Re: [Mesa-dev] [PATCH] AMDGPU: Match AMDGPU.cube intrinsic for SI.

2012-12-05 Thread Christian König 

On 04.12.2012 19:03, Michel Dänzer wrote:

From: Michel Dänzer michel.daen...@amd.com


Signed-off-by: Michel Dänzer michel.daen...@amd.com

Reviewed by: Christian König christian.koe...@amd.com


---
  lib/Target/AMDGPU/SIInstructions.td |   21 +
  1 file changed, 21 insertions(+)

diff --git a/lib/Target/AMDGPU/SIInstructions.td 
b/lib/Target/AMDGPU/SIInstructions.td
index d6f71f6..ecb874a 100644
--- a/lib/Target/AMDGPU/SIInstructions.td
+++ b/lib/Target/AMDGPU/SIInstructions.td
@@ -1269,6 +1269,27 @@ def : Pat 
(SI_KIL (V_MOV_IMM_I32 0xbf80))
  ;
  
+def : Pat 

+  (int_AMDGPU_cube VReg_128:$src),
+  (INSERT_SUBREG (INSERT_SUBREG (INSERT_SUBREG (INSERT_SUBREG (v4f32 
(IMPLICIT_DEF)),
+(V_CUBETC_F32 (EXTRACT_SUBREG VReg_128:$src, sel_x),
+  (EXTRACT_SUBREG VReg_128:$src, sel_y),
+  (EXTRACT_SUBREG VReg_128:$src, sel_z),
+  0, 0, 0, 0), sel_x),
+(V_CUBESC_F32 (EXTRACT_SUBREG VReg_128:$src, sel_x),
+  (EXTRACT_SUBREG VReg_128:$src, sel_y),
+  (EXTRACT_SUBREG VReg_128:$src, sel_z),
+  0, 0, 0, 0), sel_y),
+(V_CUBEMA_F32 (EXTRACT_SUBREG VReg_128:$src, sel_x),
+  (EXTRACT_SUBREG VReg_128:$src, sel_y),
+  (EXTRACT_SUBREG VReg_128:$src, sel_z),
+  0, 0, 0, 0), sel_z),
+(V_CUBEID_F32 (EXTRACT_SUBREG VReg_128:$src, sel_x),
+  (EXTRACT_SUBREG VReg_128:$src, sel_y),
+  (EXTRACT_SUBREG VReg_128:$src, sel_z),
+  0, 0, 0, 0), sel_w)
+;
+
  /** == **/
  /**   VOP3 Patterns**/
  /** == **/


___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 1/2] meta: Disable GL_FRAGMENT_SHADER_ATI in MESA_META_SHADER

2012-12-05 Thread Stefan Dösinger 

From b882d9e08932198e243b9cbb534d01d547661f86 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Stefan=20D=C3=B6singer?= stefandoesin...@gmx.at
Date: Mon, 3 Dec 2012 11:03:26 +0100
Subject: [PATCH 1/4] meta: Disable GL_FRAGMENT_SHADER_ATI in MESA_META_SHADER

Fixes clears in Wine on r200.
---
 src/mesa/drivers/common/meta.c |   11 +++
 1 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/src/mesa/drivers/common/meta.c b/src/mesa/drivers/common/meta.c
index d5e8af3..ad21fa8 100644
--- a/src/mesa/drivers/common/meta.c
+++ b/src/mesa/drivers/common/meta.c
@@ -133,6 +133,7 @@ struct save_state
struct gl_vertex_program *VertexProgram;
GLboolean FragmentProgramEnabled;
struct gl_fragment_program *FragmentProgram;
+   GLboolean ATIFragmentShaderEnabled;
struct gl_shader_program *VertexShader;
struct gl_shader_program *GeometryShader;
struct gl_shader_program *FragmentShader;
@@ -594,6 +595,11 @@ _mesa_meta_begin(struct gl_context *ctx, GLbitfield state)
  _mesa_set_enable(ctx, GL_FRAGMENT_PROGRAM_ARB, GL_FALSE);
   }
 
+  if (ctx-API == API_OPENGL_COMPAT  ctx-Extensions.ATI_fragment_shader) {
+ save-ATIFragmentShaderEnabled = ctx-ATIFragmentShader.Enabled;
+ _mesa_set_enable(ctx, GL_FRAGMENT_SHADER_ATI, GL_FALSE);
+  }
+
   if (ctx-Extensions.ARB_shader_objects) {
 	 _mesa_reference_shader_program(ctx, save-VertexShader,
 	ctx-Shader.CurrentVertexProgram);
@@ -914,6 +920,11 @@ _mesa_meta_end(struct gl_context *ctx)
 	 _mesa_reference_fragprog(ctx, save-FragmentProgram, NULL);
   }
 
+  if (ctx-API == API_OPENGL_COMPAT  ctx-Extensions.ATI_fragment_shader) {
+ _mesa_set_enable(ctx, GL_FRAGMENT_SHADER_ATI,
+  save-ATIFragmentShaderEnabled);
+  }
+
   if (ctx-Extensions.ARB_vertex_shader)
 	 _mesa_use_shader_program(ctx, GL_VERTEX_SHADER, save-VertexShader);
 
-- 
1.7.8.6

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 2/2] r200: Initialize swrast before setting limits

2012-12-05 Thread Stefan Dösinger 

From f572545ee0e2e17322554f03409237e40c732d1a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Stefan=20D=C3=B6singer?= stefandoesin...@gmx.at
Date: Mon, 3 Dec 2012 14:31:23 +0100
Subject: [PATCH 2/4] r200: Initialize swrast before setting limits

Otherwise the driver announces 4096 vertex shader constants and other
way too high limits.
---
 src/mesa/drivers/dri/r200/r200_context.c |   19 +--
 1 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/src/mesa/drivers/dri/r200/r200_context.c b/src/mesa/drivers/dri/r200/r200_context.c
index 360ba72..54cf33e 100644
--- a/src/mesa/drivers/dri/r200/r200_context.c
+++ b/src/mesa/drivers/dri/r200/r200_context.c
@@ -289,13 +289,20 @@ GLboolean r200CreateContext( gl_api api,
rmesa-radeon.swtcl.RenderIndex = ~0;
rmesa-radeon.hw.all_dirty = 1;
 
+   ctx = rmesa-radeon.glCtx;
+   /* Initialize the software rasterizer and helper modules.
+*/
+   _swrast_CreateContext( ctx );
+   _vbo_CreateContext( ctx );
+   _tnl_CreateContext( ctx );
+   _swsetup_CreateContext( ctx );
+   _ae_create_context( ctx );
+
/* Set the maximum texture size small enough that we can guarentee that
 * all texture units can bind a maximal texture and have all of them in
 * texturable memory at once. Depending on the allow_large_textures driconf
 * setting allow larger textures.
 */
-
-   ctx = rmesa-radeon.glCtx;
ctx-Const.MaxTextureUnits = driQueryOptioni (rmesa-radeon.optionCache,
 		 texture_units);
ctx-Const.MaxTextureImageUnits = ctx-Const.MaxTextureUnits;
@@ -345,14 +352,6 @@ GLboolean r200CreateContext( gl_api api,
 
_mesa_set_mvp_with_dp4( ctx, GL_TRUE );
 
-   /* Initialize the software rasterizer and helper modules.
-*/
-   _swrast_CreateContext( ctx );
-   _vbo_CreateContext( ctx );
-   _tnl_CreateContext( ctx );
-   _swsetup_CreateContext( ctx );
-   _ae_create_context( ctx );
-
/* Install the customized pipeline:
 */
_tnl_destroy_pipeline( ctx );
-- 
1.7.8.6

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH] [rfc] llvmpipe: EXT_transform_feedback support (v1)

2012-12-05 Thread Jose Fonseca 
115/130 sounds great.

Just one minor comment. Otherwise looks good

Reviewed-by: Jose Fonseca jfons...@vmware.com

Jose

- Original Message -
 I'd written most of this ages ago, but never finished it off.
 
 This passes 115/130 piglit tests so far. I'll look into the
 others as time permits.
 
 Signed-off-by: Dave Airlie airl...@redhat.com
 ---
  src/gallium/drivers/llvmpipe/lp_context.h  |  11 +-
  src/gallium/drivers/llvmpipe/lp_draw_arrays.c  |   9 ++
  src/gallium/drivers/llvmpipe/lp_query.c|  24 
  src/gallium/drivers/llvmpipe/lp_query.h|   2 +
  src/gallium/drivers/llvmpipe/lp_rast.c |   8 +-
  src/gallium/drivers/llvmpipe/lp_scene.c|   3 +-
  src/gallium/drivers/llvmpipe/lp_scene.h|   5 +-
  src/gallium/drivers/llvmpipe/lp_screen.c   |   3 +-
  src/gallium/drivers/llvmpipe/lp_setup.c|  12 +-
  src/gallium/drivers/llvmpipe/lp_setup.h|   4 +
  src/gallium/drivers/llvmpipe/lp_setup_context.h|   1 +
  src/gallium/drivers/llvmpipe/lp_setup_vbuf.c   |  14 +++
  src/gallium/drivers/llvmpipe/lp_state_rasterizer.c |   2 +
  src/gallium/drivers/llvmpipe/lp_state_so.c | 121
  ++---
  src/gallium/drivers/llvmpipe/lp_surface.c  |   4 +-
  15 files changed, 122 insertions(+), 101 deletions(-)
 
 diff --git a/src/gallium/drivers/llvmpipe/lp_context.h
 b/src/gallium/drivers/llvmpipe/lp_context.h
 index fe6fa3f..25cdff9 100644
 --- a/src/gallium/drivers/llvmpipe/lp_context.h
 +++ b/src/gallium/drivers/llvmpipe/lp_context.h
 @@ -81,12 +81,6 @@ struct llvmpipe_context {
 struct pipe_viewport_state viewport;
 struct pipe_vertex_buffer vertex_buffer[PIPE_MAX_ATTRIBS];
 struct pipe_index_buffer index_buffer;
 -   struct {
 -  struct llvmpipe_resource *buffer[PIPE_MAX_SO_BUFFERS];
 -  int offset[PIPE_MAX_SO_BUFFERS];
 -  int so_count[PIPE_MAX_SO_BUFFERS];
 -  int num_buffers;
 -   } so_target;
 struct pipe_resource *mapped_vs_tex[PIPE_MAX_SAMPLERS];
  
 unsigned num_samplers[PIPE_SHADER_TYPES];
 @@ -94,6 +88,11 @@ struct llvmpipe_context {
  
 unsigned num_vertex_buffers;
  
 +   struct draw_so_target *so_targets[PIPE_MAX_SO_BUFFERS];
 +   int num_so_targets;
 +   struct pipe_query_data_so_statistics so_stats;
 +   unsigned num_primitives_generated;
 +
 unsigned dirty; /** Mask of LP_NEW_x flags */
  
 unsigned active_occlusion_query;
 diff --git a/src/gallium/drivers/llvmpipe/lp_draw_arrays.c
 b/src/gallium/drivers/llvmpipe/lp_draw_arrays.c
 index 3497edf..eaa9ba3 100644
 --- a/src/gallium/drivers/llvmpipe/lp_draw_arrays.c
 +++ b/src/gallium/drivers/llvmpipe/lp_draw_arrays.c
 @@ -88,6 +88,13 @@ llvmpipe_draw_vbo(struct pipe_context *pipe, const
 struct pipe_draw_info *info)
 lp-index_buffer.index_size);
 }
  
 +   for (i = 0; i  lp-num_so_targets; i++) {
 +  void *buf =
 llvmpipe_resource(lp-so_targets[i]-target.buffer)-data;
 +  lp-so_targets[i]-mapping = buf;
 +   }
 +   draw_set_mapped_so_targets(draw, lp-num_so_targets,
 +  lp-so_targets);
 +
 llvmpipe_prepare_vertex_sampling(lp,
  
 lp-num_sampler_views[PIPE_SHADER_VERTEX],
  lp-sampler_views[PIPE_SHADER_VERTEX]);
 @@ -104,6 +111,8 @@ llvmpipe_draw_vbo(struct pipe_context *pipe,
 const struct pipe_draw_info *info)
 if (mapped_indices) {
draw_set_indexes(draw, NULL, 0);
 }
 +   draw_set_mapped_so_targets(draw, 0, NULL);
 +
 llvmpipe_cleanup_vertex_sampling(lp);
  
 /*
 diff --git a/src/gallium/drivers/llvmpipe/lp_query.c
 b/src/gallium/drivers/llvmpipe/lp_query.c
 index e302197..7a62a80 100644
 --- a/src/gallium/drivers/llvmpipe/lp_query.c
 +++ b/src/gallium/drivers/llvmpipe/lp_query.c
 @@ -138,6 +138,12 @@ llvmpipe_get_query_result(struct pipe_context
 *pipe,
  *result = os_time_get_nano();
}
break;
 +   case PIPE_QUERY_PRIMITIVES_GENERATED:
 +  *result = pq-num_primitives_generated;
 +  break;
 +   case PIPE_QUERY_PRIMITIVES_EMITTED:
 +  *result = pq-num_primitives_written;
 +  break;
 default:
assert(0);
break;
 @@ -165,6 +171,16 @@ llvmpipe_begin_query(struct pipe_context *pipe,
 struct pipe_query *q)
 memset(pq-count, 0, sizeof(pq-count));
 lp_setup_begin_query(llvmpipe-setup, pq);
  
 +   if (pq-type == PIPE_QUERY_PRIMITIVES_EMITTED) {
 +  pq-num_primitives_written = 0;
 +  llvmpipe-so_stats.num_primitives_written = 0;
 +   }
 +
 +   if (pq-type == PIPE_QUERY_PRIMITIVES_GENERATED) {
 +  pq-num_primitives_generated = 0;
 +  llvmpipe-num_primitives_generated = 0;
 +   }
 +
 if (pq-type == PIPE_QUERY_OCCLUSION_COUNTER) {
llvmpipe-active_occlusion_query = TRUE;
llvmpipe-dirty |= LP_NEW_OCCLUSION_QUERY;
 @@ -180,6 +196,14 @@ llvmpipe_end_query(struct pipe_context *pipe,
 struct pipe_query *q)

Re: [Mesa-dev] [PATCH] draw: set precalc_flat flag for AA lines too

2012-12-05 Thread Jose Fonseca 
Reviewed-by: Jose Fonseca jfons...@vmware.com

- Original Message -
 Fixes flat shading for AA lines.  demos/src/trivial/line-smooth is a
 test case which hits this.
 
 Note: This is a candidate for the stable branches.
 ---
  src/gallium/auxiliary/draw/draw_pipe_validate.c |1 +
  1 files changed, 1 insertions(+), 0 deletions(-)
 
 diff --git a/src/gallium/auxiliary/draw/draw_pipe_validate.c
 b/src/gallium/auxiliary/draw/draw_pipe_validate.c
 index 27afba5..4b0ed14 100644
 --- a/src/gallium/auxiliary/draw/draw_pipe_validate.c
 +++ b/src/gallium/auxiliary/draw/draw_pipe_validate.c
 @@ -195,6 +195,7 @@ static struct draw_stage *validate_pipeline(
 struct draw_stage *stage )
 if (rast-line_smooth  draw-pipeline.aaline) {
draw-pipeline.aaline-next = next;
next = draw-pipeline.aaline;
 +  precalc_flat = TRUE;
 }
  
 if (rast-point_smooth  draw-pipeline.aapoint) {
 --
 1.7.3.4
 
 ___
 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 automake] pipe-loader: Install pipe_* to $(libdir)/gallium-pipe.

2012-12-05 Thread Johannes Obermayr 
---
This patch is necessary because gbm and clover use same pipe_*.

In current automake they are installed only in $(libdir)/gbm and clover 
searches in $(libdir)/opencl and of course this cannot work.

The build is triggered by src/gallium/targets/gbm/Makefile.am:23 or 
src/gallium/targets/opencl/Makefile.am:12.

Successfully tested with clover.
Should also work with gbm.
---
 src/gallium/state_trackers/clover/Makefile.am |6 --
 src/gallium/targets/gbm/Makefile.am   |2 +-
 src/gallium/targets/pipe-loader/Makefile.am   |   20 ++--
 3 Dateien geändert, 15 Zeilen hinzugefügt(+), 13 Zeilen entfernt(-)

diff --git a/src/gallium/state_trackers/clover/Makefile.am 
b/src/gallium/state_trackers/clover/Makefile.am
index f068164..7cfccf6 100644
--- a/src/gallium/state_trackers/clover/Makefile.am
+++ b/src/gallium/state_trackers/clover/Makefile.am
@@ -3,7 +3,7 @@ AUTOMAKE_OPTIONS = subdir-objects
 AM_CPPFLAGS = \
$(GALLIUM_PIPE_LOADER_DEFINES) \
-DMESA_VERSION=\$(MESA_VERSION)\ \
-   -DPIPE_SEARCH_DIR=\$(OPENCL_LIB_INSTALL_DIR)\ \
+   -DPIPE_SEARCH_DIR=\$(libdir)/gallium-pipe\ \
-I$(top_srcdir)/include \
-I$(top_srcdir)/src/gallium/include \
-I$(top_srcdir)/src/gallium/drivers \
diff --git a/src/gallium/targets/gbm/Makefile.am 
b/src/gallium/targets/gbm/Makefile.am
index 28cf455..cd860d7 100644
--- a/src/gallium/targets/gbm/Makefile.am
+++ b/src/gallium/targets/gbm/Makefile.am
@@ -8,7 +8,7 @@ AM_CPPFLAGS = \
-I$(top_srcdir)/src/gbm/main \
-I$(top_srcdir)/src/gallium/winsys \
$(GALLIUM_PIPE_LOADER_DEFINES) \
-   -DPIPE_SEARCH_DIR=\$(gbmdir)\
+   -DPIPE_SEARCH_DIR=\$(libdir)/gallium-pipe\
 
 AM_CFLAGS = \
$(GALLIUM_CFLAGS) \
diff --git a/src/gallium/targets/pipe-loader/Makefile.am 
b/src/gallium/targets/pipe-loader/Makefile.am
index d02d1ae..bff92a0 100644
--- a/src/gallium/targets/pipe-loader/Makefile.am
+++ b/src/gallium/targets/pipe-loader/Makefile.am
@@ -10,8 +10,8 @@ AM_CPPFLAGS = \
-DGALLIUM_TRACE \
-DGALLIUM_GALAHAD
 
-gbmdir = $(libdir)/gbm
-gbm_LTLIBRARIES =
+pipedir = $(libdir)/gallium-pipe
+pipe_LTLIBRARIES =
 
 PIPE_LIBS = \
$(top_builddir)/src/gallium/auxiliary/libgallium.la \
@@ -27,7 +27,7 @@ endif
 
 
 if HAVE_GALLIUM_I915
-gbm_LTLIBRARIES += pipe_i915.la
+pipe_LTLIBRARIES += pipe_i915.la
 pipe_i915_la_SOURCES = pipe_i915.c
 pipe_i915_la_LIBADD = \
$(PIPE_LIBS) \
@@ -44,7 +44,7 @@ endif
 endif
 
 if HAVE_GALLIUM_NOUVEAU
-gbm_LTLIBRARIES += pipe_nouveau.la
+pipe_LTLIBRARIES += pipe_nouveau.la
 pipe_nouveau_la_SOURCES = pipe_nouveau.c
 nodist_EXTRA_pipe_nouveau_la_SOURCES = dummy.cpp
 pipe_nouveau_la_LIBADD = \
@@ -63,7 +63,7 @@ endif
 endif
 
 if HAVE_GALLIUM_R300
-gbm_LTLIBRARIES += pipe_r300.la
+pipe_LTLIBRARIES += pipe_r300.la
 pipe_r300_la_SOURCES = pipe_r300.c
 nodist_EXTRA_pipe_r300_la_SOURCES = dummy.cpp
 pipe_r300_la_LIBADD = \
@@ -81,7 +81,7 @@ endif
 endif
 
 if HAVE_GALLIUM_R600
-gbm_LTLIBRARIES += pipe_r600.la
+pipe_LTLIBRARIES += pipe_r600.la
 pipe_r600_la_SOURCES = pipe_r600.c
 pipe_r600_la_LIBADD = \
$(PIPE_LIBS) \
@@ -98,7 +98,7 @@ endif
 endif
 
 if HAVE_GALLIUM_RADEONSI
-gbm_LTLIBRARIES += pipe_radeonsi.la
+pipe_LTLIBRARIES += pipe_radeonsi.la
 pipe_radeonsi_la_SOURCES = pipe_radeonsi.c
 nodist_EXTRA_pipe_radeonsi_la_SOURCES = dummy.cpp
 pipe_radeonsi_la_LIBADD = \
@@ -115,7 +115,7 @@ endif
 endif
 
 if HAVE_GALLIUM_SVGA
-gbm_LTLIBRARIES += pipe_vmwgfx.la
+pipe_LTLIBRARIES += pipe_vmwgfx.la
 pipe_vmwgfx_la_SOURCES = pipe_vmwgfx.c
 pipe_vmwgfx_la_LIBADD = \
$(PIPE_LIBS) \
@@ -131,7 +131,7 @@ endif
 endif
 
 if HAVE_GALLIUM_LLVMPIPE
-gbm_LTLIBRARIES += pipe_swrast.la
+pipe_LTLIBRARIES += pipe_swrast.la
 pipe_swrast_la_SOURCES = pipe_swrast.c
 nodist_EXTRA_pipe_swrast_la_SOURCES = dummy.cpp
 pipe_swrast_la_LIBADD = \
@@ -141,7 +141,7 @@ pipe_swrast_la_LIBADD = \
 pipe_swrast_la_LDFLAGS = -no-undefined -avoid-version -module $(LLVM_LDFLAGS)
 else
 if HAVE_GALLIUM_SOFTPIPE
-gbm_LTLIBRARIES += pipe_swrast.la
+pipe_LTLIBRARIES += pipe_swrast.la
 pipe_swrast_la_SOURCES = pipe_swrast.c
 pipe_swrast_la_LIBADD = \
$(PIPE_LIBS) \
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 2/2] r200: Initialize swrast before setting limits

2012-12-05 Thread Alex Deucher 
Reviewed-by: Alex Deucher alexander.deuc...@amd.com

Also, should probably also make a note to apply this to the stable
branches.  While you are at it can you also check and see if radeon
needs a similar fix?

Alex
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH] st/egl/drm: only unref the udev device if needed

2012-12-05 Thread Alex Deucher 
On Tue, Dec 4, 2012 at 12:50 PM, Tobias Droste tdro...@gmx.de wrote:
 Anyone interested? ;-)

 I would just push it, but I don't have the rights to do so.

Looks reasonable to me.

Reviewed-by: Alex Deucher alexander.deuc...@amd.com


 Am Do, 29. November 2012, 17:02:28 schrieben Sie:
 Fixes compiler warning:

 drm/native_drm.c: In function ‘native_create_display’:
 drm/native_drm.c:180:21: warning: ‘device’ may be used uninitialized in this
 function [-Wmaybe-uninitialized] drm/native_drm.c:157:24: note: ‘device’
 was declared here

 Signed-off-by: Tobias Droste tdro...@gmx.de
 ---
  src/gallium/state_trackers/egl/drm/native_drm.c |9 +
  1 Datei geändert, 5 Zeilen hinzugefügt(+), 4 Zeilen entfernt(-)

 diff --git a/src/gallium/state_trackers/egl/drm/native_drm.c
 b/src/gallium/state_trackers/egl/drm/native_drm.c index 23fc137..f0c0f54
 100644
 --- a/src/gallium/state_trackers/egl/drm/native_drm.c
 +++ b/src/gallium/state_trackers/egl/drm/native_drm.c
 @@ -161,23 +161,24 @@ drm_get_device_name(int fd)
 udev = udev_new();
 if (fstat(fd, buf)  0) {
_eglLog(_EGL_WARNING, failed to stat fd %d, fd);
 -  goto out;
 +  goto outudev;
 }

 device = udev_device_new_from_devnum(udev, 'c', buf.st_rdev);
 if (device == NULL) {
_eglLog(_EGL_WARNING,
could not create udev device for fd %d, fd);
 -  goto out;
 +  goto outdevice;
 }

 tmp = udev_device_get_devnode(device);
 if (!tmp)
 -  goto out;
 +  goto outdevice;
 device_name = strdup(tmp);

 -out:
 +outdevice:
 udev_device_unref(device);
 +outudev:
 udev_unref(udev);

  #endif
 ___
 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 1/3] mesa: add set support (stores a set of pointers)

2012-12-05 Thread Brian Paul 

On 12/04/2012 04:55 PM, Jordan Justen wrote:

From: Eric Anholte...@anholt.net

From: git://people.freedesktop.org/~anholt/hash_table

Reviewed-by: Jordan Justenjordan.l.jus...@intel.com
[jordan.l.jus...@intel.com: minor rework for mesa]
Signed-off-by: Jordan Justenjordan.l.jus...@intel.com
---
  src/mesa/SConscript  |1 +
  src/mesa/main/set.c  |  349 ++
  src/mesa/main/set.h  |   94 ++
  src/mesa/sources.mak |1 +
  4 files changed, 445 insertions(+)
  create mode 100644 src/mesa/main/set.c
  create mode 100644 src/mesa/main/set.h

diff --git a/src/mesa/SConscript b/src/mesa/SConscript
index 1afa412..a2492f7 100644
--- a/src/mesa/SConscript
+++ b/src/mesa/SConscript
@@ -107,6 +107,7 @@ main_sources = [
  'main/renderbuffer.c',
  'main/samplerobj.c',
  'main/scissor.c',
+'main/set.c',
  'main/shaderapi.c',
  'main/shaderobj.c',
  'main/shader_query.cpp',
diff --git a/src/mesa/main/set.c b/src/mesa/main/set.c
new file mode 100644
index 000..c530c40
--- /dev/null
+++ b/src/mesa/main/set.c
@@ -0,0 +1,349 @@
+/*
+ * Copyright © 2009-2012 Intel Corporation
+ * Copyright © 1988-2004 Keith Packard and Bart Massey.
+ *
+ * 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
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Except as contained in this notice, the names of the authors
+ * or their institutions shall not be used in advertising or
+ * otherwise to promote the sale, use or other dealings in this
+ * Software without prior written authorization from the
+ * authors.
+ *
+ * Authors:
+ *Eric Anholte...@anholt.net
+ *Keith Packardkei...@keithp.com
+ */
+
+#includestdlib.h
+
+#include set.h
+#include ralloc.h
+
+#define ARRAY_SIZE(array) (sizeof(array) / sizeof(array[0]))
+
+/*
+ * From Knuth -- a good choice for hash/rehash values is p, p-2 where
+ * p and p-2 are both prime.  These tables are sized to have an extra 10%
+ * free to avoid exponential performance degradation as the hash table fills
+ */
+
+uint32_t deleted_key_value;
+const void *deleted_key =deleted_key_value;
+
+static const struct {
+   uint32_t max_entries, size, rehash;
+} hash_sizes[] = {
+   { 2,5,3},
+   { 4,7,5},
+   { 8,13,   11   },
+   { 16,   19,   17   },
+   { 32,   43,   41   },
+   { 64,   73,   71   },
+   { 128,  151,  149  },
+   { 256,  283,  281  },
+   { 512,  571,  569  },
+   { 1024, 1153, 1151 },
+   { 2048, 2269, 2267 },
+   { 4096, 4519, 4517 },
+   { 8192, 9013, 9011 },
+   { 16384,18043,18041},
+   { 32768,36109,36107},
+   { 65536,72091,72089},
+   { 131072,   144409,   144407   },
+   { 262144,   288361,   288359   },
+   { 524288,   576883,   576881   },
+   { 1048576,  1153459,  1153457  },
+   { 2097152,  2307163,  2307161  },
+   { 4194304,  4613893,  4613891  },
+   { 8388608,  9227641,  9227639  },
+   { 16777216, 18455029, 18455027 },
+   { 33554432, 36911011, 36911009 },
+   { 67108864, 73819861, 73819859 },
+   { 134217728,147639589,147639587},
+   { 268435456,295279081,295279079},
+   { 536870912,590559793,590559791},
+   { 1073741824,   1181116273,   1181116271   },
+   { 2147483648ul, 2362232233ul, 2362232231ul }
+};
+
+static int
+entry_is_free(struct set_entry *entry)
+{
+   return entry-key == NULL;
+}
+
+static int
+entry_is_deleted(struct set_entry *entry)
+{
+   return entry-key == deleted_key;
+}
+
+static int

Re: [Mesa-dev] [PATCH 2/3] main/syncobj: return GL_INVALID_VALUE for invalid sync objects

2012-12-05 Thread Brian Paul 

On 12/04/2012 04:55 PM, Jordan Justen wrote:

Note: The GL/GLES3 web man pages don't seem to properly
document glWaitSync's error when the sync object is invalid.

Signed-off-by: Jordan Justenjordan.l.jus...@intel.com
---
  src/mesa/main/syncobj.c |8 
  1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/mesa/main/syncobj.c b/src/mesa/main/syncobj.c
index 745e982..a2d3137 100644
--- a/src/mesa/main/syncobj.c
+++ b/src/mesa/main/syncobj.c
@@ -235,7 +235,7 @@ _mesa_DeleteSync(GLsync sync)
 }

 if (!_mesa_validate_sync(syncObj)) {
-  _mesa_error(ctx, GL_INVALID_OPERATION, glDeleteSync);
+  _mesa_error(ctx, GL_INVALID_VALUE, glDeleteSync (not a valid sync 
object));
return;
 }

@@ -303,7 +303,7 @@ _mesa_ClientWaitSync(GLsync sync, GLbitfield flags, 
GLuint64 timeout)
 ASSERT_OUTSIDE_BEGIN_END_WITH_RETVAL(ctx, GL_WAIT_FAILED);

 if (!_mesa_validate_sync(syncObj)) {
-  _mesa_error(ctx, GL_INVALID_OPERATION, glClientWaitSync);
+  _mesa_error(ctx, GL_INVALID_VALUE, glClientWaitSync (not a valid sync 
object));
return GL_WAIT_FAILED;
 }

@@ -347,7 +347,7 @@ _mesa_WaitSync(GLsync sync, GLbitfield flags, GLuint64 
timeout)
 ASSERT_OUTSIDE_BEGIN_END(ctx);

 if (!_mesa_validate_sync(syncObj)) {
-  _mesa_error(ctx, GL_INVALID_OPERATION, glWaitSync);
+  _mesa_error(ctx, GL_INVALID_VALUE, glWaitSync (not a valid sync 
object));
return;
 }

@@ -376,7 +376,7 @@ _mesa_GetSynciv(GLsync sync, GLenum pname, GLsizei bufSize, 
GLsizei *length,
 ASSERT_OUTSIDE_BEGIN_END(ctx);

 if (!_mesa_validate_sync(syncObj)) {
-  _mesa_error(ctx, GL_INVALID_OPERATION, glGetSynciv);
+  _mesa_error(ctx, GL_INVALID_VALUE, glGetSynciv (not a valid sync 
object));
return;
 }



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 3/3] mesa: validate that sync objects were created by mesa

2012-12-05 Thread Brian Paul 

On 12/04/2012 04:55 PM, Jordan Justen wrote:

Previously, the user could send in a pointer that was not created
by mesa. When we dereferenced that pointer, there would be an
exception.

Now we keep a set of pointers and verify that the pointer
exists in that set before dereferencing it.

Note: This fixes several crashing gles3conform tests.

Signed-off-by: Jordan Justenjordan.l.jus...@intel.com
---
  src/mesa/main/mtypes.h  |6 --
  src/mesa/main/shared.c  |   12 +++-
  src/mesa/main/syncobj.c |   23 +++
  3 files changed, 26 insertions(+), 15 deletions(-)




Reviewed-by: Brian Paul bri...@vmware.com
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH] silence unused code warnings

2012-12-05 Thread Fabio Pedretti 
diff --git a/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp 
b/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp
index ded4f61..ea81e53 100644
--- a/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp
+++ b/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp
@@ -67,26 +67,6 @@ static void init_colours()
   colour = _colour;
 }
 
-static const char *OpClassStr[OPCLASS_OTHER + 1] =
-{
-   MOVE,
-   LOAD,
-   STORE,
-   ARITH,
-   SHIFT,
-   SFU,
-   LOGIC,
-   COMPARE,
-   CONVERT,
-   ATOMIC,
-   TEXTURE,
-   SURFACE,
-   FLOW,
-   (INVALID),
-   PSEUDO,
-   OTHER
-};
-
 const char *operationStr[OP_LAST + 1] =
 {
nop,
diff --git a/src/gallium/drivers/nvc0/nvc0_state_validate.c 
b/src/gallium/drivers/nvc0/nvc0_state_validate.c
index 0f92614..80a8c01 100644
--- a/src/gallium/drivers/nvc0/nvc0_state_validate.c
+++ b/src/gallium/drivers/nvc0/nvc0_state_validate.c
@@ -3,6 +3,7 @@
 
 #include nvc0_context.h
 
+#if 0
 static void
 nvc0_validate_zcull(struct nvc0_context *nvc0)
 {
@@ -51,6 +52,7 @@ nvc0_validate_zcull(struct nvc0_context *nvc0)
 BEGIN_NVC0(push, NVC0_3D(ZCULL_INVALIDATE), 1);
 PUSH_DATA (push, 0);
 }
+#endif
 
 static void
 nvc0_validate_fb(struct nvc0_context *nvc0)
diff --git a/src/gallium/state_trackers/vega/path.c 
b/src/gallium/state_trackers/vega/path.c
index 43755f4..31ec719 100644
--- a/src/gallium/state_trackers/vega/path.c
+++ b/src/gallium/state_trackers/vega/path.c
@@ -1085,10 +1085,8 @@ static INLINE VGubyte normalize_coords(struct 
path_iter_data *pd,
}
   break;
case VG_SCUBIC_TO: {
-  VGfloat x0, y0, x1, y1, x2, y2, x3, y3;
+  VGfloat x1, y1, x2, y2, x3, y3;
   data_at(pd-coords, pd-path, 0, 4, data);
-  x0 = pd-ox;
-  y0 = pd-oy;
   x1 = 2*pd-ox-pd-px;
   y1 = 2*pd-oy-pd-py;
   x2 = data[0];
diff --git a/src/gallium/state_trackers/vega/vg_translate.c 
b/src/gallium/state_trackers/vega/vg_translate.c
index 7b3df27..6c1e354 100644
--- a/src/gallium/state_trackers/vega/vg_translate.c
+++ b/src/gallium/state_trackers/vega/vg_translate.c
@@ -201,8 +201,7 @@ void _vega_pack_rgba_span_float(struct vg_context *ctx,
case VG_BW_1: {
   VGshort *dst = (VGshort *)dstAddr;
   for (i = 0; i  n; ++i) {
- VGubyte r, g, b, a;
- VGubyte res;
+ VGubyte r, g, b, a, res;
  r = float_to_ubyte(rgba[i][0]);
  g = float_to_ubyte(rgba[i][1]);
  b = float_to_ubyte(rgba[i][2]);
@@ -229,8 +228,7 @@ void _vega_pack_rgba_span_float(struct vg_context *ctx,
case VG_A_4: {
   VGshort *dst = (VGshort *)dstAddr;
   for (i = 0; i  n; ++i) {
- VGubyte a;
- VGubyte res;
+ VGubyte a, res;
  a = float_to_ubyte(rgba[i][3]);
 
  res = a/4;

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH] silence unused code warnings

2012-12-05 Thread Brian Paul 

Comments below.

On 12/05/2012 07:02 AM, Fabio Pedretti wrote:

diff --git a/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp 
b/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp
index ded4f61..ea81e53 100644
--- a/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp
+++ b/src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp
@@ -67,26 +67,6 @@ static void init_colours()
   colour = _colour;
 }

-static const char *OpClassStr[OPCLASS_OTHER + 1] =
-{
-   MOVE,
-   LOAD,
-   STORE,
-   ARITH,
-   SHIFT,
-   SFU,
-   LOGIC,
-   COMPARE,
-   CONVERT,
-   ATOMIC,
-   TEXTURE,
-   SURFACE,
-   FLOW,
-   (INVALID),
-   PSEUDO,
-   OTHER
-};
-
 const char *operationStr[OP_LAST + 1] =
 {
nop,


I'd call that nv50: remove unused OpClassStr array



diff --git a/src/gallium/drivers/nvc0/nvc0_state_validate.c 
b/src/gallium/drivers/nvc0/nvc0_state_validate.c
index 0f92614..80a8c01 100644
--- a/src/gallium/drivers/nvc0/nvc0_state_validate.c
+++ b/src/gallium/drivers/nvc0/nvc0_state_validate.c
@@ -3,6 +3,7 @@

 #include nvc0_context.h

+#if 0
 static void
 nvc0_validate_zcull(struct nvc0_context *nvc0)
 {
@@ -51,6 +52,7 @@ nvc0_validate_zcull(struct nvc0_context *nvc0)
 BEGIN_NVC0(push, NVC0_3D(ZCULL_INVALIDATE), 1);
 PUSH_DATA (push, 0);
 }
+#endif

 static void
 nvc0_validate_fb(struct nvc0_context *nvc0)


This doesn't appear to apply to the master branch.



diff --git a/src/gallium/state_trackers/vega/path.c 
b/src/gallium/state_trackers/vega/path.c
index 43755f4..31ec719 100644
--- a/src/gallium/state_trackers/vega/path.c
+++ b/src/gallium/state_trackers/vega/path.c
@@ -1085,10 +1085,8 @@ static INLINE VGubyte normalize_coords(struct 
path_iter_data *pd,
}
   break;
case VG_SCUBIC_TO: {
-  VGfloat x0, y0, x1, y1, x2, y2, x3, y3;
+  VGfloat x1, y1, x2, y2, x3, y3;
   data_at(pd-coords, pd-path, 0, 4, data);
-  x0 = pd-ox;
-  y0 = pd-oy;
   x1 = 2*pd-ox-pd-px;
   y1 = 2*pd-oy-pd-py;
   x2 = data[0];


That patch is fine.  I'd suggest vega: remove unused variables as 
the name.




diff --git a/src/gallium/state_trackers/vega/vg_translate.c 
b/src/gallium/state_trackers/vega/vg_translate.c
index 7b3df27..6c1e354 100644
--- a/src/gallium/state_trackers/vega/vg_translate.c
+++ b/src/gallium/state_trackers/vega/vg_translate.c
@@ -201,8 +201,7 @@ void _vega_pack_rgba_span_float(struct vg_context *ctx,
case VG_BW_1: {
   VGshort *dst = (VGshort *)dstAddr;
   for (i = 0; i  n; ++i) {
- VGubyte r, g, b, a;
- VGubyte res;
+ VGubyte r, g, b, a, res;
  r = float_to_ubyte(rgba[i][0]);
  g = float_to_ubyte(rgba[i][1]);
  b = float_to_ubyte(rgba[i][2]);
@@ -229,8 +228,7 @@ void _vega_pack_rgba_span_float(struct vg_context *ctx,
case VG_A_4: {
   VGshort *dst = (VGshort *)dstAddr;
   for (i = 0; i  n; ++i) {
- VGubyte a;
- VGubyte res;
+ VGubyte a, res;
  a = float_to_ubyte(rgba[i][3]);

  res = a/4;



I don't really see the need for these cosmetic changes.

-Brian
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH] radeon/llvm: add a pattern for min/max

2012-12-05 Thread Vincent Lejeune 
---
 lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 53 
 lib/Target/AMDGPU/AMDGPUISelLowering.h   |  1 +
 lib/Target/AMDGPU/R600ISelLowering.cpp   |  5 +++
 lib/Target/AMDGPU/SIISelLowering.cpp |  6 
 4 files changed, 65 insertions(+)

diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp 
b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index 5bde9db..4c1a070 100644
--- a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -172,7 +172,60 @@ SDValue AMDGPUTargetLowering::LowerIntrinsicLRP(SDValue Op,
   OneSubAC);
 }
 
+/// Generate Min/Max pattern
+SDValue AMDGPUTargetLowering::LowerMinMax(SDValue Op,
+SelectionDAG DAG) const
+{
+  DebugLoc DL = Op.getDebugLoc();
+  EVT VT = Op.getValueType();
 
+  SDValue LHS = Op.getOperand(0);
+  SDValue RHS = Op.getOperand(1);
+  SDValue True = Op.getOperand(2);
+  SDValue CC = Op.getOperand(4);
+
+  ISD::CondCode CCOpcode = castCondCodeSDNode(CC)-get();
+  switch (CCOpcode) {
+  case ISD::SETOEQ:
+  case ISD::SETONE:
+  case ISD::SETUNE:
+  case ISD::SETNE:
+  case ISD::SETUEQ:
+  case ISD::SETEQ:
+  case ISD::SETFALSE:
+  case ISD::SETFALSE2:
+  case ISD::SETTRUE:
+  case ISD::SETTRUE2:
+  case ISD::SETUO:
+  case ISD::SETO:
+assert(0  Operation should already be optimised !);
+  case ISD::SETULE:
+  case ISD::SETULT:
+  case ISD::SETOLE:
+  case ISD::SETOLT:
+  case ISD::SETLE:
+  case ISD::SETLT: {
+if (LHS == True)
+  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
+else
+  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
+  }
+  case ISD::SETGT:
+  case ISD::SETGE:
+  case ISD::SETUGE:
+  case ISD::SETOGE:
+  case ISD::SETUGT:
+  case ISD::SETOGT: {
+if (LHS == True)
+  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
+else
+  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
+  }
+  case ISD::SETCC_INVALID:
+assert(0  Invalid setcc condcode !);
+  }
+  return Op;
+}
 
 SDValue AMDGPUTargetLowering::LowerUDIVREM(SDValue Op,
 SelectionDAG DAG) const
diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.h 
b/lib/Target/AMDGPU/AMDGPUISelLowering.h
index 60de190..3b60ae1 100644
--- a/lib/Target/AMDGPU/AMDGPUISelLowering.h
+++ b/lib/Target/AMDGPU/AMDGPUISelLowering.h
@@ -56,6 +56,7 @@ public:
   virtual SDValue LowerOperation(SDValue Op, SelectionDAG DAG) const;
   SDValue LowerIntrinsicIABS(SDValue Op, SelectionDAG DAG) const;
   SDValue LowerIntrinsicLRP(SDValue Op, SelectionDAG DAG) const;
+  SDValue LowerMinMax(SDValue Op, SelectionDAG DAG) const;
   virtual const char* getTargetNodeName(unsigned Opcode) const;
 
 // Functions defined in AMDILISelLowering.cpp
diff --git a/lib/Target/AMDGPU/R600ISelLowering.cpp 
b/lib/Target/AMDGPU/R600ISelLowering.cpp
index 6f1c1d7..374fbfc 100644
--- a/lib/Target/AMDGPU/R600ISelLowering.cpp
+++ b/lib/Target/AMDGPU/R600ISelLowering.cpp
@@ -764,6 +764,11 @@ SDValue R600TargetLowering::LowerSELECT_CC(SDValue Op, 
SelectionDAG DAG) const
 }
   }
 
+  if (CompareVT == VT == MVT::f32 
+  ((LHS == True  RHS == False) || (LHS == False  RHS == True))) {
+return LowerMinMax(Op, DAG);
+  }
+
   // If we make it this for it means we have no native instructions to handle
   // this SELECT_CC, so we must lower it.
   SDValue HWTrue, HWFalse;
diff --git a/lib/Target/AMDGPU/SIISelLowering.cpp 
b/lib/Target/AMDGPU/SIISelLowering.cpp
index 45f180f..98a7376 100644
--- a/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -383,6 +383,12 @@ SDValue SITargetLowering::LowerSELECT_CC(SDValue Op, 
SelectionDAG DAG) const
   EVT VT = Op.getValueType();
   DebugLoc DL = Op.getDebugLoc();
 
+  EVT CompareVT = LHS.getValueType();
+  if (CompareVT == VT == MVT::f32 
+  ((LHS == True  RHS == False) || (LHS == False  RHS == True))) {
+return LowerMinMax(Op, DAG);
+  }
+
   SDValue Cond = DAG.getNode(ISD::SETCC, DL, MVT::i1, LHS, RHS, CC);
   return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
 }
-- 
1.8.0.1

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH] radeon/llvm: add a pattern for min/max

2012-12-05 Thread Tom Stellard 
On Wed, Dec 05, 2012 at 04:58:21PM +0100, Vincent Lejeune wrote:
 ---
  lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 53 
 
  lib/Target/AMDGPU/AMDGPUISelLowering.h   |  1 +
  lib/Target/AMDGPU/R600ISelLowering.cpp   |  5 +++
  lib/Target/AMDGPU/SIISelLowering.cpp |  6 
  4 files changed, 65 insertions(+)


Hi Vincent,

This looks much better, thanks.  Just a few minor issues, but with those fixed,
this patch is:

Reviewed-by: Tom Stellard thomas.stell...@amd.com
 
 diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp 
 b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
 index 5bde9db..4c1a070 100644
 --- a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
 +++ b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
 @@ -172,7 +172,60 @@ SDValue AMDGPUTargetLowering::LowerIntrinsicLRP(SDValue 
 Op,
OneSubAC);
  }
  
 +/// Generate Min/Max pattern
 +SDValue AMDGPUTargetLowering::LowerMinMax(SDValue Op,
 +SelectionDAG DAG) const
 +{

Brace needs to be on the same line as the function here

 +  DebugLoc DL = Op.getDebugLoc();
 +  EVT VT = Op.getValueType();
  
 +  SDValue LHS = Op.getOperand(0);
 +  SDValue RHS = Op.getOperand(1);
 +  SDValue True = Op.getOperand(2);
 +  SDValue CC = Op.getOperand(4);

You should move this check into the function:

EVT CompareVT = LHS.getValueType();

if (CompareVT != VT || VT != MVT::f32 ||
(LHS != True  LHS != False) ||
(RHS != True  RHS != False)) {
  return SDValue();
}

Make sure to double check my logic here.

 +  ISD::CondCode CCOpcode = castCondCodeSDNode(CC)-get();
 +  switch (CCOpcode) {
 +  case ISD::SETOEQ:
 +  case ISD::SETONE:
 +  case ISD::SETUNE:
 +  case ISD::SETNE:
 +  case ISD::SETUEQ:
 +  case ISD::SETEQ:
 +  case ISD::SETFALSE:
 +  case ISD::SETFALSE2:
 +  case ISD::SETTRUE:
 +  case ISD::SETTRUE2:
 +  case ISD::SETUO:
 +  case ISD::SETO:
 +assert(0  Operation should already be optimised !);
 +  case ISD::SETULE:
 +  case ISD::SETULT:
 +  case ISD::SETOLE:
 +  case ISD::SETOLT:
 +  case ISD::SETLE:
 +  case ISD::SETLT: {
 +if (LHS == True)
 +  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
 +else
 +  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
 +  }
 +  case ISD::SETGT:
 +  case ISD::SETGE:
 +  case ISD::SETUGE:
 +  case ISD::SETOGE:
 +  case ISD::SETUGT:
 +  case ISD::SETOGT: {
 +if (LHS == True)
 +  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
 +else
 +  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
 +  }
 +  case ISD::SETCC_INVALID:
 +assert(0  Invalid setcc condcode !);
 +  }
 +  return Op;
 +}
  
  SDValue AMDGPUTargetLowering::LowerUDIVREM(SDValue Op,
  SelectionDAG DAG) const
 diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.h 
 b/lib/Target/AMDGPU/AMDGPUISelLowering.h
 index 60de190..3b60ae1 100644
 --- a/lib/Target/AMDGPU/AMDGPUISelLowering.h
 +++ b/lib/Target/AMDGPU/AMDGPUISelLowering.h
 @@ -56,6 +56,7 @@ public:
virtual SDValue LowerOperation(SDValue Op, SelectionDAG DAG) const;
SDValue LowerIntrinsicIABS(SDValue Op, SelectionDAG DAG) const;
SDValue LowerIntrinsicLRP(SDValue Op, SelectionDAG DAG) const;
 +  SDValue LowerMinMax(SDValue Op, SelectionDAG DAG) const;
virtual const char* getTargetNodeName(unsigned Opcode) const;
  
  // Functions defined in AMDILISelLowering.cpp
 diff --git a/lib/Target/AMDGPU/R600ISelLowering.cpp 
 b/lib/Target/AMDGPU/R600ISelLowering.cpp
 index 6f1c1d7..374fbfc 100644
 --- a/lib/Target/AMDGPU/R600ISelLowering.cpp
 +++ b/lib/Target/AMDGPU/R600ISelLowering.cpp
 @@ -764,6 +764,11 @@ SDValue R600TargetLowering::LowerSELECT_CC(SDValue Op, 
 SelectionDAG DAG) const
  }
}
  
 +  if (CompareVT == VT == MVT::f32 
 +  ((LHS == True  RHS == False) || (LHS == False  RHS == True))) {
 +return LowerMinMax(Op, DAG);
 +  }
 +

With this check in the common function, replace the above code with:

SDValue MinMax = LowerMinMax(Op, DAG)
if (MinMax.getNode()) {
  return MinMax;
}

// If we make it this for it means we have no native instructions to handle
// this SELECT_CC, so we must lower it.
SDValue HWTrue, HWFalse;
 diff --git a/lib/Target/AMDGPU/SIISelLowering.cpp 
 b/lib/Target/AMDGPU/SIISelLowering.cpp
 index 45f180f..98a7376 100644
 --- a/lib/Target/AMDGPU/SIISelLowering.cpp
 +++ b/lib/Target/AMDGPU/SIISelLowering.cpp
 @@ -383,6 +383,12 @@ SDValue SITargetLowering::LowerSELECT_CC(SDValue Op, 
 SelectionDAG DAG) const
EVT VT = Op.getValueType();
DebugLoc DL = Op.getDebugLoc();
  
 +  EVT CompareVT = LHS.getValueType();
 +  if (CompareVT == VT == MVT::f32 
 +  ((LHS == True  RHS == False) || (LHS == False  RHS == True))) {
 +return LowerMinMax(Op, DAG);
 +  }
 +

You can do the same replacement here too.

SDValue Cond = DAG.getNode(ISD::SETCC, DL, MVT::i1, LHS, RHS, CC);
return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
  }
 -- 
 1.8.0.1
 
 ___
 mesa-dev mailing list
 mesa-dev@lists.freedesktop.org

[Mesa-dev] [PATCH] radeon/llvm: add a pattern for min/max

2012-12-05 Thread Vincent Lejeune 
---
 lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 58 
 lib/Target/AMDGPU/AMDGPUISelLowering.h   |  1 +
 lib/Target/AMDGPU/R600ISelLowering.cpp   |  6 
 lib/Target/AMDGPU/SIISelLowering.cpp |  6 
 4 files changed, 71 insertions(+)

diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp 
b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index 5bde9db..41e8f44 100644
--- a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -172,7 +172,65 @@ SDValue AMDGPUTargetLowering::LowerIntrinsicLRP(SDValue Op,
   OneSubAC);
 }
 
+/// Generate Min/Max pattern
+SDValue AMDGPUTargetLowering::LowerMinMax(SDValue Op,
+SelectionDAG DAG) const {
+  DebugLoc DL = Op.getDebugLoc();
+  EVT VT = Op.getValueType();
+
+  SDValue LHS = Op.getOperand(0);
+  SDValue RHS = Op.getOperand(1);
+  SDValue True = Op.getOperand(2);
+  SDValue False = Op.getOperand(3);
+  SDValue CC = Op.getOperand(4);
+
+  if (VT != MVT::f32 ||
+  !((LHS == True  RHS == False) || (LHS == False  RHS == True))) {
+return SDValue();
+  }
 
+  ISD::CondCode CCOpcode = castCondCodeSDNode(CC)-get();
+  switch (CCOpcode) {
+  case ISD::SETOEQ:
+  case ISD::SETONE:
+  case ISD::SETUNE:
+  case ISD::SETNE:
+  case ISD::SETUEQ:
+  case ISD::SETEQ:
+  case ISD::SETFALSE:
+  case ISD::SETFALSE2:
+  case ISD::SETTRUE:
+  case ISD::SETTRUE2:
+  case ISD::SETUO:
+  case ISD::SETO:
+assert(0  Operation should already be optimised !);
+  case ISD::SETULE:
+  case ISD::SETULT:
+  case ISD::SETOLE:
+  case ISD::SETOLT:
+  case ISD::SETLE:
+  case ISD::SETLT: {
+if (LHS == True)
+  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
+else
+  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
+  }
+  case ISD::SETGT:
+  case ISD::SETGE:
+  case ISD::SETUGE:
+  case ISD::SETOGE:
+  case ISD::SETUGT:
+  case ISD::SETOGT: {
+if (LHS == True)
+  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
+else
+  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
+  }
+  case ISD::SETCC_INVALID:
+assert(0  Invalid setcc condcode !);
+  }
+  return Op;
+}
 
 SDValue AMDGPUTargetLowering::LowerUDIVREM(SDValue Op,
 SelectionDAG DAG) const
diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.h 
b/lib/Target/AMDGPU/AMDGPUISelLowering.h
index 60de190..3b60ae1 100644
--- a/lib/Target/AMDGPU/AMDGPUISelLowering.h
+++ b/lib/Target/AMDGPU/AMDGPUISelLowering.h
@@ -56,6 +56,7 @@ public:
   virtual SDValue LowerOperation(SDValue Op, SelectionDAG DAG) const;
   SDValue LowerIntrinsicIABS(SDValue Op, SelectionDAG DAG) const;
   SDValue LowerIntrinsicLRP(SDValue Op, SelectionDAG DAG) const;
+  SDValue LowerMinMax(SDValue Op, SelectionDAG DAG) const;
   virtual const char* getTargetNodeName(unsigned Opcode) const;
 
 // Functions defined in AMDILISelLowering.cpp
diff --git a/lib/Target/AMDGPU/R600ISelLowering.cpp 
b/lib/Target/AMDGPU/R600ISelLowering.cpp
index 6f1c1d7..dff29f6 100644
--- a/lib/Target/AMDGPU/R600ISelLowering.cpp
+++ b/lib/Target/AMDGPU/R600ISelLowering.cpp
@@ -764,6 +764,12 @@ SDValue R600TargetLowering::LowerSELECT_CC(SDValue Op, 
SelectionDAG DAG) const
 }
   }
 
+  // Possible Min/Max pattern
+  SDValue MinMax = LowerMinMax(Op, DAG);
+  if (MinMax.getNode()) {
+return MinMax;
+  }
+
   // If we make it this for it means we have no native instructions to handle
   // this SELECT_CC, so we must lower it.
   SDValue HWTrue, HWFalse;
diff --git a/lib/Target/AMDGPU/SIISelLowering.cpp 
b/lib/Target/AMDGPU/SIISelLowering.cpp
index 45f180f..2eb1fcc 100644
--- a/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -383,6 +383,12 @@ SDValue SITargetLowering::LowerSELECT_CC(SDValue Op, 
SelectionDAG DAG) const
   EVT VT = Op.getValueType();
   DebugLoc DL = Op.getDebugLoc();
 
+  // Possible Min/Max pattern
+  SDValue MinMax = LowerMinMax(Op, DAG);
+  if (MinMax.getNode()) {
+return MinMax;
+  }
+
   SDValue Cond = DAG.getNode(ISD::SETCC, DL, MVT::i1, LHS, RHS, CC);
   return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
 }
-- 
1.8.0.1

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH] radeon/llvm: add a pattern for min/max

2012-12-05 Thread Tom Stellard 
On Wed, Dec 05, 2012 at 06:09:46PM +0100, Vincent Lejeune wrote:
 ---
  lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 58 
 
  lib/Target/AMDGPU/AMDGPUISelLowering.h   |  1 +
  lib/Target/AMDGPU/R600ISelLowering.cpp   |  6 
  lib/Target/AMDGPU/SIISelLowering.cpp |  6 
  4 files changed, 71 insertions(+)


Looks great, thanks.

Reveiwed-by: Tom Stellard thomas.stell...@amd.com
 
 diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp 
 b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
 index 5bde9db..41e8f44 100644
 --- a/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
 +++ b/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
 @@ -172,7 +172,65 @@ SDValue AMDGPUTargetLowering::LowerIntrinsicLRP(SDValue 
 Op,
OneSubAC);
  }
  
 +/// Generate Min/Max pattern
 +SDValue AMDGPUTargetLowering::LowerMinMax(SDValue Op,
 +SelectionDAG DAG) const {
 +  DebugLoc DL = Op.getDebugLoc();
 +  EVT VT = Op.getValueType();
 +
 +  SDValue LHS = Op.getOperand(0);
 +  SDValue RHS = Op.getOperand(1);
 +  SDValue True = Op.getOperand(2);
 +  SDValue False = Op.getOperand(3);
 +  SDValue CC = Op.getOperand(4);
 +
 +  if (VT != MVT::f32 ||
 +  !((LHS == True  RHS == False) || (LHS == False  RHS == True))) {
 +return SDValue();
 +  }
  
 +  ISD::CondCode CCOpcode = castCondCodeSDNode(CC)-get();
 +  switch (CCOpcode) {
 +  case ISD::SETOEQ:
 +  case ISD::SETONE:
 +  case ISD::SETUNE:
 +  case ISD::SETNE:
 +  case ISD::SETUEQ:
 +  case ISD::SETEQ:
 +  case ISD::SETFALSE:
 +  case ISD::SETFALSE2:
 +  case ISD::SETTRUE:
 +  case ISD::SETTRUE2:
 +  case ISD::SETUO:
 +  case ISD::SETO:
 +assert(0  Operation should already be optimised !);
 +  case ISD::SETULE:
 +  case ISD::SETULT:
 +  case ISD::SETOLE:
 +  case ISD::SETOLT:
 +  case ISD::SETLE:
 +  case ISD::SETLT: {
 +if (LHS == True)
 +  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
 +else
 +  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
 +  }
 +  case ISD::SETGT:
 +  case ISD::SETGE:
 +  case ISD::SETUGE:
 +  case ISD::SETOGE:
 +  case ISD::SETUGT:
 +  case ISD::SETOGT: {
 +if (LHS == True)
 +  return DAG.getNode(AMDGPUISD::FMAX, DL, VT, LHS, RHS);
 +else
 +  return DAG.getNode(AMDGPUISD::FMIN, DL, VT, LHS, RHS);
 +  }
 +  case ISD::SETCC_INVALID:
 +assert(0  Invalid setcc condcode !);
 +  }
 +  return Op;
 +}
  
  SDValue AMDGPUTargetLowering::LowerUDIVREM(SDValue Op,
  SelectionDAG DAG) const
 diff --git a/lib/Target/AMDGPU/AMDGPUISelLowering.h 
 b/lib/Target/AMDGPU/AMDGPUISelLowering.h
 index 60de190..3b60ae1 100644
 --- a/lib/Target/AMDGPU/AMDGPUISelLowering.h
 +++ b/lib/Target/AMDGPU/AMDGPUISelLowering.h
 @@ -56,6 +56,7 @@ public:
virtual SDValue LowerOperation(SDValue Op, SelectionDAG DAG) const;
SDValue LowerIntrinsicIABS(SDValue Op, SelectionDAG DAG) const;
SDValue LowerIntrinsicLRP(SDValue Op, SelectionDAG DAG) const;
 +  SDValue LowerMinMax(SDValue Op, SelectionDAG DAG) const;
virtual const char* getTargetNodeName(unsigned Opcode) const;
  
  // Functions defined in AMDILISelLowering.cpp
 diff --git a/lib/Target/AMDGPU/R600ISelLowering.cpp 
 b/lib/Target/AMDGPU/R600ISelLowering.cpp
 index 6f1c1d7..dff29f6 100644
 --- a/lib/Target/AMDGPU/R600ISelLowering.cpp
 +++ b/lib/Target/AMDGPU/R600ISelLowering.cpp
 @@ -764,6 +764,12 @@ SDValue R600TargetLowering::LowerSELECT_CC(SDValue Op, 
 SelectionDAG DAG) const
  }
}
  
 +  // Possible Min/Max pattern
 +  SDValue MinMax = LowerMinMax(Op, DAG);
 +  if (MinMax.getNode()) {
 +return MinMax;
 +  }
 +
// If we make it this for it means we have no native instructions to handle
// this SELECT_CC, so we must lower it.
SDValue HWTrue, HWFalse;
 diff --git a/lib/Target/AMDGPU/SIISelLowering.cpp 
 b/lib/Target/AMDGPU/SIISelLowering.cpp
 index 45f180f..2eb1fcc 100644
 --- a/lib/Target/AMDGPU/SIISelLowering.cpp
 +++ b/lib/Target/AMDGPU/SIISelLowering.cpp
 @@ -383,6 +383,12 @@ SDValue SITargetLowering::LowerSELECT_CC(SDValue Op, 
 SelectionDAG DAG) const
EVT VT = Op.getValueType();
DebugLoc DL = Op.getDebugLoc();
  
 +  // Possible Min/Max pattern
 +  SDValue MinMax = LowerMinMax(Op, DAG);
 +  if (MinMax.getNode()) {
 +return MinMax;
 +  }
 +
SDValue Cond = DAG.getNode(ISD::SETCC, DL, MVT::i1, LHS, RHS, CC);
return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
  }
 -- 
 1.8.0.1
 
 ___
 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

Re: [Mesa-dev] [PATCH 2/2] r200: Initialize swrast before setting limits

2012-12-05 Thread Stefan Dösinger 
On 12/05/12 15:23, Alex Deucher wrote:
 Also, should probably also make a note to apply this to the stable
 branches.  While you are at it can you also check and see if radeon
 needs a similar fix?
From the code it looks like it needs a similar fix. I don't have a
pre-r250 card though so I can't test to be sure.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 2/3] meta: Don't use GL_TEXTURE_RECTANGLE as texture target in GLES

2012-12-05 Thread Anuj Phogat 
On Tue, Dec 4, 2012 at 11:17 AM, Eric Anholt e...@anholt.net wrote:
 Anuj Phogat anuj.pho...@gmail.com writes:

 Makes framebuffer_blit_coverage_mismatched_buffer_formats.test
 in gles3 conformance pass.

 Signed-off-by: Anuj Phogat anuj.pho...@gmail.com
 ---
  src/mesa/drivers/common/meta.c |   11 ---
  1 files changed, 8 insertions(+), 3 deletions(-)

 diff --git a/src/mesa/drivers/common/meta.c b/src/mesa/drivers/common/meta.c
 index 978eeb5..02e64ff 100644
 --- a/src/mesa/drivers/common/meta.c
 +++ b/src/mesa/drivers/common/meta.c

 -   _mesa_set_enable(ctx, tex-Target, GL_TRUE);
 +   /* glEnable() in gles2 and gles3 doesn't allow GL_TEXTURE_{1D, 2D, etc.}
 +* tokens.
 +*/
 +   if (ctx-API == API_OPENGL  ctx-API == API_OPENGLES)
 +  _mesa_set_enable(ctx, tex-Target, GL_TRUE);

 if (mask  GL_COLOR_BUFFER_BIT) {
setup_copypix_texture(tex, newTex, srcX, srcY, srcW, srcH,
 @@ -1679,7 +1683,8 @@ _mesa_meta_BlitFramebuffer(struct gl_context *ctx,
/* XXX can't easily do stencil */
 }

 -   _mesa_set_enable(ctx, tex-Target, GL_FALSE);
 +   if (ctx-API == API_OPENGL  ctx-API == API_OPENGLES)
 +  _mesa_set_enable(ctx, tex-Target, GL_FALSE);

 If BlitFramebuffer is doing fixed function, then it needs
 _mesa_set_enable() to work to enable the texture it's reading from.

 For GLES3, it sounds like we need to either make blorp handle all
 BlitFramebuffer, or make a BlitFramebuffer path using GLSL instead.  I
 don't think this halfway point of a BlitFramebuffer implementation that
 doesn't successfully blit makes sense.

Yeah. I realized it after posting this patch. I'm working on GLSL implementation
of BlitFrameBuffer.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 4/5] driconf: Add translation-generation to build system, don't track generated files

2012-12-05 Thread Chad Versace 
On 12/04/2012 03:53 PM, Carl Worth wrote:
 Another approach, (and something we did for the cairo project), would be
 to put things like lists of filenames needed for the build into a simple
 file that could be included by either Makefile.am or Android.mk. That
 would at least eliminate one source of duplication.
 
 For example, see this Makefile.sources in the cairo tree:
 
   http://cgit.freedesktop.org/cairo/tree/src/Makefile.sources

We already do this in Mesa. In some places, the file is named
Makefile.sourcce; in other places, sources.mak.

Of course, this doesn't help for generated files.

 With both Android and automake targeting a common make tool, there
 should be an opportunity to do even more sharing than cairo is able to
 do, (where the Makefile.win32 is aimed at a fairly-incompatible make
 tool on Windows).
 
 Chad, you and I should probably talk more in person the next time we're
 in the office together.

It may be possible to share rules for the generated files, but, based
on experience in maintaining the Android makefiles, I expect
such sharing to make the situation worse due to heightened complexity.
Or, it may not be possible at all. Beware of the uncanny valley:
the Android build systems looks deceptively like simple Makefiles until
you dig further in. I really think the best option is to generate the
Android makefiles with a tool.

Despite my skepticism, we should still plan to talk in person over this.
Together, we may be able to arrive at a feasible solution. Let's plan
to do that shortly after we ship the next Mesa release.

 In the meantime, the patch series that initiated this conversation adds
 a Makefile.am to the src/mesa/drivers/dri/common/xmlpool to generate an
 options.h file, (one of those complex case you mentioned of using a
 local python program to generate some code).
 
 Can you contribute whatever changes are necessary to Android.mk files so
 that I could push this series?

I'll push a branch today that adds the necessary Android changes to your
series, and inform you when I do.

-Chad

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 2/2] r200: Initialize swrast before setting limits

2012-12-05 Thread Alex Deucher 
On Wed, Dec 5, 2012 at 12:22 PM, Stefan Dösinger stefandoesin...@gmx.at wrote:
 On 12/05/12 15:23, Alex Deucher wrote:
 Also, should probably also make a note to apply this to the stable
 branches.  While you are at it can you also check and see if radeon
 needs a similar fix?
 From the code it looks like it needs a similar fix. I don't have a
 pre-r250 card though so I can't test to be sure.

It should be ok.  If you make the patches, I'll apply them unless you
have mesa commit access and want do it yourself.

Alex
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [Bug 57903] [llvmpipe] piglit fbo-blending-formats regression

2012-12-05 Thread bugzilla-daemon 
https://bugs.freedesktop.org/show_bug.cgi?id=57903

José Fonseca jfons...@vmware.com changed:

   What|Removed |Added

   Assignee|mesa-dev@lists.freedesktop. |jfons...@vmware.com
   |org |

-- 
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 1/2] meta: Disable GL_FRAGMENT_SHADER_ATI in MESA_META_SHADER

2012-12-05 Thread Ian Romanick 

On 12/05/2012 01:49 AM, Stefan Dösinger wrote:

In the future, please send patches using git-send-mail.  It makes it 
easier for people to reply with comments in-line with the patch.


In spite of that, this patch looks good.  Does it fix any known bugzilla 
entries?  If so, those should be mentioned in the commit message like:


Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=#

I also think this should get cherry-picked back to stable branches, so 
you should add the following text to the commit message:


NOTE: This is a candidate for stable release branches.

Other than that,

Reviewed-by: Ian Romanick ian.d.roman...@intel.com

___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH 4/5] driconf: Add translation-generation to build system, don't track generated files

2012-12-05 Thread Adrian M Negreanu 
On Wed, Dec 5, 2012 at 12:36 AM, Chad Versace
chad.vers...@linux.intel.com wrote:
 Taking the Android.mk's out of Mesa without replacing them with an
 Androgenizer-like tool would impose that insufferable situation on
 the Android developers.

Is maintaining Sconscripts as the only build system an option ?
It's easy to extend(it's written in python) and adding an Android.mk generator
would be easier than what androgenizer does(also, androgenizer is
unmaintained IIRC)


Thanks,


/me runs back in the cave :D
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev