https://bugs.freedesktop.org/show_bug.cgi?id=57644
Priority: medium
Bug ID: 57644
Keywords: regression
CC: srol...@vmware.com
Assignee: mesa-dev@lists.freedesktop.org
Summary: [llvmpipe]
src/gallium/auxiliary/
On 11/27/2012 11:24 AM, Chad Versace wrote:
Chris,
I applied the series to master-8f3570efc, but got build failures due to some
undefined symbols. I assume that your series depends on some other uncommitted
series. Could you push a branch somewhere that successfully builds so we can
test it?
Th
On 11/27/2012 10:04 AM, Eric Anholt wrote:
Kenneth Graunke writes:
If _Status is 0, then we don't know whether the FBO is complete or not,
and need to recheck it.
This fixes oglconform's drawBuffers-blend/basic.state.index0 test.
This should already be happening through _mesa_update_state()
On 11/27/2012 12:44 PM, Paul Berry wrote:
This should help avoid confusion now that we're using the gl_api enum
to distinguishing between core and compatibility API's. The
corresponding enum value for core API's is API_OPENGL_CORE.
--
Note: this patch is huge, so I've just included a few represe
On Tue, Nov 27, 2012 at 12:44 PM, Paul Berry wrote:
> This should help avoid confusion now that we're using the gl_api enum
> to distinguishing between core and compatibility API's. The
> corresponding enum value for core API's is API_OPENGL_CORE.
> --
I've found myself forgetting exactly what A
On 11/27/2012 12:12 AM, Eric Anholt wrote:
Fixes a ton of piglit regressions since the depthstencil fixes for gen6+.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57309
---
src/mesa/drivers/dri/i965/brw_misc_state.c |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --g
On 11/27/2012 12:12 AM, Eric Anholt wrote:
Fixes a crash in http://workshop.chromeexperiments.com/stars/ on i965,
and the new piglit test glsl-fs-clamp-5.
We were trying to emit a saturating move into a uniform, which the code
generator appropriately choked on. This was broken in the change in
3
On 11/27/2012 12:12 AM, Eric Anholt wrote:
---
src/mesa/drivers/dri/i965/brw_fs.cpp | 89 ++
src/mesa/drivers/dri/i965/brw_fs.h |3 ++
2 files changed, 92 insertions(+)
diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp
b/src/mesa/drivers/dri/i965/brw_
On 11/27/2012 12:58 PM, Brian Paul wrote:
---
src/mesa/vbo/vbo_attrib_tmp.h | 14 +++---
1 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/mesa/vbo/vbo_attrib_tmp.h b/src/mesa/vbo/vbo_attrib_tmp.h
index de1b451..adb22d7 100644
--- a/src/mesa/vbo/vbo_attrib_tmp.h
+++
https://bugs.freedesktop.org/show_bug.cgi?id=57372
--- Comment #7 from Richard Freeman ---
One other data point - the problem does not occur for xorg-server 1.13.0 if it
is running in virtualbox. So either the virtualbox video driver or the fact
that the server is running in virtualbox eliminate
Paul Berry writes:
> This should help avoid confusion now that we're using the gl_api enum
> to distinguishing between core and compatibility API's. The
> corresponding enum value for core API's is API_OPENGL_CORE.
Seems like a good idea to me.
Acked-by: Eric Anholt
pgplw0gQ9Smz2.pgp
Descri
https://bugs.freedesktop.org/show_bug.cgi?id=57372
--- Comment #6 from Richard Freeman ---
I've got some more data points here. While the segfault occurs client-side, it
appears to be triggered by a change server-side.
I can reproduce the problem if the X11 server is xorg-server 1.13.0-r1 (Gent
Matt Turner writes:
> Here's the write-up of what I was seeing earlier today for my own record.
>
> #if 1
> #elif
> #endif
>
> gives "0:2(1): preprocessor warning: ignoring illegal #elif without
> expression". gcc rejects it, but glcpp just warns since #if is true
> and it doesn't need to evaluate
On 11/27/2012 04:39 PM, Ian Romanick wrote:
> On 11/27/2012 04:16 PM, Chad Versace wrote:
>> + case 3:
>> + bool has_fragment_shader = driQueryOptionb(&screen->optionCache,
>> "fragment_shader");
>> + bool has_occlusion_query = driQueryOptionb(&screen->optionCache,
>> "stub_occlusion_
On 11/27/2012 04:16 PM, Chad Versace wrote:
Before this patch, intelInitScreen2 set DRIScreen::api_mask with the hacky
heuristic below:
if (gen >= 3)
api_mask = GL | GLES1 | GLES2;
else
api_mask = 0;
I don't understand why this hack works with gen2 (i830), or even if
Before this patch, intelInitScreen2 set DRIScreen::api_mask with the hacky
heuristic below:
if (gen >= 3)
api_mask = GL | GLES1 | GLES2;
else
api_mask = 0;
I don't understand why this hack works with gen2 (i830), or even if it
works properly at all. I don't care enough to
On 11/27/2012 01:04 PM, Ian Romanick wrote:
> On 11/21/2012 05:43 PM, Chad Versace wrote:
>> Before this patch, intelInitScreen2 set DRIScreen::api_mask with the hacky
>> heuristic below:
>>
>> if (gen >= 3)
>> api_mask = GL | GLES1 | GLES2;
>> else
>> api_mask = 0;
>>
>
On 11/27/2012 01:06 PM, Ian Romanick wrote:
> On 11/21/2012 05:43 PM, Chad Versace wrote:
>> Fixes error EGL_BAD_ATTRIBUTE in the tests below on Intel Sandybridge:
>>* piglit egl-create-context-verify-gl-flavor, testcase OpenGL ES 3.0
>>* gles3conform, revision 19700, when GL3Tests are ran
---
src/gallium/drivers/r600/r600_llvm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/gallium/drivers/r600/r600_llvm.c
b/src/gallium/drivers/r600/r600_llvm.c
index b3d4e6b..8f1ed26 100644
--- a/src/gallium/drivers/r600/r600_llvm.c
+++ b/src/gallium/drivers/r600/r600_llvm.c
@@ -11,6
---
lib/Target/AMDGPU/R600Instructions.td | 3 +++
1 file changed, 3 insertions(+)
diff --git a/lib/Target/AMDGPU/R600Instructions.td
b/lib/Target/AMDGPU/R600Instructions.td
index 1b9d462..a8be1a3 100644
--- a/lib/Target/AMDGPU/R600Instructions.td
+++ b/lib/Target/AMDGPU/R600Instructions.td
@@ -
---
lib/Target/AMDGPU/R600Instructions.td | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/Target/AMDGPU/R600Instructions.td
b/lib/Target/AMDGPU/R600Instructions.td
index d081824..1b9d462 100644
--- a/lib/Target/AMDGPU/R600Instructions.td
+++ b/lib/Target/AMDGPU/R600Ins
---
src/glsl/lower_output_reads.cpp | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/glsl/lower_output_reads.cpp b/src/glsl/lower_output_reads.cpp
index 90d71b0..a6192a5 100644
--- a/src/glsl/lower_output_reads.cpp
+++ b/src/glsl/lower_output_reads.cpp
@@ -97,6 +97,7 @@ output_read_remover:
---
lib/Target/AMDGPU/AMDGPUIntrinsics.td | 16
.../AMDGPU/MCTargetDesc/R600MCCodeEmitter.cpp | 43 +++---
lib/Target/AMDGPU/R600ISelLowering.cpp | 17 +++--
lib/Target/AMDGPU/R600Instructions.td | 42 ++---
4
Matt Turner writes:
> And add a test case to ensure that this works.
>
> See 17f9beb6 for #if change.
The code here looks fine, (and it's surprising that #if got this
treatment without #elif so long ago).
Your commit message says "add a test case to ensure this works" but the
only test case is i
On Sat, Oct 27, 2012 at 06:07:57PM +0200, Johannes Obermayr wrote:
> rtti was removed from more llvm libraries.
> Thanks to d0k for the hint via IRC #llvm on irc.oftc.net
After investigating the alternatives, I think this patch is the best way
to fix this problem. Using LLVM_CXXFLAGS introduces t
Ian Romanick writes:
> On 11/26/2012 03:12 PM, Carl Worth wrote:
> I assume this makes the piglit test pass? That should be mentioned in
> the commit message.
Yes. I've added that to the commit message now.
> Assuming that's the case, the series is
>
> Reviewed-by: Ian Romanick
Thanks. This
On Tue, Nov 27, 2012 at 12:42 AM, Kenneth Graunke wrote:
> The brw_compile structure contains the brw_instruction store and the
> brw_eu_emit.c state tracking fields. These are only useful for the
> final assembly generation pass; the earlier compilation stages doesn't
> need them.
>
> This also
On 11/27/2012 12:58 PM, Brian Paul wrote:
Reviewed-by: Ian Romanick
---
src/mesa/vbo/vbo_attrib_tmp.h | 14 +++---
1 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/mesa/vbo/vbo_attrib_tmp.h b/src/mesa/vbo/vbo_attrib_tmp.h
index de1b451..adb22d7 100644
--- a/src/m
On 11/21/2012 05:43 PM, Chad Versace wrote:
Fixes error EGL_BAD_ATTRIBUTE in the tests below on Intel Sandybridge:
* piglit egl-create-context-verify-gl-flavor, testcase OpenGL ES 3.0
* gles3conform, revision 19700, when GL3Tests are ran with -fbo
The series lives on my es3-bit-v3 branch.
On 11/21/2012 05:43 PM, Chad Versace wrote:
Before this patch, intelInitScreen2 set DRIScreen::api_mask with the hacky
heuristic below:
if (gen >= 3)
api_mask = GL | GLES1 | GLES2;
else
api_mask = 0;
I don't understand why this hack works with gen2 (i830), or even if
---
src/mesa/vbo/vbo_attrib_tmp.h | 14 +++---
1 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/mesa/vbo/vbo_attrib_tmp.h b/src/mesa/vbo/vbo_attrib_tmp.h
index de1b451..adb22d7 100644
--- a/src/mesa/vbo/vbo_attrib_tmp.h
+++ b/src/mesa/vbo/vbo_attrib_tmp.h
@@ -957,8 +957
This should help avoid confusion now that we're using the gl_api enum
to distinguishing between core and compatibility API's. The
corresponding enum value for core API's is API_OPENGL_CORE.
--
Note: this patch is huge, so I've just included a few representative
hunks. To see the full patch in con
On 11/26/2012 02:30 PM, Matt Turner wrote:
This is at least better than failing to compile the shader.
I think this is a pretty clear indication that app developers think
these macros are about as useful as the rest of us.
I think the rest of us are pretty useful. :)
Since we maintain line an
Ian Romanick writes:
> On 11/26/2012 11:19 PM, Eric Anholt wrote:
>> Ian Romanick writes:
>>
>>> From: Ian Romanick
>>>
>>> NOTE: This is a candidate for the 9.0 branch.
>>>
>>> Signed-off-by: Ian Romanick
>>
>> Has this actually been tested for the undefined-but-not-crash cases?
>> I'm guessi
Chris,
I applied the series to master-8f3570efc, but got build failures due to some
undefined symbols. I assume that your series depends on some other uncommitted
series. Could you push a branch somewhere that successfully builds so we can
test it?
Thanks,
Chad
On 11/22/2012 04:35 AM, Chris Forb
On 11/26/2012 11:19 PM, Eric Anholt wrote:
Ian Romanick writes:
From: Ian Romanick
NOTE: This is a candidate for the 9.0 branch.
Signed-off-by: Ian Romanick
Has this actually been tested for the undefined-but-not-crash cases?
I'm guessing not -- we've got lots of assumptions that a pair
On 11/26/2012 03:12 PM, Carl Worth wrote:
Simply emitting a nicely-formatted error message if any undefined macro is
encoutnered in a parser context expecting an expression.
I assume this makes the piglit test pass? That should be mentioned in
the commit message. Assuming that's the case, th
On 11/27/2012 12:12 AM, Eric Anholt wrote:
> Fixes a ton of piglit regressions since the depthstencil fixes for gen6+.
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57309
> ---
> src/mesa/drivers/dri/i965/brw_misc_state.c |4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
Google Earth v7 calls glReleaseShaderCompiler() and Mesa's GLSL
compiler is crashing.
In particular, it looks like glReleaseShaderCompiler() is getting
called between glCompileShader() and glLinkProgram(). The
glLinkProgram() call crashes when we try to reference some data
structures which
From: Michel Dänzer
Fixes 4 piglit tests and prevents many more from crashing.
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/radeonsi_shader.c |6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/radeonsi/radeonsi_shader.c
b/src/galli
- Original Message -
> Hi all,
>
> This set of patches fixes all the llvmpipe lp_test_* for PowerPC. The
> first five
> add Altivec intrinsics that fixes all arith testcases, while the
> remaining one
> deals mainly with big-endian memory loads and shift/mask algorithms.
>
> The first t
From: Michel Dänzer
This reverts commit 7bba4879bb79719e22a18b52759b1d1d839c783c.
radeonsi now supports Z16 and doesn't fail these assertions anymore.
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/si_state.c |8 ++--
1 file changed, 2 insertions(+), 6 deletions(-)
dif
From: Michel Dänzer
8 more piglits.
Signed-off-by: Michel Dänzer
---
src/gallium/drivers/radeonsi/si_state.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state.c
b/src/gallium/drivers/radeonsi/si_state.c
index 95ea860..f7c80a2 1006
Kenneth Graunke writes:
> If _Status is 0, then we don't know whether the FBO is complete or not,
> and need to recheck it.
>
> This fixes oglconform's drawBuffers-blend/basic.state.index0 test.
This should already be happening through _mesa_update_state() ->
_mesa_update_framebuffer(). Sounds
https://bugs.freedesktop.org/show_bug.cgi?id=56710
--- Comment #3 from Jon TURNEY ---
Created attachment 70672
--> https://bugs.freedesktop.org/attachment.cgi?id=70672&action=edit
kludge
Attached is a kludgy patch which jumps through the necessary hoops to include
the GLES2 prototypes to fix t
https://bugs.freedesktop.org/show_bug.cgi?id=56710
--- Comment #2 from Jon TURNEY ---
(In reply to comment #1)
> These warnings are also seen with the autotool build if you ./configure with
> --disable-asm (or build for a target where that is the default)
... and since the autotool build uses -W
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
v2: Don't use interleaved switch/case and if/else statements.
src/gallium/drivers/radeonsi/radeonsi_shader.c | 61 +++-
src/gallium/drivers/radeonsi/si_state_draw.c | 13 +++--
2 files changed, 68 insertions(+), 6
On 11/26/2012 06:14 PM, Kenneth Graunke wrote:
According to the ARB_vertex_type_2_10_10_10_rev specification:
"The error INVALID_ENUM is generated by VertexP*, NormalP*,
TexCoordP*, MultiTexCoordP*, ColorP*, or SecondaryColorP if
is not UNSIGNED_INT_2_10_10_10_REV or INT_2_10_10_10_REV."
Fix
On 11/26/2012 10:11 PM, Kenneth Graunke wrote:
I ran across this while running a glGenerateMipmap() test.
_meta_GenerateMipmap sets MESA_META_TRANSFORM, which causes
_mesa_meta_begin to try and set a default orthographic projection.
Unfortunately, if the drawbuffer isn't set up, ctx->DrawBuffer
On 11/26/2012 10:10 PM, Kenneth Graunke wrote:
If _Status is 0, then we don't know whether the FBO is complete or not,
and need to recheck it.
This fixes oglconform's drawBuffers-blend/basic.state.index0 test.
Signed-off-by: Kenneth Graunke
---
src/mesa/main/context.c | 4
1 file change
On 11/27/2012 01:09 AM, Vinson Lee wrote:
Fixes this SCons build error on Mac OS X if X11 is found.
NameError: name 'ws_xlib' is not defined:
File "SConstruct", line 144:
duplicate = 0 # http://www.scons.org/doc/0.97/HTML/scons-user/x2261.html
File "scons-2.2.0/SCons/Script/SConscript
On Fri, Nov 23, 2012 at 10:31:37PM +0100, Vincent Lejeune wrote:
> ---
> src/gallium/drivers/r600/r600_llvm.c | 5 +
> 1 file changed, 5 insertions(+)
>
> diff --git a/src/gallium/drivers/r600/r600_llvm.c
> b/src/gallium/drivers/r600/r600_llvm.c
> index b3d4e6b..2f243a2 100644
> --- a/src/ga
On Fri, Nov 23, 2012 at 10:32:33PM +0100, Vincent Lejeune wrote:
> ---
> lib/Target/AMDGPU/AMDGPUIntrinsics.td | 16 -
> .../AMDGPU/MCTargetDesc/R600MCCodeEmitter.cpp | 7 ++--
> lib/Target/AMDGPU/R600ISelLowering.cpp | 14 +---
> lib/Target/AMDGPU/R600In
The brw_compile structure contains the brw_instruction store and the
brw_eu_emit.c state tracking fields. These are only useful for the
final assembly generation pass; the earlier compilation stages doesn't
need them.
This also means that the code generator for future hardware won't have
access t
Compiling shaders requires several main steps:
1. Generating VS IR from either GLSL IR or Mesa IR
2. Optimizing the IR
3. Register allocation
4. Generating assembly code
This patch splits out step 4 into a separate class named "vec4_generator."
There are several reasons for doing so:
Final code generation should never fail. This is a bug, and there
should be no user-triggerable cases where this could occur.
Also, we're not going to have a fail() method after the split.
---
src/mesa/drivers/dri/i965/brw_vec4_emit.cpp | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-
The brw_compile structure is closely tied to the Gen4-7 hardware
encoding. However, do_vs_prog is very generic: it just calls out to
get a compiled program and then uploads it.
This isn't ultimately where we want it, but it's a step in the right
direction: it's now closer to the code generator.
-
During compilation, we allocate a bunch of things: the IR needs to last
at least until code generation...and then the program store needs to
last until after we upload the program.
For simplicity's sake, just keep it all around until we upload the
program. After that, it can all be freed.
This w
We used to steal it out of the brw_compile struct, but that won't be
initialized in time soon (and is eventually going away).
---
src/mesa/drivers/dri/i965/brw_vs.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_vs.c
b/src/mesa/drivers/dri/i
We used to steal it out of the brw_compile struct...but vec4_visitor
isn't going to have one of those in the future.
---
src/mesa/drivers/dri/i965/brw_vec4.cpp | 9 +
src/mesa/drivers/dri/i965/brw_vec4.h | 3 ++-
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 5 +++--
s
This leaves only the final code generation stage in brw_vec4_emit.cpp,
moving the payload setup, run(), and brw_vs_emit functions to brw_vec4.cpp.
The fragment shader backend puts these functions in brw_fs.cpp, so this
patch also helps with consistency.
---
src/mesa/drivers/dri/i965/brw_vec4.cpp
---
src/mesa/drivers/dri/i965/brw_fs.cpp | 89 ++
src/mesa/drivers/dri/i965/brw_fs.h |3 ++
2 files changed, 92 insertions(+)
diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp
b/src/mesa/drivers/dri/i965/brw_fs.cpp
index 2f88d92..5fd0e7b 100644
--- a/src/m
Fixes a ton of piglit regressions since the depthstencil fixes for gen6+.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57309
---
src/mesa/drivers/dri/i965/brw_misc_state.c |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/mesa/drivers/dri/i965/brw_misc_state.c
Fixes a crash in http://workshop.chromeexperiments.com/stars/ on i965,
and the new piglit test glsl-fs-clamp-5.
We were trying to emit a saturating move into a uniform, which the code
generator appropriately choked on. This was broken in the change in
32ae8d3b321185a85b73ff703d8fc26bd5f48fa7.
Bug
Fixes this SCons build error on Mac OS X if X11 is found.
NameError: name 'ws_xlib' is not defined:
File "SConstruct", line 144:
duplicate = 0 # http://www.scons.org/doc/0.97/HTML/scons-user/x2261.html
File "scons-2.2.0/SCons/Script/SConscript.py", line 614:
return method(*args, **kw)
65 matches
Mail list logo
