Jonathan
897e382d9d
gallium/u_threaded: buffer sharedness tracking
...
This fixes TC's buffer invalidation code for buffers that are shared
between contexts.
TC is unable to notify other TCs in case it replaces a shared buffer's
underlying storage when invalidating, causing those other TCs to use
wrong buffer IDs for busyness tracking, which leads to corruption due
to invalidation fast-paths being triggered when they shouldn't be.
This patch addresses this issue by tracking if a buffer is shared, and
if it is, disabling buffer storage replacement for the affected buffer.
This is achieved by tracking which TC instance first accessed a certain
buffer. If a second instance then accesses it as well, it will realize
that it isn't the only one working on the buffer and mark the buffer
accordingly.
If TC needs to invalidate a buffer for the correctness of an operation
at any point, it will fall back to doing the operation in a synchronous
fashion with this patch if the buffer is shared and currently busy.
It might be possible to later detect that a buffer has become un-shared;
however, this is outside of the scope of this bugfix patch.
v2: Do not disable buffer busyness tracking for shared buffers.
Fixes: e9c41b32marek.olsak@amd.com >
(cherry picked from commit 8f159a8576https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18349 >
2022-09-28 19:03:32 +00:00
Jonathan
8a57b219d1
gallium/u_threaded: Prepare for buffer sharedness tracking
...
tc_add_to_buffer_list and tc_bind_buffer need the tc instance too.
(cherry picked from commit aa87803069https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18349 >
2022-09-28 19:03:32 +00:00
Jesse Natalie
6b77710d2d
u_atomic: Add a helper for pointer compare-exchange
...
Notably this helps with MSVC, which complains about compiling the not-taken
branches of the ternary when called with pointer args. Using a version that
doesn't have "runtime" sizeof checks eliminates the warnings.
Reviewed-by: Jonathan Strobl <jonathan.strobl@gmx.de >
(cherry picked from commit 6718bff75bhttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18349 >
2022-09-28 19:03:32 +00:00
Feng Jiang
5383fd12d2
virgl/vtest: fix memory overwrite problem in virgl_vtest_send_get_caps()
...
Signed-off-by: Feng Jiang <jiangfeng@kylinos.cn >
(cherry picked from commit 060936fe0dhttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18071 >
2022-09-28 18:49:43 +00:00
Mike Blumenkrantz
94c771bcdb
zink: set layouts before possibly reordering image copies
...
layout-setting may change which cmdbuf can be used
Fixes: 731d7be375ajax@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18727 >
(cherry picked from commit 3ed712f748
2022-09-28 10:14:58 -07:00
Mike Blumenkrantz
63a60728ad
aux/trace: dump depth_clamp for rasterizer state
...
cc: mesa-stable
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com >
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18783 >
(cherry picked from commit 4c58b92e44
2022-09-28 10:14:58 -07:00
Pierre-Eric Pelloux-Prayer
a4222b0881
tc: don't use CPU storage for glBufferData
...
This fixes a performance regression with yquake2 cause by
the enablement of cpu_storage by default for radeonsi in
a5a8e19741a5a8e19741robdclark@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18774 >
(cherry picked from commit e06c332bf7
2022-09-28 10:14:58 -07:00
Pierre-Eric Pelloux-Prayer
9372399842
tc: do a GPU->CPU copy to initialize cpu_storage
...
If the GPU-side storage has been written to without using cpu_storage,
then we have to initialize the CPU-side storage correctly.
This requires a sync + copy but it's a one time operation so it shouldn't
affect performance much.
I don't think it fixes any existing bug, but the next commit will need
this to behave correctly.
cc: mesa-stable
Reviewed-by: Rob Clark <robdclark@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18774 >
(cherry picked from commit 8af8dc97bc
2022-09-28 10:14:58 -07:00
Lionel Landwerlin
b4c91f086a
intel/nir: disable assert on async stack id
...
This can be accessed from :
- RT shaders
- CS trampoline shader
We missed the second part here.
Fixes: 0465714790lionel.g.landwerlin@intel.com >
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16970 >
(cherry picked from commit 57593c5395
2022-09-28 10:14:58 -07:00
Lionel Landwerlin
63fc489471
intel/nir: fix potential invalid function impl ptr usage
...
We keep the nir_builder::impl value around, but we've run some passes
that might have change the main function.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: 96fde5518bcaio.oliveira@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16970 >
(cherry picked from commit 8d580de4a9
2022-09-28 10:14:58 -07:00
Lionel Landwerlin
ce21dde04d
intel/nir: fixup preserved metadata in rayquery lowering
...
Another case of not clearing the metadata correctly.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: c78be5da30caio.oliveira@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16970 >
(cherry picked from commit 1ffd28149f
2022-09-28 10:14:58 -07:00
Mike Blumenkrantz
1ab6d664c3
mesa: set normalized_coords for bindless texture buffer samplers
...
this isn't "used", but it's the default
cc: mesa-stable
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18845 >
(cherry picked from commit 7829fb9adf
2022-09-28 10:14:58 -07:00
Mike Blumenkrantz
d537057606
zink: don't flatten 64bit arrays during rewrite
...
dunno what I was thinking here
Fixes: 5b2f850425soroush.kashani@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18728 >
(cherry picked from commit 17e3df4cd2
2022-09-28 10:14:58 -07:00
Mike Blumenkrantz
70e2817f65
zink: always run optimize_nir after lower_64bit_vars
...
it's otherwise possible (and likely) that optimizations won't
happen since there's no shader key data active
Fixes: 5b2f850425soroush.kashani@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18728 >
(cherry picked from commit 0f053f0f07
2022-09-28 10:14:58 -07:00
Dylan Baker
358e054bfd
.pick_status.json: Update to 3ed712f748
2022-09-28 10:14:58 -07:00
Yonggang Luo
24af5a7f82
util: Disable usage of __attribute__((__const__)) when the compiler is clang
...
Clang didn't implement __attribute__((__const__)) properly for release build.
The issue tracker is: https://github.com/llvm/llvm-project/issues/56993
Closes #6781
Closes #6782
Closes #7170
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com >
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18798 >
2022-09-27 18:08:15 +00:00
Dylan Baker
5e73858b9e
.pick_status.json: Mark 40a235c9a8 as denominated
2022-09-27 10:49:05 -07:00
Dylan Baker
bc71af26d6
.pick_status.json: Mark cbad4adc13 as denominated
2022-09-27 10:48:22 -07:00
Dylan Baker
bef8fe631e
.pick_status.json: Mark 1aacd9492d as backported
2022-09-27 10:47:52 -07:00
Dylan Baker
b0a45e65a7
.pick_status.json: Mark 5b7c2dffc4 as denominated
2022-09-27 10:45:36 -07:00
Dylan Baker
5cba513470
.pick_status.json: Mark 49efa73ba1 as denominated
2022-09-27 10:45:35 -07:00
Erik Faye-Lund
a53554f388
docs: fixup broken rst syntax
...
Seems I messed this up when converting things, and nobody noticed until
now!
Fixes: d4397c0048luoyonggang@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18641 >
(cherry picked from commit c816081eea
2022-09-27 10:03:34 -07:00
Yonggang Luo
8a40852756
ci/docs: Stick to specific version of python packages
...
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com >
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com >
Acked-by: Eric Engestrom <eric@igalia.com >
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18833 >
(cherry picked from commit 9cbfe9a8ba
2022-09-27 10:03:32 -07:00
Mike Blumenkrantz
883c0f7b2f
zink: avoid overflow when clamping bufferviews
...
cc: mesa-stable
Reviewed-by: Dave Airlie <airlied@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18784 >
(cherry picked from commit 084f3b6664
2022-09-27 10:03:32 -07:00
Mike Blumenkrantz
01edc6f174
zink: always unset var->data.explicit_xfb_buffer
...
ensure that this isn't accidentally triggering for non-xfb shaders
Fixes: 6d40db84c9airlied@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18784 >
(cherry picked from commit 53004e57d6
2022-09-27 10:03:31 -07:00
Mike Blumenkrantz
cf2be66218
zink: free vertex_state_cache on shutdown
...
cc: mesa-stable
Reviewed-by: Dave Airlie <airlied@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18784 >
(cherry picked from commit 903aba413d
2022-09-27 10:03:31 -07:00
Mike Blumenkrantz
380ffeb205
lavapipe: propagate shader access info across libraries
...
cc: mesa-stable
Reviewed-by: Dave Airlie <airlied@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18827 >
(cherry picked from commit 6db172436a
2022-09-27 10:03:30 -07:00
Mike Blumenkrantz
23cab32e26
lavapipe: set writemask for compute shader buffers
...
I don't know how this wasn't being set?
cc: mesa-stable
Reviewed-by: Dave Airlie <airlied@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18827 >
(cherry picked from commit 6c935f3f3d
2022-09-27 10:03:29 -07:00
Dylan Baker
3c44bd1166
.pick_status.json: Update to c268ae9efe
2022-09-27 10:03:17 -07:00
Dave Airlie
73e50fdb0f
llvmpipe: bind sampler views/images properly.
...
There is some code that relies on TGSI here, and it has limits.
For now always just bind resources > 31.
Fixes
dEQP-VK.pipeline.pipeline_library.descriptor_limits*
Cc: mesa-stable
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18812 >
(cherry picked from commit 16fd0c11c6
2022-09-26 15:10:24 -07:00
Dave Airlie
1cfa6ef606
llvmpipe: memset empty images in the key
...
This fixes a bunch of valgrind warnings in
dEQP-VK.pipeline.pipeline_library.descriptor_limits.fragment_shader.storage_images_63
Cc: mesa-stable
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18812 >
(cherry picked from commit b144d3e6a7
2022-09-26 15:10:24 -07:00
Dave Airlie
6635957f5d
lavapipe: handle view mask layer count properly for begin rendering.
...
If the view mask is set the layer count should be taken from it,
not from the layerCount.
Fixes: 8a6160a354michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18811 >
(cherry picked from commit fe5636af6e
2022-09-26 15:10:23 -07:00
Dave Airlie
8f9fc8e186
lavapipe: take layers into account around render surface creation.
...
This makes the code more consistent with handling layers, but also
makes sure a new surface is created when the w/h/layers changes.
Fixes: 8a6160a354michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18811 >
(cherry picked from commit 21d39dd1de
2022-09-26 15:10:23 -07:00
Timur Kristóf
4295192f35
radv: Use a fallback for marketing name when libdrm doesn't know it.
...
Currently for GPUs which don't have a marketing name in libdrm,
RADV just prints "(null) (RADV ...)", which looks bad.
This commit replaces the "(null)" with "AMD Unknown".
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com >
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18775 >
(cherry picked from commit 25e1c3d5b3
2022-09-26 15:10:22 -07:00
Yiwei Zhang
34aa290747
mesa/st: emit string marker based on PIPE_CAP_STRING_MARKER
...
Fixes: 10ac88b72fzzyiwei@chromium.org >
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18803 >
(cherry picked from commit 37900ee2dd
2022-09-26 15:10:21 -07:00
Pierre-Eric Pelloux-Prayer
35ebfa8a63
radeonsi/gfx11: don't set VERTS_PER_SUBGRP to 0
...
It seems slower.
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Fixes: 25a66477d0https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18758 >
(cherry picked from commit 4fdf10fdaf
2022-09-26 15:10:18 -07:00
Lionel Landwerlin
a7750abce3
intel/fs: fixup a64 messages
...
And run algebraic when either int64 for float64 are not supported so
those don't end up in the generated code.
Cc: mesa-stable
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17396 >
(cherry picked from commit 139e8f4635
2022-09-26 15:10:17 -07:00
Lionel Landwerlin
8a3a0210ae
intel/nir/rt: store ray query state in scratch
...
Initially I tried to store ray query state in the RT scratch space but
got the offset wrong. In the end putting this in the scratch surface
makes more sense, especially for non RT stages.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: c78be5da30ivan.briano@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17396 >
(cherry picked from commit 838bbdcf2e
2022-09-26 15:10:17 -07:00
Lionel Landwerlin
87f6095557
intel/nir/rt: change scratch check validation
...
It's very unfortunate that we have the RT scratch being conflated with
the usual scratch. In our implementation those are 2 different buffers.
The usual scratch access are done through the scratch surface state
(delivered through thread payload), while RT scratch (which outlives
thread dispatch with shader calls) is its own buffer.
So checking the NIR scratch size makes no sense as we can have normal
scratch accesses completely unrelated to RT scratch accesses.
This change switches the validation by looking at whether the scratch
base pointer intrinsic is being used (which is what we use/abuse to
implement RT scratch).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: c78be5da30ivan.briano@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17396 >
(cherry picked from commit f7fab09a07
2022-09-26 15:10:16 -07:00
Lionel Landwerlin
318020df2b
intel/nir/rt: fix ray query proceed level
...
Initially the level is world (top level), then it's whatever level the
potential hit is.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: c78be5da30ivan.briano@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17396 >
(cherry picked from commit 259b1647e6
2022-09-26 15:10:16 -07:00
Lionel Landwerlin
a13dc0cd0e
intel/nir/rt: spill/fill the entire ray query data
...
We need the traversal stack to saved/restored along with mem hits.
Total spill/fill is 256bytes.
We can potentially optimize this but we have to be very careful about
what state the query is in.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: c78be5da30ivan.briano@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17396 >
(cherry picked from commit f843bec7de
2022-09-26 15:10:15 -07:00
Lionel Landwerlin
6094dba1da
intel/nir/rt: fixup generate hit
...
This function copies the potential hit from its memory location to the
committed hit location. A couple of fields got their bit offset wrong.
Fixes some CTS tests in dEQP-VK.ray_query.*
v2: Copy primitive/instance leaf pointers
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: 0465714790ivan.briano@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17396 >
(cherry picked from commit a88f725eea
2022-09-26 15:10:15 -07:00
Mike Blumenkrantz
7cb53bcf6f
zink: handle culldistance xfb outputs like clipdistance
...
fixes:
spec@ext_transform_feedback@builtin-varyings gl_culldistance
cc: mesa-stable
Reviewed-by: Adam Jackson <ajax@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18724 >
(cherry picked from commit 21ba47c821
2022-09-26 15:10:14 -07:00
Caio Oliveira
d600acf6d9
nir/lower_task_shader: Don't fail adding a launch when last instruction is a jump
...
Fixes: 8aff8d3dd4timur.kristof@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18442 >
(cherry picked from commit 3f4343c7cd
2022-09-26 15:10:14 -07:00
Lionel Landwerlin
6cf16955f7
anv: add missing wokraround for texture cache invalidate
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com >
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18743 >
(cherry picked from commit f9dbb65e7f
2022-09-26 15:10:13 -07:00
Connor Abbott
90001a78d1
ir3/analyze_ubo_ranges: Account for reserved consts
...
We weren't accounting for the reserved consts when calculating how much
we can upload. This led to assertion failures later if we pushed too
much.
Fixes: d3b7681df2https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18757 >
(cherry picked from commit 8cb1deded6
2022-09-26 15:10:13 -07:00
Connor Abbott
7b1f40a3db
tu: Fix maxPerStageDescriptorUpdateAfterBindInputAttachments
...
We need this to be the same as maxPerStageDescriptorInputAttachments.
Fixes: d9fcf5de55https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18757 >
(cherry picked from commit f483419c23
2022-09-26 13:11:27 -07:00
Emma Anholt
c08e78aba9
turnip: Fix busy-waiting on syncobjs with OS_TIMEOUT_INFINITE.
...
I noticed that glmark2's glFinish()es in its offscreen rendering tests
under zink were spinning. When we passed -1 as the timeout for
drmSyncobjWait(), the kernel would immediately return ETIME.
Fixes: 0a82a26a18https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18739 >
(cherry picked from commit 5e39b52e6a
2022-09-26 13:11:25 -07:00
Tapani Pälli
dba8e3f3f5
mesa/st: fix a set_sampler_views call parameter order
...
Fixes: e7b9561959tapani.palli@intel.com >
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18745 >
(cherry picked from commit d68e74955c
2022-09-26 13:11:22 -07:00
Erik Faye-Lund
f94eeaeb93
mesa/st: always use normalized coords for samplers
...
Normalized samplers is the norm, and non-normalized samplers might cause
some drivers and hardware to have to bend over backwards a bit.
We're using TXF here anyway, so it doesn't really matter what we set
this state to. So let's always use normalized samplers instead of
always using non-normalized samplers. That makes things easier for
everyone.
Fixes: e7b9561959michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17708 >
(cherry picked from commit 37733c100d
2022-09-26 13:11:22 -07:00
