On Wed, 2019-02-27 at 17:04 -0800, Ian Romanick wrote:
> On 2/27/19 4:45 AM, Iago Toral Quiroga wrote:
> > Now that we propagate constants to the first source of 2src
> > instructions we
> > see more opportunities of constant folding in the backend.
> >
> > Shader-db results on KBL:
> >
> > total
On Wed, 2019-02-27 at 15:44 -0800, Francisco Jerez wrote:
> Iago Toral Quiroga writes:
>
> > The section 'Execution Data Types' of 3D Media GPGPU volume, which
> > describes execution types, is exactly the same in BDW and SKL+.
> >
> > Also, this section states that there is a single execution t
On Wed, 2019-02-27 at 13:47 -0800, Francisco Jerez wrote:
> Iago Toral writes:
>
> > On Tue, 2019-02-26 at 14:54 -0800, Francisco Jerez wrote:
> > > Iago Toral Quiroga writes:
> > >
> > > > ---
> > > > src/intel/compiler/brw_eu_validate.c| 64 -
> > > > src/intel/compiler/test
Am Donnerstag, den 28.02.2019, 07:26 +0100 schrieb Christian Gmeiner:
> Changes v1 -> v2:
> - Avoid the GPU sampling from the resource that gets mutated by the the
>transfer map by setting DRM_ETNA_PREP_WRITE.
>
> Changes v2 -> v3:
> - make use of likely(..)
> - drop minor optimization rega
HALTI0 GPUs have broken ETC2 support (for RGB and RGBA)
formats. The blob driver does per-block patching. The patching
is quite simple: color swaping.
This patch series adds support for the needed block patching
and enabled ETC2 support for HALTI0.
Passes all tested formats of the following pigli
Changes v1 -> v2:
- Avoid the GPU sampling from the resource that gets mutated by the the
transfer map by setting DRM_ETNA_PREP_WRITE.
Changes v2 -> v3:
- make use of likely(..)
- drop minor optimization regarding rsc->layout == ETNA_LAYOUT_LINEAR
- better documentation why DRM_ETNA_PREP_WR
Saves us from calling etna_bo_map(..) and saves us from doing the
same offset calcs for map() and unmap() operations.
Signed-off-by: Christian Gmeiner
Reviewed-by: Lucas Stach
---
src/gallium/drivers/etnaviv/etnaviv_context.h | 1 +
.../drivers/etnaviv/etnaviv_transfer.c| 19 ++
Signed-off-by: Christian Gmeiner
Reviewed-by: Lucas Stach
---
src/gallium/drivers/etnaviv/etnaviv_screen.c | 12 +---
1 file changed, 1 insertion(+), 11 deletions(-)
diff --git a/src/gallium/drivers/etnaviv/etnaviv_screen.c
b/src/gallium/drivers/etnaviv/etnaviv_screen.c
index de822fc85
ETC2 is supported with HALTI0, however that implementation is buggy
in hardware. The blob driver does per-block patching to work around
this. We need to swap colors for t-mode etc2 blocks.
Changes v2 -> v3:
- Drop redundant format check
Signed-off-by: Christian Gmeiner
Acked-by: Lucas Stach
--
On 2/27/19 4:45 AM, Iago Toral Quiroga wrote:
> Now that we propagate constants to the first source of 2src instructions we
> see more opportunities of constant folding in the backend.
>
> Shader-db results on KBL:
>
> total instructions in shared programs: 14965607 -> 14855983 (-0.73%)
> instruc
Reviewed-by: Marek Olšák
Marek
On Tue, Feb 26, 2019 at 10:30 PM Timothy Arceri
wrote:
> This helps improve compile times. For example the shader-db dolphin
> shader shaders/dolphin/ubershaders/120.shader_test goes from
> ~1.69 -> ~1.57 seconds on my machine with this change.
> ---
> src/galli
On Tue, Feb 26, 2019 at 4:39 PM Chris Wilson
wrote:
> A few of the GEM drivers provide matching ioctls to allow control of
> their bo caches. Hook these up to APPLE_object_purgeable to allow
> clients to discard video memory under pressure where they are able to
> fallback to restoring content th
Reviewed-by: Marek Olšák
Marek
On Wed, Feb 27, 2019 at 2:26 AM Timothy Arceri
wrote:
> Some types of params such as some builtins are always padded. We
> need to keep track of this so we can restore the list correctly.
>
> Here we also remove a couple of cache entries that are not actually
> r
I'd rather have something more robust than an env var, like catching SIGSYS.
Marek
On Wed, Feb 27, 2019 at 6:13 PM wrote:
> From: Marc-André Lureau
>
> Since commit d877451b48a59ab0f9a4210fc736f51da5851c9a ("util/u_queue:
> add UTIL_QUEUE_INIT_SET_FULL_THREAD_AFFINITY"), mesa calls
> sched_set
Iago Toral Quiroga writes:
> The section 'Execution Data Types' of 3D Media GPGPU volume, which
> describes execution types, is exactly the same in BDW and SKL+.
>
> Also, this section states that there is a single execution type, so it
> makes sense that this is the wider of the two floating poi
From: Marek Olšák
src/gallium/drivers/radeonsi/si_state_viewport.c:196: si_emit_guardband:
Assertion `vp_as_scissor.maxx <= max_viewport_size[vp_as_scissor.quant_mode]
&& vp_as_scissor.maxy <= max_viewport_size[vp_as_scissor.quant_mode]' failed.
The comparison was unsigned, so negative maxx or m
From: Marc-André Lureau
Since commit d877451b48a59ab0f9a4210fc736f51da5851c9a ("util/u_queue:
add UTIL_QUEUE_INIT_SET_FULL_THREAD_AFFINITY"), mesa calls
sched_setaffinity syscall. Unfortunately, qemu crashes with SIGSYS
when sandboxing is enabled (by default with libvirt), as this syscall
is filt
Apparently instead of returning error when passing
a quality level different than 0 for
D3DMULTISAMPLE_NONE, we should pass.
Fixes: https://github.com/iXit/Mesa-3D/issues/340
Cc: mesa-sta...@lists.freedesktop.org
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_pipe.h | 4
Check GetWindowInfo and ignore the computed sizes
if there is an error.
Fixes the regression caused by:
commit 2318ca68bbeb4fa6e21a4d8c650cec3f64246596
"st/nine: Handle window resize when a presentation buffer is used"
when using old wine gallium nine patches
Related issues:
https://github.com/iX
From: Marek Olšák
- add *pipe
- add documentation
---
.../state_trackers/omx/vid_enc_common.c | 53 +--
1 file changed, 37 insertions(+), 16 deletions(-)
diff --git a/src/gallium/state_trackers/omx/vid_enc_common.c
b/src/gallium/state_trackers/omx/vid_enc_common.c
index 2
From: Marek Olšák
---
.../state_trackers/omx/bellagio/vid_enc.c | 3 +
.../state_trackers/omx/tizonia/h264eprc.c | 3 +
.../state_trackers/omx/vid_enc_common.c | 219 +++---
.../state_trackers/omx/vid_enc_common.h | 8 +-
4 files changed, 196 insertions(+),
From: Marek Olšák
The OpenMAX state tracker will use this.
RadeonSI is adapted to use pipe_grid_info::last_block instead of its
internal state.
---
src/gallium/auxiliary/util/u_screen.c | 3 +++
src/gallium/docs/source/screen.rst| 2 ++
src/gallium/drivers/radeonsi/si_comp
Iago Toral writes:
> On Tue, 2019-02-26 at 14:54 -0800, Francisco Jerez wrote:
>> Iago Toral Quiroga writes:
>>
>> > ---
>> > src/intel/compiler/brw_eu_validate.c| 64 -
>> > src/intel/compiler/test_eu_validate.cpp | 122
>> >
>> > 2 files changed, 185
On Mon, Feb 25, 2019 at 12:35 AM Boris Brezillon
wrote:
>
> From: Daniel Stone
>
> pipe_boxes are x/y + width/height, rather than x0/y0 -> x1/y1. This
> means that (x+width) is not included in the box.
>
> The box intersection check was seemingly written for inclusive regions,
> and would falsely
Those fuzzing guys are mean...
This stuff is subtle but I think this is correct. Assuming you've run it
through our CI and everything is good,
Reviewed-by: Jason Ekstrand
On Wed, Feb 27, 2019 at 5:05 AM Juan A. Suarez Romero
wrote:
> When emitting a branch in a block, it does not make sense
On 2/27/19 4:45 AM, Iago Toral Quiroga wrote:
> Now that we propagate constants to the first source of 2src instructions we
> see more opportunities of constant folding in the backend.
All the benefit of the series is from more constant folding? Once upon
a time, I had a patch that added another
https://bugs.freedesktop.org/show_bug.cgi?id=109532
--- Comment #49 from Ian Romanick ---
After discussing https://github.com/KhronosGroup/OpenGL-API/issues/46 in the
Khronos call today, I realized that my thinking about this bug may have been
slightly incorrect. I believe that there are two sep
https://bugs.freedesktop.org/show_bug.cgi?id=109443
--- Comment #7 from Alex Granni ---
I'd go for this way as 3.0.4 is also affected:
diff --git a/scons/gallium.py b/scons/gallium.py
index 963834a5fbc..565f19f3f1a 100755
--- a/scons/gallium.py
+++ b/scons/gallium.py
@@ -308,7 +308,10 @@ def gen
> super nitpicking: What is the purpose of this change? Should this
> define be added when it really gets used?
I mean, we don't have explicit documentation on the cmdstream; if I
don't add the define / enum / struct / whatever in, it will be forgotten
to time.
Chris Wilson writes:
> Quoting Eric Anholt (2019-02-27 02:19:32)
>> Overall, I'm hesitatant to land support for actually doing anything with
>> APPLE_object_purgeable when there are no functional tests of it. I
>> don't mean to actually have tests that force purging, but at least
>> making sure
Hi Lucas
Am Mi., 27. Feb. 2019 um 10:22 Uhr schrieb Lucas Stach :
>
> Am Dienstag, den 26.02.2019, 19:15 +0100 schrieb Christian Gmeiner:
> > ETC2 is supported with HALTI0, however that implementation is buggy
> > in hardware. The blob driver does per-block patching to work around
> > this. We nee
Hi Lucas
Am Mi., 27. Feb. 2019 um 10:19 Uhr schrieb Lucas Stach :
>
> Am Dienstag, den 26.02.2019, 19:15 +0100 schrieb Christian Gmeiner:
> > Changes v1 -> v2:
> > - Avoid the GPU sampling from the resource that gets mutated by the the
> >transfer map by setting DRM_ETNA_PREP_WRITE.
> >
> > >
https://bugs.freedesktop.org/show_bug.cgi?id=109791
asimiklit changed:
What|Removed |Added
Summary|The mesa release|The mesa release config
|
https://bugs.freedesktop.org/show_bug.cgi?id=109791
--- Comment #1 from Eero Tamminen ---
>The mesa release configuration can't be build using meson 0.45.0
Does this mean that the build actually fails, or...
> -DNDEBUG is not passed to GCC, even if buildtype=release b_ndebug=true
...just that
https://bugs.freedesktop.org/show_bug.cgi?id=109791
asimiklit changed:
What|Removed |Added
CC||andrey.simiklit.1989@gmail.
https://bugs.freedesktop.org/show_bug.cgi?id=109791
Bug ID: 109791
Summary: The mesa release configuration can't be build using
meson 0.45.0
Product: Mesa
Version: git
Hardware: Other
OS: All
Sta
Even if it is not supported by the hardware, we will fix it up
in the combine constants pass.
---
.../compiler/brw_fs_combine_constants.cpp | 37 ++---
.../compiler/brw_fs_copy_propagation.cpp | 55 +--
2 files changed, 56 insertions(+), 36 deletions(-)
diff --git
Now that we propagate constants to the first source of 2src instructions we
see more opportunities of constant folding in the backend.
Shader-db results on KBL:
total instructions in shared programs: 14965607 -> 14855983 (-0.73%)
instructions in affected programs: 3988102 -> 3878478 (-2.75%)
help
And let combine constants promote the constants if needed.
---
src/intel/compiler/brw_fs_combine_constants.cpp | 2 ++
src/intel/compiler/brw_fs_copy_propagation.cpp | 4
2 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/src/intel/compiler/brw_fs_combine_constants.cpp
b/src/int
This little series lives on top of my VK_KHR_shader_float16_int8 branch, since
it depends on having a more flexible combine constants pass, which is included
with that work.
A branch with that on this series is available here:
https://github.com/Igalia/mesa/tree/itoral/VK_KHR_shader_float16_int8_c
https://bugs.freedesktop.org/show_bug.cgi?id=109532
--- Comment #48 from asimiklit ---
(In reply to Mark Janes from comment #46)
> I confirmed that the cross file is overriding our 32 bit flags that we pass
> through the environment.
(In reply to Dylan Baker from comment #47)
> Created attachmen
https://bugs.freedesktop.org/show_bug.cgi?id=109788
Lionel Landwerlin changed:
What|Removed |Added
Resolution|--- |FIXED
Status|NEW
When emitting a branch in a block, it does not make sense to continue
processing further instructions, as they will not be reachable.
This fixes a nasty case with a loop with a branch that both then-part
and else-part exits the loop:
%1 = OpLabel
OpLoopMerge %2 %3 None
OpBranchCondition
On 27/02/2019 06:55, Kenneth Graunke wrote:
On Tuesday, February 26, 2019 9:41:07 AM PST Christian Gmeiner wrote:
Push this format to the pipe driver unchanged.
Signed-off-by: Christian Gmeiner
---
include/GL/internal/dri_interface.h | 1 +
src/gallium/state_trackers/dri/dri2.c | 2 ++
2
The section 'Execution Data Types' of 3D Media GPGPU volume, which
describes execution types, is exactly the same in BDW and SKL+.
Also, this section states that there is a single execution type, so it
makes sense that this is the wider of the two floating point types
involved in mixed float mode,
Am Dienstag, den 26.02.2019, 19:15 +0100 schrieb Christian Gmeiner:
> Signed-off-by: Christian Gmeiner
Reviewed-by: Lucas Stach
> ---
> src/gallium/drivers/etnaviv/etnaviv_screen.c | 12 +---
> 1 file changed, 1 insertion(+), 11 deletions(-)
>
> diff --git a/src/gallium/drivers/etnavi
Am Dienstag, den 26.02.2019, 19:15 +0100 schrieb Christian Gmeiner:
> ETC2 is supported with HALTI0, however that implementation is buggy
> in hardware. The blob driver does per-block patching to work around
> this. We need to swap colors for t-mode etc2 blocks.
>
> > Signed-off-by: Christian Gmei
Am Dienstag, den 26.02.2019, 19:15 +0100 schrieb Christian Gmeiner:
> Changes v1 -> v2:
> - Avoid the GPU sampling from the resource that gets mutated by the the
> transfer map by setting DRM_ETNA_PREP_WRITE.
>
> > Signed-off-by: Christian Gmeiner
> ---
> .../drivers/etnaviv/etnaviv_resource
Am Mi., 27. Feb. 2019 um 05:48 Uhr schrieb Alyssa Rosenzweig
:
>
> Signed-off-by: Alyssa Rosenzweig
> ---
> src/gallium/drivers/panfrost/include/panfrost-job.h | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/src/gallium/drivers/panfrost/include/panfrost-job.h
> b/src/gallium/drivers/p
https://bugs.freedesktop.org/show_bug.cgi?id=109535
Bug 109535 depends on bug 109594, which changed state.
Bug 109594 Summary: totem assert failure: totem:
src/intel/genxml/gen9_pack.h:72: __gen_uint: La declaración `v <= max' no se
cumple.
https://bugs.freedesktop.org/show_bug.cgi?id=109594
On Tue, 2019-02-26 at 15:50 -0800, Francisco Jerez wrote:
> Iago Toral Quiroga writes:
>
> > ---
> > src/intel/compiler/brw_eu_validate.c| 10 +-
> > src/intel/compiler/test_eu_validate.cpp | 46
> > +
> > 2 files changed, 55 insertions(+), 1 deletion(-)
> >
> >
On Tue, 2019-02-26 at 14:54 -0800, Francisco Jerez wrote:
> Iago Toral Quiroga writes:
>
> > ---
> > src/intel/compiler/brw_eu_validate.c| 64 -
> > src/intel/compiler/test_eu_validate.cpp | 122
> >
> > 2 files changed, 185 insertions(+), 1 deletion(-)
On Tue, 2019-02-26 at 13:55 -0800, Francisco Jerez wrote:
> Iago Toral Quiroga writes:
>
> > The section 'Execution Data Types' of 3D Media GPGPU volume, which
> > describes execution types, is exactly the same in BDW and SKL+.
> >
> > Also, this section states that there is a single execution t
Quoting Eric Anholt (2019-02-27 02:19:32)
> Overall, I'm hesitatant to land support for actually doing anything with
> APPLE_object_purgeable when there are no functional tests of it. I
> don't mean to actually have tests that force purging, but at least
> making sure that we don't accidentally br
54 matches
Mail list logo
