Re: [Mesa-dev] [PATCH] glxinfo: add support for creating/querying core-profile contexts (v2.1)
On Tue, Jan 15, 2013 at 7:29 PM, Mike Lothian wrote: > Silly question but what does: > > OpenGL context flags: (none) > OpenGL core profile context flags: (none) > > Actually mean? Check out context_flags_string() in the patch. The current options are forward-compatible and robust-access. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [Bug 59383] src/glsl/tests/Makefile.am missing $(top_srcdir)/include
https://bugs.freedesktop.org/show_bug.cgi?id=59383 Andreas Boll changed: What|Removed |Added Summary|src/glsl/tests/Makefile.am |src/glsl/tests/Makefile.am |missing |missing |$(top_builddir)/include |$(top_srcdir)/include -- You are receiving this mail because: You are the assignee for the bug. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] (9.0) tests: AM_CPPFLAGS must include $(top_srcdir) instead of $(top_builddir).
From: Johannes Obermayr Reviewed-by: Andreas Boll (cherry picked from commit ebcabb88cfd19b35cd1b0a327fa055fff0c1b2be) Additionally it fixes the missing $(top_srcdir)/include in src/glsl/tests/Makefile.am Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=59383 --- src/glsl/tests/Makefile.am |9 + src/glx/tests/Makefile.am | 10 +- src/mapi/shared-glapi/tests/Makefile.am |6 +++--- 3 files changed, 13 insertions(+), 12 deletions(-) diff --git a/src/glsl/tests/Makefile.am b/src/glsl/tests/Makefile.am index 957bb55..eaf5a3c 100644 --- a/src/glsl/tests/Makefile.am +++ b/src/glsl/tests/Makefile.am @@ -1,8 +1,9 @@ AM_CPPFLAGS = \ - -I$(top_builddir)/src/gtest/include \ - -I$(top_builddir)/src/mesa \ - -I$(top_builddir)/src/mapi \ - -I$(top_builddir)/src/glsl + -I$(top_srcdir)/src/gtest/include \ + -I$(top_srcdir)/src/mesa \ + -I$(top_srcdir)/src/mapi \ + -I$(top_srcdir)/src/glsl \ + -I$(top_srcdir)/include TESTS_ENVIRONMENT= \ export PYTHON2=$(PYTHON2); \ diff --git a/src/glx/tests/Makefile.am b/src/glx/tests/Makefile.am index 83626fd..d940712 100644 --- a/src/glx/tests/Makefile.am +++ b/src/glx/tests/Makefile.am @@ -1,11 +1,11 @@ if HAVE_SHARED_GLAPI AM_CFLAGS = $(PTHREAD_CFLAGS) AM_CPPFLAGS = \ - -I$(top_builddir)/src/gtest/include \ - -I$(top_builddir)/src/mapi \ - -I$(top_builddir)/src/mesa \ - -I$(top_builddir)/src/glx \ - -I$(top_builddir)/include \ + -I$(top_srcdir)/src/gtest/include \ + -I$(top_srcdir)/src/mapi \ + -I$(top_srcdir)/src/mesa \ + -I$(top_srcdir)/src/glx \ + -I$(top_srcdir)/include \ $(X11_CFLAGS) TESTS = glx-test diff --git a/src/mapi/shared-glapi/tests/Makefile.am b/src/mapi/shared-glapi/tests/Makefile.am index 23d0208..98065fc 100644 --- a/src/mapi/shared-glapi/tests/Makefile.am +++ b/src/mapi/shared-glapi/tests/Makefile.am @@ -1,9 +1,9 @@ if HAVE_SHARED_GLAPI AM_CFLAGS = $(PTHREAD_CFLAGS) AM_CPPFLAGS = \ - -I$(top_builddir)/src/gtest/include \ - -I$(top_builddir)/src/mapi \ - -I$(top_builddir)/include + -I$(top_srcdir)/src/gtest/include \ + -I$(top_srcdir)/src/mapi \ + -I$(top_srcdir)/include TESTS = shared-glapi-test check_PROGRAMS = shared-glapi-test -- 1.7.4.1 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 1/3] R600/SI: Derive all sample intrinsics from a single class.
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
lib/Target/R600/SIIntrinsics.td |8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
index c322fef..0d1064e 100644
--- a/lib/Target/R600/SIIntrinsics.td
+++ b/lib/Target/R600/SIIntrinsics.td
@@ -22,9 +22,11 @@ let TargetPrefix = "SI", isTarget = 1 in {
def int_SI_vs_load_input : Intrinsic <[llvm_v4f32_ty], [llvm_v4i32_ty,
llvm_i16_ty, llvm_i32_ty], [IntrReadMem]> ;
def int_SI_wqm : Intrinsic <[], [], []>;
- def int_SI_sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
- def int_SI_sample_bias : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty,
llvm_v4f32_ty, llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
- def int_SI_sample_lod : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty,
llvm_v4f32_ty, llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+ class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+
+ def int_SI_sample : Sample;
+ def int_SI_sample_bias : Sample;
+ def int_SI_sample_lod : Sample;
/* Interpolation Intrinsics */
--
1.7.10.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 2/3] R600/SI: Take target parameter for sample intrinsics.
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
This is a prerequisite for patch 3 (and probably more similar changes in the
future). Unfortunately, it introduces a couple of (non-fatal) error messages
such as below during the R600 target build. Tom or anyone, any ideas for
silencing these?
anonymous.105: (intrinsic_w_chain:v4f32 2651:iPTR, (imm:i32):$writemask,
VReg_128:v4f32:$coord, SReg_256:v8i32:$rsrc, SReg_128:v4i32:$sampler,
(imm:i32):$textureTarget)
Included from /home/daenzer/src/llvm-git/llvm/lib/Target/R600/AMDGPU.td:40:
Included from
/home/daenzer/src/llvm-git/llvm/lib/Target/R600/AMDGPUInstructions.td:189:
Included from
/home/daenzer/src/llvm-git/llvm/lib/Target/R600/SIInstrInfo.td:591:
/home/daenzer/src/llvm-git/llvm/lib/Target/R600/SIInstructions.td:1188:1:
error: In anonymous.105: Pattern has dead named input: $textureTarget
def : Pat <
^
lib/Target/R600/SIInstructions.td |6 +++---
lib/Target/R600/SIIntrinsics.td |2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R600/SIInstructions.td
index cac42da..01c6ec1 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -1186,21 +1186,21 @@ def : Pat <
/* int_SI_sample */
def : Pat <
- (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm:$textureTarget),
(IMAGE_SAMPLE imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
/* int_SI_sample_lod */
def : Pat <
- (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm:$textureTarget),
(IMAGE_SAMPLE_L imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
/* int_SI_sample_bias */
def : Pat <
- (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm:$textureTarget),
(IMAGE_SAMPLE_B imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
index 0d1064e..4393144 100644
--- a/lib/Target/R600/SIIntrinsics.td
+++ b/lib/Target/R600/SIIntrinsics.td
@@ -22,7 +22,7 @@ let TargetPrefix = "SI", isTarget = 1 in {
def int_SI_vs_load_input : Intrinsic <[llvm_v4f32_ty], [llvm_v4i32_ty,
llvm_i16_ty, llvm_i32_ty], [IntrReadMem]> ;
def int_SI_wqm : Intrinsic <[], [], []>;
- class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+ class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty, llvm_i32_ty], [IntrReadMem]>;
def int_SI_sample : Sample;
def int_SI_sample_bias : Sample;
--
1.7.10.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 3/3] R600/SI: Use unnormalized coordinates for sampling with the RECT target.
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
lib/Target/R600/R600Instructions.td |7 +++
lib/Target/R600/SIInstructions.td |6 ++
2 files changed, 13 insertions(+)
diff --git a/lib/Target/R600/R600Instructions.td
b/lib/Target/R600/R600Instructions.td
index 64bab18..b15ea76 100644
--- a/lib/Target/R600/R600Instructions.td
+++ b/lib/Target/R600/R600Instructions.td
@@ -342,6 +342,13 @@ def TEX_SHADOW : PatLeaf<
}]
>;
+def TEX_RECT : PatLeaf<
+ (imm),
+ [{uint32_t TType = (uint32_t)N->getZExtValue();
+return TType == 5;
+ }]
+>;
+
class EG_CF_RAT cf_inst, bits <6> rat_inst, bits<4> rat_id, dag outs,
dag ins, string asm, list pattern> :
InstR600ISA {
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R600/SIInstructions.td
index 01c6ec1..74dfacc 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -1191,6 +1191,12 @@ def : Pat <
SReg_256:$rsrc, SReg_128:$sampler)
>;
+def : Pat <
+ (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, TEX_RECT:$textureTarget),
+ (IMAGE_SAMPLE imm:$writemask, 1, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
+SReg_256:$rsrc, SReg_128:$sampler)
+>;
+
/* int_SI_sample_lod */
def : Pat <
(int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm:$textureTarget),
--
1.7.10.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] radeonsi: Pass texture type to sampling intrinsics.
From: Michel Dänzer Signed-off-by: Michel Dänzer --- src/gallium/drivers/radeonsi/radeonsi_shader.c |6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/gallium/drivers/radeonsi/radeonsi_shader.c b/src/gallium/drivers/radeonsi/radeonsi_shader.c index 5dab46e..4843330 100644 --- a/src/gallium/drivers/radeonsi/radeonsi_shader.c +++ b/src/gallium/drivers/radeonsi/radeonsi_shader.c @@ -845,12 +845,10 @@ static void tex_fetch_args( ptr, offset); /* Dimensions */ - /* XXX: We might want to pass this information to the shader at some. */ -/* emit_data->args[4] = lp_build_const_int32(bld_base->base.gallivm, + emit_data->args[4] = lp_build_const_int32(bld_base->base.gallivm, emit_data->inst->Texture.Texture); -*/ - emit_data->arg_count = 4; + emit_data->arg_count = 5; /* XXX: To optimize, we could use a float or v2f32, if the last bits of * the writemask are clear */ emit_data->dst_type = LLVMVectorType( -- 1.7.10.4 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 3/6] radeonsi: Use proper hardware format for stencil texturing.
From: Michel Dänzer Fixes piglit 'spec/ARB_depth_buffer_float/fbo-clear-formats stencil' crash. Signed-off-by: Michel Dänzer --- src/gallium/drivers/radeonsi/si_state.c |3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_state.c b/src/gallium/drivers/radeonsi/si_state.c index 46112f7..fd046c6 100644 --- a/src/gallium/drivers/radeonsi/si_state.c +++ b/src/gallium/drivers/radeonsi/si_state.c @@ -1136,10 +1136,11 @@ static uint32_t si_translate_texformat(struct pipe_screen *screen, case PIPE_FORMAT_Z24X8_UNORM: case PIPE_FORMAT_Z24_UNORM_S8_UINT: return V_008F14_IMG_DATA_FORMAT_24_8; - case PIPE_FORMAT_S8X24_UINT: case PIPE_FORMAT_X8Z24_UNORM: case PIPE_FORMAT_S8_UINT_Z24_UNORM: return V_008F14_IMG_DATA_FORMAT_8_24; + case PIPE_FORMAT_X32_S8X24_UINT: + case PIPE_FORMAT_S8X24_UINT: case PIPE_FORMAT_S8_UINT: return V_008F14_IMG_DATA_FORMAT_8; case PIPE_FORMAT_Z32_FLOAT: -- 1.7.10.4 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 1/6] radeonsi: Consolidate calculation of tile mode index.
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/si_state.c | 182 ---
1 file changed, 71 insertions(+), 111 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state.c
b/src/gallium/drivers/radeonsi/si_state.c
index 2b574f3..46112f7 100644
--- a/src/gallium/drivers/radeonsi/si_state.c
+++ b/src/gallium/drivers/radeonsi/si_state.c
@@ -1496,6 +1496,64 @@ boolean si_is_format_supported(struct pipe_screen
*screen,
return retval == usage;
}
+static unsigned si_tile_mode_index(struct r600_resource_texture *rtex,
unsigned level)
+{
+ if (util_format_is_depth_or_stencil(rtex->real_format)) {
+ if (rtex->surface.level[level].mode == RADEON_SURF_MODE_1D) {
+ return 4;
+ } else if (rtex->surface.level[level].mode ==
RADEON_SURF_MODE_2D) {
+ switch (rtex->real_format) {
+ case PIPE_FORMAT_Z16_UNORM:
+ return 5;
+ case PIPE_FORMAT_Z24X8_UNORM:
+ case PIPE_FORMAT_Z24_UNORM_S8_UINT:
+ case PIPE_FORMAT_Z32_FLOAT:
+ case PIPE_FORMAT_Z32_FLOAT_S8X24_UINT:
+ return 6;
+ default:
+ return 7;
+ }
+ }
+ }
+
+ switch (rtex->surface.level[level].mode) {
+ default:
+ assert(!"Invalid surface mode");
+ /* Fall through */
+ case RADEON_SURF_MODE_LINEAR_ALIGNED:
+ return 8;
+ case RADEON_SURF_MODE_1D:
+ return 9;
+ case RADEON_SURF_MODE_2D:
+ if (rtex->surface.flags & RADEON_SURF_SCANOUT) {
+ switch (util_format_get_blocksize(rtex->real_format)) {
+ case 1:
+ return 10;
+ case 2:
+ return 11;
+ default:
+ assert(!"Invalid block size");
+ /* Fall through */
+ case 4:
+ return 12;
+ }
+ } else {
+ switch (util_format_get_blocksize(rtex->real_format)) {
+ case 1:
+ return 14;
+ case 2:
+ return 15;
+ case 4:
+ return 16;
+ case 8:
+ return 17;
+ default:
+ return 13;
+ }
+ }
+ }
+}
+
/*
* framebuffer handling
*/
@@ -1507,10 +1565,10 @@ static void si_cb(struct r600_context *rctx, struct
si_pm4_state *pm4,
struct r600_surface *surf;
unsigned level = state->cbufs[cb]->u.tex.level;
unsigned pitch, slice;
- unsigned color_info, color_attrib;
+ unsigned color_info;
+ unsigned tile_mode_index;
unsigned format, swap, ntype, endian;
uint64_t offset;
- unsigned blocksize;
const struct util_format_description *desc;
int i;
unsigned blend_clamp = 0, blend_bypass = 0;
@@ -1518,7 +1576,6 @@ static void si_cb(struct r600_context *rctx, struct
si_pm4_state *pm4,
surf = (struct r600_surface *)state->cbufs[cb];
rtex = (struct r600_resource_texture*)state->cbufs[cb]->texture;
- blocksize = util_format_get_blocksize(rtex->real_format);
if (rtex->depth)
rctx->have_depth_fb = TRUE;
@@ -1539,46 +1596,7 @@ static void si_cb(struct r600_context *rctx, struct
si_pm4_state *pm4,
slice = slice - 1;
}
- color_attrib = S_028C74_TILE_MODE_INDEX(8);
- switch (rtex->surface.level[level].mode) {
- case RADEON_SURF_MODE_LINEAR_ALIGNED:
- color_attrib = S_028C74_TILE_MODE_INDEX(8);
- break;
- case RADEON_SURF_MODE_1D:
- color_attrib = S_028C74_TILE_MODE_INDEX(9);
- break;
- case RADEON_SURF_MODE_2D:
- if (rtex->resource.b.b.bind & PIPE_BIND_SCANOUT) {
- switch (blocksize) {
- case 1:
- color_attrib = S_028C74_TILE_MODE_INDEX(10);
- break;
- case 2:
- color_attrib = S_028C74_TILE_MODE_INDEX(11);
- break;
- case 4:
- color_attrib = S_028C74_TILE_MODE_INDEX(12);
- break;
- }
- break;
- } else switch (blocksize) {
- case 1:
-
[Mesa-dev] [PATCH 2/6] radeonsi: Enable tiling for depth/stencil resources.
From: Michel Dänzer
Enabling it for all resources still seems to cause problems, but depth/stencil
buffers are always accessed with tiling by the DB block.
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/r600_texture.c |5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/r600_texture.c
b/src/gallium/drivers/radeonsi/r600_texture.c
index de46640..6ea6220 100644
--- a/src/gallium/drivers/radeonsi/r600_texture.c
+++ b/src/gallium/drivers/radeonsi/r600_texture.c
@@ -521,14 +521,13 @@ struct pipe_resource *si_texture_create(struct
pipe_screen *screen,
unsigned array_mode = V_009910_ARRAY_LINEAR_ALIGNED;
int r;
-#if 0
if (!(templ->flags & R600_RESOURCE_FLAG_TRANSFER) &&
- !(templ->bind & PIPE_BIND_SCANOUT)) {
+ !(templ->bind & PIPE_BIND_SCANOUT) &&
+ util_format_is_depth_or_stencil(templ->format)) {
if (permit_hardware_blit(screen, templ)) {
array_mode = V_009910_ARRAY_2D_TILED_THIN1;
}
}
-#endif
r = r600_init_surface(rscreen, &surface, templ, array_mode,
templ->flags & R600_RESOURCE_FLAG_FLUSHED_DEPTH);
--
1.7.10.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 4/6] radeonsi: Set SPI_SHADER_COL_FORMAT to what the pixel shader actually exports.
From: Michel Dänzer
Instead of deriving it from the colour buffer formats only.
Fixes a number of piglit tests which export depth from the pixel shader.
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/radeonsi_pipe.h |1 -
src/gallium/drivers/radeonsi/radeonsi_shader.c | 10 ++
src/gallium/drivers/radeonsi/radeonsi_shader.h |1 +
src/gallium/drivers/radeonsi/si_state.c|7 +--
src/gallium/drivers/radeonsi/si_state_draw.c |2 ++
5 files changed, 14 insertions(+), 7 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/radeonsi_pipe.h
b/src/gallium/drivers/radeonsi/radeonsi_pipe.h
index 29e8960..686ed05 100644
--- a/src/gallium/drivers/radeonsi/radeonsi_pipe.h
+++ b/src/gallium/drivers/radeonsi/radeonsi_pipe.h
@@ -134,7 +134,6 @@ struct r600_context {
/* shader information */
unsignedsprite_coord_enable;
unsignedexport_16bpc;
- unsignedspi_shader_col_format;
struct r600_textures_info vs_samplers;
struct r600_textures_info ps_samplers;
struct si_resource *border_color_table;
diff --git a/src/gallium/drivers/radeonsi/radeonsi_shader.c
b/src/gallium/drivers/radeonsi/radeonsi_shader.c
index 4843330..52ecef9 100644
--- a/src/gallium/drivers/radeonsi/radeonsi_shader.c
+++ b/src/gallium/drivers/radeonsi/radeonsi_shader.c
@@ -478,6 +478,13 @@ static void si_llvm_init_export_args(struct
lp_build_tgsi_context *bld_base,
if (cbuf >= 0 && cbuf < 8) {
struct r600_context *rctx = si_shader_ctx->rctx;
compressed = (si_shader_ctx->key.export_16bpc >> cbuf)
& 0x1;
+
+ if (compressed)
+ si_shader_ctx->shader->spi_shader_col_format |=
+ V_028714_SPI_SHADER_FP16_ABGR << (4 *
cbuf);
+ else
+ si_shader_ctx->shader->spi_shader_col_format |=
+ V_028714_SPI_SHADER_32_ABGR << (4 *
cbuf);
}
}
@@ -759,6 +766,9 @@ static void si_llvm_emit_epilogue(struct
lp_build_tgsi_context * bld_base)
last_args[6]= uint->zero;
last_args[7]= uint->zero;
last_args[8]= uint->zero;
+
+ si_shader_ctx->shader->spi_shader_col_format |=
+ V_028714_SPI_SHADER_32_ABGR;
}
/* Specify whether the EXEC mask represents the valid mask */
diff --git a/src/gallium/drivers/radeonsi/radeonsi_shader.h
b/src/gallium/drivers/radeonsi/radeonsi_shader.h
index 6f12e76..07b2f9f 100644
--- a/src/gallium/drivers/radeonsi/radeonsi_shader.h
+++ b/src/gallium/drivers/radeonsi/radeonsi_shader.h
@@ -94,6 +94,7 @@ struct si_pipe_shader {
unsignednum_sgprs;
unsignednum_vgprs;
unsignedspi_ps_input_ena;
+ unsignedspi_shader_col_format;
unsignedsprite_coord_enable;
unsignedso_strides[4];
struct si_shader_keykey;
diff --git a/src/gallium/drivers/radeonsi/si_state.c
b/src/gallium/drivers/radeonsi/si_state.c
index fd046c6..2ec8287 100644
--- a/src/gallium/drivers/radeonsi/si_state.c
+++ b/src/gallium/drivers/radeonsi/si_state.c
@@ -1685,9 +1685,7 @@ static void si_cb(struct r600_context *rctx, struct
si_pm4_state *pm4,
max_comp_size <= 10) ||
(ntype == V_028C70_NUMBER_FLOAT && max_comp_size <= 16)) {
rctx->export_16bpc |= 1 << cb;
- rctx->spi_shader_col_format |= V_028714_SPI_SHADER_FP16_ABGR <<
(4 * cb);
- } else
- rctx->spi_shader_col_format |= V_028714_SPI_SHADER_32_ABGR <<
(4 * cb);
+ }
}
static void si_db(struct r600_context *rctx, struct si_pm4_state *pm4,
@@ -1785,7 +1783,6 @@ static void si_set_framebuffer_state(struct pipe_context
*ctx,
/* build states */
rctx->have_depth_fb = 0;
rctx->export_16bpc = 0;
- rctx->spi_shader_col_format = 0;
for (int i = 0; i < state->nr_cbufs; i++) {
si_cb(rctx, pm4, state, i);
}
@@ -1815,8 +1812,6 @@ static void si_set_framebuffer_state(struct pipe_context
*ctx,
si_pm4_set_reg(pm4, R_028200_PA_SC_WINDOW_OFFSET, 0x);
si_pm4_set_reg(pm4, R_028230_PA_SC_EDGERULE, 0x);
si_pm4_set_reg(pm4, R_02823C_CB_SHADER_MASK, shader_mask);
- si_pm4_set_reg(pm4, R_028714_SPI_SHADER_COL_FORMAT,
- rctx->spi_shader_col_format);
si_pm4_set_reg(pm4, R_028BE0_PA_SC_AA_CONFIG, 0x);
si_pm4_set_state(rctx, framebuffer, pm4);
diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c
b/src/gallium/drivers/radeonsi/si_state_draw.c
index 257bf1e..6178
[Mesa-dev] [PATCH 5/6] radeonsi: Fix Z24 texture formats.
From: Michel Dänzer About half a dozen more piglits. Signed-off-by: Michel Dänzer --- src/gallium/drivers/radeonsi/si_state.c |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_state.c b/src/gallium/drivers/radeonsi/si_state.c index 2ec8287..cc9f7ae 100644 --- a/src/gallium/drivers/radeonsi/si_state.c +++ b/src/gallium/drivers/radeonsi/si_state.c @@ -1135,10 +1135,10 @@ static uint32_t si_translate_texformat(struct pipe_screen *screen, case PIPE_FORMAT_X24S8_UINT: case PIPE_FORMAT_Z24X8_UNORM: case PIPE_FORMAT_Z24_UNORM_S8_UINT: - return V_008F14_IMG_DATA_FORMAT_24_8; + return V_008F14_IMG_DATA_FORMAT_8_24; case PIPE_FORMAT_X8Z24_UNORM: case PIPE_FORMAT_S8_UINT_Z24_UNORM: - return V_008F14_IMG_DATA_FORMAT_8_24; + return V_008F14_IMG_DATA_FORMAT_24_8; case PIPE_FORMAT_X32_S8X24_UINT: case PIPE_FORMAT_S8X24_UINT: case PIPE_FORMAT_S8_UINT: -- 1.7.10.4 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 6/6] radeonsi: Actually keep track if we are using depth textures for samplers.
From: Michel Dänzer
20-odd more piglits.
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/si_state.c | 10 +++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state.c
b/src/gallium/drivers/radeonsi/si_state.c
index cc9f7ae..f7fe603 100644
--- a/src/gallium/drivers/radeonsi/si_state.c
+++ b/src/gallium/drivers/radeonsi/si_state.c
@@ -2197,7 +2197,8 @@ static struct si_pm4_state *si_set_sampler_view(struct
r600_context *rctx,
struct si_pipe_sampler_view **resource = (struct si_pipe_sampler_view
**)views;
struct si_pm4_state *pm4 = CALLOC_STRUCT(si_pm4_state);
int i, j;
- int has_depth = 0;
+
+ rctx->have_depth_texture = FALSE;
if (!count)
goto out;
@@ -2210,8 +2211,12 @@ static struct si_pm4_state *si_set_sampler_view(struct
r600_context *rctx,
(struct pipe_sampler_view **)&samplers->views[i],
views[i]);
- if (views[i])
+ if (resource[i]) {
+ struct r600_resource_texture *rtex =
+ (struct r600_resource_texture
*)views[i]->texture;
+ rctx->have_depth_texture |= rtex->depth &&
!rtex->is_flushing_texture;
si_pm4_add_bo(pm4, resource[i]->resource,
RADEON_USAGE_READ);
+ }
for (j = 0; j < Elements(resource[i]->state); ++j) {
si_pm4_sh_data_add(pm4, resource[i] ?
resource[i]->state[j] : 0);
@@ -2226,7 +2231,6 @@ static struct si_pm4_state *si_set_sampler_view(struct
r600_context *rctx,
si_pm4_sh_data_end(pm4, user_data_reg, SI_SGPR_RESOURCE);
out:
- rctx->have_depth_texture = has_depth;
rctx->ps_samplers.n_views = count;
return pm4;
}
--
1.7.10.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 1/3] R600/SI: Derive all sample intrinsics from a single class.
On 16.01.2013 10:42, Michel Dänzer wrote:
From: Michel Dänzer
Signed-off-by: Michel Dänzer
Reviewed-by: Christian König
---
lib/Target/R600/SIIntrinsics.td |8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
index c322fef..0d1064e 100644
--- a/lib/Target/R600/SIIntrinsics.td
+++ b/lib/Target/R600/SIIntrinsics.td
@@ -22,9 +22,11 @@ let TargetPrefix = "SI", isTarget = 1 in {
def int_SI_vs_load_input : Intrinsic <[llvm_v4f32_ty], [llvm_v4i32_ty,
llvm_i16_ty, llvm_i32_ty], [IntrReadMem]> ;
def int_SI_wqm : Intrinsic <[], [], []>;
- def int_SI_sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty, llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
- def int_SI_sample_bias : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty,
llvm_v4f32_ty, llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
- def int_SI_sample_lod : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty,
llvm_v4f32_ty, llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+ class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+
+ def int_SI_sample : Sample;
+ def int_SI_sample_bias : Sample;
+ def int_SI_sample_lod : Sample;
/* Interpolation Intrinsics */
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 2/3] R600/SI: Take target parameter for sample intrinsics.
On 16.01.2013 10:42, Michel Dänzer wrote:
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
This is a prerequisite for patch 3 (and probably more similar changes in the
future). Unfortunately, it introduces a couple of (non-fatal) error messages
such as below during the R600 target build. Tom or anyone, any ideas for
silencing these?
See below.
anonymous.105: (intrinsic_w_chain:v4f32 2651:iPTR, (imm:i32):$writemask,
VReg_128:v4f32:$coord, SReg_256:v8i32:$rsrc, SReg_128:v4i32:$sampler,
(imm:i32):$textureTarget)
Included from /home/daenzer/src/llvm-git/llvm/lib/Target/R600/AMDGPU.td:40:
Included from
/home/daenzer/src/llvm-git/llvm/lib/Target/R600/AMDGPUInstructions.td:189:
Included from
/home/daenzer/src/llvm-git/llvm/lib/Target/R600/SIInstrInfo.td:591:
/home/daenzer/src/llvm-git/llvm/lib/Target/R600/SIInstructions.td:1188:1:
error: In anonymous.105: Pattern has dead named input: $textureTarget
def : Pat <
^
lib/Target/R600/SIInstructions.td |6 +++---
lib/Target/R600/SIIntrinsics.td |2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R600/SIInstructions.td
index cac42da..01c6ec1 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -1186,21 +1186,21 @@ def : Pat <
/* int_SI_sample */
def : Pat <
- (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm:$textureTarget),
(IMAGE_SAMPLE imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
Not 100% sure, but try declaring it like this:
def : Pat <
- (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, 0),
(IMAGE_SAMPLE imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
;
And I won't commit it with those warning, that definitely looks like a good
source for bugs.
Christian.
/* int_SI_sample_lod */
def : Pat <
- (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm:$textureTarget),
(IMAGE_SAMPLE_L imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
/* int_SI_sample_bias */
def : Pat <
- (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm:$textureTarget),
(IMAGE_SAMPLE_B imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
index 0d1064e..4393144 100644
--- a/lib/Target/R600/SIIntrinsics.td
+++ b/lib/Target/R600/SIIntrinsics.td
@@ -22,7 +22,7 @@ let TargetPrefix = "SI", isTarget = 1 in {
def int_SI_vs_load_input : Intrinsic <[llvm_v4f32_ty], [llvm_v4i32_ty,
llvm_i16_ty, llvm_i32_ty], [IntrReadMem]> ;
def int_SI_wqm : Intrinsic <[], [], []>;
- class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty, llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+ class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty, llvm_i32_ty], [IntrReadMem]>;
def int_SI_sample : Sample;
def int_SI_sample_bias : Sample;
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 1/6] radeonsi: Consolidate calculation of tile mode index.
On 16.01.2013 11:55, Michel Dänzer wrote:
From: Michel Dänzer
Signed-off-by: Michel Dänzer
For this series: Reviewed-by: Christian König
---
src/gallium/drivers/radeonsi/si_state.c | 182 ---
1 file changed, 71 insertions(+), 111 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state.c
b/src/gallium/drivers/radeonsi/si_state.c
index 2b574f3..46112f7 100644
--- a/src/gallium/drivers/radeonsi/si_state.c
+++ b/src/gallium/drivers/radeonsi/si_state.c
@@ -1496,6 +1496,64 @@ boolean si_is_format_supported(struct pipe_screen
*screen,
return retval == usage;
}
+static unsigned si_tile_mode_index(struct r600_resource_texture *rtex, unsigned level)
+{
+ if (util_format_is_depth_or_stencil(rtex->real_format)) {
+ if (rtex->surface.level[level].mode == RADEON_SURF_MODE_1D) {
+ return 4;
+ } else if (rtex->surface.level[level].mode ==
RADEON_SURF_MODE_2D) {
+ switch (rtex->real_format) {
+ case PIPE_FORMAT_Z16_UNORM:
+ return 5;
+ case PIPE_FORMAT_Z24X8_UNORM:
+ case PIPE_FORMAT_Z24_UNORM_S8_UINT:
+ case PIPE_FORMAT_Z32_FLOAT:
+ case PIPE_FORMAT_Z32_FLOAT_S8X24_UINT:
+ return 6;
+ default:
+ return 7;
+ }
+ }
+ }
+
+ switch (rtex->surface.level[level].mode) {
+ default:
+ assert(!"Invalid surface mode");
+ /* Fall through */
+ case RADEON_SURF_MODE_LINEAR_ALIGNED:
+ return 8;
+ case RADEON_SURF_MODE_1D:
+ return 9;
+ case RADEON_SURF_MODE_2D:
+ if (rtex->surface.flags & RADEON_SURF_SCANOUT) {
+ switch (util_format_get_blocksize(rtex->real_format)) {
+ case 1:
+ return 10;
+ case 2:
+ return 11;
+ default:
+ assert(!"Invalid block size");
+ /* Fall through */
+ case 4:
+ return 12;
+ }
+ } else {
+ switch (util_format_get_blocksize(rtex->real_format)) {
+ case 1:
+ return 14;
+ case 2:
+ return 15;
+ case 4:
+ return 16;
+ case 8:
+ return 17;
+ default:
+ return 13;
+ }
+ }
+ }
+}
+
/*
* framebuffer handling
*/
@@ -1507,10 +1565,10 @@ static void si_cb(struct r600_context *rctx, struct
si_pm4_state *pm4,
struct r600_surface *surf;
unsigned level = state->cbufs[cb]->u.tex.level;
unsigned pitch, slice;
- unsigned color_info, color_attrib;
+ unsigned color_info;
+ unsigned tile_mode_index;
unsigned format, swap, ntype, endian;
uint64_t offset;
- unsigned blocksize;
const struct util_format_description *desc;
int i;
unsigned blend_clamp = 0, blend_bypass = 0;
@@ -1518,7 +1576,6 @@ static void si_cb(struct r600_context *rctx, struct
si_pm4_state *pm4,
surf = (struct r600_surface *)state->cbufs[cb];
rtex = (struct r600_resource_texture*)state->cbufs[cb]->texture;
- blocksize = util_format_get_blocksize(rtex->real_format);
if (rtex->depth)
rctx->have_depth_fb = TRUE;
@@ -1539,46 +1596,7 @@ static void si_cb(struct r600_context *rctx, struct
si_pm4_state *pm4,
slice = slice - 1;
}
- color_attrib = S_028C74_TILE_MODE_INDEX(8);
- switch (rtex->surface.level[level].mode) {
- case RADEON_SURF_MODE_LINEAR_ALIGNED:
- color_attrib = S_028C74_TILE_MODE_INDEX(8);
- break;
- case RADEON_SURF_MODE_1D:
- color_attrib = S_028C74_TILE_MODE_INDEX(9);
- break;
- case RADEON_SURF_MODE_2D:
- if (rtex->resource.b.b.bind & PIPE_BIND_SCANOUT) {
- switch (blocksize) {
- case 1:
- color_attrib = S_028C74_TILE_MODE_INDEX(10);
- break;
- case 2:
- color_attrib = S_028C74_TILE_MODE_INDEX(11);
- break;
- case 4:
- color_attrib = S_028C74_TILE_MODE_INDEX(12);
- break;
- }
- break
[Mesa-dev] [PATCH] radeonsi/vdpau: remove nonsense state tracker dep
From: Christian König Added with automake conversion, but makes no sense at all. Signed-off-by: Christian König --- src/gallium/targets/vdpau-radeonsi/Makefile.am |1 - 1 file changed, 1 deletion(-) diff --git a/src/gallium/targets/vdpau-radeonsi/Makefile.am b/src/gallium/targets/vdpau-radeonsi/Makefile.am index 98d828e..beae309 100644 --- a/src/gallium/targets/vdpau-radeonsi/Makefile.am +++ b/src/gallium/targets/vdpau-radeonsi/Makefile.am @@ -47,7 +47,6 @@ libvdpau_radeonsi_la_LDFLAGS = \ libvdpau_radeonsi_la_LIBADD = \ $(top_builddir)/src/gallium/auxiliary/libgallium.la \ $(top_builddir)/src/gallium/drivers/radeonsi/libradeonsi.la \ - $(top_builddir)/src/gallium/state_trackers/dri/drm/libdridrm.la \ $(top_builddir)/src/gallium/state_trackers/vdpau/libvdpautracker.la \ $(top_builddir)/src/gallium/winsys/radeon/drm/libradeonwinsys.la \ $(top_builddir)/src/gallium/drivers/trace/libtrace.la \ -- 1.7.9.5 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeonsi/vdpau: remove nonsense state tracker dep
2013/1/16 Christian König : > From: Christian König > > Added with automake conversion, but makes no sense at all. > > Signed-off-by: Christian König Reviewed-by: Andreas Boll > --- > src/gallium/targets/vdpau-radeonsi/Makefile.am |1 - > 1 file changed, 1 deletion(-) > > diff --git a/src/gallium/targets/vdpau-radeonsi/Makefile.am > b/src/gallium/targets/vdpau-radeonsi/Makefile.am > index 98d828e..beae309 100644 > --- a/src/gallium/targets/vdpau-radeonsi/Makefile.am > +++ b/src/gallium/targets/vdpau-radeonsi/Makefile.am > @@ -47,7 +47,6 @@ libvdpau_radeonsi_la_LDFLAGS = \ > libvdpau_radeonsi_la_LIBADD = \ > $(top_builddir)/src/gallium/auxiliary/libgallium.la \ > $(top_builddir)/src/gallium/drivers/radeonsi/libradeonsi.la \ > - $(top_builddir)/src/gallium/state_trackers/dri/drm/libdridrm.la \ > $(top_builddir)/src/gallium/state_trackers/vdpau/libvdpautracker.la \ > $(top_builddir)/src/gallium/winsys/radeon/drm/libradeonwinsys.la \ > $(top_builddir)/src/gallium/drivers/trace/libtrace.la \ > -- > 1.7.9.5 > > ___ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/mesa-dev ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH V4 5/6] intel: implement create image from texture
On Tuesday, January 15, 2013 02:05:00 PM Eric Anholt wrote:
> > +static void
> > +intel_image_set_level_info(__DRIimage *image, struct intel_mipmap_tree
> > *mt, + int level, int slice,
> > + uint32_t mask_x, uint32_t mask_y)
> > +{
> > + image->width = mt->level[level].width;
> > + image->height = mt->level[level].height;
> > + image->level_x = mt->level[level].level_x;
> > + image->level_y = mt->level[level].level_y;
> > + image->slice_x_offset = mt->level[level].slice[slice].x_offset;
> > + image->slice_y_offset = mt->level[level].slice[slice].y_offset;
> > +
> > + image->offset = intel_region_get_aligned_offset(mt->region,
> > + image->slice_x_offset
> > & ~mask_x, +
> > image->slice_y_offset & ~mask_y, +
> >false);
> > +}
>
> I think you end up double-counting the slice_x/y_offset offset here --
> you have a tile-aligned byte offset from the base of the region (which I
> don't think you want at all), and then you reference from
> slice_x/y_offset from that offset when texturing.
>
Aren't you supposed to reference the x/y offsets (used in SURFACE_STATE[5] in
gen6 and gen7) in addition to the surface base address in tiled surfaces when
resolving to higher miplevels? Or I am probably missing something?
I agree though that the level /slice differentation definitely needs to be
dropped as you mentioned earlier. I got an update that only accounts for the
x/y offsets to the image. It works correctly as well and I'll post it soon.
Abdiel
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 2/3 v2] R600/SI: Take target parameter for sample intrinsics.
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
v2: Leave the target parameter unnamed to prevent noisy if harmless
build error messages.
lib/Target/R600/SIInstructions.td |6 +++---
lib/Target/R600/SIIntrinsics.td |2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R600/SIInstructions.td
index cac42da..0f9d572 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -1186,21 +1186,21 @@ def : Pat <
/* int_SI_sample */
def : Pat <
- (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm),
(IMAGE_SAMPLE imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
/* int_SI_sample_lod */
def : Pat <
- (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm),
(IMAGE_SAMPLE_L imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
/* int_SI_sample_bias */
def : Pat <
- (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm),
(IMAGE_SAMPLE_B imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
index 0d1064e..4393144 100644
--- a/lib/Target/R600/SIIntrinsics.td
+++ b/lib/Target/R600/SIIntrinsics.td
@@ -22,7 +22,7 @@ let TargetPrefix = "SI", isTarget = 1 in {
def int_SI_vs_load_input : Intrinsic <[llvm_v4f32_ty], [llvm_v4i32_ty,
llvm_i16_ty, llvm_i32_ty], [IntrReadMem]> ;
def int_SI_wqm : Intrinsic <[], [], []>;
- class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+ class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty, llvm_i32_ty], [IntrReadMem]>;
def int_SI_sample : Sample;
def int_SI_sample_bias : Sample;
--
1.7.10.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 3/3 v2] R600/SI: Use unnormalized coordinates for sampling with the RECT target.
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
v2: Leave the target parameter unnamed to prevent noisy if harmless
build error messages.
lib/Target/R600/R600Instructions.td |7 +++
lib/Target/R600/SIInstructions.td |6 ++
2 files changed, 13 insertions(+)
diff --git a/lib/Target/R600/R600Instructions.td
b/lib/Target/R600/R600Instructions.td
index 64bab18..b15ea76 100644
--- a/lib/Target/R600/R600Instructions.td
+++ b/lib/Target/R600/R600Instructions.td
@@ -342,6 +342,13 @@ def TEX_SHADOW : PatLeaf<
}]
>;
+def TEX_RECT : PatLeaf<
+ (imm),
+ [{uint32_t TType = (uint32_t)N->getZExtValue();
+return TType == 5;
+ }]
+>;
+
class EG_CF_RAT cf_inst, bits <6> rat_inst, bits<4> rat_id, dag outs,
dag ins, string asm, list pattern> :
InstR600ISA {
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R600/SIInstructions.td
index 0f9d572..4164c55 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -1191,6 +1191,12 @@ def : Pat <
SReg_256:$rsrc, SReg_128:$sampler)
>;
+def : Pat <
+ (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, TEX_RECT),
+ (IMAGE_SAMPLE imm:$writemask, 1, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
+SReg_256:$rsrc, SReg_128:$sampler)
+>;
+
/* int_SI_sample_lod */
def : Pat <
(int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm),
--
1.7.10.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 2/3] R600/SI: Take target parameter for sample intrinsics.
On Mit, 2013-01-16 at 12:16 +0100, Christian König wrote: > On 16.01.2013 10:42, Michel Dänzer wrote: > > From: Michel Dänzer > > > > > > Signed-off-by: Michel Dänzer > > --- > > > > This is a prerequisite for patch 3 (and probably more similar changes in the > > future). Unfortunately, it introduces a couple of (non-fatal) error messages > > such as below during the R600 target build. Tom or anyone, any ideas for > > silencing these? > > See below. > > > > > anonymous.105: (intrinsic_w_chain:v4f32 2651:iPTR, > > (imm:i32):$writemask, VReg_128:v4f32:$coord, SReg_256:v8i32:$rsrc, > > SReg_128:v4i32:$sampler, (imm:i32):$textureTarget) > > Included from /home/daenzer/src/llvm-git/llvm/lib/Target/R600/AMDGPU.td:40: > > Included from > > /home/daenzer/src/llvm-git/llvm/lib/Target/R600/AMDGPUInstructions.td:189: > > Included from > > /home/daenzer/src/llvm-git/llvm/lib/Target/R600/SIInstrInfo.td:591: > > /home/daenzer/src/llvm-git/llvm/lib/Target/R600/SIInstructions.td:1188:1: > > error: In anonymous.105: Pattern has dead named input: $textureTarget > > def : Pat < > > ^ > > > > lib/Target/R600/SIInstructions.td |6 +++--- > > lib/Target/R600/SIIntrinsics.td |2 +- > > 2 files changed, 4 insertions(+), 4 deletions(-) > > > > diff --git a/lib/Target/R600/SIInstructions.td > > b/lib/Target/R600/SIInstructions.td > > index cac42da..01c6ec1 100644 > > --- a/lib/Target/R600/SIInstructions.td > > +++ b/lib/Target/R600/SIInstructions.td > > @@ -1186,21 +1186,21 @@ def : Pat < > > > > /* int_SI_sample */ > > def : Pat < > > - (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc, > > SReg_128:$sampler), > > + (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc, > > SReg_128:$sampler, imm:$textureTarget), > > (IMAGE_SAMPLE imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord, > > SReg_256:$rsrc, SReg_128:$sampler) > > >; > > Not 100% sure, but try declaring it like this: > > def : Pat < > - (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc, > SReg_128:$sampler), > + (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc, > SReg_128:$sampler, 0), > (IMAGE_SAMPLE imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord, > SReg_256:$rsrc, SReg_128:$sampler) > >; That results in failure to select the intrinsic at runtime. I figured out how to leave the parameter unnamed, which prevents the messages. See the v2 patches. Thanks anyway, -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Debian, X and DRI developer ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 2/3 v2] R600/SI: Take target parameter for sample intrinsics.
On 16.01.2013 15:27, Michel Dänzer wrote:
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
v2: Leave the target parameter unnamed to prevent noisy if harmless
build error messages.
For both patches:
Reviewed-by Christian König
lib/Target/R600/SIInstructions.td |6 +++---
lib/Target/R600/SIIntrinsics.td |2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R600/SIInstructions.td
index cac42da..0f9d572 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -1186,21 +1186,21 @@ def : Pat <
/* int_SI_sample */
def : Pat <
- (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm),
(IMAGE_SAMPLE imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
/* int_SI_sample_lod */
def : Pat <
- (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_lod imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm),
(IMAGE_SAMPLE_L imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
/* int_SI_sample_bias */
def : Pat <
- (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler),
+ (int_SI_sample_bias imm:$writemask, VReg_128:$coord, SReg_256:$rsrc,
SReg_128:$sampler, imm),
(IMAGE_SAMPLE_B imm:$writemask, 0, 0, 0, 0, 0, 0, 0, VReg_128:$coord,
SReg_256:$rsrc, SReg_128:$sampler)
>;
diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
index 0d1064e..4393144 100644
--- a/lib/Target/R600/SIIntrinsics.td
+++ b/lib/Target/R600/SIIntrinsics.td
@@ -22,7 +22,7 @@ let TargetPrefix = "SI", isTarget = 1 in {
def int_SI_vs_load_input : Intrinsic <[llvm_v4f32_ty], [llvm_v4i32_ty,
llvm_i16_ty, llvm_i32_ty], [IntrReadMem]> ;
def int_SI_wqm : Intrinsic <[], [], []>;
- class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty, llvm_v8i32_ty, llvm_v4i32_ty], [IntrReadMem]>;
+ class Sample : Intrinsic <[llvm_v4f32_ty], [llvm_i32_ty, llvm_v4f32_ty,
llvm_v8i32_ty, llvm_v4i32_ty, llvm_i32_ty], [IntrReadMem]>;
def int_SI_sample : Sample;
def int_SI_sample_bias : Sample;
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] allow Android to use out-of-tree mesa sources
From: Adrian Marius Negreanu this helps when compiling mesa for Android, but the mesa sources resides outside of the Android tree. Signed-off-by: Adrian Marius Negreanu --- src/mapi/mapi/mapi_abi.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/mapi/mapi/mapi_abi.py b/src/mapi/mapi/mapi_abi.py index 30ffe7b..7ae8c05 100644 --- a/src/mapi/mapi/mapi_abi.py +++ b/src/mapi/mapi/mapi_abi.py @@ -29,7 +29,7 @@ import sys # make it possible to import glapi import os -GLAPI = "./%s/../glapi/gen" % (os.path.dirname(sys.argv[0])) +GLAPI = "%s/../glapi/gen" % (os.path.dirname(sys.argv[0])) sys.path.append(GLAPI) import re -- 1.8.0 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] mesa: add missing ASSERT_OUTSIDE_BEGIN_END() in _mesa_GetInternalformativ()
---
src/mesa/main/formatquery.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/src/mesa/main/formatquery.c b/src/mesa/main/formatquery.c
index f08ab66..bd895e8 100644
--- a/src/mesa/main/formatquery.c
+++ b/src/mesa/main/formatquery.c
@@ -22,6 +22,7 @@
*/
#include "mtypes.h"
+#include "context.h"
#include "glformats.h"
#include "macros.h"
#include "mfeatures.h"
@@ -37,6 +38,8 @@ _mesa_GetInternalformativ(GLenum target, GLenum
internalformat, GLenum pname,
GLsizei count = 0;
GET_CURRENT_CONTEXT(ctx);
+ ASSERT_OUTSIDE_BEGIN_END(ctx);
+
if (!ctx->Extensions.ARB_internalformat_query) {
_mesa_error(ctx, GL_INVALID_OPERATION, "glGetInternalformativ");
return;
--
1.7.3.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 1/2] st/mesa: simplify some src/dst surface setup in BlitFramebuffer
Use the renderbuffer attachment pointers that we grabbed earlier. --- src/mesa/state_tracker/st_cb_blit.c |8 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/src/mesa/state_tracker/st_cb_blit.c b/src/mesa/state_tracker/st_cb_blit.c index 603c8d4..d6194bf 100644 --- a/src/mesa/state_tracker/st_cb_blit.c +++ b/src/mesa/state_tracker/st_cb_blit.c @@ -243,16 +243,16 @@ st_BlitFramebuffer(struct gl_context *ctx, &drawFB->Attachment[BUFFER_STENCIL]; struct st_renderbuffer *srcDepthRb = - st_renderbuffer(readFB->Attachment[BUFFER_DEPTH].Renderbuffer); + st_renderbuffer(srcDepth->Renderbuffer); struct st_renderbuffer *dstDepthRb = - st_renderbuffer(drawFB->Attachment[BUFFER_DEPTH].Renderbuffer); + st_renderbuffer(dstDepth->Renderbuffer); struct pipe_surface *dstDepthSurf = dstDepthRb ? dstDepthRb->surface : NULL; struct st_renderbuffer *srcStencilRb = - st_renderbuffer(readFB->Attachment[BUFFER_STENCIL].Renderbuffer); + st_renderbuffer(srcStencil->Renderbuffer); struct st_renderbuffer *dstStencilRb = - st_renderbuffer(drawFB->Attachment[BUFFER_STENCIL].Renderbuffer); + st_renderbuffer(dstStencil->Renderbuffer); struct pipe_surface *dstStencilSurf = dstStencilRb ? dstStencilRb->surface : NULL; -- 1.7.3.4 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 2/2] st/mesa: a couple fixes for st_BlitFramebuffer()
1. Loop over multiple destination color buffers. If we set
glDrawBuffers(GL_FRONT_AND_BACK) we need to loop over multiple color
buffers, blitting to each.
2. Add checks for null src/dst surface pointers. This fixes a crash
in the piglit fbo-missing-attachment-blit test.
See bug http://bugs.freedesktop.org/show_bug.cgi?id=59450
---
src/mesa/state_tracker/st_cb_blit.c | 85 +-
1 files changed, 52 insertions(+), 33 deletions(-)
diff --git a/src/mesa/state_tracker/st_cb_blit.c
b/src/mesa/state_tracker/st_cb_blit.c
index d6194bf..c284f06 100644
--- a/src/mesa/state_tracker/st_cb_blit.c
+++ b/src/mesa/state_tracker/st_cb_blit.c
@@ -184,48 +184,67 @@ st_BlitFramebuffer(struct gl_context *ctx,
blit.mask = PIPE_MASK_RGBA;
if (srcAtt->Type == GL_TEXTURE) {
- struct st_texture_object *srcObj =
-st_texture_object(srcAtt->Texture);
- struct st_renderbuffer *dstRb =
-st_renderbuffer(drawFB->_ColorDrawBuffers[0]);
- struct pipe_surface *dstSurf = dstRb->surface;
-
- assert(srcObj->pt);
- if (!srcObj->pt) {
+ struct st_texture_object *srcObj = st_texture_object(srcAtt->Texture);
+ GLuint i;
+
+ if (!srcObj || !srcObj->pt) {
return;
}
- blit.dst.resource = dstSurf->texture;
- blit.dst.level = dstSurf->u.tex.level;
- blit.dst.box.z = dstSurf->u.tex.first_layer;
- blit.dst.format = util_format_linear(dstSurf->format);
+ for (i = 0; i < drawFB->_NumColorDrawBuffers; i++) {
+struct st_renderbuffer *dstRb =
+ st_renderbuffer(drawFB->_ColorDrawBuffers[i]);
- blit.src.resource = srcObj->pt;
- blit.src.level = srcAtt->TextureLevel;
- blit.src.box.z = srcAtt->Zoffset + srcAtt->CubeMapFace;
- blit.src.format = util_format_linear(srcObj->pt->format);
+if (dstRb) {
+ struct pipe_surface *dstSurf = dstRb->surface;
- st->pipe->blit(st->pipe, &blit);
+ blit.dst.resource = dstSurf->texture;
+ blit.dst.level = dstSurf->u.tex.level;
+ blit.dst.box.z = dstSurf->u.tex.first_layer;
+ blit.dst.format = util_format_linear(dstSurf->format);
+
+ blit.src.resource = srcObj->pt;
+ blit.src.level = srcAtt->TextureLevel;
+ blit.src.box.z = srcAtt->Zoffset + srcAtt->CubeMapFace;
+ blit.src.format = util_format_linear(srcObj->pt->format);
+
+ st->pipe->blit(st->pipe, &blit);
+}
+ }
}
else {
struct st_renderbuffer *srcRb =
st_renderbuffer(readFB->_ColorReadBuffer);
- struct st_renderbuffer *dstRb =
-st_renderbuffer(drawFB->_ColorDrawBuffers[0]);
- struct pipe_surface *srcSurf = srcRb->surface;
- struct pipe_surface *dstSurf = dstRb->surface;
+ struct pipe_surface *srcSurf;
+ GLuint i;
+
+ if (!srcRb || !srcRb->surface)
+return;
- blit.dst.resource = dstSurf->texture;
- blit.dst.level = dstSurf->u.tex.level;
- blit.dst.box.z = dstSurf->u.tex.first_layer;
- blit.dst.format = util_format_linear(dstSurf->format);
+ srcSurf = srcRb->surface;
- blit.src.resource = srcSurf->texture;
- blit.src.level = srcSurf->u.tex.level;
- blit.src.box.z = srcSurf->u.tex.first_layer;
- blit.src.format = util_format_linear(srcSurf->format);
+ for (i = 0; i < drawFB->_NumColorDrawBuffers; i++) {
+struct st_renderbuffer *dstRb =
+ st_renderbuffer(drawFB->_ColorDrawBuffers[i]);
- st->pipe->blit(st->pipe, &blit);
+if (dstRb) {
+ struct pipe_surface *dstSurf = dstRb->surface;
+
+ if (dstSurf) {
+ blit.dst.resource = dstSurf->texture;
+ blit.dst.level = dstSurf->u.tex.level;
+ blit.dst.box.z = dstSurf->u.tex.first_layer;
+ blit.dst.format = util_format_linear(dstSurf->format);
+
+ blit.src.resource = srcSurf->texture;
+ blit.src.level = srcSurf->u.tex.level;
+ blit.src.box.z = srcSurf->u.tex.first_layer;
+ blit.src.format = util_format_linear(srcSurf->format);
+
+ st->pipe->blit(st->pipe, &blit);
+ }
+}
+ }
}
}
--
1.7.3.4
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 2/2] st/mesa: a couple fixes for st_BlitFramebuffer()
For the series: Reviewed-by: Marek Olšák
Marek
On Wed, Jan 16, 2013 at 4:58 PM, Brian Paul wrote:
> 1. Loop over multiple destination color buffers. If we set
> glDrawBuffers(GL_FRONT_AND_BACK) we need to loop over multiple color
> buffers, blitting to each.
>
> 2. Add checks for null src/dst surface pointers. This fixes a crash
> in the piglit fbo-missing-attachment-blit test.
> See bug http://bugs.freedesktop.org/show_bug.cgi?id=59450
> ---
> src/mesa/state_tracker/st_cb_blit.c | 85 +-
> 1 files changed, 52 insertions(+), 33 deletions(-)
>
> diff --git a/src/mesa/state_tracker/st_cb_blit.c
> b/src/mesa/state_tracker/st_cb_blit.c
> index d6194bf..c284f06 100644
> --- a/src/mesa/state_tracker/st_cb_blit.c
> +++ b/src/mesa/state_tracker/st_cb_blit.c
> @@ -184,48 +184,67 @@ st_BlitFramebuffer(struct gl_context *ctx,
>blit.mask = PIPE_MASK_RGBA;
>
>if (srcAtt->Type == GL_TEXTURE) {
> - struct st_texture_object *srcObj =
> -st_texture_object(srcAtt->Texture);
> - struct st_renderbuffer *dstRb =
> -st_renderbuffer(drawFB->_ColorDrawBuffers[0]);
> - struct pipe_surface *dstSurf = dstRb->surface;
> -
> - assert(srcObj->pt);
> - if (!srcObj->pt) {
> + struct st_texture_object *srcObj =
> st_texture_object(srcAtt->Texture);
> + GLuint i;
> +
> + if (!srcObj || !srcObj->pt) {
> return;
> }
>
> - blit.dst.resource = dstSurf->texture;
> - blit.dst.level = dstSurf->u.tex.level;
> - blit.dst.box.z = dstSurf->u.tex.first_layer;
> - blit.dst.format = util_format_linear(dstSurf->format);
> + for (i = 0; i < drawFB->_NumColorDrawBuffers; i++) {
> +struct st_renderbuffer *dstRb =
> + st_renderbuffer(drawFB->_ColorDrawBuffers[i]);
>
> - blit.src.resource = srcObj->pt;
> - blit.src.level = srcAtt->TextureLevel;
> - blit.src.box.z = srcAtt->Zoffset + srcAtt->CubeMapFace;
> - blit.src.format = util_format_linear(srcObj->pt->format);
> +if (dstRb) {
> + struct pipe_surface *dstSurf = dstRb->surface;
>
> - st->pipe->blit(st->pipe, &blit);
> + blit.dst.resource = dstSurf->texture;
> + blit.dst.level = dstSurf->u.tex.level;
> + blit.dst.box.z = dstSurf->u.tex.first_layer;
> + blit.dst.format = util_format_linear(dstSurf->format);
> +
> + blit.src.resource = srcObj->pt;
> + blit.src.level = srcAtt->TextureLevel;
> + blit.src.box.z = srcAtt->Zoffset + srcAtt->CubeMapFace;
> + blit.src.format = util_format_linear(srcObj->pt->format);
> +
> + st->pipe->blit(st->pipe, &blit);
> +}
> + }
>}
>else {
> struct st_renderbuffer *srcRb =
> st_renderbuffer(readFB->_ColorReadBuffer);
> - struct st_renderbuffer *dstRb =
> -st_renderbuffer(drawFB->_ColorDrawBuffers[0]);
> - struct pipe_surface *srcSurf = srcRb->surface;
> - struct pipe_surface *dstSurf = dstRb->surface;
> + struct pipe_surface *srcSurf;
> + GLuint i;
> +
> + if (!srcRb || !srcRb->surface)
> +return;
>
> - blit.dst.resource = dstSurf->texture;
> - blit.dst.level = dstSurf->u.tex.level;
> - blit.dst.box.z = dstSurf->u.tex.first_layer;
> - blit.dst.format = util_format_linear(dstSurf->format);
> + srcSurf = srcRb->surface;
>
> - blit.src.resource = srcSurf->texture;
> - blit.src.level = srcSurf->u.tex.level;
> - blit.src.box.z = srcSurf->u.tex.first_layer;
> - blit.src.format = util_format_linear(srcSurf->format);
> + for (i = 0; i < drawFB->_NumColorDrawBuffers; i++) {
> +struct st_renderbuffer *dstRb =
> + st_renderbuffer(drawFB->_ColorDrawBuffers[i]);
>
> - st->pipe->blit(st->pipe, &blit);
> +if (dstRb) {
> + struct pipe_surface *dstSurf = dstRb->surface;
> +
> + if (dstSurf) {
> + blit.dst.resource = dstSurf->texture;
> + blit.dst.level = dstSurf->u.tex.level;
> + blit.dst.box.z = dstSurf->u.tex.first_layer;
> + blit.dst.format = util_format_linear(dstSurf->format);
> +
> + blit.src.resource = srcSurf->texture;
> + blit.src.level = srcSurf->u.tex.level;
> + blit.src.box.z = srcSurf->u.tex.first_layer;
> + blit.src.format = util_format_linear(srcSurf->format);
> +
> + st->pipe->blit(st->pipe, &blit);
> + }
> +}
> + }
>}
> }
>
> --
> 1.7.3.4
>
> ___
> mesa-dev mailing list
> mesa-dev@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/
Re: [Mesa-dev] [PATCH] radeonsi/vdpau: remove nonsense state tracker dep
On Wed, Jan 16, 2013 at 6:18 AM, Christian König wrote: > From: Christian König > > Added with automake conversion, but makes no sense at all. Not according to git...? See e3b2160a1ffcfba0ff60cb47c17bd1a41afb63db. It looks like it was in the old Makefile too. In fact, it looks like libdridrm has been part of vdpau-radeonsi since /you/ added the Makefile in 41625afa :) ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] i965: Transcode RGB ETC1/2 textures to DXT1 or FXT1.
Ian Romanick writes: > On 01/11/2013 08:54 PM, Eric Anholt wrote: >> The previous solution of transcoding to ARGB gave exact results, but >> below the expected performance. Given that the user has gone out of >> their way to give us compressed data for performance reasons, try to >> respect their wishes by doing the best compression we can handle. We >> expect DXT1 compression to be better quality, but it may not always be >> available, in which case FXT1 seems like a reasonable fallback. >> >> This regresses OES_compressed_ETC1_RGB8_texture/miptree (which expects >> exact results, which we no longer provide), but it improves GLBenchmark >> 2.1 offscreen performance by 3.2% +/- 2.1% (n=11/10, throttling >> outliers removed). > > There are ETC1 and ETC2 conformance tests. Do they pass? My gles2conform has continued passing, but I haven't tried this on the gles3 branch with a gles3 conformance suite. pgpwBixXW0sRO.pgp Description: PGP signature ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [Bug 59364] Mesa build fails: clientattrib.c:33:22: fatal error: indirect.h: No such file or directory
https://bugs.freedesktop.org/show_bug.cgi?id=59364 --- Comment #1 from Tom Stellard --- Note that in order to reproduce this, you will have to run make clean or even make distclean, because the indirect.h file might be left over from a previous build. -- You are receiving this mail because: You are the assignee for the bug. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] i965: Lower textureGrad() with samplerCubeShadow on pre-Haswell.
Fixes regressions since commit 899017fc54c40c969b5239b33f3a17b311878b0d
Author: Kenneth Graunke
Date: Fri Jan 4 07:53:09 2013 -0800
i965: Use Haswell's sample_d_c for textureGrad with shadow samplers.
That patch assumed that all instances were lowered. However, we weren't
lowering textureGrad() with samplerCubeShadow because I couldn't figure
out the LOD calculations. It turns out they're easy: you just have to
use 1 for the depth. This causes it to pass oglconform's four tests.
Signed-off-by: Kenneth Graunke
Cc: Anuj Phogat
Cc: Ian Romanick
---
src/mesa/drivers/dri/i965/brw_lower_texture_gradients.cpp | 15 +--
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_lower_texture_gradients.cpp
b/src/mesa/drivers/dri/i965/brw_lower_texture_gradients.cpp
index dfbb983..9e661a1 100644
--- a/src/mesa/drivers/dri/i965/brw_lower_texture_gradients.cpp
+++ b/src/mesa/drivers/dri/i965/brw_lower_texture_gradients.cpp
@@ -27,6 +27,7 @@
#include "glsl/ir.h"
#include "glsl/ir_builder.h"
+#include "program/prog_instruction.h"
using namespace ir_builder;
@@ -89,10 +90,6 @@ lower_texture_grad_visitor::visit_leave(ir_texture *ir)
if (ir->op != ir_txd || !ir->shadow_comparitor)
return visit_continue;
- /* Cubes are broken. Avoid assertion failures when swizzling. */
- if (ir->sampler->type->sampler_dimensionality == GLSL_SAMPLER_DIM_CUBE)
- return visit_continue;
-
void *mem_ctx = ralloc_parent(ir);
const glsl_type *grad_type = ir->lod_info.grad.dPdx->type;
@@ -106,8 +103,14 @@ lower_texture_grad_visitor::visit_leave(ir_texture *ir)
txs->lod_info.lod = new(mem_ctx) ir_constant(0);
ir_variable *size =
new(mem_ctx) ir_variable(grad_type, "size", ir_var_temporary);
- emit(size, expr(ir_unop_i2f,
- swizzle_for_size(txs, grad_type->vector_elements)));
+ if (ir->sampler->type->sampler_dimensionality == GLSL_SAMPLER_DIM_CUBE) {
+ base_ir->insert_before(size);
+ base_ir->insert_before(assign(size, expr(ir_unop_i2f, txs),
WRITEMASK_XY));
+ base_ir->insert_before(assign(size, new(mem_ctx) ir_constant(1.0f),
WRITEMASK_Z));
+ } else {
+ emit(size, expr(ir_unop_i2f,
+ swizzle_for_size(txs, grad_type->vector_elements)));
+ }
/* Scale the gradients by width and height. Effectively, the incoming
* gradients are s'(x,y), t'(x,y), and r'(x,y) from equation 3.19 in the
--
1.8.1
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] i965: Transcode RGB ETC1/2 textures to DXT1 or FXT1.
On Tue, Jan 15, 2013 at 3:49 PM, Ian Romanick wrote: > On 01/11/2013 08:54 PM, Eric Anholt wrote: >> >> The previous solution of transcoding to ARGB gave exact results, but >> below the expected performance. Given that the user has gone out of >> their way to give us compressed data for performance reasons, try to >> respect their wishes by doing the best compression we can handle. We >> expect DXT1 compression to be better quality, but it may not always be >> available, in which case FXT1 seems like a reasonable fallback. >> >> This regresses OES_compressed_ETC1_RGB8_texture/miptree (which expects >> exact results, which we no longer provide), but it improves GLBenchmark >> 2.1 offscreen performance by 3.2% +/- 2.1% (n=11/10, throttling >> outliers removed). > > > There are ETC1 and ETC2 conformance tests. Do they pass? In es3conform it regresses compressed_etc1_rgb8_texture.test and gl_compressed_rgb8_etc2. The first fails because of slight inaccuracies, eg (0.71, 0.698, 0.776) instead of (0.698, 0.698, 0.765). The second looks like it finds some potentially bad cases, eg (0.0 0.0 0.0 255.0) instead of (0.0 255.0 0.0 255.0). ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] mesa: Allow TexImage3D on depth/stencil textures in ES 3.0.
ES 3.0 supports 2D array textures, which are populated via TexImage3D.
Fixes es3conform's shadow_execution_frag test.
Signed-off-by: Kenneth Graunke
---
src/mesa/main/glformats.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/mesa/main/glformats.c b/src/mesa/main/glformats.c
index 7969f77..9831ab4 100644
--- a/src/mesa/main/glformats.c
+++ b/src/mesa/main/glformats.c
@@ -2068,7 +2068,7 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_COMPONENT:
- if (dimensions != 2) {
+ if (dimensions == 1) {
return GL_INVALID_OPERATION;
}
switch (type) {
@@ -2098,7 +2098,7 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_STENCIL:
- if (dimensions != 2) {
+ if (dimensions == 1) {
return GL_INVALID_OPERATION;
}
switch (type) {
--
1.8.1
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] mesa: Allow TexImage3D on depth/stencil textures in ES 3.0.
On 01/16/2013 01:49 PM, Kenneth Graunke wrote:
ES 3.0 supports 2D array textures, which are populated via TexImage3D.
Fixes es3conform's shadow_execution_frag test.
Signed-off-by: Kenneth Graunke
---
src/mesa/main/glformats.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/mesa/main/glformats.c b/src/mesa/main/glformats.c
index 7969f77..9831ab4 100644
--- a/src/mesa/main/glformats.c
+++ b/src/mesa/main/glformats.c
@@ -2068,7 +2068,7 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_COMPONENT:
- if (dimensions != 2) {
+ if (dimensions == 1) {
These checks are suck. :( Won't this let you specify depth component
textures for GL_TEXTURE_3D textures too?
return GL_INVALID_OPERATION;
}
switch (type) {
@@ -2098,7 +2098,7 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_STENCIL:
- if (dimensions != 2) {
+ if (dimensions == 1) {
return GL_INVALID_OPERATION;
}
switch (type) {
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 00/10] Support UBOs with instance names pre-series
This is the "warm up" series before my series to make UBOs with instances names (and arrays of UBOs) work. These are both required features of GLSL ES 3.00 that are not added to desktop GLSL until 1.50. This series is just a bunch of minor fixes and small refactors. This series fixes 4 gles3conform tests. There are about a dozen more patches coming in the next few day to fix the remaining 14 UBO tests. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 02/10] glsl: Track UBO block names in the symbol table.
From: Kenneth Graunke
The GLSL 1.40 spec says:
"Uniform block names and variable names declared within uniform
blocks are scoped at the program level."
Track the block name in the symbol table and emit errors when conflicts
exist.
Fixes es3conform's uniform_buffer_object_block_name_conflict test, and
fixes the piglit block-name-clashes-with-{variable,function,struct}.vert
tests.
NOTE: This is a candidate for the 9.0 branch.
Reviewed-by: Ian Romanick
---
src/glsl/ast_to_hir.cpp| 6 ++
src/glsl/glsl_symbol_table.cpp | 14 +++---
src/glsl/glsl_symbol_table.h | 1 +
3 files changed, 18 insertions(+), 3 deletions(-)
diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp
index ea5e3b7..31dd51a 100644
--- a/src/glsl/ast_to_hir.cpp
+++ b/src/glsl/ast_to_hir.cpp
@@ -4132,6 +4132,12 @@ ast_uniform_block::hir(exec_list *instructions,
struct gl_uniform_block *ubo = get_next_uniform_block(state);
ubo->Name = ralloc_strdup(state->uniform_blocks, this->block_name);
+ if (!state->symbols->add_uniform_block(ubo)) {
+ YYLTYPE loc = this->get_location();
+ _mesa_glsl_error(&loc, state, "Uniform block name `%s' already taken in "
+ "the current scope.\n", ubo->Name);
+ }
+
unsigned int num_variables = 0;
foreach_list_typed(ast_declarator_list, decl_list, link, &declarations) {
foreach_list_const(node, &decl_list->declarations) {
diff --git a/src/glsl/glsl_symbol_table.cpp b/src/glsl/glsl_symbol_table.cpp
index f934ea8..6171a27 100644
--- a/src/glsl/glsl_symbol_table.cpp
+++ b/src/glsl/glsl_symbol_table.cpp
@@ -41,13 +41,15 @@ public:
ralloc_free(entry);
}
- symbol_table_entry(ir_variable *v) : v(v), f(0), t(0) {}
- symbol_table_entry(ir_function *f) : v(0), f(f), t(0) {}
- symbol_table_entry(const glsl_type *t) : v(0), f(0), t(t) {}
+ symbol_table_entry(ir_variable *v) : v(v), f(0), t(0), u(0) {}
+ symbol_table_entry(ir_function *f) : v(0), f(f), t(0), u(0) {}
+ symbol_table_entry(const glsl_type *t) : v(0), f(0), t(t), u(0) {}
+ symbol_table_entry(struct gl_uniform_block *u) : v(0), f(0), t(t), u(u) {}
ir_variable *v;
ir_function *f;
const glsl_type *t;
+ struct gl_uniform_block *u;
};
glsl_symbol_table::glsl_symbol_table()
@@ -132,6 +134,12 @@ bool glsl_symbol_table::add_function(ir_function *f)
return _mesa_symbol_table_add_symbol(table, -1, f->name, entry) == 0;
}
+bool glsl_symbol_table::add_uniform_block(struct gl_uniform_block *u)
+{
+ symbol_table_entry *entry = new(mem_ctx) symbol_table_entry(u);
+ return _mesa_symbol_table_add_symbol(table, -1, u->Name, entry) == 0;
+}
+
void glsl_symbol_table::add_global_function(ir_function *f)
{
symbol_table_entry *entry = new(mem_ctx) symbol_table_entry(f);
diff --git a/src/glsl/glsl_symbol_table.h b/src/glsl/glsl_symbol_table.h
index 9f56027..f95fb8a 100644
--- a/src/glsl/glsl_symbol_table.h
+++ b/src/glsl/glsl_symbol_table.h
@@ -99,6 +99,7 @@ public:
bool add_variable(ir_variable *v);
bool add_type(const char *name, const glsl_type *t);
bool add_function(ir_function *f);
+ bool add_uniform_block(struct gl_uniform_block *u);
/*@}*/
/**
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 01/10] glsl: Reject row_major and column_major on non-matrix types
From: Ian Romanick
About both row_major and column_major layout qualifiers, the GLSL spec
says:
"It only affects the layout of matrices."
However, the OpenGL ES 3.0 conformance tests have taken this to mean it
is an error use it elsewhere. This seems logical given that
'layout(row_major) vec4 foo' is probably not what the programmer meant.
The only catch is dealing with structures that contain matrices. Layout
qualifiers cannot be applied directly to fields of structures, so the
only way to affect the layout of the fields is to apply a qualifier to
the structure declaration itself. There is ongoing debate about this
within Khronos, and it seems to be settling in favor of allowing the
qualifiers on structures. I light of this, I have chosen to allow the
qualifiers on structures but emit a warning since the usage may not be
portable.
Fixes gles3conform test
uniform_buffer_object_layouts_not_for_matrix_type and causes no
regressions.
Signed-off-by: Ian Romanick
---
src/glsl/ast_to_hir.cpp | 39 +--
1 file changed, 33 insertions(+), 6 deletions(-)
diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp
index f934c8e..ea5e3b7 100644
--- a/src/glsl/ast_to_hir.cpp
+++ b/src/glsl/ast_to_hir.cpp
@@ -1933,6 +1933,31 @@ is_varying_var(ir_variable *var,
_mesa_glsl_parser_targets target)
}
+/**
+ * Matrix layout qualifiers are only allowed on certain types
+ */
+static void
+validate_matrix_layout_for_type(struct _mesa_glsl_parse_state *state,
+ YYLTYPE *loc,
+ const glsl_type *type)
+{
+ if (!type->is_matrix() && !type->is_record()) {
+ _mesa_glsl_error(loc, state,
+ "uniform block layout qualifiers row_major and "
+ "column_major can only be applied to matrix and "
+ "structure types");
+ } else if (type->is_record()) {
+ /* We allow 'layout(row_major)' on structure types because it's the only
+ * way to get row-major layouts on matrices contained in structures.
+ */
+ _mesa_glsl_warning(loc, state,
+ "uniform block layout qualifiers row_major and "
+ "column_major applied to structure types is not "
+ "strictly conformant and my be rejected by other "
+ "compilers");
+ }
+}
+
static void
apply_type_qualifier_to_variable(const struct ast_type_qualifier *qual,
ir_variable *var,
@@ -2251,12 +2276,14 @@ apply_type_qualifier_to_variable(const struct
ast_type_qualifier *qual,
"members");
}
- if (!ubo_qualifiers_valid &&
- (qual->flags.q.row_major || qual->flags.q.column_major)) {
- _mesa_glsl_error(loc, state,
- "uniform block layout qualifiers row_major and "
- "column_major can only be applied to uniform block "
- "members");
+ if (qual->flags.q.row_major || qual->flags.q.column_major) {
+ if (!ubo_qualifiers_valid) {
+_mesa_glsl_error(loc, state,
+ "uniform block layout qualifiers row_major and "
+ "column_major can only be applied to uniform block "
+ "members");
+ } else
+validate_matrix_layout_for_type(state, loc, var->type);
}
}
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 04/10] linker: Use helper variable sh
From: Ian Romanick
This looks like a copy-and-paste left over.
Signed-off-by: Ian Romanick
---
src/glsl/linker.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/glsl/linker.cpp b/src/glsl/linker.cpp
index 70f3d5b..63548e0 100644
--- a/src/glsl/linker.cpp
+++ b/src/glsl/linker.cpp
@@ -885,8 +885,8 @@ link_intrastage_shaders(void *mem_ctx,
for (unsigned i = 0; i < num_shaders; i++) {
struct gl_shader *sh = shader_list[i];
- for (unsigned j = 0; j < shader_list[i]->NumUniformBlocks; j++) {
-link_assign_uniform_block_offsets(shader_list[i]);
+ for (unsigned j = 0; j < sh->NumUniformBlocks; j++) {
+link_assign_uniform_block_offsets(sh);
int index = link_cross_validate_uniform_block(mem_ctx,
&uniform_blocks,
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 03/10] glsl: Remove stale comment
From: Ian Romanick Signed-off-by: Ian Romanick --- src/glsl/ir_clone.cpp | 4 1 file changed, 4 deletions(-) diff --git a/src/glsl/ir_clone.cpp b/src/glsl/ir_clone.cpp index 4314efa..c62c1fc 100644 --- a/src/glsl/ir_clone.cpp +++ b/src/glsl/ir_clone.cpp @@ -36,10 +36,6 @@ ir_rvalue::clone(void *mem_ctx, struct hash_table *ht) const /** * Duplicate an IR variable - * - * \note - * This will probably be made \c virtual and moved to the base class - * eventually. */ ir_variable * ir_variable::clone(void *mem_ctx, struct hash_table *ht) const -- 1.7.11.7 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 05/10] mesa: Remove unused field gl_uniform_buffer_variable::Buffer
From: Ian Romanick
Signed-off-by: Ian Romanick
---
src/glsl/ast_to_hir.cpp | 1 -
src/mesa/main/mtypes.h | 1 -
2 files changed, 2 deletions(-)
diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp
index 31dd51a..de3ce90 100644
--- a/src/glsl/ast_to_hir.cpp
+++ b/src/glsl/ast_to_hir.cpp
@@ -4166,7 +4166,6 @@ ast_uniform_block::hir(exec_list *instructions,
ubo_var->Name = ralloc_strdup(state->uniform_blocks, var->name);
ubo_var->Type = var->type;
-ubo_var->Buffer = ubo - state->uniform_blocks;
ubo_var->Offset = 0; /* Assigned at link time. */
if (var->type->is_matrix() ||
diff --git a/src/mesa/main/mtypes.h b/src/mesa/main/mtypes.h
index d0c0e24..e696d30 100644
--- a/src/mesa/main/mtypes.h
+++ b/src/mesa/main/mtypes.h
@@ -2274,7 +2274,6 @@ struct gl_uniform_buffer_variable
{
char *Name;
const struct glsl_type *Type;
- unsigned int Buffer;
unsigned int Offset;
GLboolean RowMajor;
};
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 06/10] glsl: Remove unused loc parameter from generate_call
From: Ian Romanick
Signed-off-by: Ian Romanick
---
src/glsl/ast_function.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/glsl/ast_function.cpp b/src/glsl/ast_function.cpp
index b56a3c7..dc7a58b 100644
--- a/src/glsl/ast_function.cpp
+++ b/src/glsl/ast_function.cpp
@@ -183,7 +183,7 @@ verify_parameter_modes(_mesa_glsl_parse_state *state,
*/
static ir_rvalue *
generate_call(exec_list *instructions, ir_function_signature *sig,
- YYLTYPE *loc, exec_list *actual_parameters,
+ exec_list *actual_parameters,
ir_call **call_ir,
struct _mesa_glsl_parse_state *state)
{
@@ -1501,7 +1501,7 @@ ast_function_expression::hir(exec_list *instructions,
/* an error has already been emitted */
value = ir_rvalue::error_value(ctx);
} else {
-value = generate_call(instructions, sig, &loc, &actual_parameters,
+value = generate_call(instructions, sig, &actual_parameters,
&call, state);
}
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 07/10] glsl: Eliminate link_update_uniform_buffer_variables return value
From: Ian Romanick
It always returns true, so there's no point in having a return value.
Signed-off-by: Ian Romanick
---
src/glsl/link_uniforms.cpp | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/src/glsl/link_uniforms.cpp b/src/glsl/link_uniforms.cpp
index aa8a8b3..07d9c18 100644
--- a/src/glsl/link_uniforms.cpp
+++ b/src/glsl/link_uniforms.cpp
@@ -452,7 +452,7 @@ link_cross_validate_uniform_block(void *mem_ctx,
* would point at the uniform block list in one of the pre-linked
* shaders).
*/
-static bool
+static void
link_update_uniform_buffer_variables(struct gl_shader *shader)
{
foreach_list(node, shader->ir) {
@@ -478,8 +478,6 @@ link_update_uniform_buffer_variables(struct gl_shader
*shader)
}
assert(found);
}
-
- return true;
}
void
@@ -544,8 +542,7 @@ link_assign_uniform_locations(struct gl_shader_program
*prog)
if (prog->_LinkedShaders[i] == NULL)
continue;
- if (!link_update_uniform_buffer_variables(prog->_LinkedShaders[i]))
-return;
+ link_update_uniform_buffer_variables(prog->_LinkedShaders[i]);
}
/* First pass: Count the uniform resources used by the user-defined
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 08/10] mesa: Refactor getting a uniform's name to a helper function
From: Ian Romanick
We currently have a bug in this code, and I don't want to fix it in two
places.
Signed-off-by: Ian Romanick
---
src/mesa/main/uniform_query.cpp | 2 +-
src/mesa/main/uniforms.c| 12 ++--
src/mesa/main/uniforms.h| 5 +
3 files changed, 16 insertions(+), 3 deletions(-)
diff --git a/src/mesa/main/uniform_query.cpp b/src/mesa/main/uniform_query.cpp
index efff8f3..ec08ed4 100644
--- a/src/mesa/main/uniform_query.cpp
+++ b/src/mesa/main/uniform_query.cpp
@@ -59,7 +59,7 @@ _mesa_GetActiveUniform(GLhandleARB program, GLuint index,
const struct gl_uniform_storage *const uni = &shProg->UniformStorage[index];
if (nameOut) {
- _mesa_copy_string(nameOut, maxLength, length, uni->name);
+ _mesa_get_uniform_name(uni, maxLength, length, nameOut);
}
if (size) {
diff --git a/src/mesa/main/uniforms.c b/src/mesa/main/uniforms.c
index 77b195e..edb4477 100644
--- a/src/mesa/main/uniforms.c
+++ b/src/mesa/main/uniforms.c
@@ -795,7 +795,15 @@ _mesa_GetActiveUniformName(GLuint program, GLuint
uniformIndex,
}
if (uniformName) {
- _mesa_copy_string(uniformName, bufSize, length,
- shProg->UniformStorage[uniformIndex].name);
+ _mesa_get_uniform_name(& shProg->UniformStorage[uniformIndex],
+ bufSize, length, uniformName);
}
}
+
+void
+_mesa_get_uniform_name(const struct gl_uniform_storage *uni,
+ GLsizei maxLength, GLsizei *length,
+ GLchar *nameOut)
+{
+ _mesa_copy_string(nameOut, maxLength, length, uni->name);
+}
diff --git a/src/mesa/main/uniforms.h b/src/mesa/main/uniforms.h
index 287d710..f175031 100644
--- a/src/mesa/main/uniforms.h
+++ b/src/mesa/main/uniforms.h
@@ -213,6 +213,11 @@ _mesa_sampler_uniforms_are_valid(const struct
gl_shader_program *shProg,
extern const struct gl_program_parameter *
get_uniform_parameter(struct gl_shader_program *shProg, GLint index);
+extern void
+_mesa_get_uniform_name(const struct gl_uniform_storage *uni,
+ GLsizei maxLength, GLsizei *length,
+ GLchar *nameOut);
+
struct gl_builtin_uniform_element {
const char *field;
int tokens[STATE_LENGTH];
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 09/10] mesa: Array uniform names are supposed to have [0] appended
From: Ian Romanick
This is required by OpenGL ES 3.0 and desktop OpenGL 4.2. Previous
version were ambiguous. This also matches the behavior of NVIDIA's
closed-source driver (version 304.64).
Fixed gles3conformance test uniform_buffer_object_getactiveuniform.
Several piglit tests expect glGetActiveUniform to *not* include the [0]
on the end. These tests were already failing on NVIDIA, and this change
regresses them on Mesa. Patches have been sent to the piglit mailing
list to fix the tests.
Signed-off-by: Ian Romanick
---
src/mesa/main/uniforms.c | 31 +++
1 file changed, 31 insertions(+)
diff --git a/src/mesa/main/uniforms.c b/src/mesa/main/uniforms.c
index edb4477..15248fa 100644
--- a/src/mesa/main/uniforms.c
+++ b/src/mesa/main/uniforms.c
@@ -805,5 +805,36 @@ _mesa_get_uniform_name(const struct gl_uniform_storage
*uni,
GLsizei maxLength, GLsizei *length,
GLchar *nameOut)
{
+ GLsizei localLength;
+
+ if (length == NULL)
+ length = &localLength;
+
_mesa_copy_string(nameOut, maxLength, length, uni->name);
+
+ /* Page 61 (page 73 of the PDF) in section 2.11 of the OpenGL ES 3.0
+* spec says:
+*
+* "If the active uniform is an array, the uniform name returned in
+* name will always be the name of the uniform array appended with
+* "[0]"."
+*
+* The same text also appears in the OpenGL 4.2 spec. It does not,
+* however, appear in any previous spec. Previous specifications are
+* ambiguous in this regard. However, either name can later be passed
+* to glGetUniformLocation (and related APIs), so there shouldn't be any
+* harm in always appending "[0]" to uniform array names.
+*/
+ if (uni->array_elements != 0) {
+ unsigned i;
+
+ /* The comparison is strange because *length does *NOT* include the
+ * terminating NUL, but maxLength does.
+ */
+ for (i = 0; i < 3 && (*length + i + 1) < maxLength; i++)
+ nameOut[*length + i] = "[0]"[i];
+
+ nameOut[*length + i] = '\0';
+ *length += i;
+ }
}
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 10/10] mesa: Array uniform name length includes length of [0]
From: Ian Romanick This is required by OpenGL ES 3.0 and desktop OpenGL 4.2. Previous version were ambiguous. This also matches the behavior of NVIDIA's closed-source driver (version 304.64). Fixed gles3conformance test uniform_buffer_object_getactiveuniformsiv and uniform_buffer_object_structure_and_array_element_names (on my in-progress branch that fixes a bunch of other stuff...YMMV). Signed-off-by: Ian Romanick --- src/mesa/main/uniform_query.cpp | 10 ++ 1 file changed, 10 insertions(+) diff --git a/src/mesa/main/uniform_query.cpp b/src/mesa/main/uniform_query.cpp index ec08ed4..f0ab0f0 100644 --- a/src/mesa/main/uniform_query.cpp +++ b/src/mesa/main/uniform_query.cpp @@ -122,6 +122,16 @@ _mesa_GetActiveUniformsiv(GLuint program, case GL_UNIFORM_NAME_LENGTH: params[i] = strlen(uni->name) + 1; + + /* Page 61 (page 73 of the PDF) in section 2.11 of the OpenGL ES 3.0 + * spec says: + * + * "If the active uniform is an array, the uniform name returned + * in name will always be the name of the uniform array appended + * with "[0]"." + */ + if (uni->array_elements != 0) +params[i] += 3; break; case GL_UNIFORM_BLOCK_INDEX: -- 1.7.11.7 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 2/7] glsl/build: Remove dead LIBRARY_* variables
--- src/glsl/Makefile.am |2 -- src/glsl/glcpp/Makefile.am |2 -- 2 files changed, 0 insertions(+), 4 deletions(-) diff --git a/src/glsl/Makefile.am b/src/glsl/Makefile.am index 0f545d5..1d93ac4 100644 --- a/src/glsl/Makefile.am +++ b/src/glsl/Makefile.am @@ -25,9 +25,7 @@ AM_CPPFLAGS = \ -I$(top_srcdir)/include \ -I$(top_srcdir)/src/mapi \ -I$(top_srcdir)/src/mesa/ \ - $(LIBRARY_INCLUDES) \ $(DEFINES) \ - $(LIBRARY_DEFINES) \ $(API_DEFINES) AM_CFLAGS = $(VISIBILITY_CFLAGS) AM_CXXFLAGS = $(VISIBILITY_CXXFLAGS) diff --git a/src/glsl/glcpp/Makefile.am b/src/glsl/glcpp/Makefile.am index 1efc9dc..d79f865 100644 --- a/src/glsl/glcpp/Makefile.am +++ b/src/glsl/glcpp/Makefile.am @@ -29,9 +29,7 @@ AM_CFLAGS = \ -I$(top_srcdir)/src/mapi \ -I$(top_srcdir)/src/mesa/ \ $(VISIBILITY_CFLAGS) \ - $(LIBRARY_INCLUDES) \ $(DEFINES) \ - $(LIBRARY_DEFINES) \ $(API_DEFINES) AM_YFLAGS = -v -d -p "glcpp_parser_" -- 1.7.8.6 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 3/7] glsl/build: Build glsl_test only on make check
--- src/glsl/Makefile.am |3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/src/glsl/Makefile.am b/src/glsl/Makefile.am index 1d93ac4..36af90e 100644 --- a/src/glsl/Makefile.am +++ b/src/glsl/Makefile.am @@ -35,7 +35,8 @@ AM_LFLAGS = --nounistd -o$(LEX_OUTPUT_ROOT).c include Makefile.sources noinst_LTLIBRARIES = libglsl.la -noinst_PROGRAMS = glsl_compiler glsl_test +check_PROGRAMS = glsl_test +noinst_PROGRAMS = glsl_compiler libglsl_la_SOURCES = \ glsl_lexer.ll \ -- 1.7.8.6 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 1/7] xmlpool/build: generate options.h via BUILT_SOURCES
Fixes missing options.h when doing 'make check' in dri/common before 'make' has been run. --- src/mesa/drivers/dri/common/xmlpool/Makefile.am |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/mesa/drivers/dri/common/xmlpool/Makefile.am b/src/mesa/drivers/dri/common/xmlpool/Makefile.am index b99df40..b545f37 100644 --- a/src/mesa/drivers/dri/common/xmlpool/Makefile.am +++ b/src/mesa/drivers/dri/common/xmlpool/Makefile.am @@ -52,7 +52,7 @@ POT=xmlpool.pot .PHONY: all clean pot po mo -all: options.h +BUILT_SOURCES = options.h # All generated files are cleaned up. clean: -- 1.7.8.6 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 4/7] glsl/build: Don't build glsl_compiler
Use glslparsertest from piglit instead. --- Alternatively, does automake offer any options for generating rules to build a target without adding it to the default build? I'd be nice to leave glsl_compiler out of the build but still be able to do make glsl_compiler src/glsl/Makefile.am |9 - 1 files changed, 0 insertions(+), 9 deletions(-) diff --git a/src/glsl/Makefile.am b/src/glsl/Makefile.am index 36af90e..fa09c5e 100644 --- a/src/glsl/Makefile.am +++ b/src/glsl/Makefile.am @@ -36,7 +36,6 @@ include Makefile.sources noinst_LTLIBRARIES = libglsl.la check_PROGRAMS = glsl_test -noinst_PROGRAMS = glsl_compiler libglsl_la_SOURCES = \ glsl_lexer.ll \ @@ -47,14 +46,6 @@ libglsl_la_SOURCES = \ libglsl_la_LIBADD = glcpp/libglcpp.la libglsl_la_LDFLAGS = -glsl_compiler_SOURCES = \ - $(top_srcdir)/src/mesa/main/hash_table.c \ - $(top_srcdir)/src/mesa/program/prog_hash_table.c \ - $(top_srcdir)/src/mesa/program/symbol_table.c \ - $(GLSL_COMPILER_CXX_FILES) - -glsl_compiler_LDADD = libglsl.la - glsl_test_SOURCES = \ $(top_srcdir)/src/mesa/main/hash_table.c \ $(top_srcdir)/src/mesa/program/prog_hash_table.c \ -- 1.7.8.6 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 5/7] glsl/build: Don't build builtin_compiler separately if not cross compiling
Reduces the number of times that src/glsl/ is compiled when not cross compiling. Cc: Thierry Reding --- configure.ac |2 ++ src/glsl/.gitignore |1 + src/glsl/Makefile.am | 29 +++-- 3 files changed, 30 insertions(+), 2 deletions(-) diff --git a/configure.ac b/configure.ac index 99a08fd..5dd9a80 100644 --- a/configure.ac +++ b/configure.ac @@ -1970,6 +1970,8 @@ AM_CONDITIONAL(HAVE_X86_ASM, echo "$DEFINES" | grep 'X86_ASM' >/dev/null 2>&1) AM_CONDITIONAL(HAVE_X86_64_ASM, echo "$DEFINES" | grep 'X86_64_ASM' >/dev/null 2>&1) AM_CONDITIONAL(HAVE_SPARC_ASM, echo "$DEFINES" | grep 'SPARC_ASM' >/dev/null 2>&1) +AM_CONDITIONAL(CROSS_COMPILING, test "x$cross_compiling" = xyes) + AC_SUBST([VDPAU_MAJOR], 1) AC_SUBST([VDPAU_MINOR], 0) diff --git a/src/glsl/.gitignore b/src/glsl/.gitignore index 4eb4fef..a04df27 100644 --- a/src/glsl/.gitignore +++ b/src/glsl/.gitignore @@ -4,4 +4,5 @@ glsl_parser.cc glsl_parser.h glsl_parser.output builtin_function.cpp +builtincompiler glsl_test diff --git a/src/glsl/Makefile.am b/src/glsl/Makefile.am index fa09c5e..07fcf96 100644 --- a/src/glsl/Makefile.am +++ b/src/glsl/Makefile.am @@ -19,7 +19,11 @@ # FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS # IN THE SOFTWARE. +if CROSS_COMPILING SUBDIRS = builtin_compiler glcpp +else +SUBDIRS = glcpp +endif AM_CPPFLAGS = \ -I$(top_srcdir)/include \ @@ -56,6 +60,27 @@ glsl_test_SOURCES = \ glsl_test_LDADD = libglsl.la +if CROSS_COMPILING +BUILTIN_COMPILER = builtin_compiler/builtin_compiler$(BUILD_EXEEXT) +else +BUILTIN_COMPILER = builtincompiler$(BUILD_EXEEXT) + +noinst_PROGRAMS = builtincompiler +noinst_LTLIBRARIES += libglslcore.la + +libglslcore_la_SOURCES = \ + glsl_lexer.ll \ + glsl_parser.cc \ + $(LIBGLSL_FILES) +builtincompiler_SOURCES = \ + $(top_srcdir)/src/mesa/main/hash_table.c\ + $(top_srcdir)/src/mesa/program/prog_hash_table.c\ + $(top_srcdir)/src/mesa/program/symbol_table.c \ + $(top_srcdir)/src/glsl/builtin_compiler/builtin_stubs.cpp \ + $(GLSL_COMPILER_CXX_FILES) +builtincompiler_LDADD = libglslcore.la glcpp/libglcpp.la +endif + # automake <=1.11 and automake >=1.12 have different conventions for naming C++ header files # made by yacc. To work with both, we write our own rule rather than using automake's. # When (if) we require automake >=1.12 in configure.ac, this can be removed, and we can use @@ -66,5 +91,5 @@ glsl_parser.cc glsl_parser.h: glsl_parser.yy BUILT_SOURCES = glsl_parser.h CLEANFILES = glsl_lexer.cc glsl_parser.cc $(BUILT_SOURCES) builtin_function.cpp -builtin_function.cpp: $(srcdir)/builtins/profiles/* $(srcdir)/builtins/ir/* $(srcdir)/builtins/glsl/* $(srcdir)/builtins/tools/generate_builtins.py $(srcdir)/builtins/tools/texture_builtins.py builtin_compiler/builtin_compiler$(BUILD_EXEEXT) - $(AM_V_GEN) $(PYTHON2) $(PYTHON_FLAGS) $(srcdir)/builtins/tools/generate_builtins.py builtin_compiler/builtin_compiler$(BUILD_EXEEXT) > builtin_function.cpp || rm -f builtin_function.cpp +builtin_function.cpp: $(srcdir)/builtins/profiles/* $(srcdir)/builtins/ir/* $(srcdir)/builtins/glsl/* $(srcdir)/builtins/tools/generate_builtins.py $(srcdir)/builtins/tools/texture_builtins.py $(BUILTIN_COMPILER) + $(AM_V_GEN) $(PYTHON2) $(PYTHON_FLAGS) $(srcdir)/builtins/tools/generate_builtins.py $(builddir)/$(BUILTIN_COMPILER) > builtin_function.cpp || rm -f builtin_function.cpp -- 1.7.8.6 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 6/7] glsl/build: Build glcpp via the glsl Makefile
Removing the subdirectory recursion provides a small speed up. --- configure.ac|1 - src/glsl/.gitignore |4 +- src/glsl/Makefile.am| 108 ++- src/glsl/Makefile.sources |8 ++-- src/glsl/glcpp/Makefile.am | 54 --- src/glsl/glcpp/tests/glcpp-test |4 +- 6 files changed, 91 insertions(+), 88 deletions(-) delete mode 100644 src/glsl/glcpp/Makefile.am diff --git a/configure.ac b/configure.ac index 5dd9a80..b08c1c8 100644 --- a/configure.ac +++ b/configure.ac @@ -2085,7 +2085,6 @@ AC_CONFIG_FILES([Makefile src/gbm/main/gbm.pc src/glsl/Makefile src/glsl/builtin_compiler/Makefile - src/glsl/glcpp/Makefile src/glsl/tests/Makefile src/glx/Makefile src/glx/tests/Makefile diff --git a/src/glsl/.gitignore b/src/glsl/.gitignore index a04df27..2cf5b00 100644 --- a/src/glsl/.gitignore +++ b/src/glsl/.gitignore @@ -1,6 +1,6 @@ glsl_compiler -glsl_lexer.cc -glsl_parser.cc +glsl_lexer.cpp +glsl_parser.cpp glsl_parser.h glsl_parser.output builtin_function.cpp diff --git a/src/glsl/Makefile.am b/src/glsl/Makefile.am index 07fcf96..1962d3b 100644 --- a/src/glsl/Makefile.am +++ b/src/glsl/Makefile.am @@ -20,35 +20,42 @@ # IN THE SOFTWARE. if CROSS_COMPILING -SUBDIRS = builtin_compiler glcpp -else -SUBDIRS = glcpp +SUBDIRS = builtin_compiler endif AM_CPPFLAGS = \ -I$(top_srcdir)/include \ -I$(top_srcdir)/src/mapi \ -I$(top_srcdir)/src/mesa/ \ + -I$(top_srcdir)/src/glsl/glcpp \ $(DEFINES) \ $(API_DEFINES) AM_CFLAGS = $(VISIBILITY_CFLAGS) AM_CXXFLAGS = $(VISIBILITY_CXXFLAGS) -AM_LFLAGS = --nounistd -o$(LEX_OUTPUT_ROOT).c - include Makefile.sources -noinst_LTLIBRARIES = libglsl.la -check_PROGRAMS = glsl_test +TESTS = glcpp/tests/glcpp-test -libglsl_la_SOURCES = \ - glsl_lexer.ll \ - glsl_parser.cc \ - $(LIBGLSL_FILES) \ - builtin_function.cpp +noinst_LTLIBRARIES = libglcpp.la libglsl.la +check_PROGRAMS = glcpp/glcpp glsl_test -libglsl_la_LIBADD = glcpp/libglcpp.la -libglsl_la_LDFLAGS = +libglcpp_la_SOURCES = \ + glcpp/glcpp-lex.c \ + glcpp/glcpp-parse.c \ + $(LIBGLCPP_FILES) + +glcpp_glcpp_SOURCES = \ + glcpp/glcpp.c \ + $(top_srcdir)/src/mesa/program/prog_hash_table.c +glcpp_glcpp_LDADD = libglcpp.la + +libglsl_la_SOURCES = \ + glsl_lexer.cpp \ + glsl_parser.cpp \ + $(LIBGLSL_FILES)\ + builtin_function.cpp +libglsl_la_LIBADD = libglcpp.la glsl_test_SOURCES = \ $(top_srcdir)/src/mesa/main/hash_table.c \ @@ -69,8 +76,8 @@ noinst_PROGRAMS = builtincompiler noinst_LTLIBRARIES += libglslcore.la libglslcore_la_SOURCES = \ - glsl_lexer.ll \ - glsl_parser.cc \ + glsl_lexer.cpp \ + glsl_parser.cpp \ $(LIBGLSL_FILES) builtincompiler_SOURCES = \ $(top_srcdir)/src/mesa/main/hash_table.c\ @@ -78,18 +85,69 @@ builtincompiler_SOURCES = \ $(top_srcdir)/src/mesa/program/symbol_table.c \ $(top_srcdir)/src/glsl/builtin_compiler/builtin_stubs.cpp \ $(GLSL_COMPILER_CXX_FILES) -builtincompiler_LDADD = libglslcore.la glcpp/libglcpp.la +builtincompiler_LDADD = libglslcore.la libglcpp.la endif -# automake <=1.11 and automake >=1.12 have different conventions for naming C++ header files -# made by yacc. To work with both, we write our own rule rather than using automake's. -# When (if) we require automake >=1.12 in configure.ac, this can be removed, and we can use -# the automake rule. -glsl_parser.cc glsl_parser.h: glsl_parser.yy - $(AM_V_GEN) $(YACC) -v -o glsl_parser.cc -p "_mesa_glsl_" --defines=glsl_parser.h $< +# We write our own rules for yacc and lex below. We'd rather use automake, +# but automake makes it especially difficult for a number of reasons: +# +# * < automake-1.12 generates .h files from .yy and .ypp files, but +#>=automake-1.12 generates .hh and .hpp files respectively. There's no +#good way of making a project that uses C++ yacc files compatible with +#both versions of automake. Strong work automake developers. +# +# * Since we're generating code from .l/.y files in a subdirectory (glcpp/) +#we'd like the resulting generated code to also go in glcpp/ for purposes +#of distribution. Automake gives no way to do this. +# +# * Since we're building multiple yacc parsers into one library (and v
[Mesa-dev] [PATCH 7/7] glsl/build: Build tests via the glsl Makefile
--- Makefile.am |1 - configure.ac |1 - src/glsl/Makefile.am | 38 -- src/glsl/tests/Makefile.am | 35 --- src/glsl/tests/optimization-test | 10 -- 5 files changed, 44 insertions(+), 41 deletions(-) delete mode 100644 src/glsl/tests/Makefile.am diff --git a/Makefile.am b/Makefile.am index ee7..74f43da 100644 --- a/Makefile.am +++ b/Makefile.am @@ -30,7 +30,6 @@ check-local: $(MAKE) -C src/mapi/glapi/tests check $(MAKE) -C src/mapi/shared-glapi/tests check $(MAKE) -C src/mesa/main/tests check - $(MAKE) -C src/glsl/tests check $(MAKE) -C src/glx/tests check .PHONY: doxygen diff --git a/configure.ac b/configure.ac index b08c1c8..40496e7 100644 --- a/configure.ac +++ b/configure.ac @@ -2085,7 +2085,6 @@ AC_CONFIG_FILES([Makefile src/gbm/main/gbm.pc src/glsl/Makefile src/glsl/builtin_compiler/Makefile - src/glsl/tests/Makefile src/glx/Makefile src/glx/tests/Makefile src/gtest/Makefile diff --git a/src/glsl/Makefile.am b/src/glsl/Makefile.am index 1962d3b..175183a 100644 --- a/src/glsl/Makefile.am +++ b/src/glsl/Makefile.am @@ -28,6 +28,7 @@ AM_CPPFLAGS = \ -I$(top_srcdir)/src/mapi \ -I$(top_srcdir)/src/mesa/ \ -I$(top_srcdir)/src/glsl/glcpp \ + -I$(top_srcdir)/src/gtest/include \ $(DEFINES) \ $(API_DEFINES) AM_CFLAGS = $(VISIBILITY_CFLAGS) @@ -35,10 +36,43 @@ AM_CXXFLAGS = $(VISIBILITY_CXXFLAGS) include Makefile.sources -TESTS = glcpp/tests/glcpp-test +TESTS = glcpp/tests/glcpp-test \ + tests/optimization-test \ + tests/ralloc-test \ + tests/uniform-initializer-test + +TESTS_ENVIRONMENT= \ + export PYTHON2=$(PYTHON2); \ + export PYTHON_FLAGS=$(PYTHON_FLAGS); noinst_LTLIBRARIES = libglcpp.la libglsl.la -check_PROGRAMS = glcpp/glcpp glsl_test +check_PROGRAMS = \ + glcpp/glcpp \ + glsl_test \ + tests/ralloc-test \ + tests/uniform-initializer-test + +tests_uniform_initializer_test_SOURCES = \ + $(top_srcdir)/src/mesa/main/hash_table.c\ + $(top_srcdir)/src/mesa/program/prog_hash_table.c\ + $(top_srcdir)/src/mesa/program/symbol_table.c \ + tests/copy_constant_to_storage_tests.cpp\ + tests/set_uniform_initializer_tests.cpp \ + tests/uniform_initializer_utils.cpp +tests_uniform_initializer_test_CFLAGS =\ + $(PTHREAD_CFLAGS) +tests_uniform_initializer_test_LDADD = \ + $(top_builddir)/src/gtest/libgtest.la \ + $(top_builddir)/src/glsl/libglsl.la \ + $(PTHREAD_LIBS) + +tests_ralloc_test_SOURCES =\ + tests/ralloc_test.cpp \ + $(top_builddir)/src/glsl/ralloc.c +tests_ralloc_test_CFLAGS = $(PTHREAD_CFLAGS) +tests_ralloc_test_LDADD = \ + $(top_builddir)/src/gtest/libgtest.la \ + $(PTHREAD_LIBS) libglcpp_la_SOURCES = \ glcpp/glcpp-lex.c \ diff --git a/src/glsl/tests/Makefile.am b/src/glsl/tests/Makefile.am deleted file mode 100644 index eaf5a3c..000 --- a/src/glsl/tests/Makefile.am +++ /dev/null @@ -1,35 +0,0 @@ -AM_CPPFLAGS = \ - -I$(top_srcdir)/src/gtest/include \ - -I$(top_srcdir)/src/mesa \ - -I$(top_srcdir)/src/mapi \ - -I$(top_srcdir)/src/glsl \ - -I$(top_srcdir)/include - -TESTS_ENVIRONMENT= \ - export PYTHON2=$(PYTHON2); \ - export PYTHON_FLAGS=$(PYTHON_FLAGS); - -TESTS = \ - optimization-test \ - ralloc-test \ - uniform-initializer-test - -check_PROGRAMS = \ - ralloc-test \ - uniform-initializer-test - -uniform_initializer_test_SOURCES = \ - copy_constant_to_storage_tests.cpp \ - set_uniform_initializer_tests.cpp \ - uniform_initializer_utils.cpp -uniform_initializer_test_CFLAGS = \ - $(PTHREAD_CFLAGS) -uniform_initializer_test_LDADD = \ - $(top_builddir)/src/gtest/libgtest.la \ - $(top_builddir)/src/glsl/libglsl.la \ - $(top_builddir)/src/mesa/libmesa.la \ - $(PTHREAD_LIBS) - -ralloc_test_SOURCES = ralloc_test.cpp $(top_builddir)/src/glsl/ralloc.c -ralloc_test_CFLAGS = $(PTHREAD_CFLAGS) -ralloc_test_LDADD = $(top_builddir)/src/gtest/libgtest.la $(PTHREAD_LIBS) diff --git a/src/glsl/tests/optimization-test b/
Re: [Mesa-dev] [PATCH] mesa: Allow TexImage3D on depth/stencil textures in ES 3.0.
On Wed, Jan 16, 2013 at 2:38 PM, Ian Romanick wrote: > These checks are suck. :( Won't this let you specify depth component > textures for GL_TEXTURE_3D textures too? Looks like it, since packed_pixels_pixelstore is failing with this patch. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] mesa: Allow TexImage3D on depth/stencil textures in ES 3.0.
On 01/16/2013 01:49 PM, Kenneth Graunke wrote:
ES 3.0 supports 2D array textures, which are populated via TexImage3D.
Fixes es3conform's shadow_execution_frag test.
Actually, this should be handled by other checks already in
texture_error_check. Search for "additional checks for depth textures".
GL_TEXTURE_3D depth component textures aren't allowed on desktop
either, and 1D textures are never allowed in ES.
Signed-off-by: Kenneth Graunke
---
src/mesa/main/glformats.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/mesa/main/glformats.c b/src/mesa/main/glformats.c
index 7969f77..9831ab4 100644
--- a/src/mesa/main/glformats.c
+++ b/src/mesa/main/glformats.c
@@ -2068,7 +2068,7 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_COMPONENT:
- if (dimensions != 2) {
+ if (dimensions == 1) {
return GL_INVALID_OPERATION;
}
switch (type) {
@@ -2098,7 +2098,7 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_STENCIL:
- if (dimensions != 2) {
+ if (dimensions == 1) {
return GL_INVALID_OPERATION;
}
switch (type) {
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 1/2] mesa: Don't allow DEPTH_STENCIL for 3D textures
From: Ian Romanick
Just like DEPTH_COMPONENT.
Signed-off-by: Ian Romanick
---
src/mesa/main/teximage.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/mesa/main/teximage.c b/src/mesa/main/teximage.c
index eeb977e..7a15196 100644
--- a/src/mesa/main/teximage.c
+++ b/src/mesa/main/teximage.c
@@ -2006,7 +2006,8 @@ texture_error_check( struct gl_context *ctx,
}
/* additional checks for depth textures */
- if (_mesa_base_tex_format(ctx, internalFormat) == GL_DEPTH_COMPONENT) {
+ if (_mesa_base_tex_format(ctx, internalFormat) == GL_DEPTH_COMPONENT
+ || _mesa_base_tex_format(ctx, internalFormat) == GL_DEPTH_STENCIL) {
/* Only 1D, 2D, rect, array and cube textures supported, not 3D
* Cubemaps are only supported for GL version > 3.0 or with
EXT_gpu_shader4 */
if (target != GL_TEXTURE_1D &&
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 2/2] es3: Don't check dimensions in _mesa_es3_error_check_format_and_type
From: Ian Romanick
Filtering of DEPTH_COMPONENT and DEPTH_STENCIL for TEXTURE_3D is already
done in texture_error_check because these combinations aren't allowed on
desktop GL either.
Signed-off-by: Ian Romanick
---
src/mesa/main/glformats.c | 9 +
src/mesa/main/glformats.h | 3 +--
src/mesa/main/teximage.c | 3 +--
3 files changed, 3 insertions(+), 12 deletions(-)
diff --git a/src/mesa/main/glformats.c b/src/mesa/main/glformats.c
index 7969f77..3004c08 100644
--- a/src/mesa/main/glformats.c
+++ b/src/mesa/main/glformats.c
@@ -1697,8 +1697,7 @@ _mesa_es_error_check_format_and_type(GLenum format,
GLenum type,
*/
GLenum
_mesa_es3_error_check_format_and_type(GLenum format, GLenum type,
- GLenum internalFormat,
- unsigned dimensions)
+ GLenum internalFormat)
{
GLboolean type_valid = GL_TRUE;
@@ -2068,9 +2067,6 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_COMPONENT:
- if (dimensions != 2) {
- return GL_INVALID_OPERATION;
- }
switch (type) {
case GL_UNSIGNED_SHORT:
if (internalFormat != GL_DEPTH_COMPONENT16)
@@ -2098,9 +2094,6 @@ _mesa_es3_error_check_format_and_type(GLenum format,
GLenum type,
break;
case GL_DEPTH_STENCIL:
- if (dimensions != 2) {
- return GL_INVALID_OPERATION;
- }
switch (type) {
case GL_UNSIGNED_INT_24_8:
if (internalFormat != GL_DEPTH24_STENCIL8)
diff --git a/src/mesa/main/glformats.h b/src/mesa/main/glformats.h
index 4cbc82f..cdd2b94 100644
--- a/src/mesa/main/glformats.h
+++ b/src/mesa/main/glformats.h
@@ -122,8 +122,7 @@ _mesa_es_error_check_format_and_type(GLenum format, GLenum
type,
extern GLenum
_mesa_es3_error_check_format_and_type(GLenum format, GLenum type,
- GLenum internalFormat,
- unsigned dimensions);
+ GLenum internalFormat);
#ifdef __cplusplus
diff --git a/src/mesa/main/teximage.c b/src/mesa/main/teximage.c
index 7a15196..c44a4f5 100644
--- a/src/mesa/main/teximage.c
+++ b/src/mesa/main/teximage.c
@@ -1896,8 +1896,7 @@ texture_error_check( struct gl_context *ctx,
if (_mesa_is_gles(ctx)) {
if (_mesa_is_gles3(ctx)) {
err = _mesa_es3_error_check_format_and_type(format, type,
- internalFormat,
- dimensions);
+ internalFormat);
} else {
if (format != internalFormat) {
_mesa_error(ctx, GL_INVALID_OPERATION,
--
1.7.11.7
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] mesa: add missing ASSERT_OUTSIDE_BEGIN_END() in _mesa_GetInternalformativ()
On 01/16/2013 07:19 AM, Brian Paul wrote:
Ah, yes. I always forget about that. I can't wait until Eric finishes
separating the dispatch tables so that we never have to do this again.
Reviewed-by: Ian Romanick
---
src/mesa/main/formatquery.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/src/mesa/main/formatquery.c b/src/mesa/main/formatquery.c
index f08ab66..bd895e8 100644
--- a/src/mesa/main/formatquery.c
+++ b/src/mesa/main/formatquery.c
@@ -22,6 +22,7 @@
*/
#include "mtypes.h"
+#include "context.h"
#include "glformats.h"
#include "macros.h"
#include "mfeatures.h"
@@ -37,6 +38,8 @@ _mesa_GetInternalformativ(GLenum target, GLenum
internalformat, GLenum pname,
GLsizei count = 0;
GET_CURRENT_CONTEXT(ctx);
+ ASSERT_OUTSIDE_BEGIN_END(ctx);
+
if (!ctx->Extensions.ARB_internalformat_query) {
_mesa_error(ctx, GL_INVALID_OPERATION, "glGetInternalformativ");
return;
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [Bug 59450] SIGSEGV src/mesa/state_tracker/st_cb_blit.c:216
https://bugs.freedesktop.org/show_bug.cgi?id=59450 Brian Paul changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #1 from Brian Paul --- Fixed with commit 56c01d81094795286828ecb83713e043695e905a -- You are receiving this mail because: You are the assignee for the bug. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 00/10] Support UBOs with instance names pre-series
On 01/16/2013 03:03 PM, Ian Romanick wrote: This is the "warm up" series before my series to make UBOs with instances names (and arrays of UBOs) work. These are both required features of GLSL ES 3.00 that are not added to desktop GLSL until 1.50. This series is just a bunch of minor fixes and small refactors. This series fixes 4 gles3conform tests. There are about a dozen more patches coming in the next few day to fix the remaining 14 UBO tests. This series is: Reviewed-by: Kenneth Graunke ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 1/3] i965/vs: Create a 'lod_type' temporary for ir->lod_info.lod->type.
This is purely a refactor. However, in a moment, we'll want to set
lod_type to float for ir_tex, where ir->lod_info.lod is NULL.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 13 -
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
index 02feff6..594f927 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
@@ -1953,6 +1953,7 @@ vec4_visitor::visit(ir_texture *ir)
shadow_comparitor = this->result;
}
+ const glsl_type *lod_type;
src_reg lod, dPdx, dPdy;
switch (ir->op) {
case ir_txf:
@@ -1960,6 +1961,7 @@ vec4_visitor::visit(ir_texture *ir)
case ir_txs:
ir->lod_info.lod->accept(this);
lod = this->result;
+ lod_type = ir->lod_info.lod->type;
break;
case ir_txd:
ir->lod_info.grad.dPdx->accept(this);
@@ -1967,6 +1969,8 @@ vec4_visitor::visit(ir_texture *ir)
ir->lod_info.grad.dPdy->accept(this);
dPdy = this->result;
+
+ lod_type = ir->lod_info.grad.dPdx->type;
break;
case ir_tex:
case ir_txb:
@@ -2008,8 +2012,7 @@ vec4_visitor::visit(ir_texture *ir)
if (ir->op == ir_txs) {
int writemask = intel->gen == 4 ? WRITEMASK_W : WRITEMASK_X;
- emit(MOV(dst_reg(MRF, param_base, ir->lod_info.lod->type, writemask),
- lod));
+ emit(MOV(dst_reg(MRF, param_base, lod_type, writemask), lod));
} else {
int i, coord_mask = 0, zero_mask = 0;
/* Load the coordinate */
@@ -2067,12 +2070,12 @@ vec4_visitor::visit(ir_texture *ir)
mrf = param_base;
writemask = WRITEMASK_Z;
}
-emit(MOV(dst_reg(MRF, mrf, ir->lod_info.lod->type, writemask), lod));
+emit(MOV(dst_reg(MRF, mrf, lod_type, writemask), lod));
} else if (ir->op == ir_txf) {
-emit(MOV(dst_reg(MRF, param_base, ir->lod_info.lod->type, WRITEMASK_W),
+emit(MOV(dst_reg(MRF, param_base, lod_type, WRITEMASK_W),
lod));
} else if (ir->op == ir_txd) {
-const glsl_type *type = ir->lod_info.grad.dPdx->type;
+const glsl_type *type = lod_type;
if (intel->gen >= 5) {
dPdx.swizzle =
BRW_SWIZZLE4(SWIZZLE_X,SWIZZLE_X,SWIZZLE_Y,SWIZZLE_Y);
--
1.8.1
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 2/3] i965/vs: Set LOD to 0 for ordinary texture() calls.
Previously it was left undefined, causing us to select a random LOD.
NOTE: This is a candidate for stable branches (and requires the previous
commit as well).
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
index 594f927..3139245 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
@@ -1956,6 +1956,10 @@ vec4_visitor::visit(ir_texture *ir)
const glsl_type *lod_type;
src_reg lod, dPdx, dPdy;
switch (ir->op) {
+ case ir_tex:
+ lod = src_reg(0.0f);
+ lod_type = glsl_type::float_type;
+ break;
case ir_txf:
case ir_txl:
case ir_txs:
@@ -1972,7 +1976,6 @@ vec4_visitor::visit(ir_texture *ir)
lod_type = ir->lod_info.grad.dPdx->type;
break;
- case ir_tex:
case ir_txb:
break;
}
@@ -2055,7 +2058,7 @@ vec4_visitor::visit(ir_texture *ir)
}
/* Load the LOD info */
- if (ir->op == ir_txl) {
+ if (ir->op == ir_tex || ir->op == ir_txl) {
int mrf, writemask;
if (intel->gen >= 5) {
mrf = param_base + 1;
--
1.8.1
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH 3/3] i965/vs: Store texturing results into a vec4 temporary.
The sampler appears to ignore writemasks (even when correcting the
WRITEMASK_XYZW in brw_vec4_emit.cpp to the proper writemask) and just
always writes all four values.
To cope with this, just texture into a temporary, then MOV out into a
register that has the proper number of components.
NOTE: This is a candidate for stable branches.
Fixes es3conform's shadow_execution_vert.test.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
This series also fixes 18 oglconform subtests! I didn't test on a per-patch
basis to figure out which patch fixes which tests, though.
diff --git a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
index 3139245..ebf8990 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
@@ -2005,6 +2005,7 @@ vec4_visitor::visit(ir_texture *ir)
inst->mlen = inst->header_present + 1; /* always at least one */
inst->sampler = sampler;
inst->dst = dst_reg(this, ir->type);
+ inst->dst.writemask = WRITEMASK_XYZW;
inst->shadow_compare = ir->shadow_comparitor != NULL;
if (ir->offset != NULL && ir->op != ir_txf)
@@ -2123,13 +2124,16 @@ vec4_visitor::visit(ir_texture *ir)
void
vec4_visitor::swizzle_result(ir_texture *ir, src_reg orig_val, int sampler)
{
- this->result = orig_val;
-
int s = c->key.tex.swizzles[sampler];
+ this->result = src_reg(this, ir->type);
+ dst_reg swizzled_result(this->result);
+
if (ir->op == ir_txs || ir->type == glsl_type::float_type
- || s == SWIZZLE_NOOP)
+ || s == SWIZZLE_NOOP) {
+ emit(MOV(swizzled_result, orig_val));
return;
+ }
int zero_mask = 0, one_mask = 0, copy_mask = 0;
int swizzle[4];
@@ -2149,9 +2153,6 @@ vec4_visitor::swizzle_result(ir_texture *ir, src_reg
orig_val, int sampler)
}
}
- this->result = src_reg(this, ir->type);
- dst_reg swizzled_result(this->result);
-
if (copy_mask) {
orig_val.swizzle = BRW_SWIZZLE4(swizzle[0], swizzle[1], swizzle[2],
swizzle[3]);
swizzled_result.writemask = copy_mask;
--
1.8.1
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeon/r200: Fix FBO formats
No, no regress in piglit i've seen at all, only better... i even played whole Drawn 2 game: http://www.drawngame.com/games/dark-flight under wine in OpenGL mode, it works perfectly now, needs culling under fbo which is fixed now and this fix to display text and so works exactly the same as with nvidia on Windows. It can further pass 2 more test in piglit if those are ported from intel: fbo-nodepth-test http://cgit.freedesktop.org/mesa/mesa/commit/?id=4d4f2daefabdc4ca1dd778a9265475c65ef52936 i've tried that one and it pass and maybe this for fbo-cubemap - but not tried: http://cgit.freedesktop.org/mesa/mesa/commit/?id=80513ec8b4c812b9c6249cc5824337a5f04ab34c Now i checked supertuxkart game, it has some problems with alpha in XRGB format. Under non-fbo mode kart model have black background, and under fbo models are transparent. I fixed it for non fbo case with R200_TXFORMAT_ALPHA_IN_MAP for XRGB also in r200_blit, but don't know where is this format controled for fbos? I mentoioned this because text is slightly transparent also in that Drawn game, it only works with with fbo extension in opengl mode, i guess that needs the same fix for XRGB format to expect alpha under fbo, etc... But OK that can be another bug :). 2013/1/16, Marek Olšák : > On Tue, Jan 15, 2013 at 7:15 PM, Alex Deucher > wrote: >> On Fri, Jan 11, 2013 at 6:37 AM, smoki wrote: >>> Piglit passes more fbo tests on rv280, 14/28 before and now 28/33. >>> Also should fix bug: >>> https://bugs.freedesktop.org/show_bug.cgi?id=27704 >> >> Does this regress anything in piglit? It's been a while since I >> looked at this code, I think the issue may be that r1xx/r2xx asics >> only support ARGB render buffers while the texture hardware supports >> both ARGB and RGBA formats so you may run into cases where you can >> texture from a buffer, but can't render to it, at least not without an >> intermediate blit to convert the format first. > > In r300g, I decided to only support RGBA and BGRA formats and not ARGB > and ABGR because of a lack of blending support. Maybe we should do the > same for r1xx/r2xx to make things simpler. > > Marek > ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [Bug 59499] New: SIGSEGV src/mesa/swrast/s_blit.c:546
https://bugs.freedesktop.org/show_bug.cgi?id=59499 Priority: medium Bug ID: 59499 Keywords: regression CC: anuj.pho...@gmail.com Assignee: mesa-dev@lists.freedesktop.org Summary: SIGSEGV src/mesa/swrast/s_blit.c:546 Severity: critical Classification: Unclassified OS: Linux (All) Reporter: v...@freedesktop.org Hardware: x86-64 (AMD64) Status: NEW Version: git Component: Other Product: Mesa mesa: 56c01d81094795286828ecb83713e043695e905a (master) Run piglit fbo-srgb-blit on swrast. $ ./bin/fbo-srgb-blit -auto Mesa warning: failed to remap index 167 Mesa warning: failed to remap index 207 Mesa warning: failed to remap index 208 Mesa warning: failed to remap index 211 Mesa warning: failed to remap index 377 Segmentation fault (core dumped) (gdb) bt #0 0x7f9031b8738f in blit_linear (ctx=0xe0bed0, srcX0=7, srcY0=7, srcX1=9, srcY1=9, dstX0=0, dstY0=0, dstX1=8, dstY1=16) at ../../../src/mesa/swrast/s_blit.c:546 #1 0x7f9031b87f20 in _swrast_BlitFramebuffer (ctx=0xe0bed0, srcX0=7, srcY0=7, srcX1=9, srcY1=9, dstX0=0, dstY0=0, dstX1=8, dstY1=16, mask=16384, filter=9729) at ../../../src/mesa/swrast/s_blit.c:812 #2 0x7f9031a3123b in _mesa_BlitFramebuffer (srcX0=7, srcY0=7, srcX1=9, srcY1=9, dstX0=0, dstY0=0, dstX1=8, dstY1=16, mask=16384, filter=9729) at ../../../src/mesa/main/fbobject.c:3057 #3 0x00401076 in blit_rect (src_format=35907, dst_format=32856, x=-1, y=0, w=1, h=1, stretch=true) at piglit/tests/fbo/fbo-srgb-blit.c:66 #4 0x004012d6 in piglit_display () at piglit/tests/fbo/fbo-srgb-blit.c:114 #5 0x7f9035517408 in display () at piglit/tests/util/piglit-framework-gl/piglit_glut_framework.c:60 #6 0x7f9034ed1137 in fghRedrawWindow (window=0xe07e60) at freeglut_main.c:210 #7 fghcbDisplayWindow (window=0xe07e60, enumerator=0x7fffab7497a0) at freeglut_main.c:227 #8 0x7f9034ed4889 in fgEnumWindows (enumCallback=0x7f9034ed10d0 , enumerator=0x7fffab7497a0) at freeglut_structure.c:394 #9 0x7f9034ed15fa in fghDisplayAll () at freeglut_main.c:249 #10 glutMainLoopEvent () at freeglut_main.c:1450 #11 0x7f9034ed1f05 in glutMainLoop () at freeglut_main.c:1498 #12 0x7f90355175e0 in run_test (gl_fw=0x7f90357e17e0, argc=1, argv=0x7fffab749b68) at piglit/tests/util/piglit-framework-gl/piglit_glut_framework.c:127 #13 0x7f9035515726 in piglit_gl_test_run (argc=1, argv=0x7fffab749b68, config=0x7fffab749a50) at piglit/tests/util/piglit-framework-gl.c:127 #14 0x00400e9e in main (argc=2, argv=0x7fffab749b68) at piglit/tests/fbo/fbo-srgb-blit.c:41 (gdb) frame 0 #0 0x7f9031b8738f in blit_linear (ctx=0xe0bed0, srcX0=7, srcY0=7, srcX1=9, srcY1=9, dstX0=0, dstY0=0, dstX1=8, dstY1=16) at ../../../src/mesa/swrast/s_blit.c:546 546 gl_format drawFormat = _mesa_get_srgb_format_linear(drawRb->Format); (gdb) print drawRb $1 = (struct gl_renderbuffer *) 0x0 bed997dabaada5e7c8081db08f93a6f1b5932e8b is the first bad commit commit bed997dabaada5e7c8081db08f93a6f1b5932e8b Author: Anuj Phogat Date: Tue Dec 11 20:17:47 2012 -0800 mesa: Add support to allow blitting to multiple color draw buffers This patch fixes a blitting case when drawAttachment->Texture == readAttachment->Texture. It was causing an assertion failure in intel_miptree_attach_map() with gles3 conformance test case: framebuffer_blit_functionality_minifying_blit Number of changes in this file look scary. But most of them are caused by introducing a big for loop to support rendering to multiple color draw buffers. V2: Fixed a case when number of draw buffer attachments are zero. V3: Put a for loop in blit_nearest() and blit_linear() functions in to support blitting to multiple color draw buffers. V4: Remove variable declaration in for loop to avoid MSVC compilation issues. Signed-off-by: Anuj Phogat Reviewed-by: Kenneth Graunke Reviewed-by: Ian Romanick :04 04 98d5e6ea72f2fcf765cd8241682ab18c8081cc70 ee017ce78ded09098a4ded7a90e19130b53e4eea Msrc bisect run success -- You are receiving this mail because: You are the assignee for the bug. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [Bug 59501] New: mesa/main/api_exec.c:286: undefined reference to `SET_GetInternalformativ'
https://bugs.freedesktop.org/show_bug.cgi?id=59501 Priority: medium Bug ID: 59501 Keywords: regression CC: i...@freedesktop.org Assignee: mesa-dev@lists.freedesktop.org Summary: mesa/main/api_exec.c:286: undefined reference to `SET_GetInternalformativ' Severity: blocker Classification: Unclassified OS: Linux (All) Reporter: v...@freedesktop.org Hardware: x86-64 (AMD64) Status: NEW Version: git Component: Mesa core Product: Mesa mesa: 56c01d81094795286828ecb83713e043695e905a (master) $ scons platform=windows toolchain=crossmingw machine=x86_64 [...] scons: Building targets ... Linking build/windows-x86_64-debug/gallium/targets/libgl-gdi/opengl32.dll ... build/windows-x86_64-debug/mesa/libmesa.a(api_exec.o): In function `_mesa_initialize_exec_table': build/windows-x86_64-debug/mesa/main/api_exec.c:286: undefined reference to `SET_GetInternalformativ' bda540d2357f69a63dcf16550e178f803682d6ce is the first bad commit commit bda540d2357f69a63dcf16550e178f803682d6ce Author: Ian Romanick Date: Fri Nov 30 12:43:58 2012 -0800 mesa: Add dispatch and extension XML for GL_ARB_internalformat_query Signed-off-by: Ian Romanick Reviewed-by: Eric Anholt Reviewed-by: Jordan Justen :04 04 e38bd61c113bd0b4296bb47e22cd1f06d530ddd4 ecd74b726b9a188d64143c33f582cd3cd2eedfeb Msrc bisect run success -- You are receiving this mail because: You are the assignee for the bug. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 5/7] glsl/build: Don't build builtin_compiler separately if not cross compiling
On Wed, Jan 16, 2013 at 03:14:14PM -0800, Matt Turner wrote: [...] > diff --git a/src/glsl/Makefile.am b/src/glsl/Makefile.am > index fa09c5e..07fcf96 100644 > --- a/src/glsl/Makefile.am > +++ b/src/glsl/Makefile.am > @@ -19,7 +19,11 @@ > # FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER > DEALINGS > # IN THE SOFTWARE. > > +if CROSS_COMPILING > SUBDIRS = builtin_compiler glcpp > +else > +SUBDIRS = glcpp > +endif > > AM_CPPFLAGS = \ > -I$(top_srcdir)/include \ > @@ -56,6 +60,27 @@ glsl_test_SOURCES = \ > > glsl_test_LDADD = libglsl.la > > +if CROSS_COMPILING > +BUILTIN_COMPILER = builtin_compiler/builtin_compiler$(BUILD_EXEEXT) > +else > +BUILTIN_COMPILER = builtincompiler$(BUILD_EXEEXT) > + > +noinst_PROGRAMS = builtincompiler > +noinst_LTLIBRARIES += libglslcore.la > + > +libglslcore_la_SOURCES = \ > + glsl_lexer.ll \ > + glsl_parser.cc \ > + $(LIBGLSL_FILES) > +builtincompiler_SOURCES = \ > + $(top_srcdir)/src/mesa/main/hash_table.c\ > + $(top_srcdir)/src/mesa/program/prog_hash_table.c\ > + $(top_srcdir)/src/mesa/program/symbol_table.c \ > + $(top_srcdir)/src/glsl/builtin_compiler/builtin_stubs.cpp \ > + $(GLSL_COMPILER_CXX_FILES) > +builtincompiler_LDADD = libglslcore.la glcpp/libglcpp.la > +endif Can this not be moved into the builtin_compiler subdirectory? You could set the various *_FOR_BUILD variables only if CROSS_COMPILING and split up the builtin_compiler_SOURCES variables such that a native version of them is compiled if CROSS_COMPILING and make builtin_compiler link with libglcpp.la and libglslcore.la otherwise. That way there would be no duplication of these rules that need to be kept in sync. Furthermore all the cross-compiling special cases would be restricted to the builtin_compiler subdirectory. Thierry pgpqgxnZgkA0V.pgp Description: PGP signature ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 02/10] glsl: Track UBO block names in the symbol table.
On Wed, Jan 16, 2013 at 03:03:41PM -0800, Ian Romanick wrote:
> From: Kenneth Graunke
>
> The GLSL 1.40 spec says:
>
> "Uniform block names and variable names declared within uniform
> blocks are scoped at the program level."
>
> Track the block name in the symbol table and emit errors when conflicts
> exist.
>
> Fixes es3conform's uniform_buffer_object_block_name_conflict test, and
> fixes the piglit block-name-clashes-with-{variable,function,struct}.vert
> tests.
>
> NOTE: This is a candidate for the 9.0 branch.
>
> Reviewed-by: Ian Romanick
> ---
> src/glsl/ast_to_hir.cpp| 6 ++
> src/glsl/glsl_symbol_table.cpp | 14 +++---
> src/glsl/glsl_symbol_table.h | 1 +
> 3 files changed, 18 insertions(+), 3 deletions(-)
>
> diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp
> index ea5e3b7..31dd51a 100644
> --- a/src/glsl/ast_to_hir.cpp
> +++ b/src/glsl/ast_to_hir.cpp
> @@ -4132,6 +4132,12 @@ ast_uniform_block::hir(exec_list *instructions,
> struct gl_uniform_block *ubo = get_next_uniform_block(state);
> ubo->Name = ralloc_strdup(state->uniform_blocks, this->block_name);
>
> + if (!state->symbols->add_uniform_block(ubo)) {
> + YYLTYPE loc = this->get_location();
> + _mesa_glsl_error(&loc, state, "Uniform block name `%s' already taken
> in "
> + "the current scope.\n", ubo->Name);
> + }
> +
> unsigned int num_variables = 0;
> foreach_list_typed(ast_declarator_list, decl_list, link, &declarations) {
>foreach_list_const(node, &decl_list->declarations) {
> diff --git a/src/glsl/glsl_symbol_table.cpp b/src/glsl/glsl_symbol_table.cpp
> index f934ea8..6171a27 100644
> --- a/src/glsl/glsl_symbol_table.cpp
> +++ b/src/glsl/glsl_symbol_table.cpp
> @@ -41,13 +41,15 @@ public:
>ralloc_free(entry);
> }
>
> - symbol_table_entry(ir_variable *v) : v(v), f(0), t(0)
> {}
> - symbol_table_entry(ir_function *f) : v(0), f(f), t(0)
> {}
> - symbol_table_entry(const glsl_type *t) : v(0), f(0), t(t)
> {}
> + symbol_table_entry(ir_variable *v) : v(v), f(0), t(0), u(0)
> {}
> + symbol_table_entry(ir_function *f) : v(0), f(f), t(0), u(0)
> {}
> + symbol_table_entry(const glsl_type *t) : v(0), f(0), t(t), u(0)
> {}
> + symbol_table_entry(struct gl_uniform_block *u) : v(0), f(0), t(t), u(u)
> {}
I guess you did not mean to initialize 't' on itself.
-
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki
Business Identity Code: 0357606 - 4
Domiciled in Helsinki
This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
