( A, I10 ) ATTR( A, 2, \ +#define ATTRI10N_1( A, I10 ) ATTR( A, 1, GL_FLOAT, conv_i10_to_norm_float((I10) & 0x3ff), 0, 0, 1 ) +#define ATTRI10N_2( A, I10 ) ATTR( A, 2, GL_FLOAT, \ conv_i10_to_norm_float((I10) & 0x3ff), \ conv_i10_to_norm_float(((I10) >> 10) & 0x3ff), 0, 1 ) -#define ATTRI10N_3( A, I10 ) ATTR( A, 3, \ +#define ATTRI10N_3( A, I10 ) ATTR( A, 3, GL_FLOAT, \ conv_i10_to_norm_float((I10) & 0x3ff), \ conv_i10_to_norm_float(((I10) >> 10) & 0x3ff), \ conv_i10_to_norm_float(((I10) >> 20) & 0x3ff), 1 ) -#define ATTRI10N_4( A, I10 ) ATTR( A, 4, \ +#define ATTRI10N_4( A, I10 ) ATTR( A, 4, GL_FLOAT, \ conv_i10_to_norm_float((I10) & 0x3ff), \ conv_i10_to_norm_float(((I10) >> 10) & 0x3ff), \ conv_i10_to_norm_float(((I10) >> 20) & 0x3ff), \ diff --git a/src/mesa/vbo/vbo_context.h b/src/mesa/vbo/vbo_context.h index 1c49de0..6fcd47f 100644 --- a/src/mesa/vbo/vbo_context.h +++ b/src/mesa/vbo/vbo_context.h @@ -151,4 +151,47 @@ vbo_draw_method(struct vbo_context *vbo, enum draw_method method) } } +/** + * Return if format is integer. The immediate mode commands only emit floats + * for non-integer types, thus everything else is integer. + */ +static inline GLboolean +vbo_attrtype_to_integer_flag(GLenum format) +{ + switch (format) { + case GL_FLOAT: + return GL_FALSE; + case GL_INT: + case GL_UNSIGNED_INT: + return GL_TRUE; + default: + ASSERT(0); + return GL_FALSE; + } +} + + +/** + * Return default component values for the given format. + * The return type is an array of floats, because that's how we declare + * the vertex storage despite the fact we sometimes store integers in there. + */ +static inline const GLfloat * +vbo_get_default_vals_as_float(GLenum format) +{ + static const GLfloat default_float[4] = { 0, 0, 0, 1 }; + static const GLint default_int[4] = { 0, 0, 0, 1 }; + + switch (format) { + case GL_FLOAT: + return default_float; + case GL_INT: + case GL_UNSIGNED_INT: + return (const GLfloat*)default_int; + default: + ASSERT(0); + return NULL; + } +} + #endif diff --git a/src/mesa/vbo/vbo_exec.h b/src/mesa/vbo/vbo_exec.h index 4ac7d16..34a6fe7 100644 --- a/src/mesa/vbo/vbo_exec.h +++ b/src/mesa/vbo/vbo_exec.h @@ -102,6 +102,7 @@ struct vbo_exec_context struct vbo_exec_copied_vtx copied; GLubyte attrsz[VBO_ATTRIB_MAX]; + GLenum attrtype[VBO_ATTRIB_MAX]; GLubyte active_sz[VBO_ATTRIB_MAX]; GLfloat *attrptr[VBO_ATTRIB_MAX]; diff --git a/src/mesa/vbo/vbo_exec_api.c b/src/mesa/vbo/vbo_exec_api.c index 781e360..64762c7 100644 --- a/src/mesa/vbo/vbo_exec_api.c +++ b/src/mesa/vbo/vbo_exec_api.c @@ -157,11 +157,13 @@ static void vbo_exec_copy_to_current( struct vbo_exec_context *exec ) GLfloat *current = (GLfloat *)vbo->currval[i].Ptr; GLfloat tmp[4]; - COPY_CLEAN_4V(tmp, - exec->vtx.attrsz[i], - exec->vtx.attrptr[i]); + COPY_CLEAN_4V_TYPE_AS_FLOAT(tmp, + exec->vtx.attrsz[i], + exec->vtx.attrptr[i], + exec->vtx.attrtype[i]); - if (memcmp(current, tmp, sizeof(tmp)) != 0) { + if (exec->vtx.attrtype[i] != vbo->currval[i].Type || + memcmp(current, tmp, sizeof(tmp)) != 0) { memcpy(current, tmp, sizeof(tmp)); /* Given that we explicitly state size here, there is no need @@ -170,8 +172,10 @@ static void vbo_exec_copy_to_current( struct vbo_exec_context *exec ) * directly. */ vbo->currval[i].Size = exec->vtx.attrsz[i]; - assert(vbo->currval[i].Type == GL_FLOAT); vbo->currval[i]._ElementSize = vbo->currval[i].Size * sizeof(GLfloat); + vbo->currval[i].Type = exec->vtx.attrtype[i]; + vbo->currval[i].Integer = + vbo_attrtype_to_integer_flag(exec->vtx.attrtype[i]); /* This triggers rather too much recalculation of Mesa state * that doesn't get used (eg light positions). @@ -324,7 +328,9 @@ vbo_exec_wrap_upgrade_vertex(struct vbo_exec_context *exec, if (j == attr) { if (oldSize) { GLfloat tmp[4]; - COPY_CLEAN_4V(tmp, oldSize, data + old_offset); + COPY_CLEAN_4V_TYPE_AS_FLOAT(tmp, oldSize, + data + old_offset, + exec->vtx.attrtype[j]); COPY_SZ_4V(dest + new_offset, newSize, tmp); } else { GLfloat *current = (GLfloat *)vbo->currval[j].Ptr; @@ -365,8 +371,9 @@ vbo_exec_fixup_vertex(struct gl_context *ctx, GLuint attr, GLuint newSize) vbo_exec_wrap_upgrade_vertex( exec, attr, newSize ); } else if (newSize < exec->vtx.active_sz[attr]) { - static const GLfloat id[4] = { 0, 0, 0, 1 }; GLuint i; + const GLfloat *id = + vbo_get_default_vals_as_float(exec->vtx.attrtype[attr]); /* New size is smaller - just need to fill in some * zeros. Don't need to flush or wrap. @@ -390,7 +397,7 @@ vbo_exec_fixup_vertex(struct gl_context *ctx, GLuint attr, GLuint newSize) * This macro is used to implement all the glVertex, glColor, glTexCoord, * glVertexAttrib, etc functions. */ -#define ATTR( A, N, V0, V1, V2, V3 ) \ +#define ATTR( A, N, T, V0, V1, V2, V3 ) \ do { \ struct vbo_exec_context *exec = &vbo_context(ctx)->exec; \ \ @@ -406,6 +413,7 @@ do { \ if (N>1) dest[1] = V1; \ if (N>2) dest[2] = V2; \ if (N>3) dest[3] = V3; \ + exec->vtx.attrtype[A] = T; \ } \ \ if ((A) == 0) { \ @@ -1192,6 +1200,8 @@ void vbo_exec_vtx_init( struct vbo_exec_context *exec ) for (i = 0 ; i < VBO_ATTRIB_MAX ; i++) { ASSERT(i < Elements(exec->vtx.attrsz)); exec->vtx.attrsz[i] = 0; + ASSERT(i < Elements(exec->vtx.attrtype)); + exec->vtx.attrtype[i] = GL_FLOAT; ASSERT(i < Elements(exec->vtx.active_sz)); exec->vtx.active_sz[i] = 0; } @@ -1328,6 +1338,7 @@ static void reset_attrfv( struct vbo_exec_context *exec ) for (i = 0 ; i < VBO_ATTRIB_MAX ; i++) { exec->vtx.attrsz[i] = 0; + exec->vtx.attrtype[i] = GL_FLOAT; exec->vtx.active_sz[i] = 0; } @@ -1382,7 +1393,7 @@ VertexAttrib4f_nopos(GLuint index, GLfloat x, GLfloat y, GLfloat z, GLfloat w) { GET_CURRENT_CONTEXT(ctx); if (index < MAX_VERTEX_GENERIC_ATTRIBS) - ATTR(VBO_ATTRIB_GENERIC0 + index, 4, x, y, z, w); + ATTR(VBO_ATTRIB_GENERIC0 + index, 4, GL_FLOAT, x, y, z, w); else ERROR(GL_INVALID_VALUE); } diff --git a/src/mesa/vbo/vbo_exec_draw.c b/src/mesa/vbo/vbo_exec_draw.c index 77db8ec..14d22ee 100644 --- a/src/mesa/vbo/vbo_exec_draw.c +++ b/src/mesa/vbo/vbo_exec_draw.c @@ -215,9 +215,6 @@ vbo_exec_bind_arrays( struct gl_context *ctx ) assert(0); } - /* Make all active attributes (including edgeflag) available as - * arrays of floats. - */ for (attr = 0; attr < VERT_ATTRIB_MAX ; attr++) { const GLuint src = map[attr]; @@ -243,7 +240,9 @@ vbo_exec_bind_arrays( struct gl_context *ctx ) arrays[attr].Size = exec->vtx.attrsz[src]; arrays[attr].StrideB = exec->vtx.vertex_size * sizeof(GLfloat); arrays[attr].Stride = exec->vtx.vertex_size * sizeof(GLfloat); - arrays[attr].Type = GL_FLOAT; + arrays[attr].Type = exec->vtx.attrtype[src]; + arrays[attr].Integer = + vbo_attrtype_to_integer_flag(exec->vtx.attrtype[src]); arrays[attr].Format = GL_RGBA; arrays[attr].Enabled = 1; arrays[attr]._ElementSize = arrays[attr].Size * sizeof(GLfloat); diff --git a/src/mesa/vbo/vbo_save.h b/src/mesa/vbo/vbo_save.h index 0b4d563..9aadde4 100644 --- a/src/mesa/vbo/vbo_save.h +++ b/src/mesa/vbo/vbo_save.h @@ -63,6 +63,7 @@ struct vbo_save_copied_vtx { */ struct vbo_save_vertex_list { GLubyte attrsz[VBO_ATTRIB_MAX]; + GLenum attrtype[VBO_ATTRIB_MAX]; GLuint vertex_size; /* Copy of the final vertex from node->vertex_store->bufferobj. @@ -127,6 +128,7 @@ struct vbo_save_context { const struct gl_client_array *inputs[VBO_ATTRIB_MAX]; GLubyte attrsz[VBO_ATTRIB_MAX]; + GLenum attrtype[VBO_ATTRIB_MAX]; GLubyte active_sz[VBO_ATTRIB_MAX]; GLuint vertex_size; diff --git a/src/mesa/vbo/vbo_save_api.c b/src/mesa/vbo/vbo_save_api.c index a02a13d..1386d71 100644 --- a/src/mesa/vbo/vbo_save_api.c +++ b/src/mesa/vbo/vbo_save_api.c @@ -327,6 +327,7 @@ _save_compile_vertex_list(struct gl_context *ctx) /* Duplicate our template, increment refcounts to the storage structs: */ memcpy(node->attrsz, save->attrsz, sizeof(node->attrsz)); + memcpy(node->attrtype, save->attrtype, sizeof(node->attrtype)); node->vertex_size = save->vertex_size; node->buffer_offset = (save->buffer - save->vertex_store->buffer) * sizeof(GLfloat); @@ -513,7 +514,8 @@ _save_copy_to_current(struct gl_context *ctx) for (i = VBO_ATTRIB_POS + 1; i < VBO_ATTRIB_MAX; i++) { if (save->attrsz[i]) { save->currentsz[i][0] = save->attrsz[i]; - COPY_CLEAN_4V(save->current[i], save->attrsz[i], save->attrptr[i]); + COPY_CLEAN_4V_TYPE_AS_FLOAT(save->current[i], save->attrsz[i], + save->attrptr[i], save->attrtype[i]); } } } @@ -615,7 +617,8 @@ _save_upgrade_vertex(struct gl_context *ctx, GLuint attr, GLuint newsz) if (save->attrsz[j]) { if (j == attr) { if (oldsz) { - COPY_CLEAN_4V(dest, oldsz, data); + COPY_CLEAN_4V_TYPE_AS_FLOAT(dest, oldsz, data, + save->attrtype[j]); data += oldsz; dest += newsz; } @@ -652,8 +655,8 @@ save_fixup_vertex(struct gl_context *ctx, GLuint attr, GLuint sz) _save_upgrade_vertex(ctx, attr, sz); } else if (sz < save->active_sz[attr]) { - static GLfloat id[4] = { 0, 0, 0, 1 }; GLuint i; + const GLfloat *id = vbo_get_default_vals_as_float(save->attrtype[attr]); /* New size is equal or smaller - just need to fill in some * zeros. @@ -691,7 +694,7 @@ _save_reset_vertex(struct gl_context *ctx) * 3f version won't otherwise set color[3] to 1.0 -- this is the job * of the chooser function when switching between Color4f and Color3f. */ -#define ATTR(A, N, V0, V1, V2, V3) \ +#define ATTR(A, N, T, V0, V1, V2, V3) \ do { \ struct vbo_save_context *save = &vbo_context(ctx)->save; \ \ @@ -704,6 +707,7 @@ do { \ if (N>1) dest[1] = V1; \ if (N>2) dest[2] = V2; \ if (N>3) dest[3] = V3; \ + save->attrtype[A] = T; \ } \ \ if ((A) == 0) { \ diff --git a/src/mesa/vbo/vbo_save_draw.c b/src/mesa/vbo/vbo_save_draw.c index c6425ab..8e70dbd 100644 --- a/src/mesa/vbo/vbo_save_draw.c +++ b/src/mesa/vbo/vbo_save_draw.c @@ -82,16 +82,20 @@ _playback_copy_to_current(struct gl_context *ctx, GLfloat *current = (GLfloat *)vbo->currval[i].Ptr; GLfloat tmp[4]; - COPY_CLEAN_4V(tmp, - node->attrsz[i], - data); + COPY_CLEAN_4V_TYPE_AS_FLOAT(tmp, + node->attrsz[i], + data, + node->attrtype[i]); - if (memcmp(current, tmp, 4 * sizeof(GLfloat)) != 0) { + if (node->attrtype[i] != vbo->currval[i].Type || + memcmp(current, tmp, 4 * sizeof(GLfloat)) != 0) { memcpy(current, tmp, 4 * sizeof(GLfloat)); vbo->currval[i].Size = node->attrsz[i]; - assert(vbo->currval[i].Type == GL_FLOAT); vbo->currval[i]._ElementSize = vbo->currval[i].Size * sizeof(GLfloat); + vbo->currval[i].Type = node->attrtype[i]; + vbo->currval[i].Integer = + vbo_attrtype_to_integer_flag(node->attrtype[i]); if (i >= VBO_ATTRIB_FIRST_MATERIAL && i <= VBO_ATTRIB_LAST_MATERIAL) @@ -137,9 +141,11 @@ static void vbo_bind_vertex_list(struct gl_context *ctx, const GLuint *map; GLuint attr; GLubyte node_attrsz[VBO_ATTRIB_MAX]; /* copy of node->attrsz[] */ + GLenum node_attrtype[VBO_ATTRIB_MAX]; /* copy of node->attrtype[] */ GLbitfield64 varying_inputs = 0x0; memcpy(node_attrsz, node->attrsz, sizeof(node->attrsz)); + memcpy(node_attrtype, node->attrtype, sizeof(node->attrtype)); /* Install the default (ie Current) attributes first, then overlay * all active ones. @@ -178,6 +184,7 @@ static void vbo_bind_vertex_list(struct gl_context *ctx, (ctx->VertexProgram._Current->Base.InputsRead & VERT_BIT_GENERIC0)) { save->inputs[VERT_ATTRIB_GENERIC0] = save->inputs[0]; node_attrsz[VERT_ATTRIB_GENERIC0] = node_attrsz[0]; + node_attrtype[VERT_ATTRIB_GENERIC0] = node_attrtype[0]; node_attrsz[0] = 0; } break; @@ -196,7 +203,9 @@ static void vbo_bind_vertex_list(struct gl_context *ctx, arrays[attr].Size = node_attrsz[src]; arrays[attr].StrideB = node->vertex_size * sizeof(GLfloat); arrays[attr].Stride = node->vertex_size * sizeof(GLfloat); - arrays[attr].Type = GL_FLOAT; + arrays[attr].Type = node_attrtype[src]; + arrays[attr].Integer = + vbo_attrtype_to_integer_flag(node_attrtype[src]); arrays[attr].Format = GL_RGBA; arrays[attr].Enabled = 1; arrays[attr]._ElementSize = arrays[attr].Size * sizeof(GLfloat);