Re: [Mesa-dev] [PATCH 12/21] mesa: Remove GL_MESA_resize_buffers
I think you might squash the attached diff into your patch.
Andreas.
2013/6/28 Ian Romanick i...@freedesktop.org
From: Ian Romanick ian.d.roman...@intel.com
Commit bab755a made the implementation a no-op, and it was only ever
enabled by software rasterizers.
Signed-off-by: Ian Romanick ian.d.roman...@intel.com
---
docs/relnotes/9.2.html | 2 ++
src/mapi/glapi/gen/gl_API.xml | 2 +-
src/mapi/glapi/gen/mesadef.py | 1 -
src/mesa/drivers/windows/gdi/mesa.def | 1 -
src/mesa/main/extensions.c | 2 --
src/mesa/main/framebuffer.c | 10 --
src/mesa/main/framebuffer.h | 4
src/mesa/main/mtypes.h | 1 -
src/mesa/main/tests/dispatch_sanity.cpp | 3 ---
9 files changed, 3 insertions(+), 23 deletions(-)
diff --git a/docs/relnotes/9.2.html b/docs/relnotes/9.2.html
index 08e82d0..2f2c394 100644
--- a/docs/relnotes/9.2.html
+++ b/docs/relnotes/9.2.html
@@ -65,6 +65,8 @@ Note: some of the new features are only available with
certain drivers.
liRemoved d3d1x state tracker (unused, unmaintained and broken)/li
liRemoved GL_EXT_clip_volume_hint because no driver had enabled it since
2007./li
+liRemoved GL_MESA_resize_buffers because it was only really implemented
by
+the (unsupported) GDI driver./li
/ul
/div
diff --git a/src/mapi/glapi/gen/gl_API.xml b/src/mapi/glapi/gen/gl_API.xml
index a066fe2..82b908f 100644
--- a/src/mapi/glapi/gen/gl_API.xml
+++ b/src/mapi/glapi/gen/gl_API.xml
@@ -11032,7 +11032,7 @@
/category
category name=GL_MESA_resize_buffers number=196
-function name=ResizeBuffersMESA offset=assign
+function name=ResizeBuffersMESA offset=assign exec=skip
glx ignore=true/
/function
/category
diff --git a/src/mapi/glapi/gen/mesadef.py b/src/mapi/glapi/gen/mesadef.py
index f6d33cb..77cc4a3 100644
--- a/src/mapi/glapi/gen/mesadef.py
+++ b/src/mapi/glapi/gen/mesadef.py
@@ -134,7 +134,6 @@ def PrintTail():
print '\t_mesa_new_buffer_object'
print '\t_mesa_new_texture_object'
print '\t_mesa_problem'
-print '\t_mesa_ResizeBuffersMESA'
print '\t_mesa_store_compressed_teximage1d'
print '\t_mesa_store_compressed_teximage2d'
print '\t_mesa_store_compressed_teximage3d'
diff --git a/src/mesa/drivers/windows/gdi/mesa.def
b/src/mesa/drivers/windows/gdi/mesa.def
index fec7bba..92736b3 100644
--- a/src/mesa/drivers/windows/gdi/mesa.def
+++ b/src/mesa/drivers/windows/gdi/mesa.def
@@ -556,7 +556,6 @@ EXPORTS
glFogCoorddvEXT
glFogCoordPointerEXT
glBlendFuncSeparateEXT
- glResizeBuffersMESA
glWindowPos2dMESA
glWindowPos2dvMESA
glWindowPos2fMESA
diff --git a/src/mesa/main/extensions.c b/src/mesa/main/extensions.c
index 8f96a77..9c90bbe 100644
--- a/src/mesa/main/extensions.c
+++ b/src/mesa/main/extensions.c
@@ -307,7 +307,6 @@ static const struct extension extension_table[] = {
{ GL_IBM_texture_mirrored_repeat, o(dummy_true),
GLL,1998 },
{ GL_INGR_blend_func_separate,
o(EXT_blend_func_separate), GLL,1999 },
{ GL_MESA_pack_invert,o(MESA_pack_invert),
GL, 2002 },
- { GL_MESA_resize_buffers,
o(MESA_resize_buffers), GL, 1999 },
{ GL_MESA_texture_array, o(MESA_texture_array),
GLL,2007 },
{ GL_MESA_texture_signed_rgba,o(EXT_texture_snorm),
GL, 2009 },
{ GL_MESA_window_pos, o(dummy_true),
GLL,2000 },
@@ -445,7 +444,6 @@ _mesa_enable_sw_extensions(struct gl_context *ctx)
/*ctx-Extensions.EXT_transform_feedback = GL_TRUE;*/
ctx-Extensions.EXT_vertex_array_bgra = GL_TRUE;
ctx-Extensions.MESA_pack_invert = GL_TRUE;
- ctx-Extensions.MESA_resize_buffers = GL_TRUE;
ctx-Extensions.MESA_texture_array = GL_TRUE;
ctx-Extensions.MESA_ycbcr_texture = GL_TRUE;
ctx-Extensions.NV_blend_square = GL_TRUE;
diff --git a/src/mesa/main/framebuffer.c b/src/mesa/main/framebuffer.c
index d28882a..4ec4118 100644
--- a/src/mesa/main/framebuffer.c
+++ b/src/mesa/main/framebuffer.c
@@ -319,16 +319,6 @@ _mesa_resize_framebuffer(struct gl_context *ctx,
struct gl_framebuffer *fb,
}
}
-/*
- * XXX THIS IS OBSOLETE
- */
-void GLAPIENTRY
-_mesa_ResizeBuffersMESA( void )
-{
-}
-
-
-
/**
* Examine all the framebuffer's renderbuffers to update the Width/Height
* fields of the framebuffer. If we have renderbuffers with different
diff --git a/src/mesa/main/framebuffer.h b/src/mesa/main/framebuffer.h
index 1b1caab..2645664 100644
--- a/src/mesa/main/framebuffer.h
+++ b/src/mesa/main/framebuffer.h
@@ -71,10 +71,6 @@ _mesa_resize_framebuffer(struct
[Mesa-dev] [Bug 66149] [HSW] Background of application icon has garbage after update kernelmesa in order to support Graphics [8086: 0a2e]
https://bugs.freedesktop.org/show_bug.cgi?id=66149 Gordon Jin gordon@intel.com changed: What|Removed |Added Status|NEEDINFO|RESOLVED Resolution|--- |FIXED --- Comment #5 from Gordon Jin gordon@intel.com --- I think we can close this. -- 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 66149] [HSW] Background of application icon has garbage after update kernelmesa in order to support Graphics [8086: 0a2e]
https://bugs.freedesktop.org/show_bug.cgi?id=66149 --- Comment #6 from Andreas Boll andreas.boll@gmail.com --- (In reply to comment #4) The issue can't be reproduced with 9.1 branch. When will 9.1.4 release? Thanks! It's planned to be released on monday. http://lists.freedesktop.org/archives/mesa-dev/2013-June/040999.html -- 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 56710] src/mapi/glapi/glapitemp.h:1640:1: warning: no previous prototype for ‘glReadBufferNV’ [-Wmissing-prototypes]
https://bugs.freedesktop.org/show_bug.cgi?id=56710 --- Comment #14 from Jon TURNEY jon.tur...@dronecode.org.uk --- (In reply to comment #13) Could you still reproduce this bug? Appears to be fixed to me. -- You are receiving this mail because: You are the QA Contact for the bug. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] build: fix out-of-tree builds in gallium/auxiliary
The rules were writing files to e.g. util/u_indices_gen.py, but in an out-of-tree build this directory doesn't exist in the build directory. So, create the directories just in case. Note: This patch is a candidate for the 9.0 and 9.1 branches. Signed-off-by: Ross Burton ross.bur...@intel.com --- src/gallium/auxiliary/Makefile.am |4 1 file changed, 4 insertions(+) diff --git a/src/gallium/auxiliary/Makefile.am b/src/gallium/auxiliary/Makefile.am index f14279b..0c3e7ba 100644 --- a/src/gallium/auxiliary/Makefile.am +++ b/src/gallium/auxiliary/Makefile.am @@ -38,13 +38,17 @@ libgallium_la_SOURCES += \ endif indices/u_indices_gen.c: $(srcdir)/indices/u_indices_gen.py + mkdir --parents indices $(AM_V_GEN) $(PYTHON2) $ $@ indices/u_unfilled_gen.c: $(srcdir)/indices/u_unfilled_gen.py + mkdir --parents indices $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_srgb.c: $(srcdir)/util/u_format_srgb.py + mkdir --parents util $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_table.c: $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format_pack.py $(srcdir)/util/u_format_parse.py $(srcdir)/util/u_format.csv + mkdir --parents util $(AM_V_GEN) $(PYTHON2) $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format.csv $@ -- 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 01/21] i915: Remove GEN = 4 extension support
On 06/27/2013 06:20 PM, Ian Romanick wrote: From: Ian Romanick ian.d.roman...@intel.com This copy of the source file is only used for GEN = 3, so remove the dead code. Signed-off-by: Ian Romanick ian.d.roman...@intel.com --- src/mesa/drivers/dri/i915/intel_extensions.c | 79 ++-- 1 file changed, 3 insertions(+), 76 deletions(-) This series looks good to me! Unfortunately, both you and Eric removed the gen checks from the i915/i965 extensions lists, so there were some conflicts. Yours looks a bit cleaner, so I created a new branch which is master + your series + anholt's series. It's 59 commits, and available as the 'fork-i915' branch of my tree: http://cgit.freedesktop.org/~kwg/mesa/log/?h=fork-i915 I also incorporated my feedback on patch 20, so I'd appreciate an ack if you believe that's right/wrong. Unless there are any objections, I hope to push it soon. --Ken ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [Bug 56710] src/mapi/glapi/glapitemp.h:1640:1: warning: no previous prototype for ‘glReadBufferNV’ [-Wmissing-prototypes]
https://bugs.freedesktop.org/show_bug.cgi?id=56710 Andreas Boll andreas.boll@gmail.com changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #15 from Andreas Boll andreas.boll@gmail.com --- (In reply to comment #14) (In reply to comment #13) Could you still reproduce this bug? Appears to be fixed to me. Thanks for testing. -- You are receiving this mail because: You are the QA Contact for the bug. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] build: fix out-of-tree builds in gallium/auxiliary
2013/6/28 Ross Burton ross.bur...@intel.com The rules were writing files to e.g. util/u_indices_gen.py, but in an out-of-tree build this directory doesn't exist in the build directory. So, create the directories just in case. Note: This patch is a candidate for the 9.0 and 9.1 branches. I think you can drop 9.0, since gallium was converted to Automake in mesa 9.1. Signed-off-by: Ross Burton ross.bur...@intel.com --- src/gallium/auxiliary/Makefile.am |4 1 file changed, 4 insertions(+) diff --git a/src/gallium/auxiliary/Makefile.am b/src/gallium/auxiliary/Makefile.am index f14279b..0c3e7ba 100644 --- a/src/gallium/auxiliary/Makefile.am +++ b/src/gallium/auxiliary/Makefile.am @@ -38,13 +38,17 @@ libgallium_la_SOURCES += \ endif indices/u_indices_gen.c: $(srcdir)/indices/u_indices_gen.py + mkdir --parents indices Please use the autoconf variable $(MKDIR_P) instead of mkdir --parents. With these changes: Reviewed-by: Andreas Boll andreas.boll@gmail.com $(AM_V_GEN) $(PYTHON2) $ $@ indices/u_unfilled_gen.c: $(srcdir)/indices/u_unfilled_gen.py + mkdir --parents indices $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_srgb.c: $(srcdir)/util/u_format_srgb.py + mkdir --parents util $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_table.c: $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format_pack.py $(srcdir)/util/u_format_parse.py $(srcdir)/util/u_format.csv + mkdir --parents util $(AM_V_GEN) $(PYTHON2) $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format.csv $@ -- 1.7.10.4 ___ 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
[Mesa-dev] [PATCH][V2] build: fix out-of-tree builds in gallium/auxiliary
The rules were writing files to e.g. util/u_indices_gen.py, but in an out-of-tree build this directory doesn't exist in the build directory. So, create the directories just in case. Note: This patch is a candidate for the 9.0 and 9.1 branches. Signed-off-by: Ross Burton ross.bur...@intel.com --- src/gallium/auxiliary/Makefile.am |4 1 file changed, 4 insertions(+) diff --git a/src/gallium/auxiliary/Makefile.am b/src/gallium/auxiliary/Makefile.am index f14279b..670e124 100644 --- a/src/gallium/auxiliary/Makefile.am +++ b/src/gallium/auxiliary/Makefile.am @@ -38,13 +38,17 @@ libgallium_la_SOURCES += \ endif indices/u_indices_gen.c: $(srcdir)/indices/u_indices_gen.py + $(MKDIR_P) indices $(AM_V_GEN) $(PYTHON2) $ $@ indices/u_unfilled_gen.c: $(srcdir)/indices/u_unfilled_gen.py + $(MKDIR_P) indices $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_srgb.c: $(srcdir)/util/u_format_srgb.py + $(MKDIR_P) util $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_table.c: $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format_pack.py $(srcdir)/util/u_format_parse.py $(srcdir)/util/u_format.csv + $(MKDIR_P) util $(AM_V_GEN) $(PYTHON2) $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format.csv $@ -- 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 19/21] mesa: GL_ARB_texture_storage is not optional
On 06/27/2013 07:20 PM, Ian Romanick wrote:
From: Ian Romanick ian.d.roman...@intel.com
In Mesa, this extension is implemented purely in software. Drivers may
*optionally* provide optimized paths.
NOTE: This has the side effect of enabling the extension in the radeon,
r200, and nouveau drivers.
Signed-off-by: Ian Romanick ian.d.roman...@intel.com
---
docs/relnotes/9.2.html | 1 +
src/mesa/drivers/dri/i915/intel_extensions.c | 1 -
src/mesa/drivers/dri/i965/intel_extensions.c | 1 -
src/mesa/main/extensions.c | 3 +--
src/mesa/main/mtypes.h | 1 -
src/mesa/main/teximage.c | 9 +++--
src/mesa/main/texparam.c | 4
src/mesa/state_tracker/st_extensions.c | 1 -
8 files changed, 5 insertions(+), 16 deletions(-)
diff --git a/docs/relnotes/9.2.html b/docs/relnotes/9.2.html
index 2f2c394..1f49191 100644
--- a/docs/relnotes/9.2.html
+++ b/docs/relnotes/9.2.html
@@ -48,6 +48,7 @@ Note: some of the new features are only available with
certain drivers.
liGL_ARB_texture_multisample/li
liGL_ARB_texture_storage_multisample/li
liGL_ARB_texture_query_lod/li
+liEnable GL_ARB_texture_storage on radeon, r200, and nouveau/li
liAdded new freedreno gallium driver/li
liOSMesa interface for gallium llvmpipe/softpipe drivers/li
liGallium Heads-Up Display (HUD) feature for performance monitoring/li
diff --git a/src/mesa/drivers/dri/i915/intel_extensions.c
b/src/mesa/drivers/dri/i915/intel_extensions.c
index 74b304a..479217b 100644
--- a/src/mesa/drivers/dri/i915/intel_extensions.c
+++ b/src/mesa/drivers/dri/i915/intel_extensions.c
@@ -57,7 +57,6 @@ intelInitExtensions(struct gl_context *ctx)
ctx-Extensions.ARB_texture_env_combine = true;
ctx-Extensions.ARB_texture_env_crossbar = true;
ctx-Extensions.ARB_texture_env_dot3 = true;
- ctx-Extensions.ARB_texture_storage = true;
ctx-Extensions.ARB_vertex_program = true;
ctx-Extensions.ARB_vertex_shader = true;
ctx-Extensions.EXT_blend_color = true;
diff --git a/src/mesa/drivers/dri/i965/intel_extensions.c
b/src/mesa/drivers/dri/i965/intel_extensions.c
index 23b74a5..5064018 100644
--- a/src/mesa/drivers/dri/i965/intel_extensions.c
+++ b/src/mesa/drivers/dri/i965/intel_extensions.c
@@ -79,7 +79,6 @@ intelInitExtensions(struct gl_context *ctx)
ctx-Extensions.ARB_texture_non_power_of_two = true;
ctx-Extensions.ARB_texture_rg = true;
ctx-Extensions.ARB_texture_rgb10_a2ui = true;
- ctx-Extensions.ARB_texture_storage = true;
ctx-Extensions.ARB_vertex_program = true;
ctx-Extensions.ARB_vertex_shader = true;
ctx-Extensions.ARB_vertex_type_2_10_10_10_rev = true;
diff --git a/src/mesa/main/extensions.c b/src/mesa/main/extensions.c
index 73282e1..f914981 100644
--- a/src/mesa/main/extensions.c
+++ b/src/mesa/main/extensions.c
@@ -149,7 +149,7 @@ static const struct extension extension_table[] = {
{ GL_ARB_texture_rectangle, o(NV_texture_rectangle),
GL, 2004 },
{ GL_ARB_texture_rgb10_a2ui, o(ARB_texture_rgb10_a2ui),
GL, 2009 },
{ GL_ARB_texture_rg, o(ARB_texture_rg),
GL, 2008 },
- { GL_ARB_texture_storage, o(ARB_texture_storage),
GL, 2011 },
+ { GL_ARB_texture_storage, o(dummy_true),
GL, 2011 },
{ GL_ARB_texture_storage_multisample,
o(ARB_texture_storage_multisample), GL, 2012 },
{ GL_ARB_texture_swizzle, o(EXT_texture_swizzle),
GL, 2008 },
{ GL_ARB_timer_query, o(ARB_timer_query),
GL, 2010 },
@@ -403,7 +403,6 @@ _mesa_enable_sw_extensions(struct gl_context *ctx)
ctx-Extensions.ARB_texture_non_power_of_two = GL_TRUE;
ctx-Extensions.ARB_texture_rg = GL_TRUE;
ctx-Extensions.ARB_texture_compression_rgtc = GL_TRUE;
- ctx-Extensions.ARB_texture_storage = GL_TRUE;
ctx-Extensions.ARB_vertex_program = GL_TRUE;
ctx-Extensions.ARB_vertex_shader = GL_TRUE;
ctx-Extensions.ARB_sync = GL_TRUE;
diff --git a/src/mesa/main/mtypes.h b/src/mesa/main/mtypes.h
index 2879341..a19ecd6 100644
--- a/src/mesa/main/mtypes.h
+++ b/src/mesa/main/mtypes.h
@@ -3036,7 +3036,6 @@ struct gl_extensions
GLboolean ARB_texture_query_lod;
GLboolean ARB_texture_rg;
GLboolean ARB_texture_rgb10_a2ui;
- GLboolean ARB_texture_storage;
GLboolean ARB_texture_storage_multisample;
GLboolean ARB_timer_query;
GLboolean ARB_transform_feedback2;
diff --git a/src/mesa/main/teximage.c b/src/mesa/main/teximage.c
index 5226687..be03a60 100644
--- a/src/mesa/main/teximage.c
+++ b/src/mesa/main/teximage.c
@@ -1852,12 +1852,9 @@
Re: [Mesa-dev] [PATCH 00/21] Extension clean up
On 06/27/2013 07:20 PM, Ian Romanick wrote: This is my annual extension clean up blob. I don't expect much here to be controversial (or even interesting to read) except patches 12, 17, 18, and 21. LGTM. Just a minor formatting comment on #19. And I second Kenneth's comment on #20. For 5-21, Reviewed-by: Brian Paul bri...@vmware.com ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH][V2] build: fix out-of-tree builds in gallium/auxiliary
There is an open bug report about this issue: https://bugs.freedesktop.org/show_bug.cgi?id=60197 Matt, could you take a look at this? Which patch do you prefer? 2013/6/28 Ross Burton ross.bur...@intel.com The rules were writing files to e.g. util/u_indices_gen.py, but in an out-of-tree build this directory doesn't exist in the build directory. So, create the directories just in case. Note: This patch is a candidate for the 9.0 and 9.1 branches. 9.0 is still mentioned. Signed-off-by: Ross Burton ross.bur...@intel.com --- src/gallium/auxiliary/Makefile.am |4 1 file changed, 4 insertions(+) diff --git a/src/gallium/auxiliary/Makefile.am b/src/gallium/auxiliary/Makefile.am index f14279b..670e124 100644 --- a/src/gallium/auxiliary/Makefile.am +++ b/src/gallium/auxiliary/Makefile.am @@ -38,13 +38,17 @@ libgallium_la_SOURCES += \ endif indices/u_indices_gen.c: $(srcdir)/indices/u_indices_gen.py + $(MKDIR_P) indices $(AM_V_GEN) $(PYTHON2) $ $@ indices/u_unfilled_gen.c: $(srcdir)/indices/u_unfilled_gen.py + $(MKDIR_P) indices $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_srgb.c: $(srcdir)/util/u_format_srgb.py + $(MKDIR_P) util $(AM_V_GEN) $(PYTHON2) $ $@ util/u_format_table.c: $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format_pack.py $(srcdir)/util/u_format_parse.py $(srcdir)/util/u_format.csv + $(MKDIR_P) util $(AM_V_GEN) $(PYTHON2) $(srcdir)/util/u_format_table.py $(srcdir)/util/u_format.csv $@ -- 1.7.10.4 ___ 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
[Mesa-dev] [PATCH 2/2] svga: pass svga_compile_key by reference instead of value
---
src/gallium/drivers/svga/svga_tgsi.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/gallium/drivers/svga/svga_tgsi.c
b/src/gallium/drivers/svga/svga_tgsi.c
index 56529c6..29fbe84 100644
--- a/src/gallium/drivers/svga/svga_tgsi.c
+++ b/src/gallium/drivers/svga/svga_tgsi.c
@@ -266,7 +266,7 @@ svga_remap_generic_index(int8_t
remap_table[MAX_GENERIC_VARYING],
*/
static struct svga_shader_result *
svga_tgsi_translate(const struct svga_shader *shader,
-struct svga_compile_key key, unsigned unit)
+const struct svga_compile_key *key, unsigned unit)
{
struct svga_shader_result *result = NULL;
struct svga_shader_emitter emit;
@@ -281,17 +281,17 @@ svga_tgsi_translate(const struct svga_shader *shader,
emit.ptr = emit.buf;
emit.unit = unit;
- emit.key = key;
+ emit.key = *key;
tgsi_scan_shader(shader-tokens, emit.info);
emit.imm_start = emit.info.file_max[TGSI_FILE_CONSTANT] + 1;
if (unit == PIPE_SHADER_FRAGMENT)
- emit.imm_start += key.fkey.num_unnormalized_coords;
+ emit.imm_start += key-fkey.num_unnormalized_coords;
if (unit == PIPE_SHADER_VERTEX) {
- emit.imm_start += key.vkey.need_prescale ? 2 : 0;
+ emit.imm_start += key-vkey.need_prescale ? 2 : 0;
}
emit.nr_hw_float_const =
@@ -324,7 +324,7 @@ svga_tgsi_translate(const struct svga_shader *shader,
result-shader = shader;
result-tokens = (const unsigned *) emit.buf;
result-nr_tokens = (emit.ptr - emit.buf) / sizeof(unsigned);
- memcpy(result-key, key, sizeof key);
+ memcpy(result-key, key, sizeof(*key));
result-id = UTIL_BITMASK_INVALID_INDEX;
if (SVGA_DEBUG DEBUG_TGSI) {
@@ -360,7 +360,7 @@ svga_translate_fragment_program(const struct
svga_fragment_shader *fs,
memcpy(key.generic_remap_table, fs-generic_remap_table,
sizeof(fs-generic_remap_table));
- return svga_tgsi_translate(fs-base, key, PIPE_SHADER_FRAGMENT);
+ return svga_tgsi_translate(fs-base, key, PIPE_SHADER_FRAGMENT);
}
@@ -379,7 +379,7 @@ svga_translate_vertex_program(const struct
svga_vertex_shader *vs,
*/
svga_remap_generics(vkey-fs_generic_inputs, key.generic_remap_table);
- return svga_tgsi_translate(vs-base, key, PIPE_SHADER_VERTEX);
+ return svga_tgsi_translate(vs-base, key, PIPE_SHADER_VERTEX);
}
--
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][V2] build: fix out-of-tree builds in gallium/auxiliary
On 28 June 2013 14:57, Andreas Boll andreas.boll@gmail.com wrote: There is an open bug report about this issue: https://bugs.freedesktop.org/show_bug.cgi?id=60197 Matt, could you take a look at this? Which patch do you prefer? Quentin's patch is more generic as it uses $(dir), so merge that one. Ross ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 1/2] svga: use switch statement in svga_shader_type()
- Original Message -
Safer in case the PIPE_SHADER_x tokens get renumbered (as Marek
wanted to do).
Renumbering PIPE_SHADER_x seems a pure time waster -- there is much more code
that expects the current order -- and I see no benefit.
This patch looks good nevertheles.
---
src/gallium/drivers/svga/svga_state_constants.c | 15 ++-
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/src/gallium/drivers/svga/svga_state_constants.c
b/src/gallium/drivers/svga/svga_state_constants.c
index 77c9349..1c0edb4 100644
--- a/src/gallium/drivers/svga/svga_state_constants.c
+++ b/src/gallium/drivers/svga/svga_state_constants.c
@@ -46,13 +46,18 @@
/**
* Convert from PIPE_SHADER_* to SVGA3D_SHADERTYPE_*
*/
-static int
+static unsigned
svga_shader_type(unsigned shader)
{
- assert(PIPE_SHADER_VERTEX + 1 == SVGA3D_SHADERTYPE_VS);
- assert(PIPE_SHADER_FRAGMENT + 1 == SVGA3D_SHADERTYPE_PS);
- assert(shader = PIPE_SHADER_FRAGMENT);
- return shader + 1;
+ switch (shader) {
+ case PIPE_SHADER_VERTEX:
+ return SVGA3D_SHADERTYPE_VS;
+ case PIPE_SHADER_FRAGMENT:
+ return SVGA3D_SHADERTYPE_PS;
+ default:
+ assert(!Unexpected shader type);
+ return SVGA3D_SHADERTYPE_VS;
+ }
}
--
1.7.10.4
___
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 2/2] svga: pass svga_compile_key by reference instead of value
Looks good to me
- Original Message -
---
src/gallium/drivers/svga/svga_tgsi.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/gallium/drivers/svga/svga_tgsi.c
b/src/gallium/drivers/svga/svga_tgsi.c
index 56529c6..29fbe84 100644
--- a/src/gallium/drivers/svga/svga_tgsi.c
+++ b/src/gallium/drivers/svga/svga_tgsi.c
@@ -266,7 +266,7 @@ svga_remap_generic_index(int8_t
remap_table[MAX_GENERIC_VARYING],
*/
static struct svga_shader_result *
svga_tgsi_translate(const struct svga_shader *shader,
-struct svga_compile_key key, unsigned unit)
+const struct svga_compile_key *key, unsigned unit)
{
struct svga_shader_result *result = NULL;
struct svga_shader_emitter emit;
@@ -281,17 +281,17 @@ svga_tgsi_translate(const struct svga_shader *shader,
emit.ptr = emit.buf;
emit.unit = unit;
- emit.key = key;
+ emit.key = *key;
tgsi_scan_shader(shader-tokens, emit.info);
emit.imm_start = emit.info.file_max[TGSI_FILE_CONSTANT] + 1;
if (unit == PIPE_SHADER_FRAGMENT)
- emit.imm_start += key.fkey.num_unnormalized_coords;
+ emit.imm_start += key-fkey.num_unnormalized_coords;
if (unit == PIPE_SHADER_VERTEX) {
- emit.imm_start += key.vkey.need_prescale ? 2 : 0;
+ emit.imm_start += key-vkey.need_prescale ? 2 : 0;
}
emit.nr_hw_float_const =
@@ -324,7 +324,7 @@ svga_tgsi_translate(const struct svga_shader *shader,
result-shader = shader;
result-tokens = (const unsigned *) emit.buf;
result-nr_tokens = (emit.ptr - emit.buf) / sizeof(unsigned);
- memcpy(result-key, key, sizeof key);
+ memcpy(result-key, key, sizeof(*key));
result-id = UTIL_BITMASK_INVALID_INDEX;
if (SVGA_DEBUG DEBUG_TGSI) {
@@ -360,7 +360,7 @@ svga_translate_fragment_program(const struct
svga_fragment_shader *fs,
memcpy(key.generic_remap_table, fs-generic_remap_table,
sizeof(fs-generic_remap_table));
- return svga_tgsi_translate(fs-base, key, PIPE_SHADER_FRAGMENT);
+ return svga_tgsi_translate(fs-base, key, PIPE_SHADER_FRAGMENT);
}
@@ -379,7 +379,7 @@ svga_translate_vertex_program(const struct
svga_vertex_shader *vs,
*/
svga_remap_generics(vkey-fs_generic_inputs, key.generic_remap_table);
- return svga_tgsi_translate(vs-base, key, PIPE_SHADER_VERTEX);
+ return svga_tgsi_translate(vs-base, key, PIPE_SHADER_VERTEX);
}
--
1.7.10.4
___
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
[Mesa-dev] [PATCH] clover: Fix build with LLVM 3.4
PathV1.h has been removed. In theory this can go back before llvm 3.4, but I
haven't done the research to find out how far back.
Signed-off-by: Aaron Watry awa...@gmail.com
---
src/gallium/state_trackers/clover/llvm/invocation.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/src/gallium/state_trackers/clover/llvm/invocation.cpp
b/src/gallium/state_trackers/clover/llvm/invocation.cpp
index 362f02f..ee0249d 100644
--- a/src/gallium/state_trackers/clover/llvm/invocation.cpp
+++ b/src/gallium/state_trackers/clover/llvm/invocation.cpp
@@ -43,7 +43,12 @@
#include llvm/PassManager.h
#include llvm/Support/TargetSelect.h
#include llvm/Support/MemoryBuffer.h
+#if HAVE_LLVM 0x0304
#include llvm/Support/PathV1.h
+#else
+#include llvm/ADT/SmallString.h
+#include llvm/Support/Path.h
+#endif
#include llvm/Transforms/IPO.h
#include llvm/Transforms/IPO/PassManagerBuilder.h
@@ -222,9 +227,16 @@ namespace {
llvm::PassManager PM;
llvm::PassManagerBuilder Builder;
+#if HAVE_LLVM 0x0304
llvm::sys::Path libclc_path =
llvm::sys::Path(LIBCLC_LIBEXECDIR + processor +
- + triple + .bc);
+#else
+ llvm::SmallString1 libclc_path;
+ libclc_path = LIBCLC_LIBEXECDIR;
+ std::string file_name = processor + - + triple + .bc;
+ llvm::sys::path::append(libclc_path, file_name);
+#endif
// Link the kernel with libclc
#if HAVE_LLVM 0x0303
--
1.8.1.2
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 1/2] svga: use switch statement in svga_shader_type()
The renumbering only makes sense for the GLSL linker and the only
reason for doing that in gallium too is that PIPE_SHADER_x must be
equal to MESA_SHADER_x.
Marek
On Fri, Jun 28, 2013 at 4:32 PM, Jose Fonseca jfons...@vmware.com wrote:
- Original Message -
Safer in case the PIPE_SHADER_x tokens get renumbered (as Marek
wanted to do).
Renumbering PIPE_SHADER_x seems a pure time waster -- there is much more code
that expects the current order -- and I see no benefit.
This patch looks good nevertheles.
---
src/gallium/drivers/svga/svga_state_constants.c | 15 ++-
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/src/gallium/drivers/svga/svga_state_constants.c
b/src/gallium/drivers/svga/svga_state_constants.c
index 77c9349..1c0edb4 100644
--- a/src/gallium/drivers/svga/svga_state_constants.c
+++ b/src/gallium/drivers/svga/svga_state_constants.c
@@ -46,13 +46,18 @@
/**
* Convert from PIPE_SHADER_* to SVGA3D_SHADERTYPE_*
*/
-static int
+static unsigned
svga_shader_type(unsigned shader)
{
- assert(PIPE_SHADER_VERTEX + 1 == SVGA3D_SHADERTYPE_VS);
- assert(PIPE_SHADER_FRAGMENT + 1 == SVGA3D_SHADERTYPE_PS);
- assert(shader = PIPE_SHADER_FRAGMENT);
- return shader + 1;
+ switch (shader) {
+ case PIPE_SHADER_VERTEX:
+ return SVGA3D_SHADERTYPE_VS;
+ case PIPE_SHADER_FRAGMENT:
+ return SVGA3D_SHADERTYPE_PS;
+ default:
+ assert(!Unexpected shader type);
+ return SVGA3D_SHADERTYPE_VS;
+ }
}
--
1.7.10.4
___
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
[Mesa-dev] [Bug 66029] More robust way of detecting LLVM major and minor versions
https://bugs.freedesktop.org/show_bug.cgi?id=66029 --- Comment #4 from Klemens Baum klemensb...@gmail.com --- Sent: http://lists.freedesktop.org/archives/mesa-dev/2013-June/041160.html -- 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 1/6] mesa, gallium: renumber shader indices according to their placement in pipeline
- Original Message -
See my explanation in mtypes.h.
---
src/gallium/include/pipe/p_defines.h |7 ---
src/glsl/linker.cpp| 16
src/mesa/drivers/dri/i965/brw_shader.cpp |8 ++--
src/mesa/main/mtypes.h |8 ++--
src/mesa/main/shaderobj.h |4 ++--
src/mesa/main/uniform_query.cpp|2 +-
src/mesa/program/ir_to_mesa.cpp| 10 +++---
src/mesa/program/program.h |2 +-
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 10 +++---
9 files changed, 30 insertions(+), 37 deletions(-)
diff --git a/src/gallium/include/pipe/p_defines.h
b/src/gallium/include/pipe/p_defines.h
index 8af1a84..216cd2f 100644
--- a/src/gallium/include/pipe/p_defines.h
+++ b/src/gallium/include/pipe/p_defines.h
@@ -352,11 +352,12 @@ enum pipe_flush_flags {
/**
- * Shaders
+ * Shaders.
+ * These must have the same values as Mesa's MESA_SHADER_*.
Sorry for not noticing this earlier -- I haven't been able to keep up with
email traffic lately.
I'm afraid I can't agree with this. Gallium needs to be API agnostic -- it
doesn't make sense to have gallium go at the whims of particular state tracker
implementation details.
There is a lot of code out there that relies on this ordering. And
unfortunately reordering will cause it to fail, often in a silent manner (with
no compiler errors or warnings).
- Original Message -
The renumbering only makes sense for the GLSL linker and the only
reason for doing that in gallium too is that PIPE_SHADER_x must be
equal to MESA_SHADER_x.
This is an implementation detail.
PIPE_SHADER_x may have been paired with MESA_SHADER_x till now as convenience,
but now they are what they are.
Jose
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] llvmpipe: fix timer query if there's no bins
From: Roland Scheidegger srol...@vmware.com
b04a295a4a0cd2defe352b3193b5fa79ca8fc9fc removed seemingly unnecessary
code in get_query. Turns out this code could in fact be reached - while
timestamps are always binned, if there are no bins (which happens if fb
size is 0) then the rasterization query code filling this in is still
never executed.
So fix this up by filling in some timestamp, but do it at EndQuery time
not GetQuery time which should be more appropriate.
Makes piglit arb_timer_query-timestamp-get happy again.
---
src/gallium/drivers/llvmpipe/lp_setup.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/src/gallium/drivers/llvmpipe/lp_setup.c
b/src/gallium/drivers/llvmpipe/lp_setup.c
index 49b61c3..65f61ed 100644
--- a/src/gallium/drivers/llvmpipe/lp_setup.c
+++ b/src/gallium/drivers/llvmpipe/lp_setup.c
@@ -40,6 +40,7 @@
#include util/u_memory.h
#include util/u_pack_color.h
#include draw/draw_pipe.h
+#include os/os_time.h
#include lp_context.h
#include lp_memory.h
#include lp_scene.h
@@ -1263,6 +1264,15 @@ lp_setup_end_query(struct lp_setup_context *setup,
struct llvmpipe_query *pq)
pq-type == PIPE_QUERY_OCCLUSION_PREDICATE ||
pq-type == PIPE_QUERY_PIPELINE_STATISTICS ||
pq-type == PIPE_QUERY_TIMESTAMP) {
+ if (pq-type == PIPE_QUERY_TIMESTAMP
+ !(setup-scene-tiles_x | setup-scene-tiles_y)) {
+/*
+ * If there's a zero width/height framebuffer, there's no bins and
+ * hence no rast task is ever run. So fill in something here
instead.
+ */
+pq-end[0] = os_time_get_nano();
+ }
+
if (!lp_scene_bin_everywhere(setup-scene,
LP_RAST_OP_END_QUERY,
lp_rast_arg_query(pq))) {
--
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 1/6] mesa, gallium: renumber shader indices according to their placement in pipeline
On 06/28/2013 08:53 AM, Jose Fonseca wrote:
- Original Message -
See my explanation in mtypes.h.
---
src/gallium/include/pipe/p_defines.h |7 ---
src/glsl/linker.cpp| 16
src/mesa/drivers/dri/i965/brw_shader.cpp |8 ++--
src/mesa/main/mtypes.h |8 ++--
src/mesa/main/shaderobj.h |4 ++--
src/mesa/main/uniform_query.cpp|2 +-
src/mesa/program/ir_to_mesa.cpp| 10 +++---
src/mesa/program/program.h |2 +-
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 10 +++---
9 files changed, 30 insertions(+), 37 deletions(-)
diff --git a/src/gallium/include/pipe/p_defines.h
b/src/gallium/include/pipe/p_defines.h
index 8af1a84..216cd2f 100644
--- a/src/gallium/include/pipe/p_defines.h
+++ b/src/gallium/include/pipe/p_defines.h
@@ -352,11 +352,12 @@ enum pipe_flush_flags {
/**
- * Shaders
+ * Shaders.
+ * These must have the same values as Mesa's MESA_SHADER_*.
Sorry for not noticing this earlier -- I haven't been able to keep up with
email traffic lately.
I'm afraid I can't agree with this. Gallium needs to be API agnostic -- it
doesn't make sense to have gallium go at the whims of particular state tracker
implementation details.
There is a lot of code out there that relies on this ordering. And
unfortunately reordering will cause it to fail, often in a silent manner (with
no compiler errors or warnings).
- Original Message -
The renumbering only makes sense for the GLSL linker and the only
reason for doing that in gallium too is that PIPE_SHADER_x must be
equal to MESA_SHADER_x.
This is an implementation detail.
PIPE_SHADER_x may have been paired with MESA_SHADER_x till now as convenience,
but now they are what they are.
I took a quick look at the state tracker. AFAICT, there's only one
place where we obviously depend on MESA_SHADER_x == PIPE_SHADER_x, at
st_extensions.c:152
The st_atom_shader/constbuf, etc. code looks OK.
I think we could remove the MESA_SHADER_x == PIPE_SHADER_x assumption
without too much trouble.
-Brian
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] clover: Fix build with LLVM 3.4
Disregard this patch... Looks like Tom already pushed a fix last night.
--Aaron
On Fri, Jun 28, 2013 at 9:41 AM, Aaron Watry awa...@gmail.com wrote:
PathV1.h has been removed. In theory this can go back before llvm 3.4, but I
haven't done the research to find out how far back.
Signed-off-by: Aaron Watry awa...@gmail.com
---
src/gallium/state_trackers/clover/llvm/invocation.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/src/gallium/state_trackers/clover/llvm/invocation.cpp
b/src/gallium/state_trackers/clover/llvm/invocation.cpp
index 362f02f..ee0249d 100644
--- a/src/gallium/state_trackers/clover/llvm/invocation.cpp
+++ b/src/gallium/state_trackers/clover/llvm/invocation.cpp
@@ -43,7 +43,12 @@
#include llvm/PassManager.h
#include llvm/Support/TargetSelect.h
#include llvm/Support/MemoryBuffer.h
+#if HAVE_LLVM 0x0304
#include llvm/Support/PathV1.h
+#else
+#include llvm/ADT/SmallString.h
+#include llvm/Support/Path.h
+#endif
#include llvm/Transforms/IPO.h
#include llvm/Transforms/IPO/PassManagerBuilder.h
@@ -222,9 +227,16 @@ namespace {
llvm::PassManager PM;
llvm::PassManagerBuilder Builder;
+#if HAVE_LLVM 0x0304
llvm::sys::Path libclc_path =
llvm::sys::Path(LIBCLC_LIBEXECDIR + processor +
- + triple + .bc);
+#else
+ llvm::SmallString1 libclc_path;
+ libclc_path = LIBCLC_LIBEXECDIR;
+ std::string file_name = processor + - + triple + .bc;
+ llvm::sys::path::append(libclc_path, file_name);
+#endif
// Link the kernel with libclc
#if HAVE_LLVM 0x0303
--
1.8.1.2
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH 0/6] Eliminating unused built-in varyings
On 06/13/2013 05:25 AM, Marek Olšák wrote: Hi everyone, this series adds a new GLSL compiler optimization pass which eliminates unused and set-but-unused built-in varyings and adds a few improvements to the GLSL linker in the process. Before I show you how it works, I wanna say that there are patches which are related to and will most probably conflict with the geometry shader work, but they are necessary because the linkage of varyings is largely suboptimal. Also, the GL_EXT_separate_shader_objects extension must be disabled for this optimization to be enabled. The reason is a program object with both a VS and FS can be bound partially, e.g. by glUseShaderProgramEXT(GL_VERTEX_SHADER, prog), so the extension makes every program object be just a set of separate shaders. The extension is not important anyway. Could you elaborate on this a bit? The problem already exists that shader can be par-linked (e.g., just a vertex shader) and used with fixed-function. A big part of the reason that I implemented EXT_sso back in the day was to generate infrastructure, etc. for better using shaders with fixed function. Now, to illustrate how the optimization works, consider these 2 shader IR dumps: Vertex shader (8 varyings): ... (declare (shader_out ) vec4 gl_FrontColor) (declare (shader_out ) vec4 gl_FrontSecondaryColor) (declare (shader_out ) (array vec4 6) gl_TexCoord) (function main (signature void (parameters ) ( ... (assign (xyzw) (var_ref gl_FrontColor) (var_ref gl_Color) ) (assign (xyzw) (var_ref gl_FrontSecondaryColor) (var_ref gl_SecondaryColor) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (1)) ) (var_ref gl_MultiTexCoord1) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (4)) ) (var_ref gl_MultiTexCoord4) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (5)) ) (var_ref gl_MultiTexCoord5) ) )) ) Fragment shader (6 varyings): ... (declare (shader_in ) vec4 gl_SecondaryColor) (declare (shader_in ) (array vec4 5) gl_TexCoord) (function main (signature void (parameters ) ( (declare () vec4 r) (assign (xyzw) (var_ref r) ... (var_ref gl_SecondaryColor) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (1)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (2)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (3)) ) ) ) ) (declare (temporary ) vec4 assignment_tmp) (assign (xyzw) (var_ref assignment_tmp) ... (array_ref (var_ref gl_TexCoord) (constant int (4)) ) ) ) ) ... )) ) Note that only gl_TexCoord[1], gl_TexCoord[4], and gl_SecondaryColor are used by both shaders. The optimization replaces all occurences of varyings which are unused by the other stage by temporary variables. It also breaks down the gl_TexCoord array into separate vec4 variables if needed. Here's the result: This sounds similar to the way Paul's varying packing works. Is there synergy there? Also, since variables are renamed, does this interact with transform feedback? The queries of GL_ARB_program_interface_query? I suspect that it won't since this only affects built-in varyings, and built-in varyings aren't usable with those interfaces. Vertex shader (3 varyings instead of 8): ... (declare (shader_out ) vec4 gl_out_TexCoord1) (declare (shader_out ) vec4 gl_out_TexCoord4) (declare (temporary ) vec4 gl_out_TexCoord5_dummy) (declare (temporary ) vec4 gl_out_FrontColor0_dummy) (declare (shader_out ) vec4 gl_FrontSecondaryColor) (function main (signature void (parameters ) ( ... (assign (xyzw) (var_ref gl_out_FrontColor0_dummy) (var_ref gl_Color) ) (assign (xyzw) (var_ref gl_FrontSecondaryColor) (var_ref gl_SecondaryColor) ) (assign (xyzw) (var_ref gl_out_TexCoord1) (var_ref gl_MultiTexCoord1) ) (assign (xyzw) (var_ref gl_out_TexCoord4) (var_ref gl_MultiTexCoord4) ) (assign (xyzw) (var_ref gl_out_TexCoord5_dummy) (var_ref gl_MultiTexCoord5) ) )) ) Fragment shader (3 varyings instead of 6): ... (declare (shader_in ) vec4 gl_in_TexCoord1) (declare (temporary ) vec4 gl_in_TexCoord2_dummy) (declare (temporary ) vec4 gl_in_TexCoord3_dummy) (declare (shader_in ) vec4 gl_in_TexCoord4) (declare (shader_in ) vec4 gl_SecondaryColor) (function main (signature void (parameters ) ( (declare () vec4 r) (assign (xyzw) (var_ref r) ... (var_ref gl_SecondaryColor) ) ) (assign (xyzw) (var_ref r) ... (var_ref gl_in_TexCoord1) ) ) ) (assign (xyzw) (var_ref r) ... (var_ref gl_in_TexCoord2_dummy) ) ) ) (assign (xyzw) (var_ref r) ... (var_ref gl_in_TexCoord3_dummy) ) ) ) (declare (temporary ) vec4 assignment_tmp) (assign (xyzw) (var_ref assignment_tmp) ... (var_ref
Re: [Mesa-dev] [PATCH 0/6] Eliminating unused built-in varyings
On 06/28/2013 08:42 AM, Ian Romanick wrote: On 06/13/2013 05:25 AM, Marek Olšák wrote: Hi everyone, this series adds a new GLSL compiler optimization pass which eliminates unused and set-but-unused built-in varyings and adds a few improvements to the GLSL linker in the process. Before I show you how it works, I wanna say that there are patches which are related to and will most probably conflict with the geometry shader work, but they are necessary because the linkage of varyings is largely suboptimal. Also, the GL_EXT_separate_shader_objects extension must be disabled for this optimization to be enabled. The reason is a program object with both a VS and FS can be bound partially, e.g. by glUseShaderProgramEXT(GL_VERTEX_SHADER, prog), so the extension makes every program object be just a set of separate shaders. The extension is not important anyway. Could you elaborate on this a bit? The problem already exists that shader can be par-linked (e.g., just a vertex shader) and used with fixed-function. A big part of the reason that I implemented EXT_sso back in the day was to generate infrastructure, etc. for better using shaders with fixed function. Now, to illustrate how the optimization works, consider these 2 shader IR dumps: Vertex shader (8 varyings): ... (declare (shader_out ) vec4 gl_FrontColor) (declare (shader_out ) vec4 gl_FrontSecondaryColor) (declare (shader_out ) (array vec4 6) gl_TexCoord) (function main (signature void (parameters ) ( ... (assign (xyzw) (var_ref gl_FrontColor) (var_ref gl_Color) ) (assign (xyzw) (var_ref gl_FrontSecondaryColor) (var_ref gl_SecondaryColor) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (1)) ) (var_ref gl_MultiTexCoord1) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (4)) ) (var_ref gl_MultiTexCoord4) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (5)) ) (var_ref gl_MultiTexCoord5) ) )) ) Fragment shader (6 varyings): ... (declare (shader_in ) vec4 gl_SecondaryColor) (declare (shader_in ) (array vec4 5) gl_TexCoord) (function main (signature void (parameters ) ( (declare () vec4 r) (assign (xyzw) (var_ref r) ... (var_ref gl_SecondaryColor) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (1)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (2)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (3)) ) ) ) ) (declare (temporary ) vec4 assignment_tmp) (assign (xyzw) (var_ref assignment_tmp) ... (array_ref (var_ref gl_TexCoord) (constant int (4)) ) ) ) ) ... )) ) Note that only gl_TexCoord[1], gl_TexCoord[4], and gl_SecondaryColor are used by both shaders. The optimization replaces all occurences of varyings which are unused by the other stage by temporary variables. It also breaks down the gl_TexCoord array into separate vec4 variables if needed. Here's the result: This sounds similar to the way Paul's varying packing works. Is there synergy there? Also, since variables are renamed, does this interact with transform feedback? The queries of GL_ARB_program_interface_query? I suspect that it won't since this only affects built-in varyings, and built-in varyings aren't usable with those interfaces. I take part of that back. You can do transform feedback on built-in varyings. I had that crossed with not being able to do XFB on the output of fixed-function. Vertex shader (3 varyings instead of 8): ... (declare (shader_out ) vec4 gl_out_TexCoord1) (declare (shader_out ) vec4 gl_out_TexCoord4) (declare (temporary ) vec4 gl_out_TexCoord5_dummy) (declare (temporary ) vec4 gl_out_FrontColor0_dummy) (declare (shader_out ) vec4 gl_FrontSecondaryColor) (function main (signature void (parameters ) ( ... (assign (xyzw) (var_ref gl_out_FrontColor0_dummy) (var_ref gl_Color) ) (assign (xyzw) (var_ref gl_FrontSecondaryColor) (var_ref gl_SecondaryColor) ) (assign (xyzw) (var_ref gl_out_TexCoord1) (var_ref gl_MultiTexCoord1) ) (assign (xyzw) (var_ref gl_out_TexCoord4) (var_ref gl_MultiTexCoord4) ) (assign (xyzw) (var_ref gl_out_TexCoord5_dummy) (var_ref gl_MultiTexCoord5) ) )) ) Fragment shader (3 varyings instead of 6): ... (declare (shader_in ) vec4 gl_in_TexCoord1) (declare (temporary ) vec4 gl_in_TexCoord2_dummy) (declare (temporary ) vec4 gl_in_TexCoord3_dummy) (declare (shader_in ) vec4 gl_in_TexCoord4) (declare (shader_in ) vec4 gl_SecondaryColor) (function main (signature void (parameters ) ( (declare () vec4 r) (assign (xyzw) (var_ref r) ... (var_ref gl_SecondaryColor) ) ) (assign (xyzw) (var_ref r) ... (var_ref gl_in_TexCoord1) ) ) ) (assign (xyzw) (var_ref r) ... (var_ref gl_in_TexCoord2_dummy) ) ) )
Re: [Mesa-dev] [PATCH 4/6] glsl/linker: eliminate unused and set-but-unused built-in varyings
On 06/13/2013 05:25 AM, Marek Olšák wrote:
This eliminates built-in varyings such as gl_Color, gl_SecondaryColor,
gl_TexCoord, and gl_FogFragCoord if they are unused by the next stage or
not written at all (e.g. gl_TexCoord elements). The gl_TexCoord array is
broken down into separate vec4s if needed.
---
src/glsl/Makefile.sources |1 +
src/glsl/ir_optimization.h |4 +
src/glsl/link_varyings.h |4 +
src/glsl/linker.cpp| 13 +-
src/glsl/opt_dead_builtin_varyings.cpp | 468
5 files changed, 488 insertions(+), 2 deletions(-)
create mode 100644 src/glsl/opt_dead_builtin_varyings.cpp
diff --git a/src/glsl/Makefile.sources b/src/glsl/Makefile.sources
index 50bad85..cb17cf8 100644
--- a/src/glsl/Makefile.sources
+++ b/src/glsl/Makefile.sources
@@ -81,6 +81,7 @@ LIBGLSL_FILES = \
$(GLSL_SRCDIR)/opt_constant_variable.cpp \
$(GLSL_SRCDIR)/opt_copy_propagation.cpp \
$(GLSL_SRCDIR)/opt_copy_propagation_elements.cpp \
+ $(GLSL_SRCDIR)/opt_dead_builtin_varyings.cpp \
$(GLSL_SRCDIR)/opt_dead_code.cpp \
$(GLSL_SRCDIR)/opt_dead_code_local.cpp \
$(GLSL_SRCDIR)/opt_dead_functions.cpp \
diff --git a/src/glsl/ir_optimization.h b/src/glsl/ir_optimization.h
index d38d5e3..fad6f1b 100644
--- a/src/glsl/ir_optimization.h
+++ b/src/glsl/ir_optimization.h
@@ -76,6 +76,10 @@ bool do_constant_variable_unlinked(exec_list *instructions);
bool do_copy_propagation(exec_list *instructions);
bool do_copy_propagation_elements(exec_list *instructions);
bool do_constant_propagation(exec_list *instructions);
+void do_dead_builtin_varyings(struct gl_context *ctx,
+ exec_list *producer, exec_list *consumer,
+ unsigned num_tfeedback_decls,
+ class tfeedback_decl *tfeedback_decls);
bool do_dead_code(exec_list *instructions, bool uniform_locations_assigned);
bool do_dead_code_local(exec_list *instructions);
bool do_dead_code_unlinked(exec_list *instructions);
diff --git a/src/glsl/link_varyings.h b/src/glsl/link_varyings.h
index daa9d79..7f7be35 100644
--- a/src/glsl/link_varyings.h
+++ b/src/glsl/link_varyings.h
@@ -125,6 +125,10 @@ public:
return this-vector_elements * this-matrix_columns * this-size;
}
+ unsigned get_location() const {
+ return this-location;
+ }
+
private:
/**
* The name that was supplied to glTransformFeedbackVaryings. Used for
diff --git a/src/glsl/linker.cpp b/src/glsl/linker.cpp
index a8537cf..129b665 100644
--- a/src/glsl/linker.cpp
+++ b/src/glsl/linker.cpp
@@ -1887,6 +1887,9 @@ link_shaders(struct gl_context *ctx, struct
gl_shader_program *prog)
goto done;
}
+ do_dead_builtin_varyings(ctx, sh-ir, NULL,
+ num_tfeedback_decls, tfeedback_decls);
+
demote_shader_inputs_and_outputs(sh, ir_var_shader_out);
/* Eliminate code that is now dead due to unused outputs being demoted.
@@ -1895,11 +1898,13 @@ link_shaders(struct gl_context *ctx, struct
gl_shader_program *prog)
;
}
else if (first == MESA_SHADER_FRAGMENT) {
- /* If the program only contains a fragment shader, just demote
- * user-defined varyings.
+ /* If the program only contains a fragment shader...
*/
gl_shader *const sh = prog-_LinkedShaders[first];
+ do_dead_builtin_varyings(ctx, NULL, sh-ir,
+ num_tfeedback_decls, tfeedback_decls);
+
demote_shader_inputs_and_outputs(sh, ir_var_shader_in);
while (do_dead_code(sh-ir, false))
@@ -1919,6 +1924,10 @@ link_shaders(struct gl_context *ctx, struct
gl_shader_program *prog)
tfeedback_decls))
goto done;
+ do_dead_builtin_varyings(ctx, sh_i-ir, sh_next-ir,
+next == MESA_SHADER_FRAGMENT ? num_tfeedback_decls : 0,
+tfeedback_decls);
+
demote_shader_inputs_and_outputs(sh_i, ir_var_shader_out);
demote_shader_inputs_and_outputs(sh_next, ir_var_shader_in);
diff --git a/src/glsl/opt_dead_builtin_varyings.cpp
b/src/glsl/opt_dead_builtin_varyings.cpp
new file mode 100644
index 000..eb99d1e
--- /dev/null
+++ b/src/glsl/opt_dead_builtin_varyings.cpp
@@ -0,0 +1,468 @@
+/*
+ * Copyright © 2013 Marek Olšák mar...@gmail.com
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the Software),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ *
[Mesa-dev] [Bug 66149] [HSW] Background of application icon has garbage after update kernelmesa in order to support Graphics [8086: 0a2e]
https://bugs.freedesktop.org/show_bug.cgi?id=66149 --- Comment #7 from Ian Romanick i...@freedesktop.org --- Eva, can you clarify for me. 9.1 works correctly, but master does not? If the bug still exists on master, the bug should not be closed. -- 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 0/6] Eliminating unused built-in varyings
On Fri, Jun 28, 2013 at 5:42 PM, Ian Romanick i...@freedesktop.org wrote: On 06/13/2013 05:25 AM, Marek Olšák wrote: Hi everyone, this series adds a new GLSL compiler optimization pass which eliminates unused and set-but-unused built-in varyings and adds a few improvements to the GLSL linker in the process. Before I show you how it works, I wanna say that there are patches which are related to and will most probably conflict with the geometry shader work, but they are necessary because the linkage of varyings is largely suboptimal. Also, the GL_EXT_separate_shader_objects extension must be disabled for this optimization to be enabled. The reason is a program object with both a VS and FS can be bound partially, e.g. by glUseShaderProgramEXT(GL_VERTEX_SHADER, prog), so the extension makes every program object be just a set of separate shaders. The extension is not important anyway. Could you elaborate on this a bit? The problem already exists that shader can be par-linked (e.g., just a vertex shader) and used with fixed-function. A big part of the reason that I implemented EXT_sso back in the day was to generate infrastructure, etc. for better using shaders with fixed function. The only problem I have with EXT_sso is that you can link two programs, both having a vertex and fragment shader, and still mix-and-match their shaders independently as if all the shaders were separate/unlinked. For example: /* equivalent to glUseProgram(prog1); */ glUseShaderProgramEXT(GL_VERTEX_SHADER, prog1); glUseShaderProgramEXT(GL_FRAGMENT_SHADER, prog1); /* prog1 has its own fragment shader, but who cares! we don't have to bind it */ /* prog2 also has its own vertex shader */ glUseShaderProgramEXT(GL_VERTEX_SHADER, prog1); glUseShaderProgramEXT(GL_FRAGMENT_SHADER, prog2); /* more fun, we can put a geometry shader in between */ glUseShaderProgramEXT(GL_VERTEX_SHADER, prog1); glUseShaderProgramEXT(GL_GEOMETRY_SHADER, prog3); glUseShaderProgramEXT(GL_FRAGMENT_SHADER, prog1); /* assume prog3 has all 3 shaders, we can unbind the middle one */ glUseProgram(prog3); glUseShaderProgramEXT(GL_GEOMETRY_SHADER, NULL); I have no problem with linking program objects with a single shader and mixing and matching such program objects. However the ability to mix-and-match shaders no matter what program object they are part of is a real show-stopper for this optimization. Thankfully, this doesn't happen with fixed function and ARB_sso also forbids such usage. Now, to illustrate how the optimization works, consider these 2 shader IR dumps: Vertex shader (8 varyings): ... (declare (shader_out ) vec4 gl_FrontColor) (declare (shader_out ) vec4 gl_FrontSecondaryColor) (declare (shader_out ) (array vec4 6) gl_TexCoord) (function main (signature void (parameters ) ( ... (assign (xyzw) (var_ref gl_FrontColor) (var_ref gl_Color) ) (assign (xyzw) (var_ref gl_FrontSecondaryColor) (var_ref gl_SecondaryColor) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (1)) ) (var_ref gl_MultiTexCoord1) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (4)) ) (var_ref gl_MultiTexCoord4) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (5)) ) (var_ref gl_MultiTexCoord5) ) )) ) Fragment shader (6 varyings): ... (declare (shader_in ) vec4 gl_SecondaryColor) (declare (shader_in ) (array vec4 5) gl_TexCoord) (function main (signature void (parameters ) ( (declare () vec4 r) (assign (xyzw) (var_ref r) ... (var_ref gl_SecondaryColor) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (1)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (2)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (3)) ) ) ) ) (declare (temporary ) vec4 assignment_tmp) (assign (xyzw) (var_ref assignment_tmp) ... (array_ref (var_ref gl_TexCoord) (constant int (4)) ) ) ) ) ... )) ) Note that only gl_TexCoord[1], gl_TexCoord[4], and gl_SecondaryColor are used by both shaders. The optimization replaces all occurences of varyings which are unused by the other stage by temporary variables. It also breaks down the gl_TexCoord array into separate vec4 variables if needed. Here's the result: This sounds similar to the way Paul's varying packing works. Is there synergy there? Also, since variables are renamed, does this interact with transform feedback? The queries of GL_ARB_program_interface_query? I suspect that it won't since this only affects built-in varyings, and built-in varyings aren't usable with those interfaces. I haven't followed the varying packing work really, so I'm not sure, but it seemed to be only about user-defined varyings, while my work is only about built-in varyings. Concerning
Re: [Mesa-dev] [PATCH v2] r600g: implement fast color clears on evergreen+
On 12.06.2013 00:04, Grigori Goronzy wrote: Allows MSAA colorbuffers, which have a CMASK automatically and don't need any further special handling, to be fast cleared. Instead of clearing the buffer, set the clear color and the CMASK to the cleared state. Fast clear is used only when all bound colorbuffers fulfill certain conditions: a CMASK is required, we have to be able to create a clear color value for the format and the texture mustn't contain multiple images. Technically, it should be possible to support array textures and cubemaps if all images are attached to the framebuffer, but this does not appear to be common. v2: fix fast clear check Ping? Can anyone please review this? Best regards Grigori ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] prog_parameter.c ASAN Patch
Friendly ping regarding this patch :-)
--Myles
On Wed, Jun 19, 2013 at 12:47 AM, Myles C. Maxfield
myles.maxfi...@gmail.com wrote:
Any word on this?
Thanks,
Myles
On Mon, Jun 17, 2013 at 12:09 PM, Myles C. Maxfield
myles.maxfi...@gmail.com wrote:
Sure. I was under the impression that |size| couldn't be both greater
than 4 and a non-multiple of 4, but I've reworked the patch to incorporate
this and to be a little more straightforward.
Is the only way to replace ASAN with Address Sanitizer to change the
subject of this email thread?
Anyway, here's a similar but modified patch:
From: Myles C. Maxfield my...@amazon.com
Date: Mon, 17 Jun 2013 11:50:05 -0700
Subject: [PATCH] Appeasing Address Sanitizer
---
src/mesa/program/prog_parameter.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/src/mesa/program/prog_parameter.c
b/src/mesa/program/prog_parameter.c
index 95b153e..1d46476 100644
--- a/src/mesa/program/prog_parameter.c
+++ b/src/mesa/program/prog_parameter.c
@@ -155,7 +155,18 @@ _mesa_add_parameter(struct gl_program_parameter_list
*paramList,
p-Size = size;
p-DataType = datatype;
if (values) {
-COPY_4V(paramList-ParameterValues[oldNum + i], values);
+if (size = (i+1)*4) {
+COPY_4V(paramList-ParameterValues[oldNum + i], values);
+} else {
+/* silence asan */
+for (j = 0; j 4; j++) {
+if (i*4+j size) {
+paramList-ParameterValues[oldNum + i][j] =
values[i*4+j];
+} else {
+paramList-ParameterValues[oldNum + i][j].f =
0.0f;
+}
+}
+}
values += 4;
p-Initialized = GL_TRUE;
}
--
1.7.12.4 (Apple Git-37)
On Mon, Jun 17, 2013 at 8:13 AM, Brian Paul bri...@vmware.com wrote:
On 06/14/2013 05:12 PM, Myles C. Maxfield wrote:
Sorry for the triple post; I received a bounce email the first time and
got sent to the spam folder the second time, so I'm trying a third time.
Hello, all. I was running Mesa with Address Sanitizer [1] turned on,
and found one place where ASAN pointed out a read-before-initialized
problem. In particular, in _mesa_add_parameter, in prog_parameter.c,
|values| represents an array holding a variable number of values. These
values get copied out of the array 4 at a time with the COPY_4V macro,
however, the array might only contain a single element. In this case, ASAN
reports a read-before-initialize because the last 3 of the 4 elements
haven't been written to yet. I was hoping to contribute a patch that will
silence this problem that ASAN reports. I'm happy to incorporate any
feedback anyone has into this patch.
Thanks,
Myles C. Maxfield
[1]https://code.google.com/p/**address-sanitizer/https://code.google.com/p/address-sanitizer/
diff --git a/src/mesa/program/prog_**parameter.c
b/src/mesa/program/prog_**parameter.c
index 2018fa5..63915fb 100644
--- a/src/mesa/program/prog_**parameter.c
+++ b/src/mesa/program/prog_**parameter.c
@@ -158,7 +158,17 @@ _mesa_add_parameter(struct
gl_program_parameter_list *paramList,
p-DataType = datatype;
p-Flags = flags;
if (values) {
-COPY_4V(paramList-**ParameterValues[oldNum + i], values);
+if (size 3) {
+ for (j = 0; j size; j++) {
+paramList-ParameterValues[**oldNum + i][j] =
values[j];
+ }
+ /* silence asan */
+ for (j = size; j 4; j++) {
+paramList-ParameterValues[**oldNum + i][j].f = 0;
+ }
+} else {
+ COPY_4V(paramList-**ParameterValues[oldNum + i],
values);
+}
values += 4;
p-Initialized = GL_TRUE;
}
The value of 'size' can actually be greater than 4 (IIRC, and the
function comment are still correct). For example, for a matrix, size=16.
So the first for-loop should be fixed, just to be safe.
In the commit message, let's not use ASAN since it's not obvious that
it means Address Sanitizer.
-Brian
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] R600/SI: Add processor types for each CIK variant
From: Alex Deucher alexander.deuc...@amd.com Signed-off-by: Alex Deucher alexander.deuc...@amd.com --- lib/Target/R600/Processors.td |3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/lib/Target/R600/Processors.td b/lib/Target/R600/Processors.td index 81f407e..a0735d4 100644 --- a/lib/Target/R600/Processors.td +++ b/lib/Target/R600/Processors.td @@ -48,3 +48,6 @@ def : Procpitcairn, SI_Itin, [FeatureSouthernIslands]; def : Procverde, SI_Itin, [FeatureSouthernIslands]; def : Procoland, SI_Itin, [FeatureSouthernIslands]; def : Prochainan, SI_Itin, [FeatureSouthernIslands]; +def : Procbonaire,SI_Itin, [FeatureSouthernIslands]; +def : Prockabini, SI_Itin, [FeatureSouthernIslands]; +def : Prockaveri, SI_Itin, [FeatureSouthernIslands]; \ No newline at end of file -- 1.7.7.5 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] llvmpipe: fix timer query if there's no bins
Looks good.
- Original Message -
From: Roland Scheidegger srol...@vmware.com
b04a295a4a0cd2defe352b3193b5fa79ca8fc9fc removed seemingly unnecessary
code in get_query. Turns out this code could in fact be reached - while
timestamps are always binned, if there are no bins (which happens if fb
size is 0) then the rasterization query code filling this in is still
never executed.
So fix this up by filling in some timestamp, but do it at EndQuery time
not GetQuery time which should be more appropriate.
Makes piglit arb_timer_query-timestamp-get happy again.
---
src/gallium/drivers/llvmpipe/lp_setup.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/src/gallium/drivers/llvmpipe/lp_setup.c
b/src/gallium/drivers/llvmpipe/lp_setup.c
index 49b61c3..65f61ed 100644
--- a/src/gallium/drivers/llvmpipe/lp_setup.c
+++ b/src/gallium/drivers/llvmpipe/lp_setup.c
@@ -40,6 +40,7 @@
#include util/u_memory.h
#include util/u_pack_color.h
#include draw/draw_pipe.h
+#include os/os_time.h
#include lp_context.h
#include lp_memory.h
#include lp_scene.h
@@ -1263,6 +1264,15 @@ lp_setup_end_query(struct lp_setup_context *setup,
struct llvmpipe_query *pq)
pq-type == PIPE_QUERY_OCCLUSION_PREDICATE ||
pq-type == PIPE_QUERY_PIPELINE_STATISTICS ||
pq-type == PIPE_QUERY_TIMESTAMP) {
+ if (pq-type == PIPE_QUERY_TIMESTAMP
+ !(setup-scene-tiles_x | setup-scene-tiles_y)) {
+/*
+ * If there's a zero width/height framebuffer, there's no bins
and
+ * hence no rast task is ever run. So fill in something here
instead.
+ */
+pq-end[0] = os_time_get_nano();
+ }
+
if (!lp_scene_bin_everywhere(setup-scene,
LP_RAST_OP_END_QUERY,
lp_rast_arg_query(pq))) {
--
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 0/6] Eliminating unused built-in varyings
On 06/28/2013 10:55 AM, Marek Olšák wrote: On Fri, Jun 28, 2013 at 5:42 PM, Ian Romanick i...@freedesktop.org wrote: On 06/13/2013 05:25 AM, Marek Olšák wrote: Hi everyone, this series adds a new GLSL compiler optimization pass which eliminates unused and set-but-unused built-in varyings and adds a few improvements to the GLSL linker in the process. Before I show you how it works, I wanna say that there are patches which are related to and will most probably conflict with the geometry shader work, but they are necessary because the linkage of varyings is largely suboptimal. Also, the GL_EXT_separate_shader_objects extension must be disabled for this optimization to be enabled. The reason is a program object with both a VS and FS can be bound partially, e.g. by glUseShaderProgramEXT(GL_VERTEX_SHADER, prog), so the extension makes every program object be just a set of separate shaders. The extension is not important anyway. Could you elaborate on this a bit? The problem already exists that shader can be par-linked (e.g., just a vertex shader) and used with fixed-function. A big part of the reason that I implemented EXT_sso back in the day was to generate infrastructure, etc. for better using shaders with fixed function. The only problem I have with EXT_sso is that you can link two programs, both having a vertex and fragment shader, and still mix-and-match their shaders independently as if all the shaders were separate/unlinked. For example: Oh, I had forgotten all about that. I remember thinking the spec authors must have been drunk when they didn't add language to prevent that usage. That makes perfect sense, then. It sounds like I should get off my butt, land Gergory Hainaut's ARB_sso work, and gut out EXT_sso. /* equivalent to glUseProgram(prog1); */ glUseShaderProgramEXT(GL_VERTEX_SHADER, prog1); glUseShaderProgramEXT(GL_FRAGMENT_SHADER, prog1); /* prog1 has its own fragment shader, but who cares! we don't have to bind it */ /* prog2 also has its own vertex shader */ glUseShaderProgramEXT(GL_VERTEX_SHADER, prog1); glUseShaderProgramEXT(GL_FRAGMENT_SHADER, prog2); /* more fun, we can put a geometry shader in between */ glUseShaderProgramEXT(GL_VERTEX_SHADER, prog1); glUseShaderProgramEXT(GL_GEOMETRY_SHADER, prog3); glUseShaderProgramEXT(GL_FRAGMENT_SHADER, prog1); /* assume prog3 has all 3 shaders, we can unbind the middle one */ glUseProgram(prog3); glUseShaderProgramEXT(GL_GEOMETRY_SHADER, NULL); I have no problem with linking program objects with a single shader and mixing and matching such program objects. However the ability to mix-and-match shaders no matter what program object they are part of is a real show-stopper for this optimization. Thankfully, this doesn't happen with fixed function and ARB_sso also forbids such usage. Now, to illustrate how the optimization works, consider these 2 shader IR dumps: Vertex shader (8 varyings): ... (declare (shader_out ) vec4 gl_FrontColor) (declare (shader_out ) vec4 gl_FrontSecondaryColor) (declare (shader_out ) (array vec4 6) gl_TexCoord) (function main (signature void (parameters ) ( ... (assign (xyzw) (var_ref gl_FrontColor) (var_ref gl_Color) ) (assign (xyzw) (var_ref gl_FrontSecondaryColor) (var_ref gl_SecondaryColor) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (1)) ) (var_ref gl_MultiTexCoord1) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (4)) ) (var_ref gl_MultiTexCoord4) ) (assign (xyzw) (array_ref (var_ref gl_TexCoord) (constant int (5)) ) (var_ref gl_MultiTexCoord5) ) )) ) Fragment shader (6 varyings): ... (declare (shader_in ) vec4 gl_SecondaryColor) (declare (shader_in ) (array vec4 5) gl_TexCoord) (function main (signature void (parameters ) ( (declare () vec4 r) (assign (xyzw) (var_ref r) ... (var_ref gl_SecondaryColor) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (1)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (2)) ) ) ) ) (assign (xyzw) (var_ref r) ... (array_ref (var_ref gl_TexCoord) (constant int (3)) ) ) ) ) (declare (temporary ) vec4 assignment_tmp) (assign (xyzw) (var_ref assignment_tmp) ... (array_ref (var_ref gl_TexCoord) (constant int (4)) ) ) ) ) ... )) ) Note that only gl_TexCoord[1], gl_TexCoord[4], and gl_SecondaryColor are used by both shaders. The optimization replaces all occurences of varyings which are unused by the other stage by temporary variables. It also breaks down the gl_TexCoord array into separate vec4 variables if needed. Here's the result: This sounds similar to the way Paul's varying packing works. Is there synergy there? Also, since variables are renamed, does this interact with transform feedback? The queries of GL_ARB_program_interface_query? I suspect that it won't
Re: [Mesa-dev] [PATCH] R600/SI: Add processor types for each CIK variant
On Fri, Jun 28, 2013 at 03:05:04PM -0400, alexdeuc...@gmail.com wrote: From: Alex Deucher alexander.deuc...@amd.com Signed-off-by: Alex Deucher alexander.deuc...@amd.com Committed, thanks! -Tom --- lib/Target/R600/Processors.td |3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/lib/Target/R600/Processors.td b/lib/Target/R600/Processors.td index 81f407e..a0735d4 100644 --- a/lib/Target/R600/Processors.td +++ b/lib/Target/R600/Processors.td @@ -48,3 +48,6 @@ def : Procpitcairn, SI_Itin, [FeatureSouthernIslands]; def : Procverde, SI_Itin, [FeatureSouthernIslands]; def : Procoland, SI_Itin, [FeatureSouthernIslands]; def : Prochainan, SI_Itin, [FeatureSouthernIslands]; +def : Procbonaire,SI_Itin, [FeatureSouthernIslands]; +def : Prockabini, SI_Itin, [FeatureSouthernIslands]; +def : Prockaveri, SI_Itin, [FeatureSouthernIslands]; \ No newline at end of file -- 1.7.7.5 ___ 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
[Mesa-dev] [PATCH] glsl/builtins: Fix ARB_texture_cube_map_array built-in availability.
This patch adds texture() for isamplerCubeArray and usamplerCubeArray, which were entirely missing. It also makes texture() with a LOD bias fragment shader specific. The main GLSL specification explicitly says that texturing with LOD bias should not be allowed for vertex shaders. Affects Piglit's ARB_texture_cube_map_array/compiler/tex_bias-01.vert. which tries to use bias in a vertex shader. Currently, it expects this to pass (so this patch regresses the test), but I've sent a patch to reverse the expected behavior (so this patch would fix the updated test): http://lists.freedesktop.org/archives/piglit/2013-June/006123.html NOTE: This is a candidate for stable branches. Cc: Paul Berry stereotype...@gmail.com Signed-off-by: Kenneth Graunke kenn...@whitecape.org --- src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag | 6 ++ src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl | 3 ++- 2 files changed, 8 insertions(+), 1 deletion(-) create mode 100644 src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag diff --git a/src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag new file mode 100644 index 000..0d9f4f6 --- /dev/null +++ b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag @@ -0,0 +1,6 @@ +#version 130 +#extension GL_ARB_texture_cube_map_array : enable + + vec4 texture( samplerCubeArray sampler, vec4 coord, float bias); +ivec4 texture(isamplerCubeArray sampler, vec4 coord, float bias); +uvec4 texture(usamplerCubeArray sampler, vec4 coord, float bias); diff --git a/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl index 0f53212..73659b3 100644 --- a/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl +++ b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl @@ -7,7 +7,8 @@ ivec3 textureSize(usamplerCubeArray sampler, int lod); ivec3 textureSize(samplerCubeArrayShadow sampler, int lod); vec4 texture( samplerCubeArray sampler, vec4 coord); - vec4 texture( samplerCubeArray sampler, vec4 coord, float bias); +ivec4 texture(isamplerCubeArray sampler, vec4 coord); +uvec4 texture(usamplerCubeArray sampler, vec4 coord); float texture( samplerCubeArrayShadow sampler, vec4 P, float compare); vec4 textureGrad( samplerCubeArray sampler, vec4 P, vec3 dPdx, vec3 dPdy); -- 1.8.3.1 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] draw/translate: fix instancing
We were incorrectly computing the buffer offset when using the
instances. The buffer offset is always equal to:
start_instance * stride + (instance_num / instance_divisor) *
stride
We were completely ignoring the start instance quite
often producing instances that completely wrong, e.g. if
start instance = 5, instance divisor = 2, then on the first
iteration it should be:
5 * stride, not (5/2) * stride as we'd have currently, and if
start instance = 1, instance divisor = 3, then on the first
iteration it should be:
1 * stride, not 0 as we'd have.
This fixes it and adjusts all the code to the changes.
Signed-off-by: Zack Rusin za...@vmware.com
---
src/gallium/auxiliary/draw/draw_llvm.c | 15 ++---
src/gallium/auxiliary/draw/draw_pipe_vbuf.c|2 +-
src/gallium/auxiliary/draw/draw_private.h |1 +
src/gallium/auxiliary/draw/draw_pt.c |1 +
src/gallium/auxiliary/draw/draw_pt_emit.c |2 ++
src/gallium/auxiliary/draw/draw_pt_fetch.c |2 ++
src/gallium/auxiliary/draw/draw_pt_fetch_emit.c|3 ++
src/gallium/auxiliary/draw/draw_pt_so_emit.c | 23 --
src/gallium/auxiliary/draw/draw_vs_variant.c |4 +++
src/gallium/auxiliary/translate/translate.h|4 +++
.../auxiliary/translate/translate_generic.c| 17 ---
src/gallium/auxiliary/translate/translate_sse.c| 32
src/gallium/auxiliary/util/u_vbuf.c|8 ++---
src/gallium/drivers/nv30/nv30_push.c |8 ++---
src/gallium/drivers/nv50/nv50_push.c |8 ++---
src/gallium/drivers/nvc0/nvc0_push.c |8 ++---
src/gallium/drivers/nvc0/nvc0_vbo_translate.c |8 ++---
17 files changed, 106 insertions(+), 40 deletions(-)
diff --git a/src/gallium/auxiliary/draw/draw_llvm.c
b/src/gallium/auxiliary/draw/draw_llvm.c
index 97b463f..9eb5a93 100644
--- a/src/gallium/auxiliary/draw/draw_llvm.c
+++ b/src/gallium/auxiliary/draw/draw_llvm.c
@@ -674,6 +674,7 @@ generate_vs(struct draw_llvm_variant *variant,
static void
generate_fetch(struct gallivm_state *gallivm,
+ struct draw_context *draw,
LLVMValueRef vbuffers_ptr,
LLVMValueRef *res,
struct pipe_vertex_element *velem,
@@ -704,10 +705,14 @@ generate_fetch(struct gallivm_state *gallivm,
struct lp_build_if_state if_ctx;
if (velem-instance_divisor) {
- /* array index = instance_id / instance_divisor */
- index = LLVMBuildUDiv(builder, instance_id,
-lp_build_const_int32(gallivm,
velem-instance_divisor),
-instance_divisor);
+ LLVMValueRef current_instance;
+ /* array index = start_instance + (instance_num / instance_divisor) */
+ index = lp_build_const_int32(gallivm, draw-start_instance);
+ current_instance = LLVMBuildSub(builder, instance_id, index, );
+ current_instance = LLVMBuildUDiv(builder, current_instance,
+ lp_build_const_int32(gallivm,
velem-instance_divisor),
+ instance_divisor);
+ index = LLVMBuildAdd(builder, index, current_instance, instance);
}
stride = lp_build_umul_overflow(gallivm, vb_stride, index, ofbit);
@@ -1697,7 +1702,7 @@ draw_llvm_generate(struct draw_llvm *llvm, struct
draw_llvm_variant *variant,
LLVMValueRef vb_index =
lp_build_const_int32(gallivm, velem-vertex_buffer_index);
LLVMValueRef vb = LLVMBuildGEP(builder, vb_ptr, vb_index, 1, );
-generate_fetch(gallivm, vbuffers_ptr,
+generate_fetch(gallivm, draw, vbuffers_ptr,
aos_attribs[j][i], velem, vb, true_index,
system_values.instance_id);
}
diff --git a/src/gallium/auxiliary/draw/draw_pipe_vbuf.c
b/src/gallium/auxiliary/draw/draw_pipe_vbuf.c
index 578433c..d3b38eb 100644
--- a/src/gallium/auxiliary/draw/draw_pipe_vbuf.c
+++ b/src/gallium/auxiliary/draw/draw_pipe_vbuf.c
@@ -138,7 +138,7 @@ emit_vertex( struct vbuf_stage *vbuf,
/* Note: we really do want data[0] here, not data[pos]:
*/
vbuf-translate-set_buffer(vbuf-translate, 0, vertex-data[0], 0, ~0);
- vbuf-translate-run(vbuf-translate, 0, 1, 0, vbuf-vertex_ptr);
+ vbuf-translate-run(vbuf-translate, 0, 1, 0, 0, vbuf-vertex_ptr);
if (0) draw_dump_emitted_vertex(vbuf-vinfo, (uint8_t
*)vbuf-vertex_ptr);
diff --git a/src/gallium/auxiliary/draw/draw_private.h
b/src/gallium/auxiliary/draw/draw_private.h
index fd52c2d..f42cded 100644
--- a/src/gallium/auxiliary/draw/draw_private.h
+++ b/src/gallium/auxiliary/draw/draw_private.h
@@ -306,6 +306,7 @@ struct draw_context
} extra_shader_outputs;
unsigned instance_id;
+ unsigned start_instance;
#ifdef HAVE_LLVM
struct draw_llvm *llvm;
diff --git
Re: [Mesa-dev] R600/SI: Support for local memory and derivatives
On Wed, Jun 19, 2013 at 06:28:21PM +0200, Michel Dänzer wrote:
These patches implement enough of local memory support to allow radeonsi
to use that for computing derivatives, as suggested by Tom.
They also almost allow test/CodeGen/R600/local-memory.ll to generate
code for SI. Right now it still fails because it tries to copy a VGPR to
an SGPR, which is not possible.
Can you add some lit tests for these new intrinsics and also add CHECK
lines for SI to the existing local-memory.ll test.
With the tests added, these patches are:
Reviewed-by: Tom Stellard thomas.stell...@amd.com
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X and DRI developer
From f4ca359c4536aa53122b654196f2e007d50976f8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michel=20D=C3=A4nzer?= michel.daen...@amd.com
Date: Thu, 21 Feb 2013 16:12:45 +0100
Subject: [PATCH 1/6] R600/SI: Add intrinsics for texture sampling with user
derivatives
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Michel Dänzer michel.daen...@amd.com
---
lib/Target/R600/SIInstructions.td | 7 ++-
lib/Target/R600/SIIntrinsics.td | 1 +
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R600/SIInstructions.td
index 9c96c08..c9eac7d 100644
--- a/lib/Target/R600/SIInstructions.td
+++ b/lib/Target/R600/SIInstructions.td
@@ -535,7 +535,7 @@ def IMAGE_SAMPLE_B : MIMG_Sampler_Helper 0x0025,
IMAGE_SAMPLE_B;
//def IMAGE_SAMPLE_LZ : MIMG_NoPattern_ IMAGE_SAMPLE_LZ, 0x0027;
def IMAGE_SAMPLE_C : MIMG_Sampler_Helper 0x0028, IMAGE_SAMPLE_C;
//def IMAGE_SAMPLE_C_CL : MIMG_NoPattern_ IMAGE_SAMPLE_C_CL, 0x0029;
-//def IMAGE_SAMPLE_C_D : MIMG_NoPattern_ IMAGE_SAMPLE_C_D, 0x002a;
+def IMAGE_SAMPLE_C_D : MIMG_Sampler_Helper 0x002a, IMAGE_SAMPLE_C_D;
//def IMAGE_SAMPLE_C_D_CL : MIMG_NoPattern_ IMAGE_SAMPLE_C_D_CL,
0x002b;
def IMAGE_SAMPLE_C_L : MIMG_Sampler_Helper 0x002c, IMAGE_SAMPLE_C_L;
def IMAGE_SAMPLE_C_B : MIMG_Sampler_Helper 0x002d, IMAGE_SAMPLE_C_B;
@@ -1296,6 +1296,11 @@ multiclass SamplePatternsValueType addr_type {
def : SampleArrayPattern int_SI_sampleb, IMAGE_SAMPLE_B, addr_type;
def : SampleShadowPattern int_SI_sampleb, IMAGE_SAMPLE_C_B, addr_type;
def : SampleShadowArrayPattern int_SI_sampleb, IMAGE_SAMPLE_C_B,
addr_type;
+
+ def : SamplePattern int_SI_sampled, IMAGE_SAMPLE_D, addr_type;
+ def : SampleArrayPattern int_SI_sampled, IMAGE_SAMPLE_D, addr_type;
+ def : SampleShadowPattern int_SI_sampled, IMAGE_SAMPLE_C_D, addr_type;
+ def : SampleShadowArrayPattern int_SI_sampled, IMAGE_SAMPLE_C_D,
addr_type;
}
defm : SamplePatternsv2i32;
diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
index 224cd2f..d2643e0 100644
--- a/lib/Target/R600/SIIntrinsics.td
+++ b/lib/Target/R600/SIIntrinsics.td
@@ -23,6 +23,7 @@ let TargetPrefix = SI, isTarget = 1 in {
def int_SI_sample : Sample;
def int_SI_sampleb : Sample;
+ def int_SI_sampled : Sample;
def int_SI_samplel : Sample;
def int_SI_imageload : Intrinsic [llvm_v4i32_ty], [llvm_anyvector_ty,
llvm_v32i8_ty, llvm_i32_ty], [IntrNoMem];
--
1.8.3.1
From 7a0048bb2ab1b661831da2b764bf1a52f66bec15 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michel=20D=C3=A4nzer?= michel.daen...@amd.com
Date: Thu, 21 Feb 2013 18:51:38 +0100
Subject: [PATCH v3 2/6] R600/SI: Initial support for LDS/GDS instructions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Michel Dänzer michel.daen...@amd.com
---
v3: Drop vdst operand from DS_Store_Helper class, and adapt
SIInsertWaits::getHwCounts() to handle that. Unfortunately, this seems
to mess up the asm string output somehow, not sure what's going on
there.
lib/Target/R600/SIInsertWaits.cpp | 2 ++
lib/Target/R600/SIInstrFormats.td | 24
lib/Target/R600/SIInstrInfo.td | 23 +++
lib/Target/R600/SIInstructions.td | 3 +++
lib/Target/R600/SILowerControlFlow.cpp | 16
5 files changed, 68 insertions(+)
diff --git a/lib/Target/R600/SIInsertWaits.cpp
b/lib/Target/R600/SIInsertWaits.cpp
index c36e1dc..d31da45 100644
--- a/lib/Target/R600/SIInsertWaits.cpp
+++ b/lib/Target/R600/SIInsertWaits.cpp
@@ -134,6 +134,8 @@ Counters SIInsertWaits::getHwCounts(MachineInstr MI) {
if (TSFlags SIInstrFlags::LGKM_CNT) {
MachineOperand Op = MI.getOperand(0);
+if (!Op.isReg())
+ Op = MI.getOperand(1);
assert(Op.isReg() First LGKM operand must be a register!);
unsigned Reg = Op.getReg();
diff --git a/lib/Target/R600/SIInstrFormats.td
b/lib/Target/R600/SIInstrFormats.td
index 51f323d..434aa7e 100644
---
Re: [Mesa-dev] [PATCH 2/2] i965: Avoid flushing the batch for every blorp op.
Eric Anholt e...@anholt.net writes: This brings over the batch-wrap-prevention and aperture space checking code from the normal brw_draw.c path, so that we don't need to flush the batch every time. There's a risk here if the intel_emit_post_sync_nonzero_flush() call isn't high enough up in the state emit sequences -- before, we implicitly had one at the batch flush before any state was emitted, so Mesa's workaround emits didn't really matter. Improves cairo-gl performance by 13.7733% +/- 1.74876% (n=30/32) No statistically significant performance difference on unigine tropics (n=10) No statistically significant performance difference on openarena (n=755) No statistically significant performance difference on Lightsmark (n=15, though this may be an issue of test power -- looks like a ~.3% performance hit) Reduces low-resolution GLB 2.7 performance by 0.604517% +/- 0.140544% (n=132/133) --- I've got the test system running more Lightsmark now -- the bimodal distribution of its results was killing the stats, and I'd bumped the power cable and it ran out of battery and died. I'm a little mystified by the small GLB and possibly LM regressions. My theory was the first-post-swap-batch throttling, except that we've got about 5 batches per frame on GLB. Updated LM results: -2.45051% +/- 0.341284% (n=30/32). At this point I think we do need to figure out these regressions before landing the change. It's on the blorp-flush branch of my tree if someone wants to follow up while I'm gone. pgpNJU3z8Thrt.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] glsl/builtins: Fix ARB_texture_cube_map_array built-in availability.
On Sat, Jun 29, 2013 at 7:22 AM, Kenneth Graunke kenn...@whitecape.org wrote: This patch adds texture() for isamplerCubeArray and usamplerCubeArray, which were entirely missing. It also makes texture() with a LOD bias fragment shader specific. The main GLSL specification explicitly says that texturing with LOD bias should not be allowed for vertex shaders. Affects Piglit's ARB_texture_cube_map_array/compiler/tex_bias-01.vert. which tries to use bias in a vertex shader. Currently, it expects this to pass (so this patch regresses the test), but I've sent a patch to reverse the expected behavior (so this patch would fix the updated test): http://lists.freedesktop.org/archives/piglit/2013-June/006123.html No idea how I missed these. Reviewed-by: Dave Airlie airl...@redhat.com NOTE: This is a candidate for stable branches. Cc: Paul Berry stereotype...@gmail.com Signed-off-by: Kenneth Graunke kenn...@whitecape.org --- src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag | 6 ++ src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl | 3 ++- 2 files changed, 8 insertions(+), 1 deletion(-) create mode 100644 src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag diff --git a/src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag new file mode 100644 index 000..0d9f4f6 --- /dev/null +++ b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.frag @@ -0,0 +1,6 @@ +#version 130 +#extension GL_ARB_texture_cube_map_array : enable + + vec4 texture( samplerCubeArray sampler, vec4 coord, float bias); +ivec4 texture(isamplerCubeArray sampler, vec4 coord, float bias); +uvec4 texture(usamplerCubeArray sampler, vec4 coord, float bias); diff --git a/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl index 0f53212..73659b3 100644 --- a/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl +++ b/src/glsl/builtins/profiles/ARB_texture_cube_map_array.glsl @@ -7,7 +7,8 @@ ivec3 textureSize(usamplerCubeArray sampler, int lod); ivec3 textureSize(samplerCubeArrayShadow sampler, int lod); vec4 texture( samplerCubeArray sampler, vec4 coord); - vec4 texture( samplerCubeArray sampler, vec4 coord, float bias); +ivec4 texture(isamplerCubeArray sampler, vec4 coord); +uvec4 texture(usamplerCubeArray sampler, vec4 coord); float texture( samplerCubeArrayShadow sampler, vec4 P, float compare); vec4 textureGrad( samplerCubeArray sampler, vec4 P, vec3 dPdx, vec3 dPdy); -- 1.8.3.1 ___ 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
[Mesa-dev] [PATCH] i965: Delete pre-DRI2.3 viewport hacks.
The __DRI_USE_INVALIDATE extension was added in May 11th, 2010 by commit
4258e3a2e1c327. At this point, it's unlikely that anyone's using the
right mix of new and old components to hit this path. Deleting it
removes an untested code path and cleans up the driver a bit.
Cc: Kristian Høgsberg k...@bitplanet.net
Cc: Keith Packard kei...@keithp.com
---
src/mesa/drivers/dri/i965/intel_context.c | 21 -
src/mesa/drivers/dri/i965/intel_context.h | 2 --
src/mesa/drivers/dri/i965/intel_tex_image.c | 3 +--
3 files changed, 1 insertion(+), 25 deletions(-)
I'm guessing this would break if you had an early-2010 X server or libGL
and modern i965_dri.so. I'm not sure how to properly require this extension
to be present (and fail loading otherwise?). Any advice?
diff --git a/src/mesa/drivers/dri/i965/intel_context.c
b/src/mesa/drivers/dri/i965/intel_context.c
index 79420a2..491094f 100644
--- a/src/mesa/drivers/dri/i965/intel_context.c
+++ b/src/mesa/drivers/dri/i965/intel_context.c
@@ -292,21 +292,6 @@ intel_prepare_render(struct intel_context *intel)
}
}
-static void
-intel_viewport(struct gl_context *ctx, GLint x, GLint y, GLsizei w, GLsizei h)
-{
-struct intel_context *intel = intel_context(ctx);
-__DRIcontext *driContext = intel-driContext;
-
-if (intel-saved_viewport)
- intel-saved_viewport(ctx, x, y, w, h);
-
-if (_mesa_is_winsys_fbo(ctx-DrawBuffer)) {
- dri2InvalidateDrawable(driContext-driDrawablePriv);
- dri2InvalidateDrawable(driContext-driReadablePriv);
-}
-}
-
static const struct dri_debug_control debug_control[] = {
{ tex, DEBUG_TEXTURE},
{ state, DEBUG_STATE},
@@ -476,12 +461,6 @@ intelInitContext(struct intel_context *intel,
dri_ctx_error))
return false;
- /* Can't rely on invalidate events, fall back to glViewport hack */
- if (!driContextPriv-driScreenPriv-dri2.useInvalidate) {
- intel-saved_viewport = functions-Viewport;
- functions-Viewport = intel_viewport;
- }
-
if (mesaVis == NULL) {
memset(visual, 0, sizeof visual);
mesaVis = visual;
diff --git a/src/mesa/drivers/dri/i965/intel_context.h
b/src/mesa/drivers/dri/i965/intel_context.h
index 98def93..fff91db 100644
--- a/src/mesa/drivers/dri/i965/intel_context.h
+++ b/src/mesa/drivers/dri/i965/intel_context.h
@@ -252,8 +252,6 @@ struct intel_context
__DRIcontext *driContext;
struct intel_screen *intelScreen;
- void (*saved_viewport)(struct gl_context * ctx,
- GLint x, GLint y, GLsizei width, GLsizei height);
/**
* Configuration cache
diff --git a/src/mesa/drivers/dri/i965/intel_tex_image.c
b/src/mesa/drivers/dri/i965/intel_tex_image.c
index 0d0c7f1..5a10a37 100644
--- a/src/mesa/drivers/dri/i965/intel_tex_image.c
+++ b/src/mesa/drivers/dri/i965/intel_tex_image.c
@@ -310,8 +310,7 @@ intelSetTexBuffer2(__DRIcontext *pDRICtx, GLint target,
if (!intelObj)
return;
- if (dPriv-lastStamp != dPriv-dri2.stamp ||
- !pDRICtx-driScreenPriv-dri2.useInvalidate)
+ if (dPriv-lastStamp != dPriv-dri2.stamp)
intel_update_renderbuffers(pDRICtx, dPriv);
rb = intel_get_renderbuffer(fb, BUFFER_FRONT_LEFT);
--
1.8.3.1
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH V3 1/2] i965/blorp: Add bilinear filtering of samples for multisample scaled blits
+void
+brw_blorp_blit_program::manual_blend_linear(unsigned num_samples)
+{
+ if (key-tex_layout == INTEL_MSAA_LAYOUT_CMS)
+ mcs_fetch();
This won't work. The MCS value we fetch has to match up with the pixel
that we're sampling from. Since this function samples from different
pixels
in each iteration of the for loop below, the call to mcs_fetch() needs to
go inside the loop, and it needs to happen after storing the coordinates in
X and Y.
I think MCS value fetch will not be required anymore as we are anyway
getting rid of optimization which compares mcs value to zero.
MCS fetch is still needed, since the MCS value needs to be passed into the
ld2dms message that is used to read the samples from the surface.
Yes, you are right. My piglit test case for scaled blitting uses multisample
framebuffer with texture attachment. Debugging the test case showed that
multisample textures use INTEL_MSAA_LAYOUT_UMS. That's the reason
test case continued passing even after deleting mcs fetch code.
So, I modified the test case to use multisample framebuffer with renderbuffer
attachment as well for testing. Test continued failing without mcs fetch code.
Test passed after moving the mcs_fetch() inside the 'for' loop just after
computing the pixel coordinates. I think this verifies that mcs_fetch() is
now happening correctly. You can find my piglit tests (blit-scaled-glsl.cpp
and negative-blit-scaled.cpp) at:
https://github.com/aphogat/piglit.git, branch: 'blit-3'
As suggested by you, I'll also try developing a more exhaustive test to verify
mcs fetch value. In the mean time, please take a look at my piglit test cases
and suggest if they're good enough to verify the implementation and enable
the extension on intel hardware with my latest patch (V5).
+
+ /* We do this computation by performing the following operations:
+*
+* In case of 4x, 8x MSAA:
+* - Compute the pixel coordinates and sample numbers (a, b, c, d)
+* which are later used for interpolation
+* - linearly interpolate samples a and b in X
+* - linearly interpolate samples c and d in X
+* - linearly interpolate the results of last two operations in Y
+*
+* result = lrp(lrp(a + b) + lrp(c + d))
+*/
+ struct brw_reg Xp_f = retype(Xp, BRW_REGISTER_TYPE_F);
+ struct brw_reg Yp_f = retype(Yp, BRW_REGISTER_TYPE_F);
+ struct brw_reg t1_f = retype(t1, BRW_REGISTER_TYPE_F);
+ struct brw_reg t2_f = retype(t2, BRW_REGISTER_TYPE_F);
+
+ for (unsigned i = 0; i 4; ++i) {
+ assert(i ARRAY_SIZE(texture_data));
+ s_is_zero = false;
+
+ /* Compute pixel coordinates */
+ brw_ADD(func, vec16(x_sample_coords), Xp_f,
+ brw_imm_f((float)(i 0x1) * 0.5));
+ brw_ADD(func, vec16(y_sample_coords), Yp_f,
+ brw_imm_f((float)(i 0x2) * (1.0 / num_samples)));
+ brw_MOV(func, vec16(X), x_sample_coords);
+ brw_MOV(func, vec16(Y), y_sample_coords);
+
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH V5 1/2] i965/blorp: Add bilinear filtering of samples for multisample scaled blits
Current implementation of ext_framebuffer_multisample_blit_scaled in
i965/blorp uses nearest filtering for multisample scaled blits. Using
nearest filtering produces blocky artifacts and negates the benefits
of MSAA. That is the reason why extension was not enabled on i965.
This patch implements the bilinear filtering of samples in blorp engine.
Images generated with this patch are free from blocky artifacts and show
big improvement in visual quality.
Observed no piglit and gles3 regressions.
V3:
- Algorithm used for filtering assumes a rectangular grid of samples
roughly corresponding to sample locations.
- Test the boundary conditions on the edges of texture.
V4:
- Clip texcoords and use conditional MOVs.
- Send texture dimensions as push constants.
- Remove the optimization in case of scaled multisample blits.
V5:
- Move mcs_fetch() inside the 'for' loop after computing pixel coordinates.
Signed-off-by: Anuj Phogat anuj.pho...@gmail.com
---
src/mesa/drivers/dri/i965/brw_blorp.h| 16 ++
src/mesa/drivers/dri/i965/brw_blorp_blit.cpp | 278 +--
2 files changed, 273 insertions(+), 21 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_blorp.h
b/src/mesa/drivers/dri/i965/brw_blorp.h
index ffc27cc..9277d09 100644
--- a/src/mesa/drivers/dri/i965/brw_blorp.h
+++ b/src/mesa/drivers/dri/i965/brw_blorp.h
@@ -178,8 +178,15 @@ struct brw_blorp_wm_push_constants
uint32_t dst_x1;
uint32_t dst_y0;
uint32_t dst_y1;
+ /* Top right coordinates of the rectangular sample grid used for
+* multisample scaled blitting.
+*/
+ float sample_grid_x1;
+ float sample_grid_y1;
brw_blorp_coord_transform_params x_transform;
brw_blorp_coord_transform_params y_transform;
+ /* Pad out to an integral number of registers */
+ uint32_t pad[6];
};
/* Every 32 bytes of push constant data constitutes one GEN register. */
@@ -319,6 +326,15 @@ struct brw_blorp_blit_prog_key
* than one sample per pixel.
*/
bool persample_msaa_dispatch;
+
+ /* True for scaled blitting. */
+ bool blit_scaled;
+
+ /* Scale factors between the pixel grid and the grid of samples. We're
+* using grid of samples for bilinear filetring in multisample scaled blits.
+*/
+ float x_scale;
+ float y_scale;
};
class brw_blorp_blit_params : public brw_blorp_params
diff --git a/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp
b/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp
index 8694128..d39bae1 100644
--- a/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp
+++ b/src/mesa/drivers/dri/i965/brw_blorp_blit.cpp
@@ -622,7 +622,8 @@ private:
void kill_if_outside_dst_rect();
void translate_dst_to_src();
void single_to_blend();
- void manual_blend(unsigned num_samples);
+ void manual_blend_average(unsigned num_samples);
+ void manual_blend_bilinear(unsigned num_samples);
void sample(struct brw_reg dst);
void texel_fetch(struct brw_reg dst);
void mcs_fetch();
@@ -651,6 +652,11 @@ private:
struct brw_reg dst_x1;
struct brw_reg dst_y0;
struct brw_reg dst_y1;
+ /* Top right coordinates of the rectangular sample grid used for
+* multisample scaled blitting.
+*/
+ struct brw_reg sample_grid_x1;
+ struct brw_reg sample_grid_y1;
struct {
struct brw_reg multiplier;
struct brw_reg offset;
@@ -676,6 +682,16 @@ private:
*/
struct brw_reg y_coords[2];
+ /* X, Y coordinates of the pixel from which we need to fetch the specific
+* sample. These are used for multisample scaled blitting.
+*/
+ struct brw_reg x_sample_coords;
+ struct brw_reg y_sample_coords;
+
+ /* Fractional parts of the x and y coordinates, used as bilinear
interpolation coefficients */
+ struct brw_reg x_frac;
+ struct brw_reg y_frac;
+
/* Which element of x_coords and y_coords is currently in use.
*/
int xy_coord_index;
@@ -814,15 +830,17 @@ brw_blorp_blit_program::compile(struct brw_context *brw,
* that we want to texture from. Exception: if we are blending, then S is
* irrelevant, because we are going to fetch all samples.
*/
- if (key-blend) {
+ if (key-blend !key-blit_scaled) {
if (brw-intel.gen == 6) {
/* Gen6 hardware an automatically blend using the SAMPLE message */
single_to_blend();
sample(texture_data[0]);
} else {
/* Gen7+ hardware doesn't automaticaly blend. */
- manual_blend(key-src_samples);
+ manual_blend_average(key-src_samples);
}
+ } else if(key-blend key-blit_scaled) {
+ manual_blend_bilinear(key-src_samples);
} else {
/* We aren't blending, which means we just want to fetch a single sample
* from the source surface. The address that we want to fetch from is
@@ -872,18 +890,21 @@ void
brw_blorp_blit_program::alloc_push_const_regs(int base_reg)
{
#define CONST_LOC(name) offsetof(brw_blorp_wm_push_constants, name)
-#define ALLOC_REG(name) \
+#define
[Mesa-dev] [Bug 66346] New: shader_query.cpp:49: error: invalid conversion from 'void*' to 'GLuint'
https://bugs.freedesktop.org/show_bug.cgi?id=66346 Priority: medium Bug ID: 66346 Keywords: regression CC: bri...@vmware.com Assignee: mesa-dev@lists.freedesktop.org Summary: shader_query.cpp:49: error: invalid conversion from 'void*' to 'GLuint' Severity: blocker Classification: Unclassified OS: Mac OS X (All) Reporter: v...@freedesktop.org Hardware: x86-64 (AMD64) Status: NEW Version: git Component: Mesa core Product: Mesa mesa: 5a925cc5504575c22dbb7d29842d7fc5babcb5c7 (master) $ make [...] CXX shader_query.lo ../../src/mesa/main/shader_query.cpp: In function 'void _mesa_BindAttribLocation(void*, GLuint, const GLcharARB*)': ../../src/mesa/main/shader_query.cpp:49: error: invalid conversion from 'void*' to 'GLuint' ../../src/mesa/main/shader_query.cpp:49: error: initializing argument 2 of 'gl_shader_program* _mesa_lookup_shader_program_err$../../src/mesa/main/shader_query.cpp: In function 'void _mesa_GetActiveAttrib(void*, GLuint, GLsizei, GLsizei*, GLint*, GLenum*,$../../src/mesa/main/shader_query.cpp:87: error: invalid conversion from 'void*' to 'GLuint' ../../src/mesa/main/shader_query.cpp:87: error: initializing argument 2 of 'gl_shader_program* _mesa_lookup_shader_program_err$../../src/mesa/main/shader_query.cpp: In function 'GLint _mesa_GetAttribLocation(void*, const GLcharARB*)': ../../src/mesa/main/shader_query.cpp:139: error: invalid conversion from 'void*' to 'GLuint' ../../src/mesa/main/shader_query.cpp:139: error: initializing argument 2 of 'gl_shader_program* _mesa_lookup_shader_program_er$make[4]: *** [shader_query.lo] Error 1 $ gcc --version i686-apple-darwin11-llvm-gcc-4.2 (GCC) 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2336.11.00) Copyright (C) 2007 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. include/GL/glext.h 3468 #ifndef GL_ARB_shader_objects 3469 #define GL_ARB_shader_objects 1 3470 #ifdef __APPLE__ 3471 typedef void *GLhandleARB; 3472 #else 3473 typedef unsigned int GLhandleARB; 3474 #endif -- 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 V5 1/2] i965/blorp: Add bilinear filtering of samples for multisample scaled blits
Anuj, Yeah -- multisample textures on Gen7 are currently UMS for color. If you wanted to enable support for CMS, it should be reasonably straightforward, but requires some tweaks in the shader backend. This looks like a really nice quality improvement -- for the series: Acked-by: Chris Forbes chr...@ijw.co.nz ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
