On 04/23/2013 06:19 AM, Ian Romanick wrote:
On 04/23/2013 03:28 AM, Chad Versace wrote:
This allows maintainers/packagers/testers to tag the build with
information that will be reported by GL_VERSION.
If the environemt variable or make variable MESA_VERSION_STRING_EXTRA is
We accidentally set MESA_FORMAT_ETC2_RGB8_PUNCHTHROUGH_ALPHA1 twice,
rather than setting the RGB8 and SRGB8 formats.
Signed-off-by: Kenneth Graunke kenn...@whitecape.org
---
src/mesa/main/format_unpack.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
Signed-off-by: Kenneth Graunke kenn...@whitecape.org
---
src/mesa/main/format_unpack.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/src/mesa/main/format_unpack.c b/src/mesa/main/format_unpack.c
index feba386..94dd106 100644
--- a/src/mesa/main/format_unpack.c
+++
Signed-off-by: Kenneth Graunke kenn...@whitecape.org
---
src/mesa/main/format_unpack.c | 253 ++
1 file changed, 253 insertions(+)
diff --git a/src/mesa/main/format_unpack.c b/src/mesa/main/format_unpack.c
index 94dd106..279bad8 100644
---
Signed-off-by: Kenneth Graunke kenn...@whitecape.org
---
src/mesa/main/format_unpack.c | 327 ++
1 file changed, 327 insertions(+)
diff --git a/src/mesa/main/format_unpack.c b/src/mesa/main/format_unpack.c
index 279bad8..b1cf7be 100644
---
On 04/22/2013 09:17 PM, Ian Romanick wrote:
On 04/23/2013 02:39 AM, Kenneth Graunke wrote:
Otherwise, the table could be full of uninitialized garbage. That means
the NULL check at the bottom might not happen, so we wouldn't get the
_mesa_problem, and we'd then return garbage.
format_pack.c
On 04/22/2013 06:28 PM, Chad Versace wrote:
This allows maintainers/packagers/testers to tag the build with
information that will be reported by GL_VERSION.
If the environemt variable or make variable MESA_VERSION_STRING_EXTRA is
set, then its values will appear in the GL_VERSION string
On 04/22/2013 09:48 PM, Matt Turner wrote:
---
Do these comments make sense to anyone else?
src/mesa/drivers/dri/i965/brw_defines.h |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_defines.h
b/src/mesa/drivers/dri/i965/brw_defines.h
On 04/23/2013 04:28 AM, Chad Versace wrote:
This allows maintainers/packagers/testers to tag the build with
information that will be reported by GL_VERSION.
If the environemt variable or make variable MESA_VERSION_STRING_EXTRA is
set, then its values will appear in the GL_VERSION string
The commit below exposed a bug in dri2_add_config.
commit 3998f8c6b5da1a223926249755e54d8f701f81ab
Author: Ralf Jung p...@ralfj.de
Date: Tue Apr 9 14:09:50 2013 +0200
egl/x11: Fix initialisation of swap_interval
This little code snippet near the bottom of dri2_add_config,
On 04/22/2013 09:14 AM, Eric Anholt wrote:
34 files changed, 311 insertions(+), 889 deletions(-)
Also, swrast_dri.so now passes two FXT1 tests that failed before, and
i915's vertex shader texturing works. I haven't tested the radeon/nouveau
code, nor have I been as invasive as I could be,
+ /* Save op0 */
+ ir_variable *temp = new(ir) ir_variable(ir-operands[0]-
type, bfi_base,
+ ir_var_temporary);
+ this-base_ir-insert_before(temp);
+ this-base_ir-insert_before(assign(temp, ir-operands[0]));
+
+ ir-operation = ir_triop_bfi;
+
On Mon, Apr 22, 2013 at 10:06:19AM -0700, Eric Anholt wrote:
Pohjolainen, Topi topi.pohjolai...@intel.com writes:
On Tue, Apr 16, 2013 at 04:57:49PM -0700, Eric Anholt wrote:
Topi Pohjolainen topi.pohjolai...@intel.com writes:
+ if (image-regions[i]-bo)
+
https://bugs.freedesktop.org/show_bug.cgi?id=61364
--- Comment #9 from Loïc Yhuel loic.yh...@gmail.com ---
For me building mesa with --with-llvm-shared-libs fixes the crash.
--
You are receiving this mail because:
You are the assignee for the bug.
___
On Mon, Apr 22, 2013 at 10:02:21AM -0700, Eric Anholt wrote:
Pohjolainen, Topi topi.pohjolai...@intel.com writes:
On Tue, Apr 16, 2013 at 04:55:14PM -0700, Eric Anholt wrote:
Topi Pohjolainen topi.pohjolai...@intel.com writes:
Signed-off-by: Topi Pohjolainen topi.pohjolai...@intel.com
On Mon, Apr 22, 2013 at 10:08:47AM -0700, Eric Anholt wrote:
Pohjolainen, Topi topi.pohjolai...@intel.com writes:
On Tue, Apr 16, 2013 at 01:22:22PM -0700, Eric Anholt wrote:
Topi Pohjolainen topi.pohjolai...@intel.com writes:
The set introduces new target for 'eglCreateImageKHR()'
+ case ir_unop_find_msb:
+ emit(FBH(this-result, op[0]));
+ break;
This needs to change as discussed in IRC to number the bits from the LSB end.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
+ case ir_unop_find_msb:
+ emit(FBH(result_dst, op[0]));
+ break;
Same as in patch 13.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
+ bfe_to_bitops = stage != MESA_SHADER_FRAGMENT
+ ? BFE_TO_BITOPS : 0;
+ bitfield_insert = stage != MESA_SHADER_FRAGMENT
+ ? BITFIELD_INSERT_TO_BFM_BITOPS
+ : BITFIELD_INSERT_TO_BFM_BFI;
I think these
Would it make more sense to move this patch just before or after patch
11, so the disassembly is correct from the moment you start emitting
these?
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
Patches 1-3 and 7-12 are:
Reviewed-by: Chris Forbes chr...@ijw.co.nz
Have sent various comments for the others -- findMSB correctness as we
discussed in IRC, plus various minor things.
-- Chris
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
- Original Message -
Number of vertices to fetch doesn't always equal the number of input
vertices. To correctly compute the number if IA primitives we need
to use the total number of input vertices, not only those that
need to be fetched.
Signed-off-by: Zack Rusin
- Original Message -
Same approach as in the llvmpipe, if the geometry shader is
null and we have stream output then attach it to the vertex
shader right before executing the draw pipeline.
Signed-off-by: Zack Rusin za...@vmware.com
---
On Mon, Apr 22, 2013 at 11:24 PM, Tom Stellard t...@stellard.net wrote:
From: Tom Stellard thomas.stell...@amd.com
Buffer size should be in bytes not dwords.
For the series:
Reviewed-by: Alex Deucher alexander.deuc...@amd.com
---
src/gallium/drivers/r600/evergreen_compute.c | 2 +-
1
On 04/23/2013 09:20 AM, Kenneth Graunke wrote:
On 04/22/2013 09:17 PM, Ian Romanick wrote:
On 04/23/2013 02:39 AM, Kenneth Graunke wrote:
Otherwise, the table could be full of uninitialized garbage. That means
the NULL check at the bottom might not happen, so we wouldn't get the
On 04/23/2013 08:12 AM, Chad Versace wrote:
On 04/23/2013 06:19 AM, Ian Romanick wrote:
On 04/23/2013 03:28 AM, Chad Versace wrote:
This allows maintainers/packagers/testers to tag the build with
information that will be reported by GL_VERSION.
If the environemt variable or make variable
One minor nit...
On 04/23/2013 01:17 AM, Kenneth Graunke wrote:
Signed-off-by: Kenneth Graunkekenn...@whitecape.org
---
src/mesa/main/format_unpack.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/src/mesa/main/format_unpack.c b/src/mesa/main/format_unpack.c
index
Linux kernel 3.8 shipped with a bug in the prime fd passing code that
makes it unreliable. As of this writing, it seems unlikely that 3.9
will contain the fix for the issue.
This patch disable prime support when running on top of those kernels,
in order to prevent unexpected behavior when running
https://bugs.freedesktop.org/show_bug.cgi?id=63404
--- Comment #6 from Ander Conselvan de Oliveira conselv...@gmail.com ---
So this is actually a kernel bug, but I just sent a patch to mesa mailing list
to disable prime when running on top of 3.8 3.9. The latter version has not
been released yet,
https://bugs.freedesktop.org/show_bug.cgi?id=63404
Ander Conselvan de Oliveira conselv...@gmail.com changed:
What|Removed |Added
CC|
https://bugs.freedesktop.org/show_bug.cgi?id=63404
Joe Konno joe.ko...@linux.intel.com changed:
What|Removed |Added
Depends on|59299 |
--- Comment #8
From: Tom Stellard thomas.stell...@amd.com
---
src/gallium/auxiliary/gallivm/lp_bld_debug.cpp | 4
src/gallium/auxiliary/gallivm/lp_bld_misc.cpp | 4
2 files changed, 8 insertions(+)
diff --git a/src/gallium/auxiliary/gallivm/lp_bld_debug.cpp
From: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/radeon/radeon_llvm_emit.cpp | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/gallium/drivers/radeon/radeon_llvm_emit.cpp
b/src/gallium/drivers/radeon/radeon_llvm_emit.cpp
index d2dc035..5b770d0 100644
---
From: Tom Stellard thomas.stell...@amd.com
The libelf implementation that is distributed here:
http://www.mr511.de/software/english.html
requires calling elf_version() prior to calling elf_memory()
---
src/gallium/drivers/radeon/radeon_llvm_emit.cpp | 4
1 file changed, 4 insertions(+)
Looks good to me.
Jose
- Original Message -
From: Tom Stellard thomas.stell...@amd.com
---
src/gallium/auxiliary/gallivm/lp_bld_debug.cpp | 4
src/gallium/auxiliary/gallivm/lp_bld_misc.cpp | 4
2 files changed, 8 insertions(+)
diff --git
We need to handle the leading vertex information when
assembling primitives for the geometry shader otherwise
the resulting triangles will have vertices at incorrect
input locations.
Signed-off-by: Zack Rusin za...@vmware.com
---
src/gallium/auxiliary/draw/draw_gs_tmp.h |2 +-
1 file
Ok. I've moved the docs to src/gallium/docs/source/cso/rasterizer.rst , and
renamed `lower_left_origin` to `bottom_edge_rule`.
This is how it looks like:
http://people.freedesktop.org/~jrfonseca/gl_rasterization_rules/cso/rasterizer.html#other-members
Jose
- Original Message -
- Original Message -
We need to handle the leading vertex information when
assembling primitives for the geometry shader otherwise
the resulting triangles will have vertices at incorrect
input locations.
Signed-off-by: Zack Rusin za...@vmware.com
---
My follow on changes can be seen at
http://cgit.freedesktop.org/~jrfonseca/mesa/log/?h=gl-rasterization-rules
Jose
- Original Message -
Ok. I've moved the docs to src/gallium/docs/source/cso/rasterizer.rst , and
renamed `lower_left_origin` to `bottom_edge_rule`.
This is how it
On 23.04.2013 18:28, Jose Fonseca wrote:
Ok. I've moved the docs to src/gallium/docs/source/cso/rasterizer.rst , and
renamed `lower_left_origin` to `bottom_edge_rule`.
Well, that doesn't work for NV, but it's at least less invasive for
radeon since you don't have to change the state tracker
On Die, 2013-04-23 at 08:50 -0700, Tom Stellard wrote:
From: Tom Stellard thomas.stell...@amd.com
The libelf implementation that is distributed here:
http://www.mr511.de/software/english.html
will not parse sections that are marked SHT_NULL.
This patch and the corresponding Mesa patch are
On 10 April 2013 14:36, Dragomir Ivanov drago.iva...@gmail.com wrote:
Probably, the optimizer sees that there is second assignment, and skips
the first, f=1.0. It never gets to second though.
I tracked down the bug, and this is exactly right. I'll submit a fix
shortly.
---
src/glsl/ir.h | 7 +++
1 file changed, 7 insertions(+)
diff --git a/src/glsl/ir.h b/src/glsl/ir.h
index 3018e0f..878a946 100644
--- a/src/glsl/ir.h
+++ b/src/glsl/ir.h
@@ -131,6 +131,7 @@ public:
virtual class ir_swizzle * as_swizzle() { return NULL; }
virtual
Previously, the only kind of ir_jump that would terminate a basic
block was return. However, the other possible types of ir_jump
(break, continue, and discard) should terminate a basic block
too. This patch modifies basic block analysis so that it terminates a
basic block on any type of ir_jump,
This looks good to me. bottom_edge_rule should probably also be set
conditionally for Bitmap, Clear, and DrawPixels like this:
bottom_edge_rule = st_fb_orientation(ctx-DrawBuffer) == Y_0_TOP;
Marek
On Tue, Apr 23, 2013 at 6:39 PM, Jose Fonseca jfons...@vmware.com wrote:
My follow on changes
https://bugs.freedesktop.org/show_bug.cgi?id=62142
--- Comment #5 from Eric Anholt e...@anholt.net ---
Sure, it's a bug in that software, but we don't care and you're wasting your
time testing it.
--
You are receiving this mail because:
You are the assignee for the bug.
Hello,
Just to report that libdrm_radeon version (2.4.44) has not been updated
in scons/gallium.py too (still at version 2.4.42).
Regards.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
https://bugs.freedesktop.org/show_bug.cgi?id=26663
--- Comment #4 from ajax at nwnk dot net a...@nwnk.net ---
glibc 2.16 doesn't even work on pre-2.6 kernels anymore. DRI support requires
2.6.28 or better, according to the install docs, so I think we can safely drop
this entirely.
--
You are
On Tue, Apr 23, 2013 at 10:25 AM, Paul Berry stereotype...@gmail.com wrote:
Previously, the only kind of ir_jump that would terminate a basic
block was return. However, the other possible types of ir_jump
(break, continue, and discard) should terminate a basic block
too. This patch modifies
On Mon, Apr 22, 2013 at 4:42 PM, Brian Paul bri...@vmware.com wrote:
On 04/21/2013 05:25 PM, Marek Olšák wrote:
Although this might be useful for ARB_clear_buffer_object,
I need it for initializating resources in r600g.
---
src/gallium/auxiliary/util/u_blitter.c | 81
- Original Message -
This looks good to me.
Thanks.
bottom_edge_rule should probably also be set
conditionally for Bitmap, Clear, and DrawPixels like this:
bottom_edge_rule = st_fb_orientation(ctx-DrawBuffer) == Y_0_TOP;
I will do that in a follow on change.
Jose
Marek
On
We don't support pre-2.6 kernels anyway - the install docs say 2.6.28
for DRI - and apparently this confuses ld.so's sorting when multiple
libGLs are installed. Just remove it.
Signed-off-by: Adam Jackson a...@redhat.com
---
src/mapi/entry_x86-64_tls.h | 13 -
Chad Versace chad.vers...@linux.intel.com writes:
On 04/23/2013 06:19 AM, Ian Romanick wrote:
On 04/23/2013 03:28 AM, Chad Versace wrote:
This allows maintainers/packagers/testers to tag the build with
information that will be reported by GL_VERSION.
If the environemt variable or make
---
src/mesa/main/context.h | 15 +--
src/mesa/vbo/vbo_exec_api.c | 12 ++--
src/mesa/vbo/vbo_save_draw.c |2 +-
src/mesa/vbo/vbo_save_loopback.c |2 +-
4 files changed, 21 insertions(+), 10 deletions(-)
diff --git a/src/mesa/main/context.h
---
src/mesa/main/context.h | 11 +++
src/mesa/main/dlist.c |4 ++--
src/mesa/vbo/vbo_save_api.c |3 +--
3 files changed, 14 insertions(+), 4 deletions(-)
diff --git a/src/mesa/main/context.h b/src/mesa/main/context.h
index 0525184..396cbd2 100644
---
From: Jerome Glisse jgli...@redhat.com
This move the tracing timeout and printing into winsys and add
an debug environement variable for it (R600_DEBUG=trace_cs).
Lot of file touched because of winsys API changes.
Signed-off-by: Jerome Glisse jgli...@redhat.com
---
It's called from nowhere else.
---
src/mesa/main/dlist.c |6 +++---
src/mesa/main/dlist.h |2 --
2 files changed, 3 insertions(+), 5 deletions(-)
diff --git a/src/mesa/main/dlist.c b/src/mesa/main/dlist.c
index 33aeffc..fcccab4 100644
--- a/src/mesa/main/dlist.c
+++
On 23 April 2013 11:54, Chris Forbes chr...@ijw.co.nz wrote:
From low to high bits, the sample positions are packed y0,x0,y1,x1...
Fixes arb_texture_multisample-sample-position piglit.
Signed-off-by: Chris Forbes chr...@ijw.co.nz
Reviewed-by: Paul Berry stereotype...@gmail.com
---
On 04/23/2013 11:10 AM, Adam Jackson wrote:
We don't support pre-2.6 kernels anyway - the install docs say 2.6.28
for DRI - and apparently this confuses ld.so's sorting when multiple
libGLs are installed. Just remove it.
Signed-off-by: Adam Jackson a...@redhat.com
As another data point, i965
Brian Paul bri...@vmware.com writes:
It's called from nowhere else.
Reviewed-by: Eric Anholt e...@anholt.net
pgp0YB5HYcT5T.pgp
Description: PGP signature
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
On 04/23/2013 12:23 PM, Brian Paul wrote:
---
src/mesa/main/context.h | 11 +++
src/mesa/main/dlist.c |4 ++--
src/mesa/vbo/vbo_save_api.c |3 +--
3 files changed, 14 insertions(+), 4 deletions(-)
Definitely nicer!
Both are:
Reviewed-by: Kenneth Graunke
This could be used by shader-db for hopefully more accurate regression
testing.
---
src/mesa/drivers/dri/i965/brw_fs_schedule_instructions.cpp |5 +
1 file changed, 5 insertions(+)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_schedule_instructions.cpp
---
src/mesa/drivers/dri/i965/brw_wm_surface_state.c | 27 +--
src/mesa/drivers/dri/i965/gen7_wm_surface_state.c | 29 ++---
src/mesa/main/blend.c |8 ++
src/mesa/main/blend.h |4 +++
4 files
While this doesn't have the detail that the FS scheduler does, and is
ignorant of dependency control, it's still good for a 0.60% +/- 0.15%
performance improvement on GLBenchmark 2.7 (n=45/47, outliers removed)
---
src/mesa/drivers/dri/i965/Makefile.sources |1 +
The upside is less CPU overhead in fiddling with GL error handling, the
ability to use the constant color write message in most cases, and no GLSL
clear shaders appearing in MESA_GLSL=dump output. The downside is more
batch flushing and a total recompute of GL state at the end of blorp.
However,
Eric Anholt e...@anholt.net writes:
---
src/mesa/drivers/dri/i965/brw_wm_surface_state.c | 27 +--
src/mesa/drivers/dri/i965/gen7_wm_surface_state.c | 29
++---
src/mesa/main/blend.c |8 ++
src/mesa/main/blend.h
This like the fifth attempt to fix the issue.
Also with the new validating flag, we can set recalculate_inputs to FALSE
earlier in vbo_bind_arrays, because _mesa_update_state won't change it.
NOTE: This is a candidate for the stable branches.
---
src/mesa/vbo/vbo_exec.c | 20
On 04/19/2013 10:54 AM, Eric Anholt wrote:
Removes 75/78 state-dependent recompiles in GLB2.7 (the remaining 3 are
due to FBO-rendering size predictions).
AFAICT ClampColor is not deprecated in OpenGL Core. It doesn't exist in
ES, though.
If you change it to _mesa_is_desktop_gl(ctx), you
On Mon, Apr 22, 2013 at 6:28 PM, Eric Anholt e...@anholt.net wrote:
Marek Olšák mar...@gmail.com writes:
A driver doesn't have to set _NEW_BUFFER_OBJECT if it doesn't reallocate
the buffer from core Mesa's point of view, which avoids unnecessary state
validation.
Gallium drivers can be
On Sat, Apr 20, 2013 at 09:27:16AM +0200, Mathias Fröhlich wrote:
Hi Tom,
May be I need to tell where the problem really appears in real life.
OpenSceneGraph has some nifty features regarding multi channel rendering.
Assume a setup of multiple full screen views running on different
From: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/r600/r600_llvm.c | 5 +++--
src/gallium/drivers/r600/r600_llvm.h | 3 ++-
src/gallium/drivers/r600/r600_shader.c | 5 ++---
src/gallium/drivers/radeon/LLVM_REVISION.txt | 2 +-
4 files changed, 8
Hi Vincent,
I didn't see your Mesa commits from earlier today, so I wrote this patch.
You can ignore it, but in your patch you forgot to update the compute
shader path, which I think breaks the build with --enable-opencl.
-Tom
On Tue, Apr 23, 2013 at 08:49:43PM -0700, Tom Stellard wrote:
On Tue, Apr 23, 2013 at 5:41 PM, Kenneth Graunke kenn...@whitecape.org wrote:
On 04/19/2013 10:54 AM, Eric Anholt wrote:
Removes 75/78 state-dependent recompiles in GLB2.7 (the remaining 3 are
due to FBO-rendering size predictions).
AFAICT ClampColor is not deprecated in OpenGL Core. It
On 04/23/2013 09:19 PM, Eric Anholt wrote:
Chad Versace chad.vers...@linux.intel.com writes:
On 04/23/2013 06:19 AM, Ian Romanick wrote:
On 04/23/2013 03:28 AM, Chad Versace wrote:
This allows maintainers/packagers/testers to tag the build with
information that will be reported by
On 04/23/2013 09:15 PM, Jordan Justen wrote:
On Tue, Apr 23, 2013 at 5:41 PM, Kenneth Graunke kenn...@whitecape.org wrote:
On 04/19/2013 10:54 AM, Eric Anholt wrote:
Removes 75/78 state-dependent recompiles in GLB2.7 (the remaining 3 are
due to FBO-rendering size predictions).
AFAICT
On 04/23/2013 04:56 PM, Eric Anholt wrote:
This could be used by shader-db for hopefully more accurate regression
testing.
---
src/mesa/drivers/dri/i965/brw_fs_schedule_instructions.cpp |5 +
1 file changed, 5 insertions(+)
diff --git
On 04/23/2013 10:25 AM, Paul Berry wrote:
Previously, the only kind of ir_jump that would terminate a basic
block was return. However, the other possible types of ir_jump
(break, continue, and discard) should terminate a basic block
too. This patch modifies basic block analysis so that it
Seems reasonable.
-eric
On Tue, Apr 23, 2013 at 4:50 PM, Tom Stellard t...@stellard.net wrote:
From: Tom Stellard thomas.stell...@amd.com
The libelf implementation that is distributed here:
http://www.mr511.de/software/english.html
will not parse sections that are marked SHT_NULL.
---
On Mit, 2013-04-24 at 02:21 +0200, Marek Olšák wrote:
This like the fifth attempt to fix the issue.
Hang in there. :) I only have a little spelling fix to offer:
diff --git a/src/mesa/vbo/vbo_exec.c b/src/mesa/vbo/vbo_exec.c
index 5827f90..fd3a052 100644
--- a/src/mesa/vbo/vbo_exec.c
+++
On Tue, Apr 23, 2013 at 9:25 PM, Kenneth Graunke kenn...@whitecape.org wrote:
On 04/23/2013 09:15 PM, Jordan Justen wrote:
On Tue, Apr 23, 2013 at 5:41 PM, Kenneth Graunke kenn...@whitecape.org
wrote:
On 04/19/2013 10:54 AM, Eric Anholt wrote:
Removes 75/78 state-dependent recompiles in
80 matches
Mail list logo