https://bugs.freedesktop.org/show_bug.cgi?id=99677
Samuel Iglesias changed:
What|Removed |Added
CC||sigles...@igalia.com
--
You are recei
https://bugs.freedesktop.org/show_bug.cgi?id=99677
--- Comment #1 from Samuel Iglesias ---
I am not able to reproduce it with current master
(21efe2528cd88cb09ba9f69222f69a8ee47611c9), but I might have something wrong.
My CFLAGS and CXXFLAGS are "-Wall -g -O0 -fsanitize=address
-fno-omit-frame-p
https://bugs.freedesktop.org/show_bug.cgi?id=99677
--- Comment #2 from Bartosz Tomczyk ---
Hi Samuel,
I don't have setup to reproduce it right now. I will update you will more
details in the evening.
But I looked at current mesa git and the bug is definitively there.
Look at src/compiler/glsl
https://bugs.freedesktop.org/show_bug.cgi?id=99677
--- Comment #3 from Samuel Iglesias ---
(In reply to Bartosz Tomczyk from comment #2)
> Hi Samuel,
>
> I don't have setup to reproduce it right now. I will update you will more
> details in the evening.
>
> But I looked at current mesa git and
On 06.02.2017 20:57, Dave Airlie wrote:
On 26.01.2017 12:50, Nicolai Hähnle wrote:
Hi all,
this is mostly motivated by the need to support more screen depths than
just 24/32-bit RGB in Glamor. The extension is simple enough, just adding
three more enums that are accepted.
I _think_ every DR
Hi Nicolai,
On 7 February 2017 at 09:07, Nicolai Hähnle wrote:
> On 06.02.2017 20:57, Dave Airlie wrote:
>> I just saw this, EGL_MESA_drm_image is really not a good place to start,
>>
>> we have two specs instead,
>> EGL_EXT_image_dma_buf_import.
>> EGL_MESA_image_dma_buf_export
>>
>> does one of
On Tue, Feb 7, 2017 at 2:57 AM, Kenneth Graunke wrote:
> On Monday, February 6, 2017 8:54:40 PM PST Marek Olšák wrote:
>> On Mon, Feb 6, 2017 at 8:20 PM, Ernst Sjöstrand wrote:
>> > FYI glmark2 segfaults with mesa_glthread=true. Expected that some programs
>> > will segfault?
>>
>> Yes, even segf
https://bugs.freedesktop.org/show_bug.cgi?id=97879
Marek Olšák changed:
What|Removed |Added
Assignee|mesa-dev@lists.freedesktop. |dri-devel@lists.freedesktop
The get_variable_being_redeclared() function can free 'var' because
a re-declaration of an unsized array variable can establish the size, so
we set the array type to the earlier declaration and free 'var' as it is
not needed anymore.
However, the same 'var' is referenced later in ast_declarator_li
https://bugs.freedesktop.org/show_bug.cgi?id=99677
--- Comment #4 from Samuel Iglesias ---
I have sent a patch to fix this:
https://lists.freedesktop.org/archives/mesa-dev/2017-February/143329.html
Would you mind testing it with address sanitizer? I would like to confirm this
error disappears w
Hi Samuel,
Var pointer is passed by value to get_variable_being_redeclared, so it
will not fix the issue. I thinks it should be changed to pointer to pointer.
On Tue, Feb 7, 2017 at 11:45 AM, Samuel Iglesias Gonsálvez <
sigles...@igalia.com> wrote:
> The get_variable_being_redeclared() function
On Tue, 2017-02-07 at 12:01 +0100, Bartosz Tomczyk wrote:
> Hi Samuel,
>
> Var pointer is passed by value to get_variable_being_redeclared, so
> it will not fix the issue. I thinks it should be changed to pointer
> to pointer.
>
It is already a pointer. See get_variable_being_redeclared() signa
On 6 February 2017 at 22:44, Ilia Mirkin wrote:
> May I ask why
>
> commit 7d3f9ed71c71637a91ecf15f50dbe7578a65d57e
> Author: Ilia Mirkin
> Date: Wed Jan 25 22:31:58 2017 -0500
>
> st/mesa: MAX_VARYING is the max supported number of patch varyings, not
> min
>
> and
>
> commit e73f87fcbdcb
On Mon, Feb 6, 2017 at 11:48 PM, Marek Olšák wrote:
> On Mon, Feb 6, 2017 at 9:27 PM, Jan Ziak <0xe2.0x9a.0...@gmail.com> wrote:
>> Hello
>>
>> I am against application profiles - in the form of "a
>> community-maintained whitelist of apps" or in any other form
>> explicitly associating the name/I
Yep, it's pointer but should be "pointer to pointer" (**var). Changing var
inside get_variable_being_redeclared function, doesn't change it's value
outside this function. But maybe I'm missing something.
On Tue, Feb 7, 2017 at 12:11 PM, Samuel Iglesias Gonsálvez <
sigles...@igalia.com> wrote:
> O
On 17.01.2017 18:31, Kyle Brenneman wrote:
> On 01/16/2017 02:32 PM, Adam Jackson wrote:
>> On Thu, 2017-01-05 at 14:29 -0700, Kyle Brenneman wrote:
>>> ---
>>> src/egl/generate/eglFunctionList.py | 6 --
>>> 1 file changed, 4 insertions(+), 2 deletions(-)
>> Reviewed-by: Adam Jackson
>>
>>
On Tue, 2017-02-07 at 12:01 +0100, Bartosz Tomczyk wrote:
> Hi Samuel,
> Var pointer is passed by value to get_variable_being_redeclared, so
> it will not fix the issue. I thinks it should be changed to pointer
> to pointer.
>
Oh, you are right... what a day! :)
I am going to send a v2.
Thanks!
Hi,
On 07.02.2017 13:23, Jan Ziak wrote:
On Mon, Feb 6, 2017 at 11:48 PM, Marek Olšák wrote:
On Mon, Feb 6, 2017 at 9:27 PM, Jan Ziak <0xe2.0x9a.0...@gmail.com> wrote:
Hello
I am against application profiles - in the form of "a
community-maintained whitelist of apps" or in any other form
exp
On 7 February 2017 at 11:23, Jan Ziak <0xe2.0x9a.0...@gmail.com> wrote:
> On Mon, Feb 6, 2017 at 11:48 PM, Marek Olšák wrote:
>> On Mon, Feb 6, 2017 at 9:27 PM, Jan Ziak <0xe2.0x9a.0...@gmail.com> wrote:
>>> Hello
>>>
>>> I am against application profiles - in the form of "a
>>> community-maintain
On 13 January 2017 at 21:59, Jason Ekstrand wrote:
> Also, something I would like to see (maybe a follow-on patch?) would a
> change to the mesa internal API to be able to put the SHA context on the
> stack and not need to malloc it. It's not really a memory or cycle-saving
> thing so much as it
The get_variable_being_redeclared() function can free 'var' because
a re-declaration of an unsized array variable can establish the size, so
we set the array type to the 'earlier' declaration and free 'var' as it is
not needed anymore.
However, the same 'var' is referenced later in ast_declarator_
https://bugs.freedesktop.org/show_bug.cgi?id=99701
--- Comment #2 from Emil Velikov ---
(In reply to Nicolai Hähnle from comment #1)
> Created attachment 129378 [details] [review]
> always include unistd.h
>
For the patch
Reviewed-by: Emil Velikov
> The attached patch might fix this.
>
> Out
Already used as such and annotated in a number of places.
Should resolve -Wignored-qualifiers warnings, such as:
src/intel/blorp/blorp_blit.c:1857:8: warning: 'const' type qualifier on
return type has no effect [-Wignored-qualifiers]
Signed-off-by: Emil Velikov
---
src/intel/blorp/blorp_blit.c
Hi,
On 06.02.2017 22:26, Samuel Pitoiset wrote:
On 02/06/2017 04:45 PM, Eero Tamminen wrote:
Results from quick try on Ubuntu 16.04 with today's version of the game
(ARK: Survival Evolved)...
With (Ubuntu 16.04 default) Mesa 11.2, game starts to the game main
menu, but when one starts from tha
From: Emil Velikov
Replace with AS_HELP_STRING and AC_MSG_ERROR respectively, as spotted by
autoupdate.
Note that the suggested AC_CANONICAL_SYSTEM > AC_CANONICAL_TARGET change
is not addressed here since that requires very extensive testing.
Signed-off-by: Emil Velikov
---
configure.ac | 4 +
On Feb 7, 2017 12:54 PM, "Emil Velikov" wrote:
On 7 February 2017 at 11:23, Jan Ziak <0xe2.0x9a.0...@gmail.com> wrote:
> On Mon, Feb 6, 2017 at 11:48 PM, Marek Olšák wrote:
>> On Mon, Feb 6, 2017 at 9:27 PM, Jan Ziak <0xe2.0x9a.0...@gmail.com>
wrote:
>>> Hello
>>>
>>> I am against application pr
2017-02-07 14:11 GMT+01:00 Eero Tamminen :
> Hi,
>
> On 06.02.2017 22:26, Samuel Pitoiset wrote:
>>
>> On 02/06/2017 04:45 PM, Eero Tamminen wrote:
>>>
>>> Results from quick try on Ubuntu 16.04 with today's version of the game
>>> (ARK: Survival Evolved)...
>>>
>>> With (Ubuntu 16.04 default) Mesa
Hi Nicolai,
Will you push it, if I change it as described in last mail ?
On Mon, Jan 30, 2017 at 3:31 PM, Bartosz Tomczyk <
bartosz.tomczy...@gmail.com> wrote:
> It did not change anything, as we are not dereferencing iterator after
> delete.
>
> I think changing:
> delete *it;
> uses.erase(it);
From: Emil Velikov
Cc: Edward O'Callaghan
Signed-off-by: Emil Velikov
---
src/gallium/winsys/intel/drm/Android.mk | 36 --
src/gallium/winsys/intel/drm/Makefile.am| 34 --
src/gallium/winsys/intel/drm/Makefile.sources | 3 -
src/gallium/winsys/intel/drm/intel_drm_public.h
>
> Hi,
>
> On Mon, 2017-02-06 at 13:43 +0100, Jan Ziak wrote:
> >* Shadow of Mordor benchmark: 30 FPS w/o glthread -> 20 FPS with
> *>* glthread
> *>
> For what it is worth, all the Feral games have a dispatch thread that
> primarily calls GL functions.
>
> James
>
>
Hello James,
Did you have th
On Tue, Feb 7, 2017 at 4:42 AM, Timothy Arceri wrote:
> Without the radeonsi patch min fps still goes up around 4 fps in
> the Shadow of Mordor benchmark (although the stalls are still
> very noticeable).
I don't fully understand this. Can you elaborate on that?
There are 4 tests that can be don
Hi Gregory,
On Tue, 2017-02-07 at 16:04 +0100, Gregory Hainaut wrote:
> > Hi,
> >
> > On Mon, 2017-02-06 at 13:43 +0100, Jan Ziak wrote:
> > > Shadow of Mordor benchmark: 30 FPS w/o glthread -> 20 FPS with
> > > glthread
> > >
> >
> > For what it is worth, all the Feral games have a dispatch th
How does it make any sense to declare const on the return type of a
function that doesn't return a pointer??? It's not returning by reference,
it's returning a copy so const on the return type of the function means
nothing.
On Tue, Feb 7, 2017 at 4:54 AM, Emil Velikov
wrote:
> Already used as s
Patch is:
Tested-by: Bartosz Tomczyk
I can confirm it fix use-after-free issue.
On Tue, Feb 7, 2017 at 1:47 PM, Samuel Iglesias Gonsálvez <
sigles...@igalia.com> wrote:
> The get_variable_being_redeclared() function can free 'var' because
> a re-declaration of an unsized array variable can esta
https://bugs.freedesktop.org/show_bug.cgi?id=99638
--- Comment #5 from intermedi...@hotmail.com ---
Created attachment 129396
--> https://bugs.freedesktop.org/attachment.cgi?id=129396&action=edit
all blue weston
Weston blue
--
You are receiving this mail because:
You are the QA Contact for t
Reviewed-by: Bruce Cherniak
> On Feb 6, 2017, at 6:40 PM, Tim Rowley wrote:
>
> Highlights include more avx512 work and fixes for llvm>=3.8.
>
> Tim Rowley (10):
> swr: [rasterizer core] Frontend SIMD16 WIP
> swr: [rasterizer jitter] Adjust jitter header includes
> swr: [rasterizer jitter]
On Mon, Feb 6, 2017 at 9:42 PM, Timothy Arceri
wrote:
> In order to avoid costly fallback recompiles when cache items are
> created with an old version of Mesa or for a different gpu on the
> same system we want to create directories that look like this:
>
> ./{MESA_VERSION_STRING}/{GPU_ID}
>
> F
Series fix various null pointer derefeneces repored by UBSAN.
Found by running piglit tests.
Bartosz Tomczyk (5):
gallium/u_inlines: fix member access within null pointer
util/list: fix member access within null pointer
st/mesa: fix member access within null pointer
gallium/auxiliary: fix
---
src/gallium/drivers/radeon/r600_pipe_common.c | 13 +++--
src/gallium/drivers/radeon/r600_pipe_common.h | 3 ++-
2 files changed, 9 insertions(+), 7 deletions(-)
diff --git a/src/gallium/drivers/radeon/r600_pipe_common.c
b/src/gallium/drivers/radeon/r600_pipe_common.c
index 95a6a486
---
src/gallium/auxiliary/pipebuffer/pb_buffer.h | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/src/gallium/auxiliary/pipebuffer/pb_buffer.h
b/src/gallium/auxiliary/pipebuffer/pb_buffer.h
index 33c23068c2..12c9ca779a 100644
--- a/src/gallium/auxiliary/pipebuffer/pb
---
src/gallium/auxiliary/util/u_inlines.h | 65 --
1 file changed, 39 insertions(+), 26 deletions(-)
diff --git a/src/gallium/auxiliary/util/u_inlines.h
b/src/gallium/auxiliary/util/u_inlines.h
index b7b8313583..3bb3bcd6e0 100644
--- a/src/gallium/auxiliary/util/
---
configure.ac| 3 +++
src/util/list.h | 9 +
2 files changed, 12 insertions(+)
diff --git a/configure.ac b/configure.ac
index a6ceee95a3..87f635c1a3 100644
--- a/configure.ac
+++ b/configure.ac
@@ -415,6 +415,9 @@ AC_C_BIGENDIAN(
little_endian=no
)
+dnl Chek for typeof suppo
---
src/mesa/state_tracker/st_manager.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/state_tracker/st_manager.c
b/src/mesa/state_tracker/st_manager.c
index c3d8286b5a..ad69ca6eb5 100644
--- a/src/mesa/state_tracker/st_manager.c
+++ b/src/mesa/state_tracker/st_manag
On 7 February 2017 at 16:11, Jason Ekstrand wrote:
> How does it make any sense to declare const on the return type of a function
> that doesn't return a pointer??? It's not returning by reference, it's
> returning a copy so const on the return type of the function means nothing.
>
Seems like you
On Tue, Feb 7, 2017 at 10:42 AM, Emil Velikov
wrote:
> On 7 February 2017 at 16:11, Jason Ekstrand wrote:
> > How does it make any sense to declare const on the return type of a
> function
> > that doesn't return a pointer??? It's not returning by reference, it's
> > returning a copy so const o
On Tuesday 07 February 2017, Dave Airlie wrote:
> From: Dave Airlie
>
> This enables a transfer queue using the SDMA engine on
> CIK/VI/Polaris GPUs.
>
> TODO:
> decide what to do with HW limitations from radeonsi
> (fail to record?)
> add linear bounds check to the buffer->image copies
>
> dEQ
On Wed, Feb 01, 2017 at 08:07:22PM -0800, Jason Ekstrand wrote:
> The only mechanism Vulkan provides for disabling stencil writes is to set
This isn't the only mechanism for explicitly disabling stencil writes.
Stencil writes can also be disabled by disabling stencil testing (as can
be seen in thi
On Tue, Feb 7, 2017 at 12:13 PM, Nanley Chery wrote:
> On Wed, Feb 01, 2017 at 08:07:22PM -0800, Jason Ekstrand wrote:
> > The only mechanism Vulkan provides for disabling stencil writes is to set
>
> This isn't the only mechanism for explicitly disabling stencil writes.
> Stencil writes can also
From: Emil Velikov
Not much point in the const qualifier since we provide a copy to the
user. Resolves the following -Wignored-qualifiers warning.
src/intel/blorp/blorp_blit.c:1857:8: warning: 'const' type qualifier on
return type has no effect [-Wignored-qualifiers]
Cc: Jason Ekstrand
Signed-
On Tue, 7 Feb 2017 11:40:12 -0600
Aaron Watry wrote:
> On Mon, Feb 6, 2017 at 9:42 PM, Timothy Arceri
> wrote:
>
> > In order to avoid costly fallback recompiles when cache items are
> > created with an old version of Mesa or for a different gpu on the
> > same system we want to create director
On Tue, 7 Feb 2017 16:54:31 +0100
Marek Olšák wrote:
> On Tue, Feb 7, 2017 at 4:42 AM, Timothy Arceri
> wrote:
> > Without the radeonsi patch min fps still goes up around 4 fps in
> > the Shadow of Mordor benchmark (although the stalls are still
> > very noticeable).
>
> I don't fully underst
When fixing the stalls on evergreen I introduced leaking of the useinfo
structure(s). Sorry. Instead of allocating a new object to hold 3 values
where only one is actually used, rework the list to just store the node
pointer. Thus no allocating and deallocation is needed. Since use_info
and use_kin
https://bugs.freedesktop.org/show_bug.cgi?id=98263
--- Comment #8 from Rene Lindsay ---
The "No DRI3 support" is only relevant when the desktop is running on Intel,
but also gets displayed whenever NVidia is used, regardless of whether DRI3 was
enabled or not. Maybe this message should be muted
On Wed, 8 Feb 2017 08:29:29 +1100
Timothy Arceri wrote:
> On Tue, 7 Feb 2017 11:40:12 -0600
> Aaron Watry wrote:
>
> > On Mon, Feb 6, 2017 at 9:42 PM, Timothy Arceri
> > wrote:
> >
> > > In order to avoid costly fallback recompiles when cache items are
> > > created with an old version of Mes
I'm not quite sure there's really a bug here?
As far as I can tell, these functions are architected specifically to
look like that - they do not actually dereference potential null
pointers, but only take the address in the end. This change seems to add
some overhead.
Roland
Am 07.02.2017 um 19
Yeah, it would never make sense to pass NULL as the first argument to
any of the _reference() functions.
Would putting an assert(ptr) at the start of pipe_surface_reference(),
for example, silence the UBSAN warning?
-Brian
On 02/07/2017 02:45 PM, Roland Scheidegger wrote:
I'm not quite sur
From: Dave Airlie
We need to initialize dcc like we do in the subpass path.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_resolve.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/amd/vulkan/radv_meta_resolve.c
b/src/amd/vulkan/radv_meta_resolve.c
index 6
On Tue, Feb 7, 2017, at 23:05, Dave Airlie wrote:
> From: Dave Airlie
>
> We need to initialize dcc like we do in the subpass path.
>
> Signed-off-by: Dave Airlie
> ---
> src/amd/vulkan/radv_meta_resolve.c | 5 -
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/src/amd/
On Tue, 2017-02-07 at 10:56 +0100, Marek Olšák wrote:
> On Tue, Feb 7, 2017 at 2:57 AM, Kenneth Graunke g> wrote:
> > On Monday, February 6, 2017 8:54:40 PM PST Marek Olšák wrote:
> > > On Mon, Feb 6, 2017 at 8:20 PM, Ernst Sjöstrand > > > wrote:
> > > > FYI glmark2 segfaults with mesa_glthread=t
On Tue, Feb 07, 2017 at 12:25:18PM -0800, Jason Ekstrand wrote:
> On Tue, Feb 7, 2017 at 12:13 PM, Nanley Chery wrote:
>
> > On Wed, Feb 01, 2017 at 08:07:22PM -0800, Jason Ekstrand wrote:
> > > The only mechanism Vulkan provides for disabling stencil writes is to set
> >
> > This isn't the only
Signed-off-by: Micah Fedke
---
src/egl/wayland/wayland-drm/wayland-drm.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/egl/wayland/wayland-drm/wayland-drm.c
b/src/egl/wayland/wayland-drm/wayland-drm.c
index 4fc12521d2..5c2bb0369e 100644
--- a/src/egl/wayland/wayla
From: Emil Velikov
With next commits we'll require --enable-gallium-llvm (en route to a
greater good later on) for RADV. The latter is required to ensure that
as otherwise we'll fail to build.
Cc: Dave Airlie
CC: Tobias Droste
Cc: "17.0"
Signed-off-by: Emil Velikov
---
configure.ac | 36 +++
From: Emil Velikov
Signed-off-by: Emil Velikov
---
configure.ac | 1 -
1 file changed, 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index a6ceee95a3..d9170770eb 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1021,7 +1021,6 @@ radeon_llvm_check() {
llvm_set_environment_variables
Hi all,
Here is hopefully the final round of LLVM build fixes which should
address both Jose and Dave's concerns to not touch any source code.
In return the configure.ac changes ended up be twice as large, but
overall fine. Considering that most of those must land in stable.
In gist -> enable-
From: Tobias Droste
This is actually not needed because the version is checked later.
Around line 2380
if test "x$enable_gallium_llvm" == "xyes"; then
llvm_check_version_for $LLVM_REQUIRED_GALLIUM "gallium"
llvm_add_default_components "gallium"
fi
Cc: "17.0"
Cc: Tobias Droste
Signed-o
From: Emil Velikov
With this change we effectively require --enable-gallium-llvm when
building RADV. This should be perfectly safe since the gallium radeonsi
driver already explicitly requires it.
The "gallium" part in --enable-gallium-llvm is about to be removed soon
(not in stable) but until t
From: Emil Velikov
Drop the gallium prefix since we're about it use it throughout the
configure.
Note we do want to check for enable_gallium_llvm check since (as
explicitly requested) the toggle should mean --enable-llvm. We cannot do
the latter for stable so it will come with later patch.
Cc:
From: Emil Velikov
... of non-empty --with-gallium-drivers.
Since we effectively made --enable-gallium-llvm mean --enable-llvm with
earlier commits, we need to remove the, now artificial, requirement.
Otherwise we'll error (as below) when building RADV.
configure: error: --enable-gallium-llvm
From: Tobias Droste
Set FOUND_LLVM only when LLVM is present (checking for exact version/etc
is deferred) and use enable-gallium-llvm to indicate the global LLVM
status.
Renaming the latter is not appropriate for stable patches, so we'll
address it with a later commit.
Loosely based on work by
From: Emil Velikov
Earlier refactoring commits started setting the above regardless if LLVM
is used or not. Move them to the respective section to restore the
original functionality.
Since we require the preprocessor flags (includes in particular) for the
header version parsing keep those as-is.
From: Emil Velikov
The extra function brings no added benefit as of earlier commit which
made llvm_require_version (as called by radeon_llvm_check) require LLVM
(--enable-gallium-llvm).
Fixes: 5f966a96af7 "configure.ac: Mandate --enable-gallium-llvm when
checking LLVM version"
Cc: Tobias Droste
From: Emil Velikov
Already implicitly handled throughout, but keep it clear and disable
gallium-llvm. This change should be a no-op.
Cc: Tobias Droste
Signed-off-by: Emil Velikov
---
configure.ac | 1 +
1 file changed, 1 insertion(+)
diff --git a/configure.ac b/configure.ac
index 469a09e286.
From: Emil Velikov
Currently we have extra (somewhat questionable) modularity, such that
one could build some parts with LLVM while others w/o.
That is extremely fragile, error prone and requires quite noticable
amount of code throughout.
Thus lets deprecate the gallium toggle in faviour of the
As stated in [1] by the LLVM devs, the new versioning scheme will not
deploy any minor version (i.e. it will always be zero). As such the
patch should not be needed.
This reverts commit 0e9a5be7e74fa2a9bd2a634ef60822bd6600ca1d.
[1] http://blog.llvm.org/2016/12/llvms-new-versioning-scheme.html
Rev
From: Emil Velikov
Cc: Axel Davy
Signed-off-by: Emil Velikov
---
Axel, others - can someone skimg through the file and double-check that
those are still valid ?
---
src/gallium/state_trackers/nine/README | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/gallium/state_tr
From: Emil Velikov
Although it works, it's not the correct thing to do.
v2: Rebase
Signed-off-by: Emil Velikov
Reviewed-by: Tobias Droste (v1)
---
configure.ac | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/configure.ac b/configure.ac
index 1b28eb3b3b..23e8a34da1 100
From: Tobias Droste
Make sure that HAVE_LLVM compiler define is only set if LLVM is
actually used.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99010
Cc: "17.0"
Signed-off-by: Tobias Droste
v2 [Emil] fold within the existing conditional
Reviewed-by: Emil Velikov
---
configure.ac |
From: Emil Velikov
LLVM_BINDIR is completely unused while others such as LLVM_LIBDIR are
used only internally. In the latter case there's no need to AC_SUBST it.
Signed-off-by: Emil Velikov
Reviewed-by: Tobias Droste
---
configure.ac | 5 -
1 file changed, 5 deletions(-)
diff --git a/con
From: Emil Velikov
Earlier refactoring commits changed from one, dare I say it, broken
behaviour to another. Namely:
Before, as you explicitly --enable-gallium-llvm your selection was
ignored when llvm-config was not present/detected.
Today, the "auto" heuristics enables gallium llvm regardless
https://bugs.freedesktop.org/show_bug.cgi?id=99010
--- Comment #19 from Emil Velikov ---
Gents do give this series a try
https://patchwork.freedesktop.org/series/19269/
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
On Tue, Feb 7, 2017 at 10:56 AM, Marek Olšák wrote:
> On Tue, Feb 7, 2017 at 2:57 AM, Kenneth Graunke wrote:
>> On Monday, February 6, 2017 8:54:40 PM PST Marek Olšák wrote:
>>> On Mon, Feb 6, 2017 at 8:20 PM, Ernst Sjöstrand wrote:
>>> > FYI glmark2 segfaults with mesa_glthread=true. Expected t
On Tue, Feb 7, 2017 at 4:42 AM, Timothy Arceri wrote:
> This series adds support for a GLSL IR level and TGSI (OpenGL/st)
> level caches.
>
> There are a few small bug fixes and a little bit of restructuring
> to the GLSL IR patches to make things work for Gallium drivers vs
> just i965 previously
https://bugs.freedesktop.org/show_bug.cgi?id=98471
Matt Turner changed:
What|Removed |Added
Depends on|97967 |
Referenced Bugs:
https://bugs.freedeskt
https://bugs.freedesktop.org/show_bug.cgi?id=99517
Matt Turner changed:
What|Removed |Added
Depends on||97967
Referenced Bugs:
https://bugs.free
https://bugs.freedesktop.org/show_bug.cgi?id=97967
Matt Turner changed:
What|Removed |Added
Blocks|98471 |99517
Referenced Bugs:
https://bugs.free
It doesn't really matter what order the binding table entries are in so
there's no real utility to returning the final offset from
assign_common_binding_table_offsets. The only functional change in this
patch is that framebuffer fetch surfaces are now right after render
targets rather than at the
The mark_surface_used function exists for a micro-optimization that
shrinks the binding table if some of the entries are unused. However,
we already create binding tables based on num_textures, num_ubos, etc.,
so the only case where this is actually able to eliminate something is
if the back-end o
Here's the new commit message:
Vulkan doesn't have a stencilWriteEnable bit like it does for depth.
Instead, you have a stencil mask. Since the stencil mask is handled as
dynamic state, we have to handle it later during command buffer
construction. This helps Dota2 by a couple pe
On Tue, Feb 7, 2017 at 1:26 PM, Emil Velikov
wrote:
> From: Emil Velikov
>
> Not much point in the const qualifier since we provide a copy to the
> user. Resolves the following -Wignored-qualifiers warning.
>
> src/intel/blorp/blorp_blit.c:1857:8: warning: 'const' type qualifier on
> return type
Signed-off-by: Ilia Mirkin
Reviewed-by: Emil Velikov
---
configure.ac | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index 7cb2c40..7bf3680 100644
--- a/configure.ac
+++ b/configure.ac
@@ -66,7 +66,12 @@ AC_SUBST([OSMESA_VERSION])
OPENCL_
On Tue, Feb 7, 2017 at 3:19 PM, Jason Ekstrand wrote:
> On Tue, Feb 7, 2017 at 1:26 PM, Emil Velikov
> wrote:
>
>> From: Emil Velikov
>>
>> Not much point in the const qualifier since we provide a copy to the
>> user. Resolves the following -Wignored-qualifiers warning.
>>
>> src/intel/blorp/bl
From: Dave Airlie
We need to initialize dcc like we do in the subpass path.
v2: fix initial/final layouts
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_resolve.c | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_resolve.c
b/src/
Reviewed-by: Bas Nieuwenhuizen
On Wed, Feb 8, 2017, at 00:20, Dave Airlie wrote:
> From: Dave Airlie
>
> We need to initialize dcc like we do in the subpass path.
>
> v2: fix initial/final layouts
> Signed-off-by: Dave Airlie
> ---
> src/amd/vulkan/radv_meta_resolve.c | 11 ---
> 1 f
From: Emil Velikov
Not much point in the const qualifier since we provide a copy to the
user. Resolves the following -Wignored-qualifiers warning.
src/intel/blorp/blorp_blit.c:1857:8: warning: 'const' type qualifier on
return type has no effect [-Wignored-qualifiers]
v2: keep const qualifier of
On Tue, 2017-02-07 at 23:58 +0100, Matt Turner wrote:
> On Tue, Feb 7, 2017 at 4:42 AM, Timothy Arceri > wrote:
> > This series adds support for a GLSL IR level and TGSI (OpenGL/st)
> > level caches.
> >
> > There are a few small bug fixes and a little bit of restructuring
> > to the GLSL IR patc
Reviewed-by: Jason Ekstrand
On Tue, Feb 7, 2017 at 3:24 PM, Emil Velikov
wrote:
> From: Emil Velikov
>
> Not much point in the const qualifier since we provide a copy to the
> user. Resolves the following -Wignored-qualifiers warning.
>
> src/intel/blorp/blorp_blit.c:1857:8: warning: 'const' t
Just FYI: I'm working with the fd.o admins to move the mesa3d.org website
from SourceForge to freedesktop.org. You may see "Unable to verify server
identity" warnings in your browser for a bit...
Eventually, the pages on www.mesa3d.org should get automatically updated
from git check-ins to the do
On Feb 7, 2017 11:55 PM, "Matt Turner" wrote:
On Tue, Feb 7, 2017 at 10:56 AM, Marek Olšák wrote:
> On Tue, Feb 7, 2017 at 2:57 AM, Kenneth Graunke
wrote:
>> On Monday, February 6, 2017 8:54:40 PM PST Marek Olšák wrote:
>>> On Mon, Feb 6, 2017 at 8:20 PM, Ernst Sjöstrand
wrote:
>>> > FYI glmar
Reviewed-by: Tobias Droste
Am Dienstag, 7. Februar 2017, 22:44:43 CET schrieb Emil Velikov:
> From: Emil Velikov
>
> Signed-off-by: Emil Velikov
> ---
> configure.ac | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/configure.ac b/configure.ac
> index a6ceee95a3..d9170770eb 100644
> --
Reviewed-by: Tobias Droste
Am Dienstag, 7. Februar 2017, 22:44:44 CET schrieb Emil Velikov:
> From: Emil Velikov
>
> With next commits we'll require --enable-gallium-llvm (en route to a
> greater good later on) for RADV. The latter is required to ensure that
> as otherwise we'll fail to build.
1 - 100 of 127 matches
Mail list logo
