On 18/04/17 03:51 PM, gregory hainaut wrote:
>
> However, I have found others crashes on PCSX2 with those patches. It
> seems related to synchronization issue with GLX/DRI/X11. This series
> removes most of the gl sync for PCSX2. So any missing sync will trigger
> a crash. Or I got a not obvious b
On Tue, 18 Apr 2017 08:51:24 +0200
gregory hainaut wrote:
> On Mon, 17 Apr 2017 11:17:42 +0900
> Michel Dänzer wrote:
>
> > On 15/04/17 05:08 PM, gregory hainaut wrote:
> > > On Sat, 15 Apr 2017 00:50:15 +0200
> > > Dieter Nützel wrote:
> > >
> > >> Am 14.04.2017 07:53, schrieb gregory hainau
On 18/04/17 05:04 PM, gregory hainaut wrote:
> On Tue, 18 Apr 2017 08:51:24 +0200
> gregory hainaut wrote:
>
>> On Mon, 17 Apr 2017 11:17:42 +0900
>> Michel Dänzer wrote:
>>
>>> On 15/04/17 05:08 PM, gregory hainaut wrote:
On Sat, 15 Apr 2017 00:50:15 +0200
Dieter Nützel wrote:
>
Hello Michel,
As yes, I completely forgot about XInitThreads that must be it. I
don't know how Nvidia manage to solve/force it. Anyway, I will fix my
application.
Thanks you for the info.
On 4/18/17, Michel Dänzer wrote:
> On 18/04/17 05:04 PM, gregory hainaut wrote:
>> On Tue, 18 Apr 2017 08:5
Jeremy, is it OK that I mention you as a reference for MacOSX in the
releasing doc?
On Mon, 2017-03-27 at 21:48 +0300, Andres Gomez wrote:
> Signed-off-by: Andres Gomez
> Cc: Emil Velikov
> Cc: Jeremy Huddleston Sequoia
> ---
> docs/releasing.html | 5 +
> 1 file changed, 5 insertions(+)
>
Brian, is it OK that I mention you as a reference for Windows in the
releasing doc?
On Mon, 2017-03-27 at 12:03 -0700, Vinson Lee wrote:
> On Mon, Mar 27, 2017 at 11:48 AM, Andres Gomez wrote:
> > Signed-off-by: Andres Gomez
> > Cc: Emil Velikov
> > Cc: Vinson Lee
> > ---
> > docs/releasing.h
Jose, is it OK that I mention you as a reference for Windows in the
releasing doc?
On Mon, 2017-03-27 at 12:03 -0700, Vinson Lee wrote:
> On Mon, Mar 27, 2017 at 11:48 AM, Andres Gomez wrote:
> > Signed-off-by: Andres Gomez
> > Cc: Emil Velikov
> > Cc: Vinson Lee
> > ---
> > docs/releasing.ht
On 04/18/2017 04:10 AM, Timothy Arceri wrote:
On 18/04/17 11:25, Plamena Manolova wrote:
This extension provides new GLSL built-in functions
beginInvocationInterlockARB() and endInvocationInterlockARB()
that delimit a critical section of fragment shader code. For
pairs of shader invocations
On 18 April 2017 at 06:52, Timothy Arceri wrote:
> From: Timothy Arceri
>
> This shuffles constants down in the reverse of what the previous
> patch does and applies some simpilifications that may be made
> possible from doing so.
>
> Shader-db results BDW:
>
> total instructions in shared progra
On 14.04.2017 00:44, Samuel Pitoiset wrote:
Similar to st_convert_image(), will be useful for bindless. While
we are at it, rename convert_sampler() to convert_sampler_from_unit()
and make 'st' a const argument.
Signed-off-by: Samuel Pitoiset
Reviewed-by: Nicolai Hähnle
---
src/mesa/stat
On 17.04.2017 18:54, Gurchetan Singh wrote:
None of the PCI id headers have licensing information. Let's add
the standard Mesa license.
Hmm, having license text without any indication about who holds
copyrights seems weird...
Nicolai
---
include/pci_ids/i810_pci_ids.h | 22
eero, chris: michel informs me that this is a known bug in
xserver-xorg-core 1.19.0 and that upgrading to at least 1.19.1 or
above fixes the problem. i'll need to wait until another unscheduled
reboot but will keep you informed, and will try out DRI2 as well as
DRI3.
l.
__
On 18.04.2017 02:09, Timothy Arceri wrote:
V2: typo know -> known
---
docs/envvars.html | 3 +++
src/mesa/main/context.c | 3 +++
2 files changed, 6 insertions(+)
diff --git a/docs/envvars.html b/docs/envvars.html
index 6537365..a064f56 100644
--- a/docs/envvars.html
+++ b/docs/envvars.ht
On 18.04.2017 02:09, Timothy Arceri wrote:
V2: restore lost comment, add static to validate_uniform(),
simplify array offset logic.
---
src/mesa/main/uniform_query.cpp | 74 -
1 file changed, 51 insertions(+), 23 deletions(-)
diff --git a/src/mesa/mai
On 18.04.2017 02:09, Timothy Arceri wrote:
V2: some bug fixes and suggestions addressed as suggested by Brian and
Nicolai.
This series adds some initial support for the KHR_no_error.
For now it can only be enabled with the MESA_NO_ERROR environment
variable. To start with I've added support
From: Emil Velikov
The macro is introduced with pkgconfig v0.28 which isn't universally
available. Thus it will error at configure stage.
Cc: Brian Paul
Reported-by: Brian Paul
Fixes: ce562f9e3fa ("EGL: Implement the libglvnd interface for EGL (v3)")
Signed-off-by: Emil Velikov
---
Pardon abo
On 13.04.2017 20:10, Bartosz Tomczyk wrote:
v2: fix attribute name, it is count_scale not scale_count
R-b and pushed, thanks!
---
src/mapi/glapi/gen/ARB_viewport_array.xml | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/src/mapi/glapi/gen/ARB_viewport_array.xml
On 17 April 2017 at 19:11, Brian Paul wrote:
> On 04/17/2017 07:34 AM, Emil Velikov wrote:
>>
>> Module: Mesa
>> Branch: master
>> Commit: ce562f9e3fab769d64b0e5453ec2b4f8710a31ce
>> URL:
>> https://urldefense.proofpoint.com/v2/url?u=http-3A__cgit.freedesktop.org_mesa_mesa_commit_-3Fid-3Dce562f9e3
On 18.04.2017 03:34, Rob Clark wrote:
So atexit() is horrible and 4aea8fe7 is probably not a good idea. But
add an extra layer of duct-tape to the problem. Otherwise we hit a
situation where app using an atexit() handler that runs later than ours
doesn't hang when trying to tear down a context.
On 18.04.2017 07:19, Dave Airlie wrote:
On 18 April 2017 at 15:07, Bas Nieuwenhuizen wrote:
On Tue, Apr 18, 2017 at 5:57 AM, Dave Airlie wrote:
From: Dave Airlie
In practice this will probably just drop draw id in a few places.
Signed-off-by: Dave Airlie
---
src/amd/common/ac_nir_to_llvm
On 04/11/2017 07:11 PM, Chris Wilson wrote:
Decode the EGL/DRI priority enum into the [-1023, 1023] range as
interpreted by the kernel and call DRM_I915_GEM_CONTEXT_SETPARAM to
adjust the priority. We use 0 as the default medium priority (also the
kernel default) and so only need adjust up or d
On 17.04.2017 14:07, Edward O'Callaghan wrote:
Signed-off-by: Edward O'Callaghan
---
docs/features.txt | 2 +-
src/mesa/state_tracker/st_extensions.c | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/docs/features.txt b/docs/features.txt
index 5f63632..40
On 17.04.2017 14:07, Edward O'Callaghan wrote:
Hi folks,
This is my early draft on the threaded shader compile "suggestion"
thread. The requirements according the spec are rather weak as
it is up to the implementation, which frankly can even just ignore
the requested thread count. However, afaik
On 18.04.2017 03:25, Plamena Manolova wrote:
This extension provides new GLSL built-in functions
beginInvocationInterlockARB() and endInvocationInterlockARB()
that delimit a critical section of fragment shader code. For
pairs of shader invocations with "overlapping" coverage in a
given pixel, the
You can list me as fallback contact, but Brian Paul is the right person
for Windows support overall. As lately I do little more than help out
with odd Windows build related issues.
Jose
On 18/04/17 10:40, Andres Gomez wrote:
Jose, is it OK that I mention you as a reference for Windows in the
Support supertiled textures on hardware that has the appropriate
feature flag SUPERTILED_TEXTURE.
Most of the scaffolding was already in place in etna_layout_multiple:
case ETNA_LAYOUT_SUPER_TILED:
*paddingX = 64;
*paddingY = 64;
*halign = TEXTURE_HALIGN_SUPER_TILED;
So this
On Fri, Apr 14, 2017 at 10:37:53AM -0700, Jason Ekstrand wrote:
> From: Chad Versace
>
> This is a complete but trivial implementation. It's trivial becasue We
because we
___
mesa-dev mailin
On Fri, Apr 14, 2017 at 10:37:57AM -0700, Jason Ekstrand wrote:
> This commit just exposes the memory handle type. There's interesting we
> need to do here for images. So long as the user doesn't set any crazy
Sentence beginning with "There's..." looks a little clumsy?
__
On 04/18/2017 09:01 PM, Nicolai Hähnle wrote:
> On 17.04.2017 14:07, Edward O'Callaghan wrote:
>> Signed-off-by: Edward O'Callaghan
>> ---
>> docs/features.txt | 2 +-
>> src/mesa/state_tracker/st_extensions.c | 1 +
>> 2 files changed, 2 insertions(+), 1 deletion(-)
>>
>>
On Wed, 2017-04-12 at 08:50 -0700, Jason Ekstrand wrote:
> On Wed, Apr 12, 2017 at 2:47 AM, Juan A. Suarez Romero
> wrote:
> > Reviewed-by: Juan A. Suarez Romero
> >
>
> Is this patch sufficient to fix the bug you were seeing yesterday?
>
Sorry for answering late. I took some days off and c
On Fri, Apr 14, 2017 at 10:38:08AM -0700, Jason Ekstrand wrote:
> ---
> src/intel/vulkan/anv_batch_chain.c | 69
> src/intel/vulkan/anv_device.c | 2 +
> src/intel/vulkan/anv_private.h | 8
> src/intel/vulkan/anv_queue.c | 93
> ++
On Tue, Apr 18, 2017 at 2:21 AM, Michel Dänzer wrote:
> On 18/04/17 01:02 PM, Ilia Mirkin wrote:
>> On Tue, Apr 18, 2017 at 12:00 AM, Ilia Mirkin wrote:
>>> val_bool and val_int are in a union. val_bool gets the first byte, which
>>> happens to work on LE when setting via the int, but breaks on B
On Mon, 27 Feb 2017 13:26:11 +
Emil Velikov wrote:
> Hi Julien,
>
> On 27 February 2017 at 12:08, Julien Isorce wrote:
> > Hi,
> >
> > Since 2012 commit ccff74971203b533bf16b46b49a9e61753f75e6c it is said:
> > "sample must be initialized, or else the result is undefined" in the
> > descript
https://bugs.freedesktop.org/show_bug.cgi?id=100668
Jose Fonseca changed:
What|Removed |Added
CC||bri...@vmware.com,
|
On Fri, Apr 14, 2017 at 10:38:02AM -0700, Jason Ekstrand wrote:
> ---
> src/intel/vulkan/anv_batch_chain.c | 59
> ++
> 1 file changed, 35 insertions(+), 24 deletions(-)
Reviewed-by: Topi Pohjolainen
>
> diff --git a/src/intel/vulkan/anv_batch_chain.c
> b/
Hi all,
Another re-send with the issue mentioned by Nils fixed, since I said I
would do so, and stable tags added, since 17.1 was branched in the
meantime. That said, all patches have been looked at, so if nobody
objects, I'll push them out tomorrow.
Thanks,
Nicolai
_
From: Nicolai Hähnle
Found by inspection.
Cc: mesa-sta...@lists.freedesktop.org
Reviewed-by: Timothy Arceri
---
src/mesa/main/api_validate.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/main/api_validate.c b/src/mesa/main/api_validate.c
index af4f7cb..8f83432 10
From: Nicolai Hähnle
The same logic needs to be applied to glMultiDrawArrays.
Cc: mesa-sta...@lists.freedesktop.org
Reviewed-by: Timothy Arceri
---
src/mesa/main/api_validate.c | 48 ++--
1 file changed, 28 insertions(+), 20 deletions(-)
diff --git a/sr
From: Nicolai Hähnle
When any count[i] is negative, we must skip all draws.
Moving to vbo makes the subsequent change easier.
v2:
- provide the function in all contexts, including GLES
- adjust validation accordingly to include the xfb check
v3:
- fix mix-up of pre- and post-xfb prim count
Cc:
From: Nicolai Hähnle
Fixes a bug in
KHR-GL45.shader_draw_parameters_tests.ShaderMultiDrawArraysParameters.
Cc: mesa-sta...@lists.freedesktop.org
Reviewed-by: Marek Olšák
---
src/mesa/vbo/vbo_exec_array.c | 21 +++--
1 file changed, 15 insertions(+), 6 deletions(-)
diff --git a
All GL calls that might use libX11 must not be asynchronous within glthread.
Marek
On Apr 18, 2017 10:43 AM, "Gregory Hainaut"
wrote:
Hello Michel,
As yes, I completely forgot about XInitThreads that must be it. I
don't know how Nvidia manage to solve/force it. Anyway, I will fix my
applicatio
On 04/18/2017 04:43 AM, Emil Velikov wrote:
From: Emil Velikov
The macro is introduced with pkgconfig v0.28 which isn't universally
available. Thus it will error at configure stage.
Cc: Brian Paul
Reported-by: Brian Paul
Fixes: ce562f9e3fa ("EGL: Implement the libglvnd interface for EGL (v3)
On Mon, 2017-04-10 at 12:15 +0200, Juan A. Suarez Romero wrote:
> On Fri, 2017-04-07 at 19:38 +0100, Emil Velikov wrote:
> > On 7 April 2017 at 12:30, Juan A. Suarez Romero wrote:
> > > Both scripts does not use a file with the commits to ignore. So if we
> > > have handled one of the suggested co
On 18 April 2017 at 16:16, Andres Gomez wrote:
> On Mon, 2017-04-10 at 12:15 +0200, Juan A. Suarez Romero wrote:
>> On Fri, 2017-04-07 at 19:38 +0100, Emil Velikov wrote:
>> > On 7 April 2017 at 12:30, Juan A. Suarez Romero
>> > wrote:
>> > > Both scripts does not use a file with the commits to
That should be fine. If nothing else, I can help route questions to
appropriate folks. =)
Reviewed-by: Jeremy Sequoia
> On Apr 18, 2017, at 02:34, Andres Gomez wrote:
>
> Jeremy, is it OK that I mention you as a reference for MacOSX in the
> releasing doc?
>
> On Mon, 2017-03-27 at 21:48 +0
https://bugs.freedesktop.org/show_bug.cgi?id=100668
--- Comment #2 from Brian Paul ---
Hi Charles,
I compiled your test program and ran it with Mesa in a Windows VM and
wglChoosePixelFormatARB failed to find a pixel format. But I found the exact
same failure when I ran it with NVIDIA's OpenGL d
For 1-3:
Reviewed-by: Marek Olšák
Marek
On Apr 18, 2017 3:35 PM, "Nicolai Hähnle" wrote:
> From: Nicolai Hähnle
>
> When any count[i] is negative, we must skip all draws.
>
> Moving to vbo makes the subsequent change easier.
>
> v2:
> - provide the function in all contexts, including GLES
>
We also have asynchronous compilation for optimized shaders (linking
separate shaders etc.) using the same threaded queue as the initial
compilations, and we can't let apps suppress the optimized ones.
Marek
On Apr 18, 2017 1:08 PM, "Nicolai Hähnle" wrote:
> On 17.04.2017 14:07, Edward O'Callag
On 11.04.2017 18:48, Samuel Pitoiset wrote:
The ARB_bindless_texture spec says:
"Modify Section 5.4.1, Conversion and Scalar Constructors, p. 60"
(add the following constructors:)
// In the following four constructors, the low 32 bits of the sampler
// type correspond to the
On 11.04.2017 18:48, Samuel Pitoiset wrote:
The ARB_bindless_texture spec says:
"If both bindless_sampler and bound_sampler, or bindless_image
and bound_image, are declared at global scope in any
compilation unit, a link- time error will be generated."
Signed-off-by: Samuel Pitoiset
On 11.04.2017 18:48, Samuel Pitoiset wrote:
Bindless sampler/image types are really different from the existing
sampler/image types. They are considered 64-bit unsigned integers,
they can be declared as temporary, shader inputs/outputs and are
non-opaque types.
For these reasons, it looks more c
On 18.04.2017 18:57, Marek Olšák wrote:
We also have asynchronous compilation for optimized shaders (linking
separate shaders etc.) using the same threaded queue as the initial
compilations, and we can't let apps suppress the optimized ones.
Good point. I think if the app says it doesn't want p
On 11.04.2017 18:48, Samuel Pitoiset wrote:
This also adds the extension to the standalone GLSL compiler.
Signed-off-by: Samuel Pitoiset
Patches 1 & 2:
Reviewed-by: Nicolai Hähnle
---
src/compiler/glsl/glsl_parser_extras.cpp | 1 +
src/compiler/glsl/glsl_parser_extras.h | 7 ++
On Fri, Apr 14, 2017 at 12:47 PM, Eric Anholt wrote:
> Android.mk was setting the flag across the entire driver, so we didn't
> have non-NEON versions getting built. This was going to be a problem with
> the next commit, when I start auto-detecting NEON support and use the
> non-NEON version when
Before this commit, when a group with count="0" is found, only one field
is added to the struct representing the instruction. This causes only
one entry to be printed by aubinator, for variable length groups.
With this commit we "detect" that there's a variable length group
(count="0") and store t
From: Emil Velikov
If VDPAU is installed in the non-default location, we'll fail to find
the headers and error at build time.
../../src/gallium/include/state_tracker/vdpau_dmabuf.h:37:25: fatal error:
vdpau/vdpau.h: No such file or directory
#include
^
Fixes: faba96b
Without it the compiler will barf at us with dozens of errors like
the following.
In file included from llvm/codegen/bitcode.cpp:34:0:
./llvm/codegen.hpp:52:22: error: ‘<::’ cannot begin a template-argument list
[-fpermissive]
./llvm/codegen.hpp:52:22: note: ‘<:’ is an alternate spelling for ‘[’.
Fixes: 1ac40173c2a ("configure.ac: simplify EGL requirements for drivers
dependent on EGL")
Signed-off-by: Emil Velikov
---
configure.ac | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index e1c60d842b5..957d15df8ca 100644
--- a/configure.ac
+++
From: Emil Velikov
Resolves build issues like the following:
src/gallium/winsys/sw/dri/dri_sw_winsys.c:203:31: error: pointer of type ‘void
*’ used in arithmetic [-Werror=pointer-arith]
data = dri_sw_dt->data + (dri_sw_dt->stride * box->y) + box->x * blsize;
On 04/18/2017 08:14 PM, Nicolai Hähnle wrote:
On 11.04.2017 18:48, Samuel Pitoiset wrote:
Bindless sampler/image types are really different from the existing
sampler/image types. They are considered 64-bit unsigned integers,
they can be declared as temporary, shader inputs/outputs and are
non-
On 04/18/2017 08:10 PM, Nicolai Hähnle wrote:
On 11.04.2017 18:48, Samuel Pitoiset wrote:
The ARB_bindless_texture spec says:
"Modify Section 5.4.1, Conversion and Scalar Constructors, p. 60"
(add the following constructors:)
// In the following four constructors, the low 32 bi
Emil Velikov writes:
> Without it the compiler will barf at us with dozens of errors like
> the following.
>
> In file included from llvm/codegen/bitcode.cpp:34:0:
> ./llvm/codegen.hpp:52:22: error: ‘<::’ cannot begin a template-argument list
> [-fpermissive]
> ./llvm/codegen.hpp:52:22: note: ‘<
On 19 April 2017 at 05:30, Samuel Pitoiset wrote:
>
>
> On 04/18/2017 08:14 PM, Nicolai Hähnle wrote:
>>
>> On 11.04.2017 18:48, Samuel Pitoiset wrote:
>>>
>>> Bindless sampler/image types are really different from the existing
>>> sampler/image types. They are considered 64-bit unsigned integers,
From: Dave Airlie
In practice this will probably just drop draw id in a few places.
v2: just do draw_id for now. (Bas)
it might be possible to do something more if we need it in the
future. (nha)
Signed-off-by: Dave Airlie
---
src/amd/common/ac_nir_to_llvm.c | 26 ++
On Tue, Apr 18, 2017 at 05:56:14PM +0100, Tvrtko Ursulin wrote:
> +enum drm_i915_gem_engine_class {
> + DRM_I915_ENGINE_CLASS_OTHER = 0,
> + DRM_I915_ENGINE_CLASS_RENDER = 1,
> + DRM_I915_ENGINE_CLASS_COPY = 2,
> + DRM_I915_ENGINE_CLASS_VIDEO_DECODE = 3,
> + DRM_I915_ENGINE_CLAS
On Tue, Apr 18, 2017 at 05:56:15PM +0100, Tvrtko Ursulin wrote:
> From: Tvrtko Ursulin
>
> Building on top of the previous patch which exported the concept
> of engine classes and instances, we can also use this instead of
> the current awkward engine selection uAPI.
>
> This is primarily intere
On 18.04.2017 21:49, Dave Airlie wrote:
On 19 April 2017 at 05:30, Samuel Pitoiset wrote:
On 04/18/2017 08:14 PM, Nicolai Hähnle wrote:
On 11.04.2017 18:48, Samuel Pitoiset wrote:
Bindless sampler/image types are really different from the existing
sampler/image types. They are considered
We need to emit BLEND_STATE, which size is 1 + 2 * nr_draw_buffers
dwords (on gen8+), but the BLEND_STATE struct length is always 17. By
marking it size 1, which is actually the size of the struct minus the
BLEND_STATE_ENTRY's, we can emit a BLEND_STATE of variable number of
entries.
For gen6 and
If the 'dwords' dict is empty, max(dwords.keys()) throws an exception.
This case could happen when we have an instruction that is only an array
of other structs, with variable length.
v2:
- Add another clause for empty dwords and make it work with python 3
(Dylan)
- Set the length to 0 if
'start' parameter from Group.emit_pack_function() is useless.
Signed-off-by: Rafael Antognolli
---
src/intel/genxml/gen_pack_header.py | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/intel/genxml/gen_pack_header.py
b/src/intel/genxml/gen_pack_header.py
index 2a70945.
On Tue, Apr 18, 2017 at 5:26 PM, Nicolai Hähnle wrote:
> On 18.04.2017 21:49, Dave Airlie wrote:
>>
>> On 19 April 2017 at 05:30, Samuel Pitoiset
>> wrote:
>>>
>>>
>>>
>>> On 04/18/2017 08:14 PM, Nicolai Hähnle wrote:
On 11.04.2017 18:48, Samuel Pitoiset wrote:
>
>
> Bi
Here is an early version of basic compute shader for a5xx. I've already
pushed a bunch of cleanup/prep work for this series. In the later driver
specific parts of this series I am still fixing a couple things. And
I completely lie about image support (required for ARB_compute_shader),
so it is h
This is equivalent to what mesa/st does in glsl_to_tgsi. For most hw
there isn't a particularly good reason to treat these differently.
Signed-off-by: Rob Clark
---
src/compiler/Makefile.sources| 1 +
src/compiler/nir/nir.h | 1 +
src/compiler/nir/nir_l
The generation-independent support for binding shader buffer objects.
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/freedreno_context.h | 8
src/gallium/drivers/freedreno/freedreno_draw.c | 6 +++
src/gallium/drivers/freedreno/freedreno_resource.c | 9
src/gallium/
Signed-off-by: Rob Clark
---
src/mesa/state_tracker/st_glsl_to_nir.cpp | 14 ++
src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 3 ++-
src/mesa/state_tracker/st_program.c| 13 +
src/mesa/state_tracker/st_program.h| 3 +++
4 files changed, 32 insertions(+),
TODO cwabbott pointed out a write-after-read hazzard, which effects both
this and arrays. A write needs to depend on *all* reads since the last
write, not just the last read.
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/ir3/ir3.c| 74 +++-
src/gallium/drivers/free
To simplify things for now, since all the gfx shader stages share a
single SSBO state block, only advertise SSBO support for fragment shader
(and compute when we have that). We could possibly use a fixed-
partitioning of the SSBO index space to support SSBOs on other stages
without having to resor
Sync to the same files from freedreno.git to correct decoding of ldgb/
stgb instructions.
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/a2xx/disasm-a2xx.c | 2 +-
src/gallium/drivers/freedreno/disasm.h | 3 +
src/gallium/drivers/freedreno/ir3/disasm-a3xx.c | 83
Hidden under FD_MESA_DEBUG=compute flag for now. In particular because
we completely lie about having image support.
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/Makefile.sources | 2 +
src/gallium/drivers/freedreno/a5xx/fd5_compute.c | 175 +++
src/gallium
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/freedreno_context.h | 14 +++-
src/gallium/drivers/freedreno/freedreno_draw.c| 47
src/gallium/drivers/freedreno/freedreno_gmem.c| 7 ++
src/gallium/drivers/freedreno/freedreno_gmem.h| 1 +
src/gallium/drivers/f
Signed-off-by: Rob Clark
---
src/gallium/drivers/freedreno/disasm.h | 1 +
.../drivers/freedreno/ir3/ir3_compiler_nir.c | 81 +-
src/gallium/drivers/freedreno/ir3/ir3_shader.c | 76 +++-
src/gallium/drivers/freedreno/ir3/ir3_shader.h
2017-04-18 8:56 GMT+02:00 Andres Gomez :
> Mauro, is it OK that I mention you as a reference for Android in the
> releasing doc?
>
Looking at my, somehow limited, coding skills, I think that my role would
be defined better
as an active android-x86 debugging contributor/build tester, with
experien
Thanks for making the changes I requested.
Reviewed-by: Dylan Baker
Quoting Rafael Antognolli (2017-04-18 14:50:20)
> If the 'dwords' dict is empty, max(dwords.keys()) throws an exception.
> This case could happen when we have an instruction that is only an array
> of other structs, with variabl
On 18/04/17 20:48, Nicolai Hähnle wrote:
On 18.04.2017 02:09, Timothy Arceri wrote:
V2: some bug fixes and suggestions addressed as suggested by Brian and
Nicolai.
This series adds some initial support for the KHR_no_error.
For now it can only be enabled with the MESA_NO_ERROR environmen
From: Dave Airlie
This removes the vertex buffer, and just generates the values
in the shader.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_decompress.c | 75 +++
1 file changed, 6 insertions(+), 69 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_
From: Dave Airlie
Avoids having to setup vertex buffers.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_fast_clear.c | 74 +++
1 file changed, 6 insertions(+), 68 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_fast_clear.c
b/src/amd/vulkan/radv_me
From: Dave Airlie
Some of the shaders could just generate the vertex data in the
shader, so add helpers to allow us to move to doing that.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta.c | 39 +++
src/amd/vulkan/radv_meta.h | 7 +++
2 files ch
From: Dave Airlie
Instead of passing in the same 1.0, -1.0 combinations via
vertex buffers, we can just use vertex id to have the vertex
shader build them. This function introduces the generator
code needed, later patches will use this.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta.c
From: Dave Airlie
This generates the position info using the vertex shader.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_blit.c | 67 ++---
1 file changed, 16 insertions(+), 51 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_blit.c b/src/amd/v
From: Dave Airlie
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_resolve.c | 74 --
1 file changed, 7 insertions(+), 67 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_resolve.c
b/src/amd/vulkan/radv_meta_resolve.c
index 52f7246..70dcc30 100644
From: Dave Airlie
For depth clears we have to pass the depth in the 2nd
component, we can use push constants for some of this
later to drop the vertex buffer completely
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta.c | 9 --
src/amd/vulkan/radv_meta.h | 2 +-
src/am
From: Dave Airlie
Generate the position vertices.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_blit2d.c | 35 +++
1 file changed, 7 insertions(+), 28 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_blit2d.c
b/src/amd/vulkan/radv_meta_blit2d.c
ind
From: Dave Airlie
The vs vertex generate and fs noop shaders are used in a few places,
so refactor them out.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta.c| 35 ++
src/amd/vulkan/radv_meta.h| 2 ++
src/amd/vulkan/radv_meta_decompr
v2: Handle out of pool memory error.
Signed-off-by: Bas Nieuwenhuizen
---
src/amd/vulkan/radv_descriptor_set.c | 69 +++-
src/amd/vulkan/radv_private.h| 4 +++
2 files changed, 57 insertions(+), 16 deletions(-)
diff --git a/src/amd/vulkan/radv_descriptor
Signed-off-by: Bas Nieuwenhuizen
---
src/amd/vulkan/radv_descriptor_set.c | 16
1 file changed, 4 insertions(+), 12 deletions(-)
diff --git a/src/amd/vulkan/radv_descriptor_set.c
b/src/amd/vulkan/radv_descriptor_set.c
index 48cb8c2a30a..26f89152d52 100644
--- a/src/amd/vulkan/r
In cases where it is used it is always 1.
Signed-off-by: Bas Nieuwenhuizen
---
src/amd/vulkan/radv_descriptor_set.c | 21 ++---
src/amd/vulkan/radv_descriptor_set.h | 1 -
src/amd/vulkan/radv_private.h| 1 -
3 files changed, 10 insertions(+), 13 deletions(-)
diff --git
With this we don't have any operations on a pool with non-freeable
descriptors left that have O(#descriptors) complexity.
Signed-off-by: Bas Nieuwenhuizen
---
src/amd/vulkan/radv_descriptor_set.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/amd/vulkan/radv_descrip
Signed-off-by: Nanley Chery
---
src/intel/isl/isl.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/intel/isl/isl.c b/src/intel/isl/isl.c
index f89f351c15..09dafcbab6 100644
--- a/src/intel/isl/isl.c
+++ b/src/intel/isl/isl.c
@@ -1573,8 +1573,11 @@ isl_surf_get_ccs_sur
The section of the PRM mentioned in the code comment above this table
says that this format supports the render target write message. Internal
documentation says that this format also supports alpha blending. As a
side effect, this allows CCS_D buffers to be created for images with
this format.
Si
According to the Linux kernel sources, the ioctl in anv_gem_mmap() will
not fail if the size isn't a multiple of 4KB.
Signed-off-by: Nanley Chery
---
src/intel/vulkan/anv_image.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/src/intel/vulkan/anv_image.c b/src/intel/vul
1 - 100 of 143 matches
Mail list logo
