On Wed, Aug 13, 2014 at 9:52 PM, Ilia Mirkin wrote:
> I left all the variants as separate operations in the glsl ir. However for
> gallium I only added the fine version, as it seems like DDX can do pretty much
> whatever it wants. I was on the fence about adding coarse versions as well and
> then
Reviewed-by: Matt Turner
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Wed, Aug 13, 2014 at 9:52 PM, Ilia Mirkin wrote:
> Signed-off-by: Ilia Mirkin
> ---
> src/mesa/main/extensions.c | 1 +
> src/mesa/main/mtypes.h | 1 +
> 2 files changed, 2 insertions(+)
>
> diff --git a/src/mesa/main/extensions.c b/src/mesa/main/extensions.c
> index 8658ca8..3dcb199 1006
Signed-off-by: Ilia Mirkin
---
Oops :) The piglit tests caught these...
v1 -> v2:
- add constant expression handling
- add name strings for new opcodes
- add new opcodes into ir_expression constructor
src/glsl/builtin_functions.cpp | 48 +
src/glsl/g
I left all the variants as separate operations in the glsl ir. However for
gallium I only added the fine version, as it seems like DDX can do pretty much
whatever it wants. I was on the fence about adding coarse versions as well and
then using the FragmentShaderDerivative hint to select one or the
The quadop-based method we currently use on all chipsets already
provides the fine version of the derivatives.
Signed-off-by: Ilia Mirkin
---
docs/GL3.txt | 2 +-
src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp | 4
src/gallium/drivers/
Signed-off-by: Ilia Mirkin
---
src/mesa/state_tracker/st_extensions.c | 3 ++-
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 9 -
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/src/mesa/state_tracker/st_extensions.c
b/src/mesa/state_tracker/st_extensions.c
index eace3
Signed-off-by: Ilia Mirkin
---
src/gallium/auxiliary/tgsi/tgsi_info.c | 3 +++
src/gallium/auxiliary/tgsi/tgsi_util.c | 2 ++
src/gallium/docs/source/screen.rst | 2 ++
src/gallium/docs/source/tgsi.rst | 12 ++--
src/gallium/drivers/fre
Signed-off-by: Ilia Mirkin
---
src/glsl/builtin_functions.cpp | 48 +
src/glsl/glcpp/glcpp-parse.y| 3 +++
src/glsl/glsl_parser_extras.cpp | 1 +
src/glsl/glsl_parser_extras.h | 2 ++
src/glsl/ir.h | 4
src/glsl/ir_validate
Signed-off-by: Ilia Mirkin
---
src/mesa/main/extensions.c | 1 +
src/mesa/main/mtypes.h | 1 +
2 files changed, 2 insertions(+)
diff --git a/src/mesa/main/extensions.c b/src/mesa/main/extensions.c
index 8658ca8..3dcb199 100644
--- a/src/mesa/main/extensions.c
+++ b/src/mesa/main/extensions.c
Layered rendering is part of OpenGL 3.2; GL_ARB_draw_instanced is part
of OpenGL 3.1. As such, all drivers supporting layered rendering
already support gl_InstanceID.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/common/meta.c | 21 +
1 file changed, 5 insertions(+), 1
So we can use it in meta.c.
Signed-off-by: Kenneth Graunke
---
src/mesa/main/varray.h| 4
src/mesa/vbo/vbo_exec_array.c | 6 ++
2 files changed, 10 insertions(+)
diff --git a/src/mesa/main/varray.h b/src/mesa/main/varray.h
index f94ebac..d5d8b36 100644
--- a/src/mesa/main/varra
Ilia Mirkin writes:
> Fixes the LUMINANCE_ALPHA formats of fbo-clear-formats piglit test.
>
> Signed-off-by: Ilia Mirkin
> ---
> src/mesa/drivers/dri/nouveau/nouveau_driver.c | 10 +-
> src/mesa/drivers/dri/nouveau/nouveau_util.h | 9 +
> 2 files changed, 18 insertions(+), 1
Carl Worth writes:
> Normally, I would have completed my own testing of the branch prior to
> pushing the candidate. This time, I'm a little late, but I plan to
> follow up tomorrow with my own testing results, (which hopefully will
> not lead to any changes in the branch).
I've completed that te
---
I'd squash this in at minimum. The changes are
- Whitespace
- Removal of unnecessary destructor
- Renaming "one" and "two" to "a" and "b" (one->value.u[c0] <
two->value.u[c0]...)
- continue -> break
- assert(!...) -> unreachable
- Not doing assignments in if conditionals
- Marking swiz
On Wed, Aug 13, 2014 at 1:22 PM, Matt Turner wrote:
> From: Kenneth Graunke
>
> The pass breaks live ranges of virtual registers by allocating new
> registers when it sees an assignment to a virtual GRF it's already seen
> written.
>
> total instructions in shared programs: 4337879 -> 4335014 (-0
Signed-off-by: Ilia Mirkin
---
src/egl/drivers/dri2/egl_dri2.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/egl/drivers/dri2/egl_dri2.c b/src/egl/drivers/dri2/egl_dri2.c
index 5602ec3..dda13a3 100644
--- a/src/egl/drivers/dri2/egl_dri2.c
+++ b/src/egl/drivers/dri2/egl_d
https://bugs.freedesktop.org/show_bug.cgi?id=82327
--- Comment #3 from Ian Romanick ---
Those tests make heavy use of sed and tr. Is there some difference between the
versions on those platforms? What sed and tr (or other commands used in those
tests) exist on OS X and FreeBSD?
--
You are rec
On 07/29/2014 02:36 AM, Petri Latvala wrote:
> Add an optimization pass that drops min/max expression operands that
> can be proven to not contribute to the final result. The algorithm is
> similar to alpha-beta pruning on a minmax search, from the field of
> AI.
>
> This optimization pass can opt
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #27 from Michel Dänzer ---
(In reply to comment #17)
> Program received signal SIGSEGV, Segmentation fault.
> PatchJump (label=..., jump=...) at
When it says 'PatchJump (label=..., jump=...) at [...]', it's not a crash but
normal Jav
On 08/11/2014 05:29 PM, Kristian Høgsberg wrote:
> Right now we decide which kernels to use and the GRF start offsets in
> one place and emit the kernel pointers later. The logic of how to map
> 8, 16 and 32 kernels to kernel start pointers follows the same logic as which
> GRF start offsets to us
On 08/13/2014 06:16 PM, Ian Romanick wrote:
> On 08/11/2014 05:29 PM, Kristian Høgsberg wrote:
>> Right now we decide which kernels to use and the GRF start offsets in
>> one place and emit the kernel pointers later. The logic of how to map
>> 8, 16 and 32 kernels to kernel start pointers follows
On Jun 26, 2014, at 7:15 AM, Francisco Jerez wrote:
> Tom Stellard writes:
>
>> v2:
>> - Report correct values for CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE and
>>CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE.
>> - Only define cl_khr_fp64 if the extension is supported.
>> - Remove trailing space fr
Fixes the LUMINANCE_ALPHA formats of fbo-clear-formats piglit test.
Signed-off-by: Ilia Mirkin
---
src/mesa/drivers/dri/nouveau/nouveau_driver.c | 10 +-
src/mesa/drivers/dri/nouveau/nouveau_util.h | 9 +
2 files changed, 18 insertions(+), 1 deletion(-)
diff --git a/src/mesa/
Patches 7, 8, and 9 are
Reviewed-by: Ian Romanick
I made a few minor comments on 7, but they should be trivial to resolve.
Patches 10 through 13 are
Acked-by: Ian Romanick
I'd like to see some feed back on those last four from Ken and / or
Matt. I'd also like, if possible, for this to land
On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> It can be used to implement, for example, threaded glCompileShader and
> glLinkProgram.
>
> v2: allow tasks to "complete" other tasks
>
> Signed-off-by: Chia-I Wu
> ---
> src/glsl/Makefile.am | 12 +-
> src/glsl/Makefile.sources
On Wed, Aug 13, 2014 at 5:18 PM, Roland Scheidegger
wrote:
> Reviewed-by: Roland Scheidegger
>
> Not sure though why you moved the function, it's declared elsewhere anyway.
>
Heh, I guess you're right. I'll un-move the function to reduce churn.
> (And I bet there's cases where transfer ops w
Reviewed-by: Roland Scheidegger
Not sure though why you moved the function, it's declared elsewhere anyway.
(And I bet there's cases where transfer ops wouldn't actually be
required which aren't caught by this check, like for instance RGB
formats with only alpha pixel scale, but that was already
On Tuesday, July 29, 2014 12:36:33 PM Petri Latvala wrote:
> Move the IR sexp builder helpers and test script creation parts of
> tests/lower_jumps/create_test_cases.py into tests/test_case_generator.py
>
> No functional changes.
>
> Signed-off-by: Petri Latvala
> ---
> src/glsl/tests/lower_jum
On Tuesday, July 29, 2014 12:36:39 PM Petri Latvala wrote:
> tests/minmax/create_test_cases.py generates the following tests:
>
> multiple_min*.opt_test:
> Construct a tree of min expressions for all permutations of a var_ref
> and three constants. They should all optimize to a single min with
>
On Tuesday, July 29, 2014 12:36:34 PM Petri Latvala wrote:
> Sort expression operands when possible so that building expected IR
> sexps doesn't need to know which ordering will be produced by an
> optimization pass.
>
> Signed-off-by: Petri Latvala
> ---
> src/glsl/tests/compare_ir | 4 ++--
>
https://bugs.freedesktop.org/show_bug.cgi?id=82539
--- Comment #7 from Emil Velikov ---
OK apart from hacking a ton of printfs into configure.ac there is one more
thing which you can try:
Please revert 2af28040d639dddbb7c258981a00eaf3dfcbcf03 on top of master,
cleanup everything (make clean && gi
On 08/13/2014 11:18 AM, Emil Velikov wrote:
This reverts commit 2af28040d639dddbb7c258981a00eaf3dfcbcf03.
The commit was resolving an issue where libtool will not setup the
environment correctly when one explicitly provides --enable-{32,64}-bit
at configure time. It was caused due to the "-m32,6
https://bugs.freedesktop.org/show_bug.cgi?id=82539
--- Comment #6 from Vinson Lee ---
Created attachment 104593
--> https://bugs.freedesktop.org/attachment.cgi?id=104593&action=edit
Makefile
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freedesktop.org/show_bug.cgi?id=82539
--- Comment #5 from Vinson Lee ---
$ make V=1 -C src/gallium/winsys/svga/drm
make: Entering directory `mesa/src/gallium/winsys/svga/drm'
/bin/bash ../../../../../libtool --tag=CC --mode=compile gcc
-DPACKAGE_NAME=\"Mesa\" -DPACKAGE_TARNAME=\"m
On Wed, Aug 13, 2014 at 6:15 PM, Emil Velikov wrote:
> On 13/08/14 23:01, Ilia Mirkin wrote:
>> On Wed, Aug 13, 2014 at 5:56 PM, Emil Velikov
>> wrote:
>>> On 13/08/14 22:22, Ilia Mirkin wrote:
On Wed, Aug 13, 2014 at 5:16 PM, Emil Velikov
wrote:
> Signed-off-by: Emil Velikov
>>
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #26 from Marek Olšák ---
(In reply to comment #25)
> Recently tried:
>
> exec=env R600_DEBUG=nosb firefox
>
> After exiting firefox crashed too.
>
> P.S. Let me guess, you need output (do you ?)
Not really. Your previous backtrace
On 13/08/14 23:01, Ilia Mirkin wrote:
> On Wed, Aug 13, 2014 at 5:56 PM, Emil Velikov
> wrote:
>> On 13/08/14 22:22, Ilia Mirkin wrote:
>>> On Wed, Aug 13, 2014 at 5:16 PM, Emil Velikov
>>> wrote:
Signed-off-by: Emil Velikov
---
Unless someone object I would like to squash
On Wed, Aug 13, 2014 at 5:56 PM, Emil Velikov wrote:
> On 13/08/14 22:22, Ilia Mirkin wrote:
>> On Wed, Aug 13, 2014 at 5:16 PM, Emil Velikov
>> wrote:
>>> Signed-off-by: Emil Velikov
>>> ---
>>>
>>> Unless someone object I would like to squash this patch with the
>>> previous one.
>>>
>>> -Emi
On 13/08/14 22:22, Ilia Mirkin wrote:
> On Wed, Aug 13, 2014 at 5:16 PM, Emil Velikov
> wrote:
>> Signed-off-by: Emil Velikov
>> ---
>>
>> Unless someone object I would like to squash this patch with the
>> previous one.
>>
>> -Emil
>>
>> docs/autoconf.html | 16 ++--
>> 1 file chan
On Mon, Aug 11, 2014 at 10:35 PM, Kristian Høgsberg wrote:
> On Mon, Aug 11, 2014 at 08:08:33PM -0700, Ben Widawsky wrote:
>> On Mon, Aug 11, 2014 at 05:29:32PM -0700, Kristian Høgsberg wrote:
>> > const unsigned *assembly = NULL;
>> > if (brw->gen >= 8) {
>> >gen8_fs_generator g(b
Looks good. Thanks Emil!
Reviewed-by: Matt Turner
___
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=81680
--- Comment #25 from Eugene ---
Recently tried:
exec=env R600_DEBUG=nosb firefox
After exiting firefox crashed too.
P.S. Let me guess, you need output (do you ?)
--
You are receiving this mail because:
You are the assignee for the bug.
_
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #24 from Marek Olšák ---
No, just try to reproduce the bug while the environment variable is set.
--
You are receiving this mail because:
You are the assignee for the bug.
___
mesa-dev mai
On Wed, Aug 13, 2014 at 5:16 PM, Emil Velikov wrote:
> Signed-off-by: Emil Velikov
> ---
>
> Unless someone object I would like to squash this patch with the
> previous one.
>
> -Emil
>
> docs/autoconf.html | 16 ++--
> 1 file changed, 14 insertions(+), 2 deletions(-)
>
> diff --git
Signed-off-by: Emil Velikov
---
Unless someone object I would like to squash this patch with the
previous one.
-Emil
docs/autoconf.html | 16 ++--
1 file changed, 14 insertions(+), 2 deletions(-)
diff --git a/docs/autoconf.html b/docs/autoconf.html
index f27838f..c225659 100644
-
Signed-off-by: Jan Vesely
---
src/mapi/glapi/gen/gl_gentable.py | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/mapi/glapi/gen/gl_gentable.py
b/src/mapi/glapi/gen/gl_gentable.py
index 7577b66..c49f9a5 100644
--- a/src/mapi/glapi/gen/gl_gentable.py
+++ b/src/mapi/glapi
Signed-off-by: Jan Vesely
---
My llvm-config --version is
3.6.0git-svn-r215564-cd35a3b3
This patch assumes that the interesting part consists of only digits and dots.
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index 4ff87eb..dc
https://bugs.freedesktop.org/show_bug.cgi?id=82546
--- Comment #6 from Emil Velikov ---
My plan is to revert the offending commit and remove the enable-32,64-bit hacks
that the commit was fixing.
I'm also adding a note how to properly do cross-compile/multilib builds :)
--
You are receiving thi
On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> Inherit gl_shader_program and add save/restore functions to save precompile
> results in the shader programs. When DeferLinkProgram is set, we will save
> the precompile results instead of uploading them immediately because we may be
> on a different thre
https://bugs.freedesktop.org/show_bug.cgi?id=82546
Emil Velikov changed:
What|Removed |Added
CC||aaronbotte...@gmail.com
--- Comment #5 fr
https://bugs.freedesktop.org/show_bug.cgi?id=82581
Emil Velikov changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> @@ -3489,6 +3508,18 @@ struct gl_constants
> GLfloat MaxFragmentInterpolationOffset;
>
> GLboolean FakeSWMSAA;
> +
> + /*
> +* Defer certain operations to a thread pool.
> +*
> +* When DeferLinkProgram is set, these functions must b
This series fell off my radar. Sorry about the delays. :(
Patches 1 through 5 are
Reviewed-by: Ian Romanick
I'm still working my way through the rest. I'd like to see these land
as soon as reasonably possible.
On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> We are about to change mesa to spawn th
* --enable-{32,64}-bit is done. Use --build and --host instead.
* Configure does not add "-g -O2" to C{,XX}FLAGS.
* Pkg-config has been mandatory for a while now.
* Avoid using LDFLAGS, refer to pkg-config.
* --with-expat is deprecated. Use pkg-config.
Signed-off-by: Emil Velikov
---
Matt h
https://bugs.freedesktop.org/show_bug.cgi?id=82581
Aaron B changed:
What|Removed |Added
Summary|Build fails.|Build fails on OSMesa
|
https://bugs.freedesktop.org/show_bug.cgi?id=82581
Priority: medium
Bug ID: 82581
Assignee: mesa-dev@lists.freedesktop.org
Summary: Build fails.
Severity: major
Classification: Unclassified
OS: Linux (All)
Reporter:
On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> Add _mesa_enable_glsl_threadpool to enable the thread pool for a context, and
> add ctx->Const.DeferCompileShader and ctx->Const.DeferLinkProgram to
> fine-control what gets threaded.
>
> Setting DeferCompileShader to true will make _mesa_glsl_compile_sha
On Monday, August 11, 2014 05:29:40 PM Kristian Høgsberg wrote:
> GetTexParamterfv() doesnt change texture state, so instead of
> _mesa_lock_texture() we can use _mesa_lock_context_textures(),
> which doesn't increase the texture stamp. With this change,
> _mesa_update_state_locked() is now only c
From: Kenneth Graunke
The pass breaks live ranges of virtual registers by allocating new
registers when it sees an assignment to a virtual GRF it's already seen
written.
total instructions in shared programs: 4337879 -> 4335014 (-0.07%)
instructions in affected programs: 343865 -> 341000 (-0
https://bugs.freedesktop.org/show_bug.cgi?id=82539
--- Comment #4 from Emil Velikov ---
Hmm I still don't see how this can even remotely happen. To make things even
"better" it builds like a charm on my system :\
Can you attach src/gallium/winsys/svga/drm/Makefile and pastebin the output of
$ m
On Wednesday, August 13, 2014 12:03:10 PM Ian Romanick wrote:
> On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> > There may be two contexts compiling shaders at the same time, and we want
> > the
> > anonymous struct id to be globally unique.
> >
> > Signed-off-by: Chia-I Wu
> > ---
> > src/glsl/gls
On Wed, Aug 13, 2014 at 11:51 AM, Christian König
wrote:
> Am 13.08.2014 um 17:13 schrieb Ilia Mirkin:
>
>> On Wed, Aug 13, 2014 at 11:07 AM, Christian König
>> wrote:
>>>
>>> From: Christian König
>>>
>>> This fixes an issue with flash where it tries to destroy a decoder
>>> after already destr
On Wed, Aug 13, 2014 at 3:30 PM, Andreas Boll
wrote:
> accel_working2 returns 3 if the new firmware is used.
>
> The comment wasn't updated in v3 of commit:
> 36771dc winsys/radeon: fix nop packet padding for hawaii
>
> Signed-off-by: Andreas Boll
Reviewed-by: Alex Deucher
> ---
> src/gallium
On 07/09/2014 07:42 AM, Brian Paul wrote:
> On 07/09/2014 01:47 AM, Chia-I Wu wrote:
>> It can be used to implement, for example, threaded glCompileShader and
>> glLinkProgram.
>>
>> v2: allow tasks to "complete" other tasks
>>
>> Signed-off-by: Chia-I Wu
>> ---
>> src/glsl/Makefile.am
On Thu, Aug 07, 2014 at 12:14:24PM +0200, Bruno Jiménez wrote:
> This patch adds a new struct: r600_transfer_global. It will
> act as a wrapper around an r600_resource_global and an r600_transfer.
>
> It will be used for calling r600_compute_global_transfer_unmap when
> transfer_unmap is called. A
On Tue, Aug 05, 2014 at 07:06:51PM +0200, David Heidelberger wrote:
> fixes piglit glean "do-loop with continue and break" on RS690
>
> It's based on Tom Stellard patch and improved to handle CMP instruction.
>
> [v2] handle CMP instruction
>
> Signed-off-by: David Heidelberger
I've pushed thi
On 08/13/2014 12:20 PM, Matt Turner wrote:
On Wed, Aug 13, 2014 at 11:03 AM, Emil Velikov wrote:
On 13/08/14 18:38, Matt Turner wrote:
On Wed, Aug 13, 2014 at 10:18 AM, Emil Velikov wrote:
These two were added ages ago, with an explicit comment "Hacks ..."
They have been insufficient for yea
https://bugs.freedesktop.org/show_bug.cgi?id=82534
--- Comment #8 from Vinson Lee ---
(In reply to comment #7)
> Created attachment 104581 [details] [review]
> Jose's patch
>
> Can you give this a bash ? I'm guessing that some of the formatting has been
> causing problems.
attachment 104581 fix
https://bugs.freedesktop.org/show_bug.cgi?id=82534
--- Comment #7 from Emil Velikov ---
Created attachment 104581
--> https://bugs.freedesktop.org/attachment.cgi?id=104581&action=edit
Jose's patch
Can you give this a bash ? I'm guessing that some of the formatting has been
causing problems.
-
accel_working2 returns 3 if the new firmware is used.
The comment wasn't updated in v3 of commit:
36771dc winsys/radeon: fix nop packet padding for hawaii
Signed-off-by: Andreas Boll
---
src/gallium/winsys/radeon/drm/radeon_drm_cs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --
https://bugs.freedesktop.org/show_bug.cgi?id=82534
--- Comment #6 from Vinson Lee ---
(In reply to comment #4)
> You can also try Jose's patch
> http://patchwork.freedesktop.org/patch/31651/
This patch does not apply.
--
You are receiving this mail because:
You are the assignee for the bug.
__
https://bugs.freedesktop.org/show_bug.cgi?id=82534
--- Comment #5 from Vinson Lee ---
(In reply to comment #3)
> Can you give http://patchwork.freedesktop.org/patch/31650/ a try ?
>
> It brings us back to the original behaviour by dropping the headers as scons
> parses through the list. It will
Thanks for fixing!
Tested-by: Andreas Boll
2014-08-11 22:46 GMT+02:00 Marek Olšák :
> From: Marek Olšák
>
> Somebody forgot to do this. It was uncovered by recent st/mesa changes.
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=82139
>
> Cc: mesa-sta...@lists.freedesktop.org
> ---
> s
https://bugs.freedesktop.org/show_bug.cgi?id=82539
--- Comment #3 from Vinson Lee ---
drm.h is at /usr/include/drm/drm.h
$ pkg-config --cflags libdrm
-I/usr/include/drm
--
You are receiving this mail because:
You are the assignee for the bug.
___
mes
On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> There may be two contexts compiling shaders at the same time, and we want the
> anonymous struct id to be globally unique.
>
> Signed-off-by: Chia-I Wu
> ---
> src/glsl/glsl_parser_extras.cpp | 10 --
> 1 file changed, 8 insertions(+), 2 deletio
On 07/09/2014 12:47 AM, Chia-I Wu wrote:
> The compiler may be used by multiple contexts simultaneously and needs to be
> thread-safe.
>
> Signed-off-by: Chia-I Wu
> ---
> src/glsl/strtod.cpp | 29 +++--
> 1 file changed, 15 insertions(+), 14 deletions(-)
>
> diff --git
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #23 from Eugene ---
(In reply to comment #22)
> What happens if you set this environment variable and test firefox?
>
> R600_DEBUG=nosb
Whan do you mean 'test firefox' ? Try to debug it again ?
--
You are receiving this mail becau
https://bugs.freedesktop.org/show_bug.cgi?id=82539
Vinson Lee changed:
What|Removed |Added
Keywords||bisected
--- Comment #2 from Vinson Lee --
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #22 from Marek Olšák ---
What happens if you set this environment variable and test firefox?
R600_DEBUG=nosb
--
You are receiving this mail because:
You are the assignee for the bug.
___
https://bugs.freedesktop.org/show_bug.cgi?id=80561
--- Comment #1 from Christian König ---
Created attachment 104577
--> https://bugs.freedesktop.org/attachment.cgi?id=104577&action=edit
Possible fix
The attached patch should fix the issue, please test.
--
You are receiving this mail because
On Wed, Aug 13, 2014 at 11:03 AM, Emil Velikov wrote:
> On 13/08/14 18:38, Matt Turner wrote:
>> On Wed, Aug 13, 2014 at 10:18 AM, Emil Velikov
>> wrote:
>>> These two were added ages ago, with an explicit comment "Hacks ..."
>>> They have been insufficient for years and maintainers needed to
>>
The _mesa_swizzle_and_convert path can't do transfer ops, so we should bail
if they're needed.
Signed-off-by: Jason Ekstrand
---
src/mesa/main/texstore.c | 63 +---
1 file changed, 33 insertions(+), 30 deletions(-)
diff --git a/src/mesa/main/texstore.
On 13/08/14 18:38, Matt Turner wrote:
> On Wed, Aug 13, 2014 at 10:18 AM, Emil Velikov
> wrote:
>> These two were added ages ago, with an explicit comment "Hacks ..."
>> They have been insufficient for years and maintainers needed to
>> explicitly handle the build themselves.
>>
>> Rather than ly
Yeah, I think you're right about that. I'll send out a quick fix-up
patch. I guess we don't have piglit tests for that either.
--Jason
On Wed, Aug 13, 2014 at 10:56 AM, Roland Scheidegger
wrote:
> Oh and forgot to mention, on a quick looked it seemed to me like the
> texstore_swizzle path cou
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #21 from Eugene ---
Recently tried another way. Please, look at attachment. May be this way result
will be something helpfull (?)
--
You are receiving this mail because:
You are the assignee for the bug.
Oh and forgot to mention, on a quick looked it seemed to me like the
texstore_swizzle path could potentially forget about transfer ops, or
can't this happen? Of course at least the state tracker would not be
affected by this (as it implements the transfer ops on its own).
Roland
Am 13.08.2014 19:
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #20 from Eugene ---
Created attachment 104576
--> https://bugs.freedesktop.org/attachment.cgi?id=104576&action=edit
Debugging output
--
You are receiving this mail because:
You are the assignee for the bug.
___
Ha one minute faster :-).
FWIW interestingly this one also fixes the other conform failures I was
seeing (with GL_BYTE/GL_RGB data) when unconditionally disabling
texstore_swizzle so I guess the results of the swizzle path aren't quite
identical to the fallback one, which is a bit worrying.
Roland
The _mesa_swizzle_and_convert function can't handle them. Now we fall back
to the more generic path which can handle it instead of failing an assert.
Signed-off-by: Jason Ekstrand
---
src/mesa/main/texstore.c | 4
1 file changed, 4 insertions(+)
diff --git a/src/mesa/main/texstore.c b/src
On Wed, Aug 13, 2014 at 10:18 AM, Emil Velikov wrote:
> These two were added ages ago, with an explicit comment "Hacks ..."
> They have been insufficient for years and maintainers needed to
> explicitly handle the build themselves.
>
> Rather than lying and pretending that it works, just kill this
Roland,
I just sent the exact same patch.
Reviewed-by: Jason Ekstrand
On Wed, Aug 13, 2014 at 10:33 AM, wrote:
> From: Roland Scheidegger
>
> This got broken by 3dbf5bf6571e0c9d3e4febce01dea82be190d9d2.
> GL_COLOR_INDEX data is still supported (in legacy contexts), but the new
> texstore_swi
From: Roland Scheidegger
This got broken by 3dbf5bf6571e0c9d3e4febce01dea82be190d9d2.
GL_COLOR_INDEX data is still supported (in legacy contexts), but the new
texstore_swizzle path cannot handle it (and didn't detect this).
Unfortunately there's no piglit test trying to specify textures with a
GL
On Tue, Aug 12, 2014 at 8:52 PM, Roland Scheidegger
wrote:
> Am 06.08.2014 11:28, schrieb Michel Dänzer:
> > On 06.08.2014 03:08, Jason Ekstrand wrote:
> >> Module: Mesa
> >> Branch: master
> >> Commit: 850fb0d1dca616179d3239a7b7bd94fe1979604c
> >> URL:
> https://urldefense.proofpoint.com/v1/url?
These two were added ages ago, with an explicit comment "Hacks ..."
They have been insufficient for years and maintainers needed to
explicitly handle the build themselves.
Rather than lying and pretending that it works, just kill this hack and
let maintainers build things the way it should be done
This reverts commit 2af28040d639dddbb7c258981a00eaf3dfcbcf03.
The commit was resolving an issue where libtool will not setup the
environment correctly when one explicitly provides --enable-{32,64}-bit
at configure time. It was caused due to the "-m32,64" C{,XX}FLAGS being
set too late relative to
From: Ville Syrjälä
i915 fragment programs utilize the texture coordinate registers
for both texture coordinates and varyings. Unfortunately the
code doesn't check if the same index might be in use for both.
It just naively uses the index to pick a texture unit, which
could lead to collisions.
A
https://bugs.freedesktop.org/show_bug.cgi?id=81680
--- Comment #19 from Eugene ---
Yes, it seems several crash happens during debugging. But Firefox not closed.
And at last I reached the crash that happens after closing Firefox. But if
several crashes happens before the crash closing Firefox, how
On Fri, Aug 08, 2014 at 05:28:59PM +0300, Pekka Paalanen wrote:
> From: Pekka Paalanen
>
> The EGL_EXT_image_dma_buf_import specification was revised (according to
> its revision history) on Dec 5th, 2013, for EGL to not take ownership of
> the file descriptors.
>
> Do not close the file descrip
https://bugs.freedesktop.org/show_bug.cgi?id=82546
--- Comment #4 from Tobias Klausmann ---
git bisect shows
2af28040d639dddbb7c258981a00eaf3dfcbcf03 is the first bad commit
commit 2af28040d639dddbb7c258981a00eaf3dfcbcf03
Author: Alexandre Demers
Date: Fri Nov 22 20:06:20 2013 -0500
conf
1 - 100 of 118 matches
Mail list logo
