We are reading out some of the parameters from IR data structure those
have been written previously, on some platforms L3 is not coherent, so
explicitly add those flushes.
Cc: mesa-stable
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36952>
(cherry picked from commit 2cd564c1de)
There is a FW issue when using constrained intra prediction with rate
control enabled, causing unexpected quality degradation.
Disable it until FW fix is available.
Cc: mesa-stable
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36837>
(cherry picked from commit 51d5c0d537)
Images aren't always coherent with L2 and AMD generations have
different rules, see radv_image_is_l2_coherent() for the full picture.
This fixes a rendering issue on GFX9 because depth/stencil images
aren't coherent, but this also affects color images.
This also fixes a cache coherency issue with an ongoing extension.
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12274
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36815>
(cherry picked from commit 99b287bde5)
f2f640f35 introduces base variants, so we get 2 NIR shaders:
glsl-to-nir -> base_serialized_nir
-> serialized_nir
Then, depending on who is using the shader the right one would be picked:
* draw uses base_serialized_nir
* hw driver uses serialized_nir
ef0c9231a7 made sure that base wasn't used when the shader is loaded from
the cache because it's missing, so in this case, glsl-to-nir does:
glsl-to-nir -> from-cache -> serialized_nir
The problem is that if draw tries to use this shader it may fail, for
the same reason as the referenced commits were introduced: draw may not
be compatible with some NIR passes used by the driver.
To fix this we need to serialize both NIR shaders, and pick the right
one depending on the user.
Fixes: ef0c9231a7 ("mesa/st: don't use serialized_nir for cached shaders")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36411>
(cherry picked from commit 0e3ec9e82c)
Multiple contexts can use those causing deadlocks if e.g. fence_get_fd
gets called before fence_server_signal on another thread on the same
pipe_fence_handle.
Cc: mesa-stable
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36819>
(cherry picked from commit d9c3bbb08c)
This is an issue found in testing multiple mediafoundation MFT
concurrently. Thanks to Jesse for the fix.
cc: mesa-stable
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36923>
(cherry picked from commit e76e95e084)
An AHB with IMPLEMENTATION_DEFINED format is commonly backed by NV12 or
XBGR8888. The former is the usual pick for camera <-> GPU interop, while
the latter is mostly only seen in Android CTS. Ideally, we can rely on
the queried fourcc to resolve everything instead of being on the
fallback path, but keeping this a minimal fix is easy for porting.
Cc: mesa-stable
Reviewed-by: Lucas Fryzek <lfryzek@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36866>
(cherry picked from commit 25a8e124e0)
After the commit mentioned below the `extra` variable only holds the
last added extra sampler. For 3-plane formats this results in one extra
sampler being leaked.
Add the bits for each extra sampler instead.
Fixes: abcd02a07d (gallium: Properly handle non-contiguous used sampler view indexes)
Signed-off-by: Robert Mader <robert.mader@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Matthias Reichl <hias@horus.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36870>
(cherry picked from commit 25fe82630a)
Descriptor prefetches may be generated for instructions in control flow.
This means we cannot simply emit prefetches at the end of the preamble
because that may not be dominated by all their sources. This commit uses
the helpers introduced by e7ac1094f6 ("ir3: rematerialize preamble defs
in block dominated by sources") to find the correct block to insert
prefetches.
Fixes NIR validation errors in Dying Light 2.
Signed-off-by: Job Noorman <jnoorman@igalia.com>
Fixes: 4e2a0a5ad0 ("ir3: Add descriptor prefetching optimization on a7xx")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36885>
(cherry picked from commit 24cdb0b636)
This might actually fixes a couple of things because needed dynamic
states are computed before radv_emit_graphics_pipeline(), so dirtying
them too late doesn't make much sense.
This doesn't fix anything known.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36900>
(cherry picked from commit d40e841cc4)
We were not implementing vkGetImageOpaqueCaptureDescriptorDataEXT,
relying on the common implementation that does nothing. That works well
enough for regular images because the fixed address needed for
capture/replay is handled by the memory allocation path, but for sparse
images we initialize the sparse bindings at image creation time.
Here we implement the function to retrieve the addresses of all the
used bindings for the image, then use all of them at creation time.
Also, set the correct alloc_flags for this to work.
Fixes: 43b57ee8a5 ("anv: add capture/replay support for image with descriptor buffers")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35872>
(cherry picked from commit 20f546d6c1)
In some setups the rust compiler emits errors like the following:
-----------------------------------------------------------------------
error: hiding a lifetime that's elided elsewhere is confusing
--> ../subprojects/proc-macro2-1.0.86/src/parse.rs:125:25
|
125 | fn block_comment(input: Cursor) -> PResult<&str> {
| ^^^^^^ -------------
| | | |
| | | the same lifetime is elided here
| | the same lifetime is hidden here
| the lifetime is hidden here
|
= help: the same lifetime is referred to in inconsistent ways, making the signature confusing
= note: `-D mismatched-lifetime-syntaxes` implied by `-D warnings`
= help: to override `-D warnings` add `#[allow(mismatched_lifetime_syntaxes)]`
help: use `'_` for type paths
|
125 | fn block_comment(input: Cursor<'_>) -> PResult<'_, &str> {
| ++++ +++
-----------------------------------------------------------------------
Follow the solution suggested by the compiler to silence the errors, for
all the observed occurrences.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Mel Henning <mhenning@darkrefraction.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36849>
(cherry picked from commit 8f84ae7de7)
In case of emulated ASTC on supported platforms, currently returning
0 for linear tiled images causes vpGetPhysicalDeviceProfileSupport
failure during AndroidBaselineProfile test. The patch handles it
similar to linearly-tiled images that are used for transfers.
Fixes android.graphics.cts.VulkanFeaturesTest#testAndroidBaselineProfile2021Support.
Cc: mesa-stable
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36798>
(cherry picked from commit f67edacf8b)
this avoids a scenario where a non-subdata UNSYNCHRONIZED unmap triggers through
tc at the same time the frontend calls an UNSYNCHRONIZED subdata call
in the main thread, which desynchronizes the cmdbuf and hits an assert
Fixes: 8ee0d6dd71 ("zink: add a third cmdbuf for unsynchronized (not reordered) ops")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36846>
(cherry picked from commit cf5d41575b)
If vkCmdSetRenderingAttachmentLocations() isn't setting all color
attachment locations (ie. MAX_RTS), the remapping might be wrong
because MESA_VK_ATTACHMENT_UNUSED is used to trim the unused locations
Found by inspection while implementing a new extension.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36778>
(cherry picked from commit 45c91edd18)
This change fixes the gds implementation of
atomic_counter_comp_swap which requires three arguments.
This update is based on 4e3b43f180 "r600/atomic: fix
ATOMCAS instruction." which was the tgsi implementation.
Note: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36554
is required for this change to work properly on cayman.
This change was tested on palm, cypress and barts. Here is the test fixed:
khr-gl4[5-6]/shader_atomic_counter_ops_tests/shaderatomiccounteropsexchangetestcase: fail pass
Cc: mesa-stable
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36254>
(cherry picked from commit 521b848ea8)
Otherwise the SPIR-V parser prints a warning the first time the driver
is loaded after a fresh compile.
Fixes: 91b62e9868 ("anv: Use spirv_capabilities for the float64 shader")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36783>
(cherry picked from commit 94931fd4f4)
The evauation of loading the AR register was off by one, so that
splitting an ALU group could actually happen after AR was loaded
resulting in a failure to lower to assembly.
Fixes: d617052db6 ("r600/sfn: take address loads into account when scheduling")
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36742>
(cherry picked from commit 5d0f212d81)
Without this update a very long ALU block may not be splitted as
required and lowering to assembly may fail because the maximum
supported length of a ALU CF is overrun.
Fixes: 6aafa2bb49 ("r600/sfn: Split ALU blocks in scheduler to fit into 128 slots")
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36742>
(cherry picked from commit 63c801e8c2)
Calling vkCmdBindDescriptorBuffersEXT() does not invalidate previously
set descriptor sets. Move the state dirtying to
vkCmdSetDescriptorBufferOffets.
Fixes: ab7641b8dc ("anv: implement descriptor buffer binding")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36768>
(cherry picked from commit 35190aac91)
If an application switches back and forth between descriptor sets and
descriptor buffers before executing a draw/dispatch, we could end up in
a wrong state due to pending_db_mode not getting updated.
Fixes: ab7641b8dc ("anv: implement descriptor buffer binding")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36768>
(cherry picked from commit 382b8e3aa2)
layout->push_constant_mask is only the DGC push constant mask (ie. the
tokens that are specified), but with IES all push constants are emitted
from the DGC shader. So it should be the total range of push constant.
This used to work by luck due to the preprocess buffer alignment.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36753>
(cherry picked from commit 3359386145)
