Previously we disabled the guardband when the viewport was smaller than
the framebuffer on Gen7/7.5, to prevent portions of primitives from
being draw outside of the viewport. On Gen8+, we relied on the viewport
extents test to effectively scissor this away for us.
We can simply always enable sci
On 01/12/2017 02:12 AM, Marek Olšák wrote:
On Thu, Jan 12, 2017 at 12:33 AM, Ilia Mirkin wrote:
On Wed, Jan 11, 2017 at 4:00 PM, Roland Scheidegger wrote:
Am 11.01.2017 um 21:08 schrieb Samuel Pitoiset:
On 01/11/2017 07:00 PM, Roland Scheidegger wrote:
I don't think there's any glsl, es
On 11/01/17 08:41 PM, Andy Furniss wrote:
>
> Pure luck noticing this because I haven't tested modesetting driver for
> ages, but -
>
> These patches also break full screen vdpau playback when using that.
>
> Result is a screen of mostly junk with a hint of the vid - looks like
> when direct sca
This is also:
Reviewed-by: Timothy Arceri
On Wed, 2017-01-11 at 01:29 +0100, Bas Nieuwenhuizen wrote:
> Port of c5d664f9dc2d281c74844cef36ecb9f5862a8f6a
> "anv/pipeline: Call nir_lower_constant_initializers"
>
> Signed-off-by: Bas Nieuwenhuizen
> Cc:
> ---
> src/amd/vulkan/radv_pipeline.c |
On Thu, Jan 12, 2017 at 12:19:41AM -0800, Kenneth Graunke wrote:
> Previously we disabled the guardband when the viewport was smaller than
> the framebuffer on Gen7/7.5, to prevent portions of primitives from
> being draw outside of the viewport. On Gen8+, we relied on the viewport
> extents test
Hi Emil,
thanks for your review!
2017-01-09 21:00 GMT+01:00 Emil Velikov :
> Hi Christian,
>
> There's a few nitpicks. Note that neither is a blocker so feel free to
> send patches in the week(s) to come.
>
> On 23 December 2016 at 22:04, Christian Gmeiner
> wrote:
>
>> --- a/configure.ac
>> ++
Hi Emil,
thanks for your review!
2017-01-10 14:22 GMT+01:00 Emil Velikov :
> Hi Christian,
>
> Similar to 2/3 there's a few trivial nitpicks which can be addressed
> at a later stage.
>
I fixed them and will squash them.
> On 23 December 2016 at 22:04, Christian Gmeiner
> wrote:
>> Changes fro
Hi Emil,
2017-01-10 14:28 GMT+01:00 Emil Velikov :
> Hi Christian,
>
> On 23 December 2016 at 22:04, Christian Gmeiner
> wrote:
>> As the original patchstack is now about 300 patches, I have choosen to
>> squash the patches together into three different parts.
>>
>> - renderonly library
>> A li
On 12/01/17 12:09 AM, Cheng, Tony wrote:
> Vblank interrupt fires as soon as the last line of active region is
> scanned out.
> VSync interrupt fires at the vsync.
> VUpdate interrupt fires HW is ready to scan out a new frame, this include
> latch on double buffer registers, starting memory request
On Thursday, January 12, 2017 8:46:43 AM PST Chris Wilson wrote:
> On Thu, Jan 12, 2017 at 12:19:41AM -0800, Kenneth Graunke wrote:
> > Previously we disabled the guardband when the viewport was smaller than
> > the framebuffer on Gen7/7.5, to prevent portions of primitives from
> > being draw outs
On Wed, 2017-01-11 at 10:25 -0800, Matt Turner wrote:
> Coverity thinks there are some bad shifts introduced by commit c2acf97.
>
> I think it thinks that first can be 64 because 64 is the greatest
> value that can be returned by ffsll(), but that's not actually
> possible? If that's correct, mayb
On Wed, 2017-01-11 at 16:28 -0800, Jordan Justen wrote:
> On 2017-01-11 09:53:15, Juan A. Suarez Romero wrote:
> > ---
> > docs/envvars.html | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/docs/envvars.html b/docs/envvars.html
> > index 9eee8db..4f05d7f 100644
> > --- a/docs/envvar
From: Nicolai Hähnle
Add a proper check for feature support, and raise an invalid enum for
GL_CLAMP_VERTEX/FRAGMENT_COLOR unconditionally in core profiles, since
those enums were explicitly removed after the extension was promoted
to core functionality (not in the profile sense) with OpenGL 3.0.
On 11.01.2017 18:11, Samuel Pitoiset wrote:
Currently, we can store up to 256 immediates in a static array,
but this is not always enough. Instead, allocate a dynamic array
like what we currently do for temps.
This fixes a segfault with
dEQP-GLES31.functional.ssbo.layout.random.all_shared_buffer
On 12.01.2017 00:58, Grazvydas Ignotas wrote:
b838f642 "ac/debug: Move sid_tables.h generation to common code." moved
sid_tables.h but forgot the corresponding .gitignore.
Signed-off-by: Grazvydas Ignotas
---
no commit access
Pushed, thanks!
src/amd/common/.gitignore | 1 +
On 12.01.2017 01:14, Mauro Rossi wrote:
Hi,
I'm sending a patch to support recent changes in radeonsi and amd/common,
in order to fix android building errors due to files moved to amd/common.
The patch has been implemented as a merge of the two ports,
in order to fully test the build.
Please c
On 11.01.2017 15:01, Emil Velikov wrote:
> On 6 December 2016 at 14:55, Marek Olšák wrote:
>> Hi,
>>
>> I'd like to announce that this release doesn't fix the worst GPU
>> hangs/freezes it has. I'm talking about all Gallium drivers here.
>> There was a bug recently discovered in shared code that l
On 12.01.2017 09:24, Samuel Pitoiset wrote:
On 01/12/2017 02:12 AM, Marek Olšák wrote:
On Thu, Jan 12, 2017 at 12:33 AM, Ilia Mirkin
wrote:
On Wed, Jan 11, 2017 at 4:00 PM, Roland Scheidegger
wrote:
Am 11.01.2017 um 21:08 schrieb Samuel Pitoiset:
On 01/11/2017 07:00 PM, Roland Scheidegg
shader-db results BDW:
total instructions in shared programs: 13039631 -> 13039536 (-0.00%)
instructions in affected programs: 24328 -> 24233 (-0.39%)
helped: 86
HURT: 0
total cycles in shared programs: 256345938 -> 256344020 (-0.00%)
cycles in affected programs: 643654 -> 641736 (-0.30%)
helped:
This fixes a defect detected by Coverity Scan.
---
src/mesa/drivers/dri/i965/brw_draw_upload.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/drivers/dri/i965/brw_draw_upload.c
b/src/mesa/drivers/dri/i965/brw_draw_upload.c
index b7527f2cd9..abf3859614 100644
--- a/src/mesa/drivers/d
Hi,
On 11.01.2017 20:07, Lionel Landwerlin wrote:
On 11/01/17 17:02, Eero Tamminen wrote:
On 09.01.2017 07:26, Kenneth Graunke wrote:
Signed-off-by: Kenneth Graunke
---
src/intel/vulkan/anv_device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
I'm not clear whether we want to do th
I think v_mad always flushes denorms.
I would just ignore this failure. It's not required to fix every silly test
on the planet. If you opencode v_max, you'll have the same problem, and
then you'd have to fix v_cmp. It's just silly.
Marek
On Jan 12, 2017 11:59 AM, "Nicolai Hähnle" wrote:
> On
On Mon, 2017-01-09 at 18:09 +0100, Juan A. Suarez Romero wrote:
> Hi,
>
> This series implements the support for Haswell 64bit vertex attributes. With
> it,
> we can enable OpenGL 4.2 in Haswell.
>
> This work covers partially https://bugs.freedesktop.org/show_bug.cgi?id=94442.
>
> These patche
We no longer need to use lp_build_tgsi_soa_context.
No regressions founds with full piglit run.
Signed-off-by: Samuel Pitoiset
---
src/gallium/drivers/radeonsi/si_shader.c | 112 ++---
src/gallium/drivers/radeonsi/si_shader_internal.h | 2 +-
.../drivers/radeonsi/si
The plan is to replace si_shader_context::soa with its parent
structure (ie. bld_base).
Signed-off-by: Samuel Pitoiset
---
src/gallium/drivers/radeonsi/si_shader.c| 4 ++--
src/gallium/drivers/radeonsi/si_shader_internal.h | 2 ++
src/gallium/drivers/radeonsi/si_shader_tgsi_setup
The plan is to replace si_shader_context::soa with its parent
structure (ie. bld_base).
Signed-off-by: Samuel Pitoiset
---
src/gallium/drivers/radeonsi/si_shader.c | 20 +++
.../drivers/radeonsi/si_shader_tgsi_setup.c| 29 +-
2 files changed, 26
On 01/12/2017 12:55 PM, Marek Olšák wrote:
I think v_mad always flushes denorms.
I would just ignore this failure. It's not required to fix every silly
test on the planet. If you opencode v_max, you'll have the same problem,
and then you'd have to fix v_cmp. It's just silly.
Your call.
That
Thinking about it some more I think I can come up with something
hopefully much better than this and my other patch "nir: shuffle fmuls
to allow const evaluation" combined.
On Thu, 2017-01-12 at 15:20 +1100, Timothy Arceri wrote:
> If one of the inputs to the multiplcation in ffma is the result
Reviewed-by: Iago Toral Quiroga
On Thu, 2017-01-12 at 12:24 +0100, Juan A. Suarez Romero wrote:
> This fixes a defect detected by Coverity Scan.
> ---
> src/mesa/drivers/dri/i965/brw_draw_upload.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_draw_upload
On 01/06/2017 03:34 AM, Alexandre Courbot wrote:
On 12/23/2016 08:15 AM, Samuel Pitoiset wrote:
This series makes use of the scheduling control code in order to improve the
instruction pipelining on Maxwell GPUs.
Tested this on Jetson TX1. The performance improvement on glmark2 was
only marg
Just pushed the series before the branchpoint. :-)
If someone want to do benchmarks, make sure to use Linux 4.10 with
pstate 0f. The sched control codes are enabled by default but they can
be disabled by setting NV50_PROG_SCHED=0 (for comparison purposes and
debugging eventually).
Thanks!
O
Hi Ben,
On 3 January 2017 at 02:36, Ben Widawsky wrote:
> static uint32_t
> gbm_dri_bo_get_stride(struct gbm_bo *_bo, int plane)
> {
> - return _bo->stride;
> + struct gbm_dri_device *dri = gbm_dri_device(_bo->gbm);
> + struct gbm_dri_bo *bo = gbm_dri_bo(_bo);
> + __DRIimage *image;
>
On 12 January 2017 at 14:32, Daniel Stone wrote:
> If allocated, this image is just leaked, along with its reference on
> the BO. Same problem in gbm_dri_bo_get_handle_for_plane.
... and gbm_dri_bo_get_offset.
Cheers,
Daniel
___
mesa-dev mailing list
m
Hi Ben,
On 3 January 2017 at 02:36, Ben Widawsky wrote:
> Same as v1 with the requested fixes and tags added. v1 is here:
> https://lists.freedesktop.org/archives/intel-gfx/2016-December/113639.html
>
> I haven't yet updated kmscube to use Kristian's GET_PLANE2, but I will be
> doing
> it this w
On Mon, Jan 2, 2017 at 9:36 PM, Ben Widawsky wrote:
> Same as v1 with the requested fixes and tags added. v1 is here:
> https://lists.freedesktop.org/archives/intel-gfx/2016-December/113639.html
>
> I haven't yet updated kmscube to use Kristian's GET_PLANE2, but I will be
> doing
> it this week,
On 12.01.2017 14:07, Samuel Pitoiset wrote:
We no longer need to use lp_build_tgsi_soa_context.
No regressions founds with full piglit run.
Yay! For the series:
Reviewed-by: Nicolai Hähnle
Signed-off-by: Samuel Pitoiset
---
src/gallium/drivers/radeonsi/si_shader.c | 112 ++
On 12 January 2017 at 07:38, Liu Zhiquan wrote:
> when dri2_add_config, driver_configs may add to exist dri2_conf,
> the config id should not increase in this case.
> In the code, when ConfigID equal to count+1, it's mean a new config,
> config_count will increase. otherwise it's a exist config.
>
---
src/egl/wayland/wayland-egl/wayland-egl-priv.h | 27 ++
1 file changed, 27 insertions(+)
diff --git a/src/egl/wayland/wayland-egl/wayland-egl-priv.h
b/src/egl/wayland/wayland-egl/wayland-egl-priv.h
index c91f9cd..92c31d9 100644
--- a/src/egl/wayland/wayland-egl/waylan
From: Nicolai Hähnle
Reviewed-by: Bas Nieuwenhuizen
---
src/gallium/drivers/radeonsi/si_shader_internal.h | 4 -
src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c | 196 --
2 files changed, 200 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_shader_internal.h
b
From: Nicolai Hähnle
Code is taken from a combination of radv (for the more basic functions,
to avoid gallivm dependencies) and radeonsi (for the new and improved
derivative calculations).
v2: add 0.5 offset to tex coords only after derivative calculation
v3:
- really only touch the first three
From: Nicolai Hähnle
As remarked by the comment in the original code, the old algorithm fails when
(tc + deriv) points at a different cube face. Instead, simply project the
derivative directly to the plane of the selected cube face.
The new code is based on exactly differentiating (using the cha
From: Nicolai Hähnle
Reviewed-by: Bas Nieuwenhuizen
---
src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c | 30 +--
1 file changed, 1 insertion(+), 29 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c
b/src/gallium/drivers/radeonsi/si_shader_tgsi_
From: Nicolai Hähnle
Sourcing coords_arg[4] is actually never correct, since bias is handled
differently in tex_fetch_args anyway.
Reviewed-by: Bas Nieuwenhuizen
---
src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c | 13 +
1 file changed, 1 insertion(+), 12 deletions(-)
diff --gi
From: Nicolai Hähnle
Reviewed-by: Bas Nieuwenhuizen
---
src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c | 74 +--
1 file changed, 42 insertions(+), 32 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c
b/src/gallium/drivers/radeonsi/si_shader_tgs
From: Nicolai Hähnle
... by straight-forward search & replace, and eliminate
emit_llvm_intrinsic.
Reviewed-by: Bas Nieuwenhuizen
---
src/amd/common/ac_nir_to_llvm.c | 120 ++--
1 file changed, 41 insertions(+), 79 deletions(-)
diff --git a/src/amd/common/ac
Hi all,
once more without the change to the offset, and with a small fix on
patch #5: while munging radv and radeonsi versions of prepare_cube_coords
together, I accidentally went back to touching all four coordinates.
This is fixed now.
With Bas' testing on radv and more thorough piglit/cts test
From: Nicolai Hähnle
... and eliminate emit_fdiv and nir_to_llvm_context::fpmath_md_*, which
are now unused.
Reviewed-by: Bas Nieuwenhuizen
---
src/amd/common/ac_nir_to_llvm.c | 28 ++--
1 file changed, 6 insertions(+), 22 deletions(-)
diff --git a/src/amd/common/ac_ni
From: Nicolai Hähnle
... and eliminate the non-ac copies. Mostly straight-forward
search & replace.
Reviewed-by: Bas Nieuwenhuizen
---
src/amd/common/ac_nir_to_llvm.c | 89 +++--
1 file changed, 24 insertions(+), 65 deletions(-)
diff --git a/src/amd/common/
Ben's work is the only copyrightable piece I can see here, so:
Reviewed-by: Daniel Stone
Cheers,
Daniel
On 12 January 2017 at 15:38, Micah Fedke wrote:
> ---
> src/egl/wayland/wayland-egl/wayland-egl-priv.h | 27
> ++
> 1 file changed, 27 insertions(+)
>
> diff --git a
On 12 January 2017 at 07:14, Liu Zhiquan wrote:
> This reverts commit 4d6d55deef291b489af4d7870c6f5eb223c8da5d.
>
> SurfaceType added EGL_PBUFFER_BIT when dri_single_config is null.
> This will fix in "egl: correct surface_type when add config" patch.
I'm struggling to parse what you're saying her
Am 12.01.2017 um 02:12 schrieb Marek Olšák:
> On Thu, Jan 12, 2017 at 12:33 AM, Ilia Mirkin wrote:
>> On Wed, Jan 11, 2017 at 4:00 PM, Roland Scheidegger
>> wrote:
>>> Am 11.01.2017 um 21:08 schrieb Samuel Pitoiset:
On 01/11/2017 07:00 PM, Roland Scheidegger wrote:
> I don't t
[Adding Tomasz]
On 12 January 2017 at 07:14, Liu Zhiquan wrote:
> When add config, dri_config is double or single. Should only add
> EGL_WINDOW_BIT to surface_type for double dri_config, Should only add
> EGL_PBUFFER_BIT EGL_PIXMAP_BIT to surface_type for single dri_config.
> This avoid crash whe
On Thu, Jan 12, 2017 at 11:13 AM, Roland Scheidegger wrote:
> Am 12.01.2017 um 02:12 schrieb Marek Olšák:
>> On Thu, Jan 12, 2017 at 12:33 AM, Ilia Mirkin wrote:
>>> On Wed, Jan 11, 2017 at 4:00 PM, Roland Scheidegger
>>> wrote:
Am 11.01.2017 um 21:08 schrieb Samuel Pitoiset:
>
>
>
Am 12.01.2017 um 17:13 schrieb Roland Scheidegger:
> Am 12.01.2017 um 02:12 schrieb Marek Olšák:
>> On Thu, Jan 12, 2017 at 12:33 AM, Ilia Mirkin wrote:
>>> On Wed, Jan 11, 2017 at 4:00 PM, Roland Scheidegger
>>> wrote:
Am 11.01.2017 um 21:08 schrieb Samuel Pitoiset:
>
>
> On 01
Am 12.01.2017 um 17:20 schrieb Ilia Mirkin:
> On Thu, Jan 12, 2017 at 11:13 AM, Roland Scheidegger
> wrote:
>> Am 12.01.2017 um 02:12 schrieb Marek Olšák:
>>> On Thu, Jan 12, 2017 at 12:33 AM, Ilia Mirkin wrote:
On Wed, Jan 11, 2017 at 4:00 PM, Roland Scheidegger
wrote:
> Am 11.0
On 12 January 2017 at 09:55, Nicolai Hähnle wrote:
> From: Nicolai Hähnle
>
> Add a proper check for feature support, and raise an invalid enum for
> GL_CLAMP_VERTEX/FRAGMENT_COLOR unconditionally in core profiles, since
> those enums were explicitly removed after the extension was promoted
> to
On Thu, Jan 12, 2017 at 11:39 AM, Emil Velikov wrote:
> On 12 January 2017 at 09:55, Nicolai Hähnle wrote:
>> From: Nicolai Hähnle
>>
>> Add a proper check for feature support, and raise an invalid enum for
>> GL_CLAMP_VERTEX/FRAGMENT_COLOR unconditionally in core profiles, since
>> those enums
On 12.01.2017 17:39, Emil Velikov wrote:
On 12 January 2017 at 09:55, Nicolai Hähnle wrote:
From: Nicolai Hähnle
Add a proper check for feature support, and raise an invalid enum for
GL_CLAMP_VERTEX/FRAGMENT_COLOR unconditionally in core profiles, since
those enums were explicitly removed aft
On 12.01.2017 17:41, Ilia Mirkin wrote:
On Thu, Jan 12, 2017 at 11:39 AM, Emil Velikov wrote:
On 12 January 2017 at 09:55, Nicolai Hähnle wrote:
From: Nicolai Hähnle
Add a proper check for feature support, and raise an invalid enum for
GL_CLAMP_VERTEX/FRAGMENT_COLOR unconditionally in core
On Thu, Jan 12, 2017 at 11:46 AM, Nicolai Hähnle wrote:
> On 12.01.2017 17:41, Ilia Mirkin wrote:
>>
>> On Thu, Jan 12, 2017 at 11:39 AM, Emil Velikov
>> wrote:
>>>
>>> On 12 January 2017 at 09:55, Nicolai Hähnle wrote:
From: Nicolai Hähnle
Add a proper check for feature sup
The doc wasn't update since we moved the glsl compiler to src/compiler/glsl.
I also updated the description of the standalone compiler.
Signed-off-by: Elie Tournier
---
docs/shading.html | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/docs/shading.html b/docs/shading
On Wed, Jan 11, 2017 at 09:31:10PM -0800, Jason Ekstrand wrote:
> I'm done reviewing things for the evening. I've got a little more looking
> to do but it all looks fantastic so far. All of my comments have been
> pretty cosmetic. I'll finish tomorrow and we can go over things in the
> office if
On 12/01/17 15:09, Elie Tournier wrote:
> The doc wasn't update since we moved the glsl compiler to src/compiler/glsl.
> I also updated the description of the standalone compiler.
>
> Signed-off-by: Elie Tournier
> ---
> docs/shading.html | 9 ++---
> 1 file changed, 6 insertions(+), 3 deleti
On Thu, Jan 12, 2017 at 3:24 AM, Juan A. Suarez Romero
wrote:
> This fixes a defect detected by Coverity Scan.
> ---
> src/mesa/drivers/dri/i965/brw_draw_upload.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_draw_upload.c
> b/src/mesa/drivers/dri/i965/br
Reviewed-by: Matt Turner
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Thu, Jan 12, 2017 at 12:37 PM, Nicolai Hähnle
wrote:
> On 12.01.2017 17:57, Ilia Mirkin wrote:
>>
>> On Thu, Jan 12, 2017 at 11:46 AM, Nicolai Hähnle
>> wrote:
>>>
>>> On 12.01.2017 17:41, Ilia Mirkin wrote:
On Thu, Jan 12, 2017 at 11:39 AM, Emil Velikov
wrote:
>
On 12 January 2017 at 16:43, Nicolai Hähnle wrote:
> On 12.01.2017 17:39, Emil Velikov wrote:
>>
>> On 12 January 2017 at 09:55, Nicolai Hähnle wrote:
>>>
>>> From: Nicolai Hähnle
>>>
>>> Add a proper check for feature support, and raise an invalid enum for
>>> GL_CLAMP_VERTEX/FRAGMENT_COLOR unc
Reviewed-by: Marek Olšák
Marek
On Thu, Jan 12, 2017 at 11:14 AM, Nicolai Hähnle wrote:
> On 11.01.2017 18:11, Samuel Pitoiset wrote:
>>
>> Currently, we can store up to 256 immediates in a static array,
>> but this is not always enough. Instead, allocate a dynamic array
>> like what we currentl
On Wed, Jan 11, 2017 at 5:55 PM, Nanley Chery wrote:
> Signed-off-by: Nanley Chery
> ---
> src/intel/vulkan/TODO| 1 -
> src/intel/vulkan/anv_image.c | 19 ---
> 2 files changed, 16 insertions(+), 4 deletions(-)
>
> diff --git a/src/intel/vulkan/TODO b/src/intel/vulkan/
Hi,
On 10 January 2017 at 10:46, Daniel Stone wrote:
> On 10 January 2017 at 05:49, Ben Widawsky wrote:
>>> I ask because I genuinely don't know how this stuff is supposed to work.
>>> Returning a default of 1 seems reasonable.
>>
>> I'm not entirely sure about a reasonable default, 1 seemed rig
Hi,
On 3 January 2017 at 02:36, Ben Widawsky wrote:
> static uint32_t
> gbm_dri_bo_get_stride(struct gbm_bo *_bo, int plane)
> {
> - return _bo->stride;
> + struct gbm_dri_device *dri = gbm_dri_device(_bo->gbm);
> + struct gbm_dri_bo *bo = gbm_dri_bo(_bo);
> + __DRIimage *image;
> +
Hi,
On 3 January 2017 at 02:36, Ben Widawsky wrote:
> +static uint32_t
> +gbm_dri_bo_get_offset(struct gbm_bo *_bo, int plane)
> +{
> + struct gbm_dri_device *dri = gbm_dri_device(_bo->gbm);
> + struct gbm_dri_bo *bo = gbm_dri_bo(_bo);
> + int offset = 0;
> +
> + if (!dri->image || dri->
On 12.01.2017 17:57, Ilia Mirkin wrote:
On Thu, Jan 12, 2017 at 11:46 AM, Nicolai Hähnle wrote:
On 12.01.2017 17:41, Ilia Mirkin wrote:
On Thu, Jan 12, 2017 at 11:39 AM, Emil Velikov
wrote:
On 12 January 2017 at 09:55, Nicolai Hähnle wrote:
From: Nicolai Hähnle
Add a proper check for
This should prove benefitial in the common case of inserting
and not rewriting anything.
---
src/compiler/nir/nir_instr_set.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/src/compiler/nir/nir_instr_set.c b/src/compiler/nir/nir_instr_set.c
index 9cb9ed43e8..4c45d051f8
Walking the whole hash table, inserting entries by hashing them first
is just a really really bad idea. We can simply memcpy the whole thing.
This eliminates hash_table_insert from perf's top 15 costly functions.
---
src/compiler/glsl/opt_copy_propagation.cpp | 17 -
.../glsl/o
We have quite a few places where we are doing hashing multiple times.
These are some cases I found, but there are likely more.
Thomas Helland (4):
glsl: Hash only once if inserting new elemnt in refcount hash table
nir: Prehash in instr_set to avoid unnecessary hashing
util: Implement a hash
---
src/util/hash_table.c | 22 ++
src/util/hash_table.h | 2 ++
2 files changed, 24 insertions(+)
diff --git a/src/util/hash_table.c b/src/util/hash_table.c
index 9e643af8b2..702f465382 100644
--- a/src/util/hash_table.c
+++ b/src/util/hash_table.c
@@ -140,6 +140,28 @@ _mesa
_mesa_hash_table_search is one of our hottest functions according to perf.
Callgrind shows the refcounting as one of the major users of the
searching functions. We can reduce the pain by prehashing, so that we
avoid hash two times when inserting in the table.
On a short shader-db run (with validat
On Thu, Jan 12, 2017 at 4:39 PM, Nicolai Hähnle wrote:
> Hi all,
>
> once more without the change to the offset, and with a small fix on
> patch #5: while munging radv and radeonsi versions of prepare_cube_coords
> together, I accidentally went back to touching all four coordinates.
> This is fixe
On 17-01-09 15:45:47, Jason Ekstrand wrote:
On Wed, Jan 4, 2017 at 8:43 PM, Ben Widawsky wrote:
v2: Make the error return be -1 instead of 0 because I think 0 is
actually valid.
Cc: Daniel Stone
Signed-off-by: Ben Widawsky
---
src/gbm/backends/dri/gbm_dri.c | 28 +++
On 17-01-09 15:53:06, Jason Ekstrand wrote:
On Mon, Jan 9, 2017 at 3:50 PM, Jason Ekstrand wrote:
On Mon, Jan 2, 2017 at 6:36 PM, Ben Widawsky wrote:
Signed-off-by: Ben Widawsky
Reviewed-by: Eric Engestrom
Acked-by: Daniel Stone
---
src/gbm/backends/dri/gbm_dri.c | 26 ++
Hi,
On 5 January 2017 at 04:43, Ben Widawsky wrote:
> diff --git a/src/gbm/backends/dri/gbm_dri.c b/src/gbm/backends/dri/gbm_dri.c
> index c61d56b44a..f9c1afd8cb 100644
> --- a/src/gbm/backends/dri/gbm_dri.c
> +++ b/src/gbm/backends/dri/gbm_dri.c
> @@ -622,6 +622,33 @@ gbm_dri_bo_get_planes(struc
https://bugs.freedesktop.org/show_bug.cgi?id=99388
Bug ID: 99388
Summary: RHEL 7.3, Matlab 2016a Segmentation fault - glEnable
Product: Mesa
Version: 11.2
Hardware: x86-64 (AMD64)
OS: Linux (All)
Status: NEW
On 17-01-12 14:33:01, Daniel Stone wrote:
On 12 January 2017 at 14:32, Daniel Stone wrote:
If allocated, this image is just leaked, along with its reference on
the BO. Same problem in gbm_dri_bo_get_handle_for_plane.
... and gbm_dri_bo_get_offset.
Cheers,
Daniel
Does this work (compile tes
12.01.2017 22:23, Thomas Helland пишет:
---
src/util/hash_table.c | 22 ++
src/util/hash_table.h | 2 ++
2 files changed, 24 insertions(+)
diff --git a/src/util/hash_table.c b/src/util/hash_table.c
index 9e643af8b2..702f465382 100644
--- a/src/util/hash_table.c
+++ b/sr
On Thu, Jan 12, 2017 at 4:39 PM, Nicolai Hähnle wrote:
> From: Nicolai Hähnle
>
> Reviewed-by: Bas Nieuwenhuizen
> ---
> src/gallium/drivers/radeonsi/si_shader_tgsi_alu.c | 74
> +--
> 1 file changed, 42 insertions(+), 32 deletions(-)
>
> diff --git a/src/gallium/drivers/ra
For the series:
Reviewed-by: Marek Olšák
Marek
On Thu, Jan 12, 2017 at 2:07 PM, Samuel Pitoiset
wrote:
> We no longer need to use lp_build_tgsi_soa_context.
>
> No regressions founds with full piglit run.
>
> Signed-off-by: Samuel Pitoiset
> ---
> src/gallium/drivers/radeonsi/si_shader.c
On Thu, Dec 1, 2016 at 1:51 PM, Jason Ekstrand wrote:
> This fixes a bug in code motion that occurred when the best block is the
> same as the schedule early block. In this case, because we're checking
> (lca != def->parent_instr->block) at the top of the loop, we never get to
> the check for loo
On Thu, Jan 12, 2017 at 4:39 PM, Nicolai Hähnle wrote:
> From: Nicolai Hähnle
>
> As remarked by the comment in the original code, the old algorithm fails when
> (tc + deriv) points at a different cube face. Instead, simply project the
> derivative directly to the plane of the selected cube face.
I commented on patches 1 & 2. Other than that, patches 1-6 are:
Reviewed-by: Marek Olšák
Marek
On Thu, Jan 12, 2017 at 4:39 PM, Nicolai Hähnle wrote:
> From: Nicolai Hähnle
>
> Reviewed-by: Bas Nieuwenhuizen
> ---
> src/gallium/drivers/radeonsi/si_shader_internal.h | 4 -
> src/gallium/dr
Walking the whole hash table, inserting entries by hashing them first
is just a really really bad idea. We can simply memcpy the whole thing.
---
src/compiler/glsl/opt_copy_propagation.cpp | 17 -
.../glsl/opt_copy_propagation_elements.cpp | 29 --
2
V2: Don't rzalloc; we are about to rewrite the whole thing (Vladislav)
---
src/util/hash_table.c | 22 ++
src/util/hash_table.h | 2 ++
2 files changed, 24 insertions(+)
diff --git a/src/util/hash_table.c b/src/util/hash_table.c
index 9e643af8b2..603f694ccb 100644
--- a/src/u
2017-01-11 19:09 GMT+01:00 Jason Ekstrand :
> Another reason why I'm not a huge fan is that there is some momentum in the
> industry to make GLSL better defined with respect to NaN. I don't know that
> anything will ever come of it (because it may break apps) but if something
> does, we may find o
On 17-01-09 17:03:48, Jason Ekstrand wrote:
On Mon, Jan 2, 2017 at 6:37 PM, Ben Widawsky wrote:
The idea behind modifiers like this is that the user of GBM will have
some mechanism to query what properties the hardware supports for its BO
or surface. This information is directly passed in (and
https://bugs.freedesktop.org/show_bug.cgi?id=97102
--- Comment #11 from Bruce Cherniak ---
As Tim suggests, pruning empty nodes is probably the best solution for the
crash.
For performance, however, I'm not sure how many cores to expose in your case.
cpuinfo shows that there are 4 threads acros
Am 12.01.2017 um 22:03 schrieb Matteo Bruni:
> 2017-01-11 19:09 GMT+01:00 Jason Ekstrand :
>> Another reason why I'm not a huge fan is that there is some momentum in the
>> industry to make GLSL better defined with respect to NaN. I don't know that
>> anything will ever come of it (because it may
Didn't turn out as useful as I'd hoped, but it will help alot more on
i965 by reducing regressions when we drop brw_do_channel_expressions()
and brw_do_vector_splitting().
I'm not sure how much sense 'is_not_used_by_conditional' makes on
platforms other than i965 but since this is a new opt it at
Hi,
We're currently generating a few vulkan entry point files by parsing
the API files using regular expressions. This is limited for a couple
of reasons :
- function pointer types might not always be declared on a single
line
- if function pointers are declared under conditional fla
Signed-off-by: Lionel Landwerlin
---
include/vulkan/vk_platform.h | 2 +-
include/vulkan/vulkan.h | 247 ++-
2 files changed, 244 insertions(+), 5 deletions(-)
diff --git a/include/vulkan/vk_platform.h b/include/vulkan/vk_platform.h
index 0fa62ee23d
Signed-off-by: Lionel Landwerlin
---
src/intel/vulkan/Makefile.am| 15 ++--
src/intel/vulkan/anv_entrypoints_gen.py | 125 +---
2 files changed, 71 insertions(+), 69 deletions(-)
diff --git a/src/intel/vulkan/Makefile.am b/src/intel/vulkan/Makefile.am
ind
1 - 100 of 181 matches
Mail list logo
