Instead of doing a memcpy, this moves us to start with a blank
instruction (memset to zero) and copy the fields over one at a time.
Cc: mesa-sta...@lists.freedesktop.org
---
src/intel/compiler/brw_eu_emit.c | 95 +++-
1 file changed, 94 insertions(+), 1
Instead of the state stack that's based on copying a dummy instruction
around, we start using a logical stack of brw_insn_states. This uses a
bit less memory and is way less conceptually bogus.
---
src/intel/compiler/brw_eu.c | 90 +---
This is much cleaner than everything that wants a default value poking
at the bits of p->current directly.
---
src/intel/compiler/brw_eu.c | 29 ++
src/intel/compiler/brw_eu.h | 3 +
src/intel/compiler/brw_eu_emit.c| 98 +++--
Prior to gen8, the flag [sub]register number is in a different spot on
3src instructions than on other instructions. Starting with Broadwell,
they made it consistent. This commit fixes bugs that occur when a
conditional modifier gets propagated into a 3src instruction such as a
MAD.
Cc:
On Tue, May 29, 2018 at 7:41 AM, Eric Engestrom
wrote:
> Cc: Emil Velikov
> Cc: Daniel Stone
> Cc: Andres Gomez
> Cc: Dylan Baker
> Signed-off-by: Eric Engestrom
> ---
> docs/relnotes/18.2.0.html | 8
> 1 file changed, 8 insertions(+)
>
> diff --git a/docs/relnotes/18.2.0.html
Nevermind, Jason used a git revert trick to get it to work. The v2 will
have the diff and preserve the git file history.
On Fri, May 25, 2018 at 7:58 PM, Laura Ekstrand
wrote:
> I specifically tried forcing a rename earlier, but it doesn't work. Git
> sees too much change. The only way I
Reviewed-by: Marek Olšák
Marek
On Mon, May 28, 2018 at 7:20 AM, Timothy Arceri
wrote:
> This is required for tessellation shader Compat profile support.
> ---
> src/mesa/main/dlist.c | 78 +++
> 1 file changed, 78 insertions(+)
>
> diff --git
https://bugs.freedesktop.org/show_bug.cgi?id=106619
Jan Vesely changed:
What|Removed |Added
Blocks||99553
Resolution|---
On Tuesday, May 29, 2018 9:28:07 AM PDT Chris Wilson wrote:
> Commit 92f01fc5f914 ("i965: Emit VF cache invalidates for 48-bit
> addressing bugs with softpin.") tried to only emit the VF invalidate if
> the high bits changed, but it accidentally always set need_invalidate to
> true; causing it to
On Tue, 2018-05-29 at 13:08 +1000, Dave Airlie wrote:
> This is a rebased version of Marc-Andre's patches, I'll put my
> r-b on them but I thought it would be nice to have others have a look.
>
> The only difference are I added a patch on the end, and the interface
> version is now 4 instead of
Reviewed-by: Marek Olšák
Marek
On Tue, May 29, 2018 at 7:24 AM, Tapani Pälli
wrote:
> Patch skips useless and possibly dangerous calls down to the driver
> in case invalid arguments were given. I noticed this would be happening
> with demo of Darwinia game. AFAIK this does not fix anything
On Tue, 2018-05-29 at 09:54 -0700, Dylan Baker wrote:
> Quoting Adam Jackson (2018-05-29 06:50:46)
> > GL_LIB expands to GLX_mesa, but applications should not link against
> > that. -lGL is never wrong, just hardcode it.
>
> Actually There is this really stupid option in the autotools build
For the series:
Reviewed-by: Marek Olšák
Marek
On Mon, May 28, 2018 at 9:26 PM, Dave Airlie wrote:
> From: Dave Airlie
>
> This just renames this as we want to add an shm handle which
> isn't really drm related.
>
> Originally by: Marc-André Lureau
> (airlied: I used this sed script
On Mon, May 28, 2018 at 3:52 AM, Nicolai Hähnle wrote:
> From: Nicolai Hähnle
>
> The only effect this has is that written cache lines are immediately
> freed in L1$. We're not going to read the data again, so it's better
> to leave room for other things in the cache.
> ---
>
On Tue, 2018-05-29 at 18:35 +0100, Emil Velikov wrote:
> On 29 May 2018 at 15:29, Daniel Stone wrote:
> > On 29 May 2018 at 15:17, Eric Engestrom wrote:
> > > On Tuesday, 2018-05-29 09:50:46 -0400, Adam Jackson wrote:
> > > > GL_LIB expands to GLX_mesa, but applications should not link against
>
On 29 May 2018 at 15:29, Daniel Stone wrote:
> On 29 May 2018 at 15:17, Eric Engestrom wrote:
>> On Tuesday, 2018-05-29 09:50:46 -0400, Adam Jackson wrote:
>>> GL_LIB expands to GLX_mesa, but applications should not link against
>>> that. -lGL is never wrong, just hardcode it.
>>
>> Agreed, and
On 25 May 2018 at 10:05, Juan A. Suarez Romero wrote:
> On Tue, 2018-05-22 at 10:48 -0700, Dylan Baker wrote:
>> This looks good to me. I'm also opened to doing the 18.1.x releases if Emil
>> would rather not do them (and I have been updating my 18.1-proposed branch
>> either way).
>>
>
> I'm
On 21 May 2018 at 18:34, Dylan Baker wrote:
> ---
> docs/releasing.html | 19 +++
> 1 file changed, 19 insertions(+)
>
> diff --git a/docs/releasing.html b/docs/releasing.html
> index 07f100caae1..20fc4579a32 100644
> --- a/docs/releasing.html
> +++ b/docs/releasing.html
> @@
Jason Ekstrand writes:
> On Mon, May 28, 2018 at 10:50 AM, Francisco Jerez
> wrote:
>
>> Jason Ekstrand writes:
>>
>> > On Sat, May 26, 2018 at 6:33 PM, Francisco Jerez
>> > wrote:
>> >
>> >> Jason Ekstrand writes:
>> >>
>> >> > On Sat, May 26, 2018 at 3:36 PM, Francisco Jerez <
>>
Daniel Stone writes:
> On 26 May 2018 at 00:47, Mark Janes wrote:
>
>> I am eager to make Intel CI results more visible. Can I push our
>> artifacts to an fd.o directory, and have GitLab display them to users as
>> if the build ran on fd.o? A typical build produces ~500MB of junit test
>>
Does this fix something? If I understand correctly, the first blob_write
to fail will set out_of_memory and all subsequent blob_writes will fail.
On Tue, May 29, 2018 at 9:12 AM, Eric Engestrom
wrote:
> Signed-off-by: Eric Engestrom
> ---
> src/intel/vulkan/anv_pipeline_cache.c | 38
Quoting Adam Jackson (2018-05-29 06:50:46)
> GL_LIB expands to GLX_mesa, but applications should not link against
> that. -lGL is never wrong, just hardcode it.
Actually There is this really stupid option in the autotools build called
--gl-lib-name. We should remove that.
Emil and I had also
EvaluateDepthValuesARB()/ResolveDepthValuesNV() is a hint for the
driver to decompress the depth buffer if needed. This can be needed
because the decompressed result can depend on the current sample
locations.
Fiddling around with the current state of the patches, I could not
find a case where it
https://bugs.freedesktop.org/show_bug.cgi?id=106673
--- Comment #2 from Alexandre Demers ---
(In reply to Timothy Arceri from comment #1)
> I assume you mean when using the NIR backend? Anyway I've pushed the revert
> of this change as it was breaking other drivers also.
Obviously. I'll test it
Commit 92f01fc5f914 ("i965: Emit VF cache invalidates for 48-bit
addressing bugs with softpin.") tried to only emit the VF invalidate if
the high bits changed, but it accidentally always set need_invalidate to
true; causing it to emit unconditionally emit the pipe control before
every primitive.
3-4 Reviewed-by: Jordan Justen
On 2018-05-27 21:35:43, Kenneth Graunke wrote:
> This isn't strictly necessary, but anyone running Cannonlake will
> already have Kernel 4.5 or later, so there's no reason to support
> the relocation model on Gen10+.
>
> This will let us avoid dealing with them
Changes in v4:
- Move code to nv50_ir_dump.cpp
- Dump headers of nvc0 programs
- Use CRC-32 instead of a truncated SHA1
- Set prog->maxGPR to targ->getFileSize() - 1 and set prog->tlsSize
- Don't compile the program if a replacement is offered
This has the consequence that a program is not
Signed-off-by: Eric Engestrom
---
src/intel/vulkan/anv_pipeline_cache.c | 38 +--
1 file changed, 19 insertions(+), 19 deletions(-)
diff --git a/src/intel/vulkan/anv_pipeline_cache.c
b/src/intel/vulkan/anv_pipeline_cache.c
index
On Tuesday, 2018-05-29 16:34:48 +0100, Marathe, Yogesh wrote:
> > -Original Message-
> > From: Engestrom, Eric
> > Sent: Tuesday, May 29, 2018 7:48 PM
> > To: Marathe, Yogesh
> > Cc: Vinson Lee ; Daniel Stone
> > ; Emil Velikov ; Eric
> > Engestrom ; ML mesa-dev >
Hi,
On 29.05.2018 18:58, Eero Tamminen wrote:
On 25.05.2018 00:55, Jason Ekstrand wrote:
This patch series adds back-end compiler support for SIMD32 fragment
shaders. Support is added and everything works but it's currently hidden
behind INTEL_DEBUG=do32. We know that it improves performance
Hi,
On 25.05.2018 00:55, Jason Ekstrand wrote:
This patch series adds back-end compiler support for SIMD32 fragment
shaders. Support is added and everything works but it's currently hidden
behind INTEL_DEBUG=do32. We know that it improves performance in some
cases but we do not yet have a
> -Original Message-
> From: Engestrom, Eric
> Sent: Tuesday, May 29, 2018 7:48 PM
> To: Marathe, Yogesh
> Cc: Vinson Lee ; Daniel Stone
> ; Emil Velikov ; Eric
> Engestrom ; ML mesa-dev d...@lists.freedesktop.org>
> Subject: Re: [Mesa-dev] [PATCH] egl/x11: Move dri2_format_for_depth
>
On Mon, May 28, 2018 at 10:50 AM, Francisco Jerez
wrote:
> Jason Ekstrand writes:
>
> > On Sat, May 26, 2018 at 6:33 PM, Francisco Jerez
> > wrote:
> >
> >> Jason Ekstrand writes:
> >>
> >> > On Sat, May 26, 2018 at 3:36 PM, Francisco Jerez <
> curroje...@riseup.net>
> >> > wrote:
> >> >
> >>
On Tue, May 29, 2018, at 16:50, Thierry Reding wrote:
> On Tue, May 29, 2018 at 12:46:35PM +0200, Daniel Kolesa wrote:
> > Hello,
> >
> > On 2018/04/19 18:31, Thierry Reding wrote:
> > > From: Thierry Reding
> > >
> > > This code path is no longer required with framebuffer modifier support.
> >
On Tue, May 29, 2018 at 12:46:35PM +0200, Daniel Kolesa wrote:
> Hello,
>
> On 2018/04/19 18:31, Thierry Reding wrote:
> > From: Thierry Reding
> >
> > This code path is no longer required with framebuffer modifier support.
>
> This patch series fixes Xorg 1.20 startup (since 1.20, it does not
Cc: Emil Velikov
Cc: Daniel Stone
Cc: Andres Gomez
Cc: Dylan Baker
Signed-off-by: Eric Engestrom
---
docs/relnotes/18.2.0.html | 8
1 file changed, 8 insertions(+)
diff --git a/docs/relnotes/18.2.0.html b/docs/relnotes/18.2.0.html
index 00b253c076fd0f5bef00..94bc17e0082ce6f22fa2
Cc: Emil Velikov
Cc: Daniel Stone
Signed-off-by: Eric Engestrom
---
A couple things worth mentioning:
- I chose to add libwayland-egl as a separate dependency for EGL rather
than bumping the libwayland required version, so that old libwayland
can still be used if EGL is not needed. I'm
Cc: Emil Velikov
Cc: Daniel Stone
Signed-off-by: Eric Engestrom
---
src/egl/wayland/wayland-egl/Makefile.am | 24 --
src/egl/wayland/wayland-egl/meson.build | 50
.../wayland-egl/wayland-egl-abi-check.c | 235 --
https://bugs.freedesktop.org/show_bug.cgi?id=106472
Michel Dänzer changed:
What|Removed |Added
Version|unspecified |18.0
QA
On 29 May 2018 at 15:17, Eric Engestrom wrote:
> On Tuesday, 2018-05-29 09:50:46 -0400, Adam Jackson wrote:
>> GL_LIB expands to GLX_mesa, but applications should not link against
>> that. -lGL is never wrong, just hardcode it.
>
> Agreed, and Archlinux has had this patch since basically forever:
On Tuesday, 2018-05-29 05:21:02 +0100, Marathe, Yogesh wrote:
> Hi Vinson, Eric,
>
> > -Original Message-
> > From: mesa-dev [mailto:mesa-dev-boun...@lists.freedesktop.org] On Behalf
> > Of Vinson Lee
> > Sent: Sunday, May 27, 2018 1:34 PM
> > To: Engestrom, Eric
> > Cc: Daniel Stone ;
On Tuesday, 2018-05-29 09:50:46 -0400, Adam Jackson wrote:
> GL_LIB expands to GLX_mesa, but applications should not link against
> that. -lGL is never wrong, just hardcode it.
Agreed, and Archlinux has had this patch since basically forever:
Build mesa 7798 completed
Commit e6a1aca0b2 by Eric Engestrom on 5/29/2018 1:47 PM:
docs: add missing html closing tag\n\nSigned-off-by: Eric Engestrom
Configure your notification preferences
___
mesa-dev
GL_LIB expands to GLX_mesa, but applications should not link against
that. -lGL is never wrong, just hardcode it.
Signed-off-by: Adam Jackson
---
src/mesa/gl.pc.in | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/gl.pc.in b/src/mesa/gl.pc.in
index
Hey,
On 27 May 2018 at 16:35, Rob Clark wrote:
> On Sun, May 27, 2018 at 10:47 AM, Jason Ekstrand wrote:
>> Given the number of people who have said they still like the mailing list,
>> that's probably a discussion for another email thread.
>
> It would be kinda clever if gitlab could
Hi,
On 26 May 2018 at 07:42, Marek Olšák wrote:
> On Thu, May 24, 2018 at 6:46 AM, Daniel Stone wrote:
>> cgit and anongit will not be orphaned: they remain as push mirrors so
>> are updated simultaneously with GItLab pushes, as will the GitHub
>> mirrors. Realistically, we can't deprecate
Hi Mark,
On 26 May 2018 at 00:47, Mark Janes wrote:
> Daniel Stone writes:
>> We had a go at using Jenkins for some of this: Intel's been really
>> quite successful at doing it internally, but our community efforts
>> have been a miserable failure. After a few years I've concluded that
>> it's
Build mesa 7797 failed
Commit 56792a0876 by Karol Herbst on 5/10/2018 8:20 AM:
nir/print: fix printing of 8/16 bit constant variables\n\nv2 (Jose Maria Casanova Crespo ): add float16 support\n\nSigned-off-by: Karol Herbst \nReviewed-by: Jose Maria Casanova
El 29/05/18 a las 02:14, Karol Herbst escribió:
> v2 (Chema Casanova ): add float16 support
>
> Signed-off-by: Karol Herbst
> ---
> src/compiler/nir/nir_print.c | 31 +++
> 1 file changed, 31 insertions(+)
>
> diff --git a/src/compiler/nir/nir_print.c
Patch skips useless and possibly dangerous calls down to the driver
in case invalid arguments were given. I noticed this would be happening
with demo of Darwinia game. AFAIK this does not fix anything but makes
this path safer and more like how other API functions are implemented.
Signed-off-by:
ping
On Thu, May 24, 2018 at 2:02 PM, vadym.shovkoplias <
vadim.shovkopl...@gmail.com> wrote:
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104388
> Signed-off-by: Andriy Khulap
> ---
> src/mesa/drivers/dri/i965/genX_state_upload.c | 11 +++
> 1 file changed, 11
Hello,
On 2018/04/19 18:31, Thierry Reding wrote:
From: Thierry Reding
This code path is no longer required with framebuffer modifier support.
This patch series fixes Xorg 1.20 startup (since 1.20, it does not
modeset with upstream Mesa tegra driver unless xserver commit c8c276c9
is
https://bugs.freedesktop.org/show_bug.cgi?id=106696
Dave Gilbert changed:
What|Removed |Added
Resolution|NOTABUG |---
Status|RESOLVED
Reviewed-by: Jordan Justen
On 2018-05-27 21:35:40, Kenneth Graunke wrote:
> This introduces a new fast virtual memory allocator integrated with our
> BO cache bucketing. For larger objects, it falls back to the simple
> free-list allocator (util_vma).
>
> This puts the allocators in place but
Signed-off-by: Tapani Pälli
---
src/intel/vulkan/anv_extensions.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/intel/vulkan/anv_extensions.py
b/src/intel/vulkan/anv_extensions.py
index b5bee0881c..9128d62fa5 100644
--- a/src/intel/vulkan/anv_extensions.py
+++
Signed-off-by: Tapani Pälli
---
src/intel/vulkan/anv_android.c | 97 ++
src/intel/vulkan/anv_device.c | 48 -
src/intel/vulkan/anv_private.h | 18
3 files changed, 162 insertions(+), 1 deletion(-)
diff --git
Hi;
Here's RFC for the extension. I've tested the import/export functionality
with RGBA images on Android Celadon with a custom NDK app that draws a
textured cube using AHardwareBuffer contents. Export feature I tested
by creating just a regular image (LunarG cube texture) and then exporting
Since we don't know the exact format at creation time, some initialization
is done only when bound with memory in vkBindImageMemory.
Signed-off-by: Tapani Pälli
---
src/intel/vulkan/anv_android.c | 40 ++
src/intel/vulkan/anv_image.c | 115
TODO - need to figure out implementation-defined external format identifier
- YUV support
Signed-off-by: Tapani Pälli
---
src/intel/vulkan/anv_android.c | 91 ++
1 file changed, 91 insertions(+)
diff --git a/src/intel/vulkan/anv_android.c
Signed-off-by: Tapani Pälli
---
src/intel/vulkan/vk_format_info.h | 43 +++
1 file changed, 43 insertions(+)
diff --git a/src/intel/vulkan/vk_format_info.h
b/src/intel/vulkan/vk_format_info.h
index a1cc6952c8..0ae0a2d43e 100644
---
This will be utilized later by GetAndroidHardwareBufferPropertiesANDROID.
Signed-off-by: Tapani Pälli
---
src/intel/vulkan/anv_formats.c | 22 +++---
src/intel/vulkan/anv_private.h | 5 +
2 files changed, 16 insertions(+), 11 deletions(-)
diff --git
If the application asks for the maximum number of fragment input
components (128), use all of them plus some builtins that are
passed in the VUE, then we exceed the maximum number of used VUE
slots (32) and we break one assert that checks this limit.
Also, with separate shader objects, we add
https://bugs.freedesktop.org/show_bug.cgi?id=106696
Christian König changed:
What|Removed |Added
Resolution|--- |NOTABUG
Status|NEW
Line number of a predefined macro should be set
as where it is referenced rather than declared
Signed-off-by: zhaowei yuan
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106590
---
src/compiler/glsl/glcpp/glcpp-lex.l | 1 +
src/compiler/glsl/glcpp/glcpp-parse.y | 55
64 matches
Mail list logo
