On Sun, Jul 12, 2015 at 7:30 PM, Ilia Mirkin wrote:
> I asked this on IRC, but figured I'd get wider distribution for the
> question. The situation is that nv50 doesn't support RGBA32 MS8
> textures. However QuerySamplesForFormat will try its hardest to find
> supported sample counts, which means
Acked-by: Marek Olšák
Marek
On Wed, Jul 15, 2015 at 7:15 AM, Mario Kleiner
wrote:
> The amdgpu_device for a device node needs its own dup'ed fd, instead
> of using the original fd passed in for a screen, to make multi-x-screen
> ZaphodHeads configurations work on amdgpu.
>
&g
On Thu, Jul 16, 2015 at 3:06 PM, Emil Velikov wrote:
> On 12 July 2015 at 20:10, Marek Olšák wrote:
>> From: Marek Olšák
>>
>> v2: The fix for the darkness in Ubuntu Unity is in the hunk
>> with the 4-line comment.
>> ---
>> src/gallium/state_trac
On Thu, Jul 16, 2015 at 3:14 PM, Emil Velikov wrote:
> On 12 July 2015 at 20:10, Marek Olšák wrote:
>> From: Anatoli Antonovitch
>>
>> ---
>> src/gallium/state_trackers/dri/dri_query_renderer.c | 14 ++
>> 1 file changed, 14 insertions(
kov wrote:
> On 16/07/15 14:38, Marek Olšák wrote:
>> On Thu, Jul 16, 2015 at 3:06 PM, Emil Velikov
>> wrote:
>>> On 12 July 2015 at 20:10, Marek Olšák wrote:
>>>> From: Marek Olšák
>>>>
>>>> v2: The fix for the darkness in Ubuntu Uni
From: Marek Olšák
---
src/gallium/auxiliary/tgsi/tgsi_dump.c| 19 +++
src/gallium/auxiliary/tgsi/tgsi_dump.h| 5 +
src/gallium/auxiliary/util/u_dump_state.c | 7 +++
3 files changed, 23 insertions(+), 8 deletions(-)
diff --git a/src/gallium/auxiliary/tgsi
From: Marek Olšák
---
src/mesa/state_tracker/st_atom_sampler.c | 9 -
src/mesa/state_tracker/st_cb_drawpixels.c | 9 -
2 files changed, 8 insertions(+), 10 deletions(-)
diff --git a/src/mesa/state_tracker/st_atom_sampler.c
b/src/mesa/state_tracker/st_atom_sampler.c
index
From: Marek Olšák
There is no need for this.
---
src/gallium/auxiliary/cso_cache/cso_context.c | 44 ---
src/gallium/auxiliary/cso_cache/cso_context.h | 13
src/mesa/state_tracker/st_atom_clip.c | 2 +-
3 files changed, 1 insertion(+), 58 deletions
From: Marek Olšák
None of the draw states are used here.
This fixes a crash in piglit: ext_framebuffer_blit/blit-early
Calling st_manager_validate_framebuffers is the minimum requirement here.
Cc: mesa-sta...@lists.freedesktop.org
---
src/mesa/state_tracker/st_cb_blit.c | 4 ++--
1 file
From: Marek Olšák
Not needed for other shader stages.
---
src/gallium/auxiliary/cso_cache/cso_context.c | 103 +-
src/gallium/auxiliary/cso_cache/cso_context.h | 4 +-
src/gallium/auxiliary/hud/hud_context.c | 4 +-
src/gallium/auxiliary/postprocess/pp_run.c
From: Marek Olšák
---
src/gallium/auxiliary/cso_cache/cso_context.c | 28 +++
src/gallium/auxiliary/cso_cache/cso_context.h | 4 ++--
src/gallium/auxiliary/hud/hud_context.c | 4 ++--
src/gallium/auxiliary/postprocess/pp_run.c| 4 ++--
src/gallium/auxiliary
From: Marek Olšák
---
src/gallium/auxiliary/cso_cache/cso_context.c | 39 +++
src/gallium/auxiliary/cso_cache/cso_context.h | 6 ++---
2 files changed, 12 insertions(+), 33 deletions(-)
diff --git a/src/gallium/auxiliary/cso_cache/cso_context.c
b/src/gallium/auxiliary
From: Marek Olšák
---
src/gallium/auxiliary/util/u_dump.h | 20 ++-
src/gallium/auxiliary/util/u_dump_defines.c | 41 +
src/gallium/auxiliary/util/u_dump_state.c | 233 ++--
3 files changed, 241 insertions(+), 53 deletions(-)
diff --git a/src/gallium
From: Marek Olšák
Drivers can do this better, because they can skip redundant state changes
at per-slot granularity.
---
src/gallium/auxiliary/cso_cache/cso_context.c | 28 ++-
1 file changed, 2 insertions(+), 26 deletions(-)
diff --git a/src/gallium/auxiliary/cso_cache
From: Marek Olšák
Broken by f1be3d8cdde17a9b9ae283e1bab2f46b992d3bf3, which returns NULL
if no commands have been submitted.
---
src/gallium/include/pipe/p_context.h | 5 -
src/gallium/state_trackers/dri/dri2.c | 14 ++
2 files changed, 6 insertions(+), 13 deletions(-)
diff
From: Marek Olšák
---
src/gallium/tests/trivial/quad-tex.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/gallium/tests/trivial/quad-tex.c
b/src/gallium/tests/trivial/quad-tex.c
index daae577..c019c7bb 100644
--- a/src/gallium/tests/trivial/quad-tex.c
+++ b/src
From: Marek Olšák
---
src/gallium/auxiliary/postprocess/pp_colors.c | 4 ++--
src/gallium/auxiliary/postprocess/pp_mlaa.c | 23 ++-
src/gallium/auxiliary/util/u_blit.c | 6 --
3 files changed, 20 insertions(+), 13 deletions(-)
diff --git a/src/gallium
From: Marek Olšák
and make si_init_config static
---
src/gallium/drivers/radeonsi/si_pipe.c | 13 ++---
src/gallium/drivers/radeonsi/si_state.c | 6 +-
src/gallium/drivers/radeonsi/si_state.h | 1 -
3 files changed, 7 insertions(+), 13 deletions(-)
diff --git a/src/gallium
From: Marek Olšák
Copied from r600g. pipe_resource can be shared by multiple threads, so we
shouldn't change it.
---
src/gallium/drivers/radeon/r600_pipe_common.h | 1 -
src/gallium/drivers/radeonsi/si_blit.c| 201 +++---
src/gallium/drivers/radeonsi/si_st
From: Marek Olšák
This has been a no-op due to performance concerns. From now on, drivers
should decide when they don't want to unmap, not the winsys.
---
src/gallium/winsys/radeon/drm/radeon_drm_bo.c | 29 +--
src/gallium/winsys/radeon/drm/radeon_drm_bo.h | 1
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_compute.c | 2 +-
src/gallium/drivers/radeonsi/si_shader.c | 8 +++-
src/gallium/drivers/radeonsi/si_shader.h | 3 +--
3 files changed, 5 insertions(+), 8 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_compute.c
b/src
From: Marek Olšák
Cc: 10.5 10.6
---
src/gallium/drivers/radeonsi/si_shader.c| 49 ++---
src/gallium/drivers/radeonsi/si_shader.h| 1 +
src/gallium/drivers/radeonsi/si_state_shaders.c | 8 +---
3 files changed, 29 insertions(+), 29 deletions(-)
diff --git
From: Marek Olšák
buffer_unmap is currently a no-op on radeon and done correctly on amdgpu.
I plan to fix it for radeon, but before that, all occurences of buffer_unmap
that can negatively affect performance in the future must be removed.
There are 2 reasons for removing buffer_unmap calls
From: Marek Olšák
There is no need for this.
v2: handle redundant clip state changes in st/mesa
---
src/gallium/auxiliary/cso_cache/cso_context.c | 44 ---
src/gallium/auxiliary/cso_cache/cso_context.h | 13
src/mesa/state_tracker/st_atom_clip.c | 7
On Thu, Jul 16, 2015 at 8:05 PM, Marek Olšák wrote:
> From: Marek Olšák
>
> There is no need for this.
>
> v2: handle redundant clip state changes in st/mesa
> ---
> src/gallium/auxiliary/cso_cache/cso_context.c | 44
> ---
> src/
On Sat, Jun 20, 2015 at 12:08 AM, Kenneth Graunke wrote:
> On Wednesday, June 17, 2015 01:01:21 AM Marek Olšák wrote:
>> From: Fabian Bieler
>>
>> Similar to gl_ClipDistance -> gl_ClipDistanceMESA
>> ---
>> src/glsl/Makefile.sources |
On Tue, Jun 23, 2015 at 2:04 AM, Kenneth Graunke wrote:
> On Wednesday, June 17, 2015 01:01:24 AM Marek Olšák wrote:
>> From: Marek Olšák
>>
>> There is no way to lower them, because the array sizes are unknown
>> at compile time.
>>
>> Based on a p
From: Marek Olšák
There is no way to lower them, because the array sizes are unknown
at compile time.
Based on a patch from: Fabian Bieler
v2: add comments
---
src/glsl/ir_optimization.h | 5 +-
src/glsl/lower_variable_index_to_cond_assign.cpp | 58
Thanks for the reviewing.
The patches that are missing Rbs are:
23-24, 26-43, 45-46.
The patches have been on the mailing list for one month. It looks like
I won't get any more reviews. :(
Marek
On Fri, Jun 19, 2015 at 12:46 PM, Kenneth Graunke wrote:
> I made some comments, but assuming th
From: Marek Olšák
An alternative (and ugly) solution to the current clover issue.
---
src/gallium/drivers/radeon/r600_pipe_common.c | 12
src/gallium/drivers/radeon/r600_pipe_common.h | 3 +++
src/gallium/drivers/radeonsi/si_hw_context.c | 10 --
3 files changed, 23
It looks good. Would you push it please?
Reviewed-by: Marek Olšák
Marek
On Fri, Jul 17, 2015 at 11:05 AM, Michel Dänzer wrote:
> On 17.07.2015 06:03, Marek Olšák wrote:
>> From: Marek Olšák
>>
>> An alternative (and ugly) solution to the current clover issue.
>
&g
On Thu, Jul 9, 2015 at 9:17 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This adds linker support for subroutine uniforms, they
> have some subtle differences from real uniforms, we also hide
> them and they are given internal uniform names.
>
> This also adds the subroutine locations and subro
On Thu, Jul 9, 2015 at 9:17 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This fleshes out the ARB_program_query support for the
> APIs that ARB_shader_subroutine introduces, leaving
> some TODOs for later addition.
>
> Signed-off-by: Dave Airlie
> ---
> src/mesa/main/shader_query.cpp | 46
>
If the extension is core only, we can rip out the checks, but the
checks that test ctx->API == API_OPEGL_CORE should stay (if they are
missing, they should be added).
Marek
On Thu, Jul 9, 2015 at 3:27 PM, Roland Scheidegger wrote:
> Should expose that only if hw has glsl 130 support?
>
> Roland
On Tue, Jul 21, 2015 at 3:56 PM, Brian Paul wrote:
> Just a couple minor nits...
>
> On 07/16/2015 10:54 AM, Marek Olšák wrote:
>>
>> From: Marek Olšák
>>
>> ---
>> src/gallium/auxiliary/util/u_dump.h | 20 ++-
>> src/gallium/auxilia
Acked-by: Marek Olšák
What's next? boolean -> bool? :)
Marek
On Tue, Jul 21, 2015 at 2:20 AM, Ilia Mirkin wrote:
> Generated by running:
> git grep -l INLINE src/gallium/ | xargs sed -i 's/\bINLINE\b/inline/g'
> git grep -l INLINE src/mesa/state_tracker/ | xargs sed
I don't think resource_copy_region is a good match for all possible
cases that can occur with ARB_copy_image and I extensively explained
this to Ilia on IRC. The format swizzling is the problem.
Assume you have a copy from RGBA8 to R32F. This is equivalent to
copying RGBA8 to RGBA8, because it's j
FYI, I will push tessellation after Dave lands shader subroutines.
Marek
On Thu, Jul 16, 2015 at 10:27 PM, Marek Olšák wrote:
> Thanks for the reviewing.
>
> The patches that are missing Rbs are:
>
> 23-24, 26-43, 45-46.
>
> The patches have been on the mailing list for one
For the series:
Reviewed-by: Marek Olšák
Marek
On Wed, Jul 22, 2015 at 1:06 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This is also picked from Marek's tess work, it reduces
> the conflicts with my GS5 work.
>
> Signed-off-by: Dave Airlie
> ---
>
Reviewed-by: Marek Olšák
Marek
On Wed, Jul 22, 2015 at 7:04 AM, Vinson Lee wrote:
> radeon_fbo.c: In function 'radeon_map_renderbuffer_s8z24':
> radeon_fbo.c:162:9: warning: variable 'ret' set but not used
> [-Wunused-but-set-variable]
> int ret;
>
On Wed, Jul 22, 2015 at 12:51 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This is prep work for using it in the interpolation code
> later.
>
> Also add storage for the input interpolation mode so we
> can pick it up later.
>
> Signed-off-by: Dave Airlie
> ---
> src/gallium/drivers/radeonsi/
On Wed, Jul 22, 2015 at 12:51 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This adds support for fine derivatives and enables
> ARB_derivative_control on radeonsi.
>
> (just fell out of my working out interpolation)
>
> Signed-off-by: Dave Airlie
> ---
> docs/GL3.txt
Patches 1 & 2:
Reviewed-by: Marek Olšák
I can't review 3 & 4 before I know more about derivatives.
Marek
On Wed, Jul 22, 2015 at 12:51 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This is prep work for reusing this in the interpolation
> code later.
>
On Wed, Jul 22, 2015 at 3:27 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This adds the frontend support, however the llvm
> backend produces the wrong pattern, however
> we can conditionalise enabling ARB_gpu_shader5
> on whatever version of llvm we fix this in.
The driver can assume that thi
On Wed, Jul 22, 2015 at 3:27 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> This is required as part of ARB_gpu_shader5.
>
> no backend changes are required for this, or if
> any are, it's the same ones as for samplers.
>
> Signed-off-by: Dave Airlie
> ---
> docs/GL3.txt
Reviewed-by: Marek Olšák
Marek
On Thu, Jul 23, 2015 at 3:57 PM, Brian Paul wrote:
> ---
> src/mesa/state_tracker/st_atom_texture.c | 8 +++-
> 1 file changed, 3 insertions(+), 5 deletions(-)
>
> diff --git a/src/mesa/state_tracker/st_atom_texture.c
> b/src/
Hi Brian,
Like the commit message says, I have no idea how Firefox does it. My
commit only changed the situation from "crash on all drivers" to
"undefined behavior on drivers where sampler views are not sharable by
contexts".
Marek
On Thu, Jul 23, 2015 at 12:32 AM, Brian Paul wrote:
> Hi Marek,
Hi Jose,
FYI, I had to revert this, because it broke glxgears on radeonsi.
Marek
On Wed, Jul 15, 2015 at 3:25 PM, Jose Fonseca wrote:
> From: Tom Hughes
>
> https://bugs.freedesktop.org/show_bug.cgi?id=90817
>
> Signed-off-by: Jose Fonseca
> ---
> src/glx/dri_common.c | 59
> +++
s SW rendering with X servers running
> something else.)
>
> Jose
>
> On 23/07/15 20:54, Marek Olšák wrote:
>>
>> Hi Jose,
>>
>> FYI, I had to revert this, because it broke glxgears on radeonsi.
>>
>> Marek
>>
>> On Wed, Jul 15, 2015
Hi,
This is a continuation of the previous series. It allows drivers to have only 1
shader variant for every user shader in st/mesa, not counting glDrawPixels and
glBitmap variants.
In such case, the shader variant is created in LinkShader or
ProgramStringNotify.
Please review.
Marek
___
From: Marek Olšák
I'll let drivers figure out how to do it.
---
src/gallium/docs/source/screen.rst | 2 ++
src/gallium/drivers/freedreno/freedreno_screen.c | 1 +
src/gallium/drivers/i915/i915_screen.c | 1 +
src/gallium/drivers/ilo/ilo_screen.c | 1 +
From: Marek Olšák
---
src/mesa/state_tracker/st_cb_program.c | 5 +++--
src/mesa/state_tracker/st_context.c| 14 ++
src/mesa/state_tracker/st_context.h| 7 +++
3 files changed, 24 insertions(+), 2 deletions(-)
diff --git a/src/mesa/state_tracker/st_cb_program.c
b/src
From: Marek Olšák
---
src/mesa/state_tracker/st_atom_shader.c | 10 +-
src/mesa/state_tracker/st_cb_bitmap.c | 2 +-
src/mesa/state_tracker/st_cb_drawpixels.c | 2 +-
src/mesa/state_tracker/st_context.c | 3 ++-
src/mesa/state_tracker/st_context.h | 1 +
src/mesa
From: Marek Olšák
---
src/gallium/auxiliary/tgsi/tgsi_ureg.c | 26 ++
src/gallium/auxiliary/util/u_inlines.h | 22 ++
2 files changed, 24 insertions(+), 24 deletions(-)
diff --git a/src/gallium/auxiliary/tgsi/tgsi_ureg.c
b/src/gallium/auxiliary/tgsi
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_compute.c | 4 ++--
src/gallium/drivers/radeonsi/si_shader.c| 4 ++--
src/gallium/drivers/radeonsi/si_shader.h| 2 +-
src/gallium/drivers/radeonsi/si_state_shaders.c | 2 +-
4 files changed, 6 insertions(+), 6 deletions
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_shader.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/gallium/drivers/radeonsi/si_shader.c
b/src/gallium/drivers/radeonsi/si_shader.c
index 109a805..789b1b7 100644
--- a/src/gallium/drivers/radeonsi/si_shader.c
+++ b/src
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_state_shaders.c | 84 +
1 file changed, 17 insertions(+), 67 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state_shaders.c
b/src/gallium/drivers/radeonsi/si_state_shaders.c
index 9d05cb5..cc053bb 100644
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_shader.c | 76 ++--
1 file changed, 34 insertions(+), 42 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_shader.c
b/src/gallium/drivers/radeonsi/si_shader.c
index 32a702f..109a805 100644
--- a/src
Nothing special here other than cleanups. One patch disables NaNs for LS and
HS, and there's also one GS shader leak fix.
Please review.
Marek
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-d
From: Marek Olšák
They're disabled for all other shaders except compute, but I forgot
to do this for tess stages.
---
src/gallium/drivers/radeonsi/si_state_shaders.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state_shaders.c
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_state_shaders.c | 39 +
1 file changed, 13 insertions(+), 26 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state_shaders.c
b/src/gallium/drivers/radeonsi/si_state_shaders.c
index c1d61d5..9395c31 100644
From: Marek Olšák
Cc: mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/radeonsi/si_shader.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/radeonsi/si_shader.c
b/src/gallium/drivers/radeonsi/si_shader.c
index 0e98915..012d708 100644
--- a
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_state_shaders.c | 23 ---
1 file changed, 8 insertions(+), 15 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state_shaders.c
b/src/gallium/drivers/radeonsi/si_state_shaders.c
index 9395c31..71349a5 100644
From: Marek Olšák
The shader specifies the processor type, so use that instead.
---
src/gallium/drivers/radeonsi/si_state_shaders.c | 49 +
1 file changed, 9 insertions(+), 40 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state_shaders.c
b/src/gallium
From: Marek Olšák
The "current" shader pointer is moved from the CSO to the context, so that
the CSO is mostly immutable.
The only drawback is that the "current" pointer isn't saved when unbinding
a shader and it must be looked up when the shader is bound again.
This i
From: Marek Olšák
This is only supported in the compatibility profile (without GS and tess).
---
src/gallium/drivers/radeonsi/si_pipe.c | 2 +-
src/gallium/drivers/radeonsi/si_shader.c| 42 +
src/gallium/drivers/radeonsi/si_shader.h| 8
Hi,
This patch series implements all features needed for st/mesa to send shaders to
the driver immediately.
The good thing about thread-safe shader CSOs is that multithreaded shader
compilation suddenly seems easy.
Please review.
Marek
___
mesa-dev
From: Marek Olšák
using the shader key for now.
---
src/gallium/drivers/radeonsi/si_pipe.c | 2 +-
src/gallium/drivers/radeonsi/si_shader.c| 13 +
src/gallium/drivers/radeonsi/si_shader.h| 1 +
src/gallium/drivers/radeonsi/si_state.c | 2 +-
src
From: Marek Olšák
The previous version has precision issues. This can be a problem
with tessellation. Sadly, I can't find the article where I read it
anymore. I'm not sure if the unsafe-fp-math flag would be enough to revert
this.
---
src/gallium/auxiliary/gallivm/lp_bld_tgsi_act
From: Marek Olšák
This removes "v_mov v0, 0" which typically occurs before exports.
Totals:
SGPRS: 345216 -> 344552 (-0.19 %)
VGPRS: 197684 -> 197132 (-0.28 %)
Code Size: 7390408 -> 7375376 (-0.20 %) bytes
LDS: 91 -> 91 (0.00 %) blocks
Scratch: 1842176 -> 1679360
From: Marek Olšák
This is useful only when emit functions use it.
The new radeonsi min/max opcode implementation requires this.
---
src/gallium/auxiliary/gallivm/lp_bld_tgsi.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/src/gallium/auxiliary/gallivm/lp_bld_tgsi.c
b/src/gallium
From: Marek Olšák
---
src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
b/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
index 23ea23a..c22ea7c 100644
--- a/src/gallium/drivers
From: Marek Olšák
No difference according to shader-db. (with the new S_ABS_I32 pattern)
---
.../drivers/radeon/radeon_setup_tgsi_llvm.c| 60 ++
1 file changed, 50 insertions(+), 10 deletions(-)
diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
b/src
From: Marek Olšák
No difference according to shader-db.
---
src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
b/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
From: Marek Olšák
The increase in VGPRs in unfortunate, but the decrease in the scratch size
is always welcome.
Totals:
SGPRS: 344552 -> 344368 (-0.05 %)
VGPRS: 197132 -> 197552 (0.21 %)
Code Size: 7375376 -> 7366304 (-0.12 %) bytes
LDS: 91 -> 91 (0.00 %) blocks
Scratch: 167936
From: Marek Olšák
Include compute shaders too, which includes OpenGL, but not OpenCL.
LLVM doesn't use this much according to shader-db:
Totals:
SGPRS: 344944 -> 344944 (0.00 %)
VGPRS: 197024 -> 197024 (0.00 %)
Code Size: 7325688 -> 7325624 (-0.00 %) bytes
LDS: 91 -> 91 (0.00
Hi,
This patch series improves IR generation for radeonsi. Most of it removes uses
of AMDGPU intrinsics.
There is one piglit regression caused by aggressive handling of "undef" in
LLVM, breaking piglit/glsl-routing. I have a lit test which I'll send later.
Complete stats from shader-db are bel
From: Marek Olšák
Required for 1/sqrt ==> rsq.
We should finally fix the hang instead of running away from the issue. This
assumes the bug is in LLVM and we have time to fix it before the release.
Include compute shaders as well, which only affects TGSI and thus OpenGL.
Totals:
SGPRS: 344
From: Marek Olšák
Intel and Nouveau use IEEE opcodes, so we should too.
If there is a bug caused by not using the clamped RSQ variant, there must
be another way to fix it. I don't think the RSQ behavior matters much now
that NaNs are disabled.
Nine and Wine should implement nece
On Sun, Oct 11, 2015 at 4:22 AM, Roland Scheidegger wrote:
> Am 11.10.2015 um 03:29 schrieb Marek Olšák:
>> From: Marek Olšák
>>
>> This is useful only when emit functions use it.
>> The new radeonsi min/max opcode implementation requires this.
>> ---
On Sun, Oct 11, 2015 at 4:29 AM, Roland Scheidegger wrote:
> FWIW I'm still baffled by this shader bit.
> NaNs are absolutely required to be generated and handled as NaNs in
> shaders (albeit conversion to ints will make them 0) by DX10 (there's
> plenty of tests which actually check for this). An
On Sun, Oct 11, 2015 at 4:12 AM, Connor Abbott wrote:
> FWIW, this isn't quite correct with ARB_shader_precision or GL4.1 --
> it specifies that infinities should be correctly generated through
> division by 0, which unsafe-fp-math doesn't guarantee. At least,
> that's assuming this is similar to
On Sun, Oct 11, 2015 at 6:48 PM, Matt Arsenault wrote:
>
> On Oct 10, 2015, at 6:29 PM, Marek Olšák wrote:
>
> The increase in VGPRs in unfortunate, but the decrease in the scratch size
> is always welcome.
Oh sorry, the stats are actually for the previous patch replacing the
On Sun, Oct 11, 2015 at 11:15 PM, Roland Scheidegger wrote:
> So why do you need to set the info back after action->emit? If you want
> to set that always so that information can be used, looks fine to me.
> But if you have to set it back afterwards that screams hack (and I see
> no reason for suc
On Sun, Oct 11, 2015 at 9:20 AM, Timothy Arceri wrote:
> On Sat, 2015-10-10 at 18:06 +0200, Marek Olšák wrote:
>> Hi Timothy,
>>
>> One of these 3 commits breaks compilation for Talos shaders with
>> gallium. My piglit patch "glsl-1.30/sampler-bug: ..." con
On Sun, Oct 11, 2015 at 5:38 PM, Bas Nieuwenhuizen
wrote:
> Signed-off-by: Bas Nieuwenhuizen
> ---
> src/gallium/drivers/radeon/r600_texture.c | 5 +
> src/gallium/drivers/radeonsi/cik_sdma.c | 3 ++-
> src/gallium/drivers/radeonsi/si_blit.c| 3 ++-
> src/gallium/drivers/radeonsi/si_dm
On Sun, Oct 11, 2015 at 5:38 PM, Bas Nieuwenhuizen
wrote:
> Signed-off-by: Bas Nieuwenhuizen
> ---
> src/gallium/drivers/radeon/r600_pipe_common.h | 1 +
> src/gallium/drivers/radeon/r600_texture.c | 2 ++
> src/gallium/drivers/radeon/r600d_common.h | 1 +
> src/gallium/drivers/radeon
On Sun, Oct 11, 2015 at 5:38 PM, Bas Nieuwenhuizen
wrote:
> Needed for various DCC related operations. As invalidating the L2
> cache seems unnecesary, this introduces a new flag to flush the cache
> without invalidating the L2 cache.
>
> Signed-off-by: Bas Nieuwenhuizen
> ---
> src/gallium/driv
i've reviewed patches 1-4. I'll try to review the rest tomorrow.
Marek
On Sun, Oct 11, 2015 at 5:38 PM, Bas Nieuwenhuizen
wrote:
> These patches enable delta color compression (DCC) for Vulcanic
> Islands GPU's. This should reduce memory bandwidth to increase
> performance.
>
> In this second ve
On Sun, Oct 11, 2015 at 5:38 PM, Bas Nieuwenhuizen
wrote:
> This is currently not needed but will be necessary when we have
> features that do not work with DCC enabled, such as image stores
> and sharing non-scanout surfaces.
For shared non-scanout surfaces, the plan is:
- allocate DCC as part o
Patches 5-7 reviewed as well.
Marek
On Mon, Oct 12, 2015 at 2:46 AM, Marek Olšák wrote:
> i've reviewed patches 1-4. I'll try to review the rest tomorrow.
>
> Marek
>
> On Sun, Oct 11, 2015 at 5:38 PM, Bas Nieuwenhuizen
> wrote:
>> These patches enable
On Sun, Oct 11, 2015 at 5:38 PM, Bas Nieuwenhuizen
wrote:
> Add a mask to track if a texture level is still in its decompressed
> state or that we would need to decompress again.
I think you can drop this patch. I described to plan the support DCC
with shared textures at patch 6.
For image store
On Mon, Oct 12, 2015 at 5:05 PM, Bas Nieuwenhuizen
wrote:
> Hi Marek,
>
> Thanks for the quick review.
>
> I do not think I understand the sharing semantics. We currently have
> fast clear for scanout surfaces with the CMASK and eliminate it on
> flush resource. I would think we could do that simi
On Mon, Oct 12, 2015 at 10:27 PM, Axel Davy wrote:
> On 12/10/2015 17:05, Bas Nieuwenhuizen wrote:
>>
>> Hi Marek,
>>
>> Thanks for the quick review.
>>
>> I do not think I understand the sharing semantics. We currently have
>> fast clear for scanout surfaces with the CMASK and eliminate it on
>>
On Tue, Oct 13, 2015 at 10:13 AM, Timothy Arceri wrote:
> On Mon, 2015-10-12 at 01:06 +0200, Marek Olšák wrote:
>> On Sun, Oct 11, 2015 at 9:20 AM, Timothy Arceri <
>> t_arc...@yahoo.com.au> wrote:
>> > On Sat, 2015-10-10 at 18:06 +0200, Marek Olšák wrote:
>>
Reviewed-by: Marek Olšák
Marek
On Tue, Oct 13, 2015 at 12:26 PM, Emil Velikov wrote:
> Rather than accepting a void pointer, only to down and up cast around
> it, convert the function to take the base (struct gl_program) pointer.
>
> Cc: Marek Olšák
> Signed-off-by: Emil Veliko
Reviewed-by: Marek Olšák
Marek
On Tue, Oct 13, 2015 at 6:40 AM, Dave Airlie wrote:
> This adds support to the parser to accept hex values as floats,
> and then adds support to the dumper to allow the user to select
> to dump float as 32-bit hex numbers.
>
> This is required
s_hex = FALSE;
>
> tgsi_iterate_shader( tokens, &ctx.base.iter );
> +
> + return (ctx.nospace == true) ? -1 : 0;
Why not just return bool meaning success/failure?
Anyway:
Reviewed-by: Marek Olšák
Marek
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Reviewed-by: Marek Olšák
Marek
On Tue, Oct 13, 2015 at 4:45 AM, Brian Paul wrote:
> The variable 'i' is a value in [0, MAT_ATTRIB_MAX-1] so subtracting
> VERT_ATTRIB_GENERIC0 gave a bogus value and we executed the default
> switch clause for all loop iterations.
>
> T
For the series:
Reviewed-by: Marek Olšák
Marek
On Tue, Oct 13, 2015 at 4:45 AM, Brian Paul wrote:
> ---
> src/mesa/vbo/vbo_context.c | 9 +++--
> 1 file changed, 3 insertions(+), 6 deletions(-)
>
> diff --git a/src/mesa/vbo/vbo_context.c b/src/mesa/vbo/vbo_context.c
201 - 300 of 12391 matches
Mail list logo
