Based on the same logic in the i965 driver 2f225f61451abd51 and
16060c5adcd4.
perf reports st_finalize_texture() going from 0.60% -> 0.16% with
this change when running the Xonotic benchmark from PTS.
---
A full run of piglit on radeonsi produced no regressions. No other drivers
have been teste
Hi Eric,
On 06/09/2017 08:59 PM, Eric Engestrom wrote:
On Friday, 2017-06-09 20:13:34 +0530, Harish Krupo wrote:
This patch adds support for the EGL_KHR_partial_update extension for
android platform. It passes 36/37 tests in dEQP for EGL_KHR_partial_update.
1 test not supported.
v2: add fallba
Some drivers require that the vertex buffers be unmapped prior to
drawing. This change unmaps the stream_uploader buffer after we've
uploaded the zero-stride attributes (unless the driver supports
rendering with mapped buffers).
This fixes a regression in the VMware driver since 17f776c27be266f2.
---
src/gallium/auxiliary/util/u_upload_mgr.c | 54 +--
1 file changed, 29 insertions(+), 25 deletions(-)
diff --git a/src/gallium/auxiliary/util/u_upload_mgr.c
b/src/gallium/auxiliary/util/u_upload_mgr.c
index 9528495..4bb14d6 100644
--- a/src/gallium/auxiliary/util/
Sure.
On Fri, Jun 9, 2017 at 5:04 PM, Bas Nieuwenhuizen
wrote:
> Merge this with patch 14?
>
> On Sat, Jun 10, 2017 at 1:47 AM, Connor Abbott
> wrote:
>> From: Connor Abbott
>>
>> To match si_shader_context.
>>
>> Signed-off-by: Connor Abbott
>> ---
>> src/amd/common/ac_llvm_build.c | 2 ++
>>
On Fri, Jun 9, 2017 at 5:04 PM, Bas Nieuwenhuizen
wrote:
> On Sat, Jun 10, 2017 at 1:50 AM, Connor Abbott
> wrote:
>> From: Connor Abbott
>>
>> Signed-off-by: Connor Abbott
>> ---
>> src/amd/common/ac_nir_to_llvm.c | 75
>> +
>> src/amd/vulkan/radv_devi
Merge this with patch 14?
On Sat, Jun 10, 2017 at 1:47 AM, Connor Abbott
wrote:
> From: Connor Abbott
>
> To match si_shader_context.
>
> Signed-off-by: Connor Abbott
> ---
> src/amd/common/ac_llvm_build.c | 2 ++
> src/amd/common/ac_llvm_build.h | 2 ++
> 2 files changed, 4 insertions(+)
>
>
On Sat, Jun 10, 2017 at 1:50 AM, Connor Abbott
wrote:
> From: Connor Abbott
>
> Signed-off-by: Connor Abbott
> ---
> src/amd/common/ac_nir_to_llvm.c | 75
> +
> src/amd/vulkan/radv_device.c| 8 +
> src/amd/vulkan/radv_pipeline.c | 2 ++
> 3 fi
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_build.c | 30 ++
src/amd/common/ac_llvm_build.h | 6 ++
src/gallium/drivers/radeonsi/si_shader.c | 24 +++-
3 files changed, 39 insertions(+), 21
From: Connor Abbott
Ported from radeonsi. Needed for emitting optimization barriers, which
contain inline asm.
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_util.c | 4
1 file changed, 4 insertions(+)
diff --git a/src/amd/common/ac_llvm_util.c b/src/amd/common/ac_llvm_util.c
in
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/amd/common/ac_nir_to_llvm.c | 75 +
src/amd/vulkan/radv_device.c| 8 +
src/amd/vulkan/radv_pipeline.c | 2 ++
3 files changed, 85 insertions(+)
diff --git a/src/amd/common/ac_nir_to_llvm
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_build.c | 1 +
src/amd/common/ac_llvm_build.h | 1 +
2 files changed, 2 insertions(+)
diff --git a/src/amd/common/ac_llvm_build.c b/src/amd/common/ac_llvm_build.c
index c4cf7ce..472a1b8 100644
--- a/src/amd/common/ac_l
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_build.c | 26 ++
src/amd/common/ac_llvm_build.h | 4
src/gallium/drivers/radeonsi/si_shader.c | 38 +---
3 files changed, 36 insertions(+), 32 de
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_build.c | 43 ++
src/amd/common/ac_llvm_build.h | 2 ++
src/gallium/drivers/radeonsi/si_shader.c | 45 ++--
3 files changed, 47 insertions(+),
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_build.c | 1 +
src/amd/common/ac_llvm_build.h | 1 +
2 files changed, 2 insertions(+)
diff --git a/src/amd/common/ac_llvm_build.c b/src/amd/common/ac_llvm_build.c
index 8145bcb..efbeb65 100644
--- a/src/amd/common/ac_l
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_build.c | 24 +++
src/amd/common/ac_llvm_build.h | 2 ++
src/gallium/drivers/radeonsi/si_shader.c | 41 +++-
3 files changed, 34 insertions(+), 33 deletio
From: Connor Abbott
To match si_shader_context.
Signed-off-by: Connor Abbott
---
src/amd/common/ac_llvm_build.c | 2 ++
src/amd/common/ac_llvm_build.h | 2 ++
2 files changed, 4 insertions(+)
diff --git a/src/amd/common/ac_llvm_build.c b/src/amd/common/ac_llvm_build.c
index 237e929..c411880 1
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/compiler/spirv/nir_spirv.h | 2 +
src/compiler/spirv/spirv_to_nir.c | 80 ++
src/compiler/spirv/vtn_variables.c | 28 +
3 files changed, 110 insertions(+)
diff --git a/src/compiler/sp
From: Connor Abbott
For SPIR-V, Khronos decided to make the SubGroup*Mask system values
consist of 4 32-bit components rather than one 64-bit component.
Although we'll lower away the difference in nir_lower_system_values so
drivers won't have to deal with them, adding these system values makes
it
From: Connor Abbott
We're about to bump the number of system values above 32. The
system_values_read bitfield itself is 64 bits, but some users weren't
taking that into account. Fix the ones I could find by grepping for
"system_values_read". This prevents regressions at least with radeonsi
and ot
From: Connor Abbott
Lower them to the GL version.
Signed-off-by: Connor Abbott
---
src/compiler/nir/nir_lower_system_values.c | 38 ++
1 file changed, 38 insertions(+)
diff --git a/src/compiler/nir/nir_lower_system_values.c
b/src/compiler/nir/nir_lower_system_valu
From: Connor Abbott
These are properties of the instruction that must be respected when
moving it around, in addition to the usual SSA dominance guarantee.
Previously, we only had special handling for fddx and fddy, in a very
ad-hoc way. But with arb_shader_ballot and arb_shader_group_vote, we'll
From: Connor Abbott
Signed-off-by: Connor Abbott
---
src/compiler/nir/nir_opt_gcm.c | 72 ++
1 file changed, 23 insertions(+), 49 deletions(-)
diff --git a/src/compiler/nir/nir_opt_gcm.c b/src/compiler/nir/nir_opt_gcm.c
index 879a77a..0fdf587 100644
---
From: Connor Abbott
These optimizations happened to work with derivatives, but they won't
with upcoming shader_ballot and group_vote instructions.
Signed-off-by: Connor Abbott
---
src/compiler/nir/nir_instr_set.c | 22 ++
src/compiler/nir/nir_opt_peephole_select.c
From: Connor Abbott
This series adds all the bits to enable EXT_shader_subgroup_vote and
EXT_shader_subgroup_ballot for radv. It's based on my previous series to
fix some 64-bit bugs in radv and anv, since nothing would work without
them.
Patches 1-4 are a resend of my previous series to add ARB
From: Connor Abbott
v2: make every instruction an intrinsic, add missing subgroup_size,
also add system value translation stuff
Signed-off-by: Connor Abbott
---
src/compiler/nir/nir.c| 28
src/compiler/nir/nir_intrinsics.h | 30 +
This patch replaces the old register livetime estimation with the
new approach.
---
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 17 +++--
1 file changed, 15 insertions(+), 2 deletions(-)
diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
b/src/mesa/state_tracker/st_glsl_to_tgsi
This patch adds new classes and tests to implement a tracker for the
life time of temporary registers for the register renaming stage of
glsl_to_tgsi. The tracker aims at estimating the shortest possible
life time for each register. The code base requires c++11, the flag is
propagated from the LLVM
To prepare the implementation of a temp register lifetime tracker
some of the classes are moved into seperate header/implementation
files to make them accessible from other files.
---
src/mesa/Makefile.sources | 2 +
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 2
Dear all,
as I wrote before, I was looking into the temporary register renaming.
This series of patches implements a new approach that achieves a tigher
estimation of the life time of the temporaries, and as a result the Piano
and Voloplosion benchmarks implemented in gputest [1] now work. Before
The existence of this software fallback makes me sad 😥.
On June 9, 2017 7:05:33 AM Topi Pohjolainen wrote:
Signed-off-by: Topi Pohjolainen
---
src/mesa/drivers/dri/i965/intel_mipmap_tree.c | 60 ---
1 file changed, 35 insertions(+), 25 deletions(-)
diff --git a/src/
Reviewed-by: Jason Ekstrand
On June 9, 2017 9:01:38 AM "Juan A. Suarez Romero" wrote:
Otherwise, `make distcheck` will fail.
---
src/mesa/drivers/dri/i965/Makefile.am | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/drivers/dri/i965/Makefile.am
b/src/mesa/drivers/dri/i965/Makef
On June 9, 2017 2:41:46 PM Anuj Phogat wrote:
On Thu, Jun 8, 2017 at 5:07 PM, Jason Ekstrand wrote:
On Mon, May 15, 2017 at 10:05 AM, Anuj Phogat wrote:
On Sat, May 13, 2017 at 9:43 AM, Jason Ekstrand
wrote:
On May 12, 2017 4:41:36 PM Anuj Phogat wrote:
Signed-off-by: Anuj Phogat
https://bugs.freedesktop.org/show_bug.cgi?id=101334
John changed:
What|Removed |Added
CC||airl...@freedesktop.org
--- Comment #5 from John
https://bugs.freedesktop.org/show_bug.cgi?id=101334
--- Comment #4 from John ---
Alright, after bisecting here's the problematic commit:
https://cgit.freedesktop.org/mesa/mesa/commit/?id=bcae3274692954ad2cd6dfc253579ec98d50856f
Thanks!
--
You are receiving this mail because:
You are the QA Co
On Thu, Jun 8, 2017 at 5:23 PM, Jason Ekstrand wrote:
> I think I've now reviewed everything except 2 patches. For the "Update a
> few assertions" patch, you said you would run a test but never reported back
> the results. The other is the patch for re-enabling sRGB fast-clears. That
> one isn'
On Thu, Jun 8, 2017 at 5:07 PM, Jason Ekstrand wrote:
> On Mon, May 15, 2017 at 10:05 AM, Anuj Phogat wrote:
>>
>>
>>
>> On Sat, May 13, 2017 at 9:43 AM, Jason Ekstrand
>> wrote:
>>>
>>> On May 12, 2017 4:41:36 PM Anuj Phogat wrote:
>>>
Signed-off-by: Anuj Phogat
---
src/intel/
The resolve code looks at the current color draw buffers. These are not
valid until intel_prepare_render() is called. You can end up with one
color buffer bound, but where the renderbuffer has zero width/height and
no miptree allocated.
You can get a call chain like: _mesa_Clear -> _mesa_update_
https://bugs.freedesktop.org/show_bug.cgi?id=101338
--- Comment #5 from Lloyd Wood ---
Brian,
Thanks for checking this out. Which versions of Geomview and SaVi did you
install?
Seeing a solid blue sphere on your Intel system suggests either that
texturemapping is not used by default in SaVi (in
https://bugs.freedesktop.org/show_bug.cgi?id=101338
--- Comment #4 from Roland Scheidegger ---
The original Raspberry Pi IIRC doesn't even support NEON. Theoretically, llvm
should still support all the vector instructions by decomposing them into
scalar ones, but if that really works correctly? I
On Fri, 2017-06-09 at 13:44 -0500, Aaron Watry wrote:
> On Fri, Jun 9, 2017 at 12:36 PM, Jan Vesely wrote:
> > On Fri, 2017-06-09 at 10:12 -0500, Aaron Watry wrote:
> > > On Fri, Jun 9, 2017 at 8:20 AM, Jan Vesely wrote:
> > > > This is a verbatim copy of the code. The functions can be cleaned up
On Fri, Jun 9, 2017 at 12:36 PM, Jan Vesely wrote:
> On Fri, 2017-06-09 at 10:12 -0500, Aaron Watry wrote:
>> On Fri, Jun 9, 2017 at 8:20 AM, Jan Vesely wrote:
>> > This is a verbatim copy of the code. The functions can be cleaned up since
>> > r600 does not use all the stuff that gcn does.
>> >
Reviewed-by: Connor Abbott
On Tue, Jun 6, 2017 at 4:25 PM, Grazvydas Ignotas wrote:
> This will allow to constify other things.
>
> Signed-off-by: Grazvydas Ignotas
> ---
> src/compiler/nir/nir.h | 25 +
> src/compiler/nir/nir_lower_io.c | 2 +-
> 2 files chan
On Fri, 2017-06-09 at 10:12 -0500, Aaron Watry wrote:
> On Fri, Jun 9, 2017 at 8:20 AM, Jan Vesely wrote:
> > This is a verbatim copy of the code. The functions can be cleaned up since
> > r600 does not use all the stuff that gcn does.
> > The symbol names have been changed since we still use ac_b
On Friday, 2017-06-09 20:23:04 +0300, Grazvydas Ignotas wrote:
> Ping. Boring patch, should be easy to review or NAK.
Don't know anything about NIR, so no clue if it's a good idea, but
I personally like `const` (I think it should've been the default :P)
and this patch looks good to me:
Reviewed-by
This is a verbatim copy of the code. The functions can be cleaned up since
r600 does not use all the stuff that gcn does.
The symbol names have been changed since we still use ac_binary.h header
(for struct definition)
v2: Add ifdef guard around r600_binary_clean call (Aaron)
Remove stray comm
https://bugs.freedesktop.org/show_bug.cgi?id=101334
--- Comment #3 from Eric Engestrom ---
(In reply to John from comment #2)
> Would anything else help?
It would be really helpful if you could bisect the issue.
This means picking an app (game) that was working and doesn't work anymore, and
runn
Ping. Boring patch, should be easy to review or NAK.
On Wed, Jun 7, 2017 at 2:25 AM, Grazvydas Ignotas wrote:
> This will allow to constify other things.
>
> Signed-off-by: Grazvydas Ignotas
> ---
> src/compiler/nir/nir.h | 25 +
> src/compiler/nir/nir_lower_io.
Otherwise, `make distcheck` will fail.
---
src/gallium/drivers/r600/Makefile.am | 4
1 file changed, 4 insertions(+)
diff --git a/src/gallium/drivers/r600/Makefile.am
b/src/gallium/drivers/r600/Makefile.am
index 44fd51d..2b1ffb5 100644
--- a/src/gallium/drivers/r600/Makefile.am
+++ b/src/ga
Otherwise, `make distcheck` will fail.
---
src/mesa/drivers/dri/i965/Makefile.am | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/drivers/dri/i965/Makefile.am
b/src/mesa/drivers/dri/i965/Makefile.am
index 762aefc..e2d5992 100644
--- a/src/mesa/drivers/dri/i965/Makefile.am
+++ b/src/me
On Wed, Jun 7, 2017 at 11:12 PM, Aaron Watry wrote:
> On Wed, Jun 7, 2017 at 9:15 PM, Michel Dänzer wrote:
>> On 08/06/17 03:42 AM, Marek Olšák wrote:
>>> On Wed, Jun 7, 2017 at 4:10 PM, Aaron Watry wrote:
On Mon, Jun 5, 2017 at 3:07 PM, Marek Olšák wrote:
>
> Can you make the chan
Actually, these are the only patches that are required. We're trying to
run the Android Studio emulator using the host's GLES implementation. The
emulator uses the image extension in that case:
https://android.googlesource.com/platform/sdk/+/emu-2.4-
release/emulator/opengl/host/libs/libOpenglRe
On Thu, Jun 8, 2017 at 5:19 PM, Jason Ekstrand wrote:
> I sent out a Vulkan patch (and CCd you) that needs to land before this one
> so that we don't accidentally start advertising Vulkan support once the PCI
> IDs land.
>
I'll land the vulkan patch before this one. Thanks.
> This seems to match t
On Friday, 2017-06-09 20:13:34 +0530, Harish Krupo wrote:
> This patch adds support for the EGL_KHR_partial_update extension for
> android platform. It passes 36/37 tests in dEQP for EGL_KHR_partial_update.
> 1 test not supported.
>
> v2: add fallback for eglSetDamageRegionKHR (Tapani)
>
> v3: Th
On Fri, Jun 9, 2017 at 8:20 AM, Jan Vesely wrote:
> This is a verbatim copy of the code. The functions can be cleaned up since
> r600 does not use all the stuff that gcn does.
> The symbol names have been changed since we still use ac_binary.h header
> (for struct definition)
>
> Signed-off-by: Ja
https://bugs.freedesktop.org/show_bug.cgi?id=101338
--- Comment #3 from Brian Paul ---
I don't have a Raspberry Pi to test with. I installed geomview/savi on my
Intel deskside system and tested with both NVIDIA's driver and llvmpipe. With
both I see a shaded blue sphere with an orbit ring and r
On Fri, Jun 9, 2017 at 4:31 PM, Samuel Pitoiset
wrote:
>
>
> On 06/09/2017 04:26 PM, Marek Olšák wrote:
>>
>> On Fri, Jun 9, 2017 at 4:00 PM, Samuel Pitoiset
>> wrote:
>>>
>>>
>>>
>>> On 06/09/2017 03:49 PM, Marek Olšák wrote:
From: Marek Olšák
---
src/gallium/dr
On 9 June 2017 at 14:16, Rowley, Timothy O wrote:
> With this patch series applied, the build fails for me on ubuntu 16.04.
>
Thanks Tim - I can see what's going wrong. I'll double-check things
and send v2 in a bit.
Emil
___
mesa-dev mailing list
mesa-d
Hi Emil,
Did you test the upstream versions or the distribution ones which tend
> to be patched?
>
Both. I build 17.1.1 against the system supplied zlib-devel packages for
1.2.3 in EL6 and 1.2.7 on EL7. I then swapped out the zlib version at
runtime via LD_LIBRARY_PATH with ones build from the
On Thu, Jun 8, 2017 at 5:10 PM, Jason Ekstrand wrote:
> This will prevent the driver from even trying to work on Cannon Lake
> until we get actual support added.
>
> Cc: Anuj Phogat
> ---
> src/intel/vulkan/anv_device.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src
https://bugs.freedesktop.org/show_bug.cgi?id=101326
Brian Paul changed:
What|Removed |Added
Resolution|--- |FIXED
Status|NEW
Let's not impose behavior that would burden drivers (for example: "idiv
by zero must result in ~0u"). Just updating the docs to say the results
of div/mod by zero is undefined (unless we know something specific is
needed) would be fine. As it is now, some div/mod operations are
documented to
This patch adds support for the EGL_KHR_partial_update extension for
android platform. It passes 36/37 tests in dEQP for EGL_KHR_partial_update.
1 test not supported.
v2: add fallback for eglSetDamageRegionKHR (Tapani)
v3: The native_window_set_surface_damage call is available only from
Andro
On 06/09/2017 04:26 PM, Marek Olšák wrote:
On Fri, Jun 9, 2017 at 4:00 PM, Samuel Pitoiset
wrote:
On 06/09/2017 03:49 PM, Marek Olšák wrote:
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_state.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/galli
Reviewed-by: Marek Olšák
Marek
On Fri, Jun 9, 2017 at 3:56 PM, Brian Paul wrote:
> ---
> src/gallium/docs/source/tgsi.rst | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/src/gallium/docs/source/tgsi.rst
> b/src/gallium/docs/source/tgsi.rst
> index eceaa6d..7fb963
On Fri, Jun 9, 2017 at 4:00 PM, Samuel Pitoiset
wrote:
>
>
> On 06/09/2017 03:49 PM, Marek Olšák wrote:
>>
>> From: Marek Olšák
>>
>> ---
>> src/gallium/drivers/radeonsi/si_state.c | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/src/gallium/drivers/radeonsi/si_sta
On Fri, Jun 9, 2017 at 3:35 PM, Samuel Pitoiset
wrote:
> This adds some new helper functions to know if the current draw
> call (or dispatch compute) is using bindless samplers/images,
> based on TGSI analysis.
>
> v3: - add si_context::uses_bindless_{samplers,images}
> - add si_bind_shader_co
Signed-off-by: Topi Pohjolainen
---
src/mesa/drivers/dri/i965/intel_mipmap_tree.c | 60 ---
1 file changed, 35 insertions(+), 25 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/intel_mipmap_tree.c
b/src/mesa/drivers/dri/i965/intel_mipmap_tree.c
index f8fdde7..a4b2aeb
On gen < 6 one doesn't have level or layer specifiers available
for render and depth targets. In order to support rendering to
specific level/layer, driver needs to manually offset the surface
to the desired slice.
There are, however, alignment restrictions to respect as well and
in come cases the
Signed-off-by: Topi Pohjolainen
---
src/mesa/drivers/dri/i965/intel_mipmap_tree.c | 39 ++-
1 file changed, 20 insertions(+), 19 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/intel_mipmap_tree.c
b/src/mesa/drivers/dri/i965/intel_mipmap_tree.c
index 4b58268..f8fdde7
On 06/09/2017 03:49 PM, Marek Olšák wrote:
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_state.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/radeonsi/si_state.c
b/src/gallium/drivers/radeonsi/si_state.c
index 53f66ac..a8255f2 100644
--
On Fri, Jun 9, 2017 at 3:35 PM, Samuel Pitoiset
wrote:
> This implements the Gallium interface. Decompression of resident
> textures/images will follow in the next patches.
>
> v3: - do not unmap bindless descriptors
> - remove unnecessary util_copy_image_view()
> - use READON_USAGE_READWR
---
src/gallium/docs/source/tgsi.rst | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/gallium/docs/source/tgsi.rst b/src/gallium/docs/source/tgsi.rst
index eceaa6d..7fb963f 100644
--- a/src/gallium/docs/source/tgsi.rst
+++ b/src/gallium/docs/source/tgsi.rst
@@ -3298,8 +
https://bugs.freedesktop.org/show_bug.cgi?id=101326
--- Comment #3 from Brian Paul ---
Thanks for the info. The patch looks good. I'll push it soon with minor
reformatting.
--
You are receiving this mail because:
You are the assignee for the bug.
You are the QA Contact for the bug.___
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_state_draw.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c
b/src/gallium/drivers/radeonsi/si_state_draw.c
index 8508259..ec564c1 100644
--- a/src/gallium/drivers/radeonsi/si_state_draw.c
+++
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_state.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/radeonsi/si_state.c
b/src/gallium/drivers/radeonsi/si_state.c
index 53f66ac..a8255f2 100644
--- a/src/gallium/drivers/radeonsi/si_state.c
+++ b
From: Marek Olšák
---
src/gallium/drivers/noop/noop_state.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/gallium/drivers/noop/noop_state.c
b/src/gallium/drivers/noop/noop_state.c
index 46d99ab..80cfae8 100644
--- a/src/gallium/drivers/noop/noop_state.c
+++ b/src/gallium/drivers/no
From: Marek Olšák
---
src/gallium/drivers/radeonsi/si_blit.c| 4 ++--
src/gallium/drivers/radeonsi/si_descriptors.c | 28 +--
src/gallium/drivers/radeonsi/si_pipe.h| 2 +-
src/gallium/drivers/radeonsi/si_state.h | 2 +-
4 files changed, 18 inserti
From: Marek Olšák
---
src/mesa/state_tracker/st_atom.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/mesa/state_tracker/st_atom.c b/src/mesa/state_tracker/st_atom.c
index cbac762..bcfbcf8 100644
--- a/src/mesa/state_tracker/st_atom.c
+++ b/src/mesa/state_tracker/
From: Marek Olšák
---
src/gallium/docs/source/screen.rst | 9 +
1 file changed, 9 insertions(+)
diff --git a/src/gallium/docs/source/screen.rst
b/src/gallium/docs/source/screen.rst
index 852c31b..288fb5c 100644
--- a/src/gallium/docs/source/screen.rst
+++ b/src/gallium/docs/source/scre
This has only been tested on RX480.
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák (v2)
---
docs/features.txt | 2 +-
docs/relnotes/17.2.0.html | 1 +
src/gallium/drivers/radeonsi/si_pipe.c | 4 +++-
3 files changed, 5 insertions(+), 2 deletions(-)
di
When a buffer becomes resident, check if it has been invalidated,
if so update the descriptor and the dirty flag.
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák (v2)
---
src/gallium/drivers/radeonsi/si_descriptors.c | 34 +++
1 file changed, 34 insertions(+)
di
This will be used in order to initialize resident descriptors
for bindless textures/images.
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák
---
src/gallium/drivers/radeonsi/si_descriptors.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/src/gallium/drivers/radeonsi
v2: - removed use of the Bindless flags after rebasing
- make use of tgsi_is_bindless_image_file()
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák (v2)
---
src/gallium/drivers/radeonsi/si_shader_tgsi_mem.c | 28 +--
1 file changed, 21 insertions(+), 7 deletions(-
v2: - removed use of the Bindless flags after rebasing
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák (v2)
---
src/gallium/drivers/radeonsi/si_shader_tgsi_mem.c | 15 ---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_shader_t
softpipe throws integer division by zero exceptions on windows
when using % with integers in a geometry shader.
v2: Made error results consistent with existing div/mod zero handling in
tgsi. 64 bit signed integer division by zero returns zero like in
micro_idiv, unsigned returns ~0u like i
Resident buffers have to be added to every new command stream.
Though, this could be slightly improved when current shaders
don't use any bindless textures/images but usually applications
tend to use bindless for almost every draw call, and the winsys
thread might help when buffers are added early.
To share some common code between bound and bindless textures.
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák
---
src/gallium/drivers/radeonsi/si_descriptors.c | 95 +++
1 file changed, 52 insertions(+), 43 deletions(-)
diff --git a/src/gallium/drivers/radeonsi
When texture buffers are invalidated the addr in the resident
descriptor has to be updated but we can't create a new descriptor
because the resident handle has to be the same.
Instead, use the WRITE_DATA packet which allows to update memory
directly but graphics/compute have to be idle in case the
Similar to the existing decompression code path except that it
loops over the list of resident textures/images.
v3: - move the blitter check back to si_decompress_textures()
- decompress resident textures directly in si_decompress_textures()
- depth_texture -> needs_depth_decompress
-
This won't help much except for applications that use a ton
of resident handles. Though, this will reduce the winsys
overhead a little bit.
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák (v2)
---
src/gallium/drivers/radeonsi/si_descriptors.c | 6 ++
1 file changed, 6 insertions(+)
The ARB_bindless_texture spec say:
"If ARB_seamless_cubemap (or OpenGL 4.0, which includes it) is
supported, the per-context seamless cubemap enable is ignored
and treated as disabled when using texture handles."
"If AMD_seamless_cubemap_per_texture is supported, the seamless
cu
When the current bound shaders don't use any bindless textures
or images, it's useless to decompress the resident resources.
v3: - rebased and updated check for compute
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák (v2)
---
src/gallium/drivers/radeonsi/si_blit.c | 10 --
1 fi
This adds some new helper functions to know if the current draw
call (or dispatch compute) is using bindless samplers/images,
based on TGSI analysis.
v3: - add si_context::uses_bindless_{samplers,images}
- add si_bind_shader_common() to limit code duplication
Signed-off-by: Samuel Pitoiset
-
Analogous to bound textures/images. We should also update the
resident descriptors and disable COMPRESSION_EN for avoiding
useless DCC fetches, but I postpone this optimization for a
separate series.
v3: - remove use of si_update_check_render_feedback()
v2: - store pipe_sampler_view instead of si_
v2: - remove "texture" parameter from create_texture_handle()
Signed-off-by: Samuel Pitoiset
Reviewed-by: Nicolai Hähnle (v1)
Reviewed-by: Marek Olšák (v2)
---
src/mesa/state_tracker/st_cb_texture.c | 84 ++
1 file changed, 84 insertions(+)
diff --git a/src/mes
For each texture/image handles, we need to allocate a new
buffer for the bindless descriptor. But when the number of
buffers added to the current CS becomes high, the overhead
in the winsys (and in the kernel) is important.
To reduce this bottleneck, the idea is to suballocate the
bindless descrip
To share some common code between bound and bindless images.
v3: - use vi_dcc_enabled() instead of dcc_offset for images
Signed-off-by: Samuel Pitoiset
Reviewed-by: Marek Olšák (v2)
---
src/gallium/drivers/radeonsi/si_descriptors.c | 79 ---
1 file changed, 47 insertion
This implements the Gallium interface. Decompression of resident
textures/images will follow in the next patches.
v3: - do not unmap bindless descriptors
- remove unnecessary util_copy_image_view()
- use READON_USAGE_READWRITE because of the WRITE_DATA packet
- replace util_dynarray_de
Signed-off-by: Samuel Pitoiset
Reviewed-by: Nicolai Hähnle
Reviewed-by: Marek Olšák
---
src/mesa/state_tracker/st_atom_image.c | 33 ++---
src/mesa/state_tracker/st_texture.h| 5 +
2 files changed, 27 insertions(+), 11 deletions(-)
diff --git a/src/mesa/sta
1 - 100 of 186 matches
Mail list logo
