On October 13, 2018 19:50:00 "Keith Packard" wrote:
Jason Ekstrand writes:
Actually, I think anv is doing the right thing too. Now I have no idea why
Keith was having problems.
anv is happily returning a valid pointer and radv is not?
In any case, I've switched to using vkGetInstanceProc
Jason Ekstrand writes:
> Actually, I think anv is doing the right thing too. Now I have no idea why
> Keith was having problems.
anv is happily returning a valid pointer and radv is not?
In any case, I've switched to using vkGetInstanceProcAddr for the
VkPhysicalDevice function and it works fi
https://bugs.freedesktop.org/show_bug.cgi?id=108263
Hi-Angel changed:
What|Removed |Added
CC||hi-an...@yandex.ru
--
You are receiving thi
There's no reason why we need generate trampoline functions for instance
functions or carry N copies of the instance dispatch table around for
every hardware generation. Splitting the tables and being more
conservative shaves about 34K off .text and about 4K off .data when
built with clang.
Befor
I ran the Vulkan CTS both with and without this patch in the hopes that the
massive pile of redundant shaders in the CTS would be helped. It shaved 6
minutes of a 4.5 hour run. Quite possibly in the noise.
--Jason
On Fri, Oct 12, 2018 at 7:08 PM Jason Ekstrand wrote:
> This patch series adds
On Sat, Oct 13, 2018 at 11:27 AM Jason Ekstrand
wrote:
> On Sat, Oct 13, 2018 at 11:24 AM Bas Nieuwenhuizen <
> b...@basnieuwenhuizen.nl> wrote:
>
>> On Sat, Oct 13, 2018 at 6:12 PM Jason Ekstrand
>> wrote:
>> >
>> > On Sat, Oct 13, 2018 at 10:58 AM Bas Nieuwenhuizen <
>> b...@basnieuwenhuizen.n
---
src/amd/vulkan/radv_device.c | 10 ++
src/amd/vulkan/radv_extensions.py | 1 +
src/amd/vulkan/radv_private.h | 2 ++
3 files changed, 13 insertions(+)
diff --git a/src/amd/vulkan/radv_device.c b/src/amd/vulkan/radv_device.c
index 88786c999de..c418176978e 100644
--- a/src/am
On Sat, Oct 13, 2018 at 11:24 AM Bas Nieuwenhuizen
wrote:
> On Sat, Oct 13, 2018 at 6:12 PM Jason Ekstrand
> wrote:
> >
> > On Sat, Oct 13, 2018 at 10:58 AM Bas Nieuwenhuizen <
> b...@basnieuwenhuizen.nl> wrote:
> >>
> >> On Fri, Oct 12, 2018 at 10:38 PM Keith Packard
> wrote:
> >> >
> >> > Acc
On Sat, Oct 13, 2018 at 6:12 PM Jason Ekstrand wrote:
>
> On Sat, Oct 13, 2018 at 10:58 AM Bas Nieuwenhuizen
> wrote:
>>
>> On Fri, Oct 12, 2018 at 10:38 PM Keith Packard wrote:
>> >
>> > According to the Vulkan spec:
>> >
>> > "Vulkan 1.0 initially required all new physical-device-level extens
On Sat, Oct 13, 2018 at 10:58 AM Bas Nieuwenhuizen
wrote:
> On Fri, Oct 12, 2018 at 10:38 PM Keith Packard wrote:
> >
> > According to the Vulkan spec:
> >
> > "Vulkan 1.0 initially required all new physical-device-level extension
> > functionality to be structured within an instance extension.
On Fri, Oct 12, 2018 at 10:38 PM Keith Packard wrote:
>
> According to the Vulkan spec:
>
> "Vulkan 1.0 initially required all new physical-device-level extension
> functionality to be structured within an instance extension. In order
> to avoid using an instance extension, which often requires
https://bugs.freedesktop.org/show_bug.cgi?id=108353
--- Comment #1 from Axel Davy ---
You can use the drirc configuration file to force applications to automatically
use the AMD card.
udevadm info /dev/dri/card0 | grep ID_PATH_TAG
udevadm info /dev/dri/card1 | grep ID_PATH_TAG
This will give yo
https://bugs.freedesktop.org/show_bug.cgi?id=108353
Bug ID: 108353
Summary: Request: Control Center for AMD GPU
Product: Mesa
Version: unspecified
Hardware: x86-64 (AMD64)
OS: Linux (All)
Status: NEW
Sever
These were missing in the last couple of spec updates.
---
include/vulkan/vulkan.h | 4 +++
include/vulkan/vulkan_fuchsia.h | 58 +
2 files changed, 62 insertions(+)
create mode 100644 include/vulkan/vulkan_fuchsia.h
diff --git a/include/vulkan/vulkan.h b
Well, you beat me to it.
Reviewed-by: Bas Nieuwenhuizen
for patch 2-4 (1 is acked with comments).
On Sat, Oct 13, 2018 at 3:49 PM Jason Ekstrand wrote:
>
> This patch exposes support for the following two extensions:
>
> * VK_GOOGLE_decorate_string
> * VK_GOOGLE_hlsl_functionality1
>
> There
I don't think we actually need this? The SPV extensions are a bit older already.
Then again,
Acked-by: Bas Nieuwenhuizen
On Sat, Oct 13, 2018 at 3:49 PM Jason Ekstrand wrote:
>
> This corresponds to commit 801cca8104245c07e8cc532 on GitHub.
> ---
> src/compiler/spirv/spirv.core.grammar.json |
https://bugs.freedesktop.org/show_bug.cgi?id=107971
--- Comment #3 from Jason Ekstrand ---
Patches are on the list. Mind giving them a try?
https://patchwork.freedesktop.org/series/50971/
--
You are receiving this mail because:
You are the assignee for the bug.
This corresponds to commit 801cca8104245c07e8cc532 on GitHub.
---
src/compiler/spirv/spirv.core.grammar.json | 533 -
src/compiler/spirv/spirv.h | 84
2 files changed, 604 insertions(+), 13 deletions(-)
diff --git a/src/compiler/spirv/spirv.core.grammar.j
---
src/compiler/spirv/spirv_to_nir.c | 8
1 file changed, 8 insertions(+)
diff --git a/src/compiler/spirv/spirv_to_nir.c
b/src/compiler/spirv/spirv_to_nir.c
index 2ad83196e46..7c501f3e71d 100644
--- a/src/compiler/spirv/spirv_to_nir.c
+++ b/src/compiler/spirv/spirv_to_nir.c
@@ -495,15
This extension adds two new decorations which carry meaning only for
HLSL shaders. They are expected to be handled by higher level layers
and can be ignored by implementations. However, it does save the client
a bit of work if the implementation safely ignores them instead of the
client having to
This patch exposes support for the following two extensions:
* VK_GOOGLE_decorate_string
* VK_GOOGLE_hlsl_functionality1
There's nothing for the driver to do; it's all handled in spirv_to_nir.
---
src/amd/vulkan/radv_extensions.py | 2 ++
src/intel/vulkan/anv_extensions.py | 2 ++
2 files cha
This is trivial for us to implement because Intel integrated GPUs have
always been at :00:02.0. One of these days, when we actually ship a
discrete GPU, we'll have to implement it for real but right now we can
just hard code it.
---
src/intel/vulkan/anv_device.c | 11 +++
src/int
---
src/intel/vulkan/genX_query.c | 21 -
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/src/intel/vulkan/genX_query.c b/src/intel/vulkan/genX_query.c
index 7533ec05095..4fffa6e7221 100644
--- a/src/intel/vulkan/genX_query.c
+++ b/src/intel/vulkan/genX_query.c
@@
---
src/intel/vulkan/anv_private.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/intel/vulkan/anv_private.h b/src/intel/vulkan/anv_private.h
index b897d679533..f30ffb36179 100644
--- a/src/intel/vulkan/anv_private.h
+++ b/src/intel/vulkan/anv_private.h
@@ -1681,7 +1681,
---
include/vulkan/vulkan.h | 4 +
include/vulkan/vulkan_core.h| 244 +-
include/vulkan/vulkan_fuchsia.h | 58
src/vulkan/registry/vk.xml | 253
4 files changed, 524 insertions(+), 35 deletions(-)
create mod
---
src/intel/vulkan/anv_extensions.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/intel/vulkan/anv_extensions.py
b/src/intel/vulkan/anv_extensions.py
index d4915c95013..4d089adfaa2 100644
--- a/src/intel/vulkan/anv_extensions.py
+++ b/src/intel/vulkan/anv_extensions.py
@@ -123,6 +123
This series adds support for the new VK_EXT_transform_feedback extension.
As far as I know, everything works but it's still all a bit experimental as
we don't have very good tests yet. CTS tests are in-progress and we hope
to have decent testing soon. We likely won't be landing them in master
unt
---
src/compiler/nir/nir_lower_io_to_scalar.c | 8
1 file changed, 8 insertions(+)
diff --git a/src/compiler/nir/nir_lower_io_to_scalar.c
b/src/compiler/nir/nir_lower_io_to_scalar.c
index f0c2a6a95d6..a7373d0e791 100644
--- a/src/compiler/nir/nir_lower_io_to_scalar.c
+++ b/src/compiler/
Annoyingly, this requires that we implement integer division on the
command streamer. Fortunately, we're only ever dividing by constants so
we can use the mulh+add+shift trick and it's not as bad as it sounds.
---
src/intel/vulkan/anv_device.c | 2 +-
src/intel/vulkan/genX_cmd_buffer.c | 1
---
src/intel/vulkan/anv_cmd_buffer.c | 29 +++
src/intel/vulkan/anv_device.c | 24 ++
src/intel/vulkan/anv_extensions.py | 2 +-
src/intel/vulkan/anv_pipeline.c| 10 ++-
src/intel/vulkan/anv_private.h | 13 +++
src/intel/vulkan/genX_cmd_buffer.c | 125 +
---
src/intel/genxml/gen10.xml | 32
src/intel/genxml/gen11.xml | 32
src/intel/genxml/gen7.xml | 32
src/intel/genxml/gen75.xml | 32
src/intel/genxml/gen8.xml | 3
---
src/intel/vulkan/anv_device.c| 2 +-
src/intel/vulkan/genX_pipeline.c | 1 +
src/intel/vulkan/genX_query.c| 72 +++-
3 files changed, 73 insertions(+), 2 deletions(-)
diff --git a/src/intel/vulkan/anv_device.c b/src/intel/vulkan/anv_device.c
index acca97b
---
src/compiler/nir/nir_xfb_info.h | 2 +-
src/intel/vulkan/anv_blorp.c | 3 +-
src/intel/vulkan/anv_pipeline.c | 4 +--
src/intel/vulkan/anv_pipeline_cache.c | 48 +++
src/intel/vulkan/anv_private.h| 6
5 files changed, 53 insertions(
This is different from the GL_ARB_spirv pass because it generates a much
simpler data structure that isn't tied to OpenGL and mtypes.h.
---
src/compiler/Makefile.sources | 4 +-
src/compiler/nir/meson.build | 2 +
src/compiler/nir/nir_gather_xfb_info.c | 150
From: Samuel Pitoiset
If we have a transform feedback output like:
float[2] x2_out (VARYING_SLOT_VAR1.x, 0, 0)
which is lowered by nir_lower_io_arrays_to_elements to,
float x2_out (VARYING_SLOT_VAR1.x, 0, 0)
float x2_out@5 (VARYING_SLOT_VAR2.x, 0, 0)
We have to update the destination offset t
From: Samuel Pitoiset
When a xfb buffer is explicitely declared on a varying
variable, we shouldn't remove it at link time.
Signed-off-by: Samuel Pitoiset
---
src/compiler/nir/nir_linking_helpers.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/compiler/nir/nir_linking_helpers.c
b
Signed-off-by: Samuel Pitoiset
---
src/amd/common/sid.h | 1 +
src/amd/vulkan/radv_cmd_buffer.c | 385 +-
src/amd/vulkan/radv_device.c | 22 ++
src/amd/vulkan/radv_extensions.py | 1 +
src/amd/vulkan/radv_pipeline.c| 19 ++
src/amd/vulkan/r
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_private.h | 3 +++
src/amd/vulkan/radv_shader.h | 18 ++
src/amd/vulkan/radv_shader_info.c | 39 +++
3 files changed, 60 insertions(+)
diff --git a/src/amd/vulkan/radv_private.h b/src/amd/vul
From: Jason Ekstrand
This is different from the GL_ARB_spirv pass because it generates a much
simpler data structure that isn't tied to OpenGL and mtypes.h.
---
src/compiler/Makefile.sources | 4 +-
src/compiler/nir/meson.build | 2 +
src/compiler/nir/nir_gather_xfb_info.c
This is required for GS multiple streams support.
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_nir_to_llvm.c | 22 ++
1 file changed, 14 insertions(+), 8 deletions(-)
diff --git a/src/amd/vulkan/radv_nir_to_llvm.c
b/src/amd/vulkan/radv_nir_to_llvm.c
index c8b8670f
For multiple streams support we have to set the different ring
buffer sizes correctly. This relies on the number of output
components per stream.
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_pipeline.c | 25 +++--
1 file changed, 19 insertions(+), 6 deletions(-)
di
This will be used for splitting the GS->VS ring buffer. The
stream ID is always 0 for now.
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_shader.h | 1 +
src/amd/vulkan/radv_shader_info.c | 14 ++
2 files changed, 15 insertions(+)
diff --git a/src/amd/vulkan/radv_shade
This will be also used for splitting the GS->VS ring buffer.
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_shader.h | 1 +
src/amd/vulkan/radv_shader_info.c | 2 ++
2 files changed, 3 insertions(+)
diff --git a/src/amd/vulkan/radv_shader.h b/src/amd/vulkan/radv_shader.h
index f02c
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_nir_to_llvm.c | 102 ++
1 file changed, 76 insertions(+), 26 deletions(-)
diff --git a/src/amd/vulkan/radv_nir_to_llvm.c
b/src/amd/vulkan/radv_nir_to_llvm.c
index 46c96dfac0..5ea1755b2a 100644
--- a/src/amd/vulka
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_nir_to_llvm.c | 137 ++
1 file changed, 137 insertions(+)
diff --git a/src/amd/vulkan/radv_nir_to_llvm.c
b/src/amd/vulkan/radv_nir_to_llvm.c
index 0ebd3bbd34..46c96dfac0 100644
--- a/src/amd/vulkan/radv_nir_to_ll
For all streams. We basically just need to update the
base address and compute a stride for every stream.
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_nir_to_llvm.c | 78 ++-
1 file changed, 57 insertions(+), 21 deletions(-)
diff --git a/src/amd/vulkan/radv
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_nir_to_llvm.c | 55 ++-
src/amd/vulkan/radv_shader.h | 3 +-
2 files changed, 56 insertions(+), 2 deletions(-)
diff --git a/src/amd/vulkan/radv_nir_to_llvm.c
b/src/amd/vulkan/radv_nir_to_llvm.c
index 07dc6a
Same as the previous patch, except that is only the number of
components.
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_pipeline.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/src/amd/vulkan/radv_pipeline.c b/src/amd/vulkan/radv_pipeline.c
index 338d58f9d
Hi,
This series implements VK_EXT_transform_feedback for RADV. We tested it
quite a lot with DXVK and also with RenderDoc, it should be stable enough.
Please review, thanks!
Jason Ekstrand (1):
nir: Add a pass for gathering transform feedback info
Samuel Pitoiset (14):
vulkan: Update the XM
We shouldn't try to compact any varyings known as always
active IO, especially XFB outputs. For example, if one
component of an xfb output is also used as input varying
in the next stage, it shouldn't be compacted.
Because we look at the input varyings from the consumer
stage, we don't know if one
To only emit outputs for the given stream.
Signed-off-by: Samuel Pitoiset
---
src/amd/vulkan/radv_shader.h | 1 +
src/amd/vulkan/radv_shader_info.c | 2 ++
2 files changed, 3 insertions(+)
diff --git a/src/amd/vulkan/radv_shader.h b/src/amd/vulkan/radv_shader.h
index cf7bb580b5..dc6c48d306
Signed-off-by: Samuel Pitoiset
---
include/vulkan/vulkan_core.h | 244 -
src/vulkan/registry/vk.xml | 253 ++-
2 files changed, 462 insertions(+), 35 deletions(-)
diff --git a/include/vulkan/vulkan_core.h b/include/vulkan/vulkan_c
https://bugs.freedesktop.org/show_bug.cgi?id=35268
Sora Lee changed:
What|Removed |Added
Attachment #142010|0 |1
is obsolete|
https://bugs.freedesktop.org/show_bug.cgi?id=35268
--- Comment #25 from Sora Lee ---
Created attachment 142011
--> https://bugs.freedesktop.org/attachment.cgi?id=142011&action=edit
dlopen_test.c
--
You are receiving this mail because:
You are the assignee for the bug._
https://bugs.freedesktop.org/show_bug.cgi?id=35268
Sora Lee changed:
What|Removed |Added
Attachment #44777|0 |1
is obsolete|
https://bugs.freedesktop.org/show_bug.cgi?id=35268
--- Comment #23 from Sora Lee ---
(In reply to Rich Felker from comment #22)
> Ping.
>
> Looking at this again, I see the original proposed patch introduced a lot of
> extra asm to save/restore registers that might be clobbered by
> __tls_get_ad
56 matches
Mail list logo
