25d4943481
nir: make use_interpolated_input_intrinsics a nir_lower_io parameter
...
This will need to be set to true when the GLSL linker lowers IO, which
can later be unlowered by st/mesa, and then drivers can lower it again
without load_interpolated_input. Therefore, it can't be a global
immutable option.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32229 >
2024-11-20 02:45:37 +00:00
190fbbe1d6
nak/hw_test: Use std::mem::offset_of!()
...
It got stabilized with Rust 1.77.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31510 >
2024-11-18 22:29:13 +00:00
45c1280d2c
nir_lower_mem_access_bit_sizes: pass access to callback
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31904 >
2024-11-13 12:59:26 +00:00
61752152f7
nir_lower_mem_access_bit_sizes: add nir_mem_access_shift_method
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com >
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31904 >
2024-11-13 12:59:26 +00:00
234b9c72f9
nvk/ci: document flakes seen recently
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32080 >
2024-11-13 12:26:50 +00:00
68cdcf5bbc
nvk: Fix two typos in comments
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32086 >
2024-11-12 23:26:49 +00:00
05770374a3
nvk: Remove params for dirty_cbufs_for_descriptors
...
dyn_start/dyn_end are unused now.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32086 >
2024-11-12 23:26:49 +00:00
dc12c78235
nvk: Fix invalidation of NVK_CBUF_TYPE_DYNAMIC_UBO
...
Because dyn_start and dyn_end are indices into
nvk_root_descriptor_table->dynamic_buffers, we would need to offset
cbuf->dynamic_idx by
nvk_root_descriptor_table->set_dynamic_buffer_start[cbuf->desc_set]
in order to do those comparisons correctly.
We could do that, but it's simpler and no less precise to sinply
re-use the same comparison that we do in the other cases here.
This fixes a rendering artifact in Baldur's Gate 3 (Vulkan), which
regressed with the commit listed below.
Fixes: 091a945b57faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32086 >
2024-11-12 23:26:49 +00:00
64f17c1391
nvk/cmd_buffer: Pass count to set_root_array
...
Previously, we were passing the end index which was incorrect.
Also, improve the macros so that they can take an expression for
the count.
Fixes: b2d85ca36ffaith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32086 >
2024-11-12 23:26:49 +00:00
47a1565c3d
nv/codegen: Do not use a zero immediate for tex instructions
...
They aren't always legal for tex instructions, specifically for TXQ when
an actual source is needed.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11999
Fixes: 85a31fa1fchttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32043 >
2024-11-08 09:18:54 +00:00
73db4d350a
nvk/ci: document flakes seen lately
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32007 >
2024-11-06 14:14:38 +00:00
831eab2375
nvk/ci: add flakes seen recently
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31825 >
2024-10-24 16:50:44 +00:00
654683143b
nvk/ci: add back a crash that was mistakenly removed from the expectations
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31825 >
2024-10-24 16:50:44 +00:00
8d1abd4996
treewide: use nir_src_is_divergent() rather than checking the divergence of the SSA
...
Without LCSSA, divergence between src and def might differ.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30787 >
2024-10-24 10:06:17 +00:00
c25c63ebc0
nir/divergence: separately indicate whether loops have divergent continues or breaks
...
bool nir_loop_is_divergent(nir_loop *)
replaces the previous loop->divergent indicator.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30787 >
2024-10-24 10:06:17 +00:00
9fda96db5b
ci: uprev vkd3d-proton to 59d6d4b5ed23766e69fe252408a3401d2fd52ce8
...
It contains few fixes for recent DGC tests.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31796 >
2024-10-23 15:47:54 +00:00
7cd7631554
ci: uprev vkd3d-proton to 65b81403435576d882d9141ae3eb4a29373fba0e
...
For EXT DGC implementation.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31754 >
2024-10-22 14:12:32 +00:00
5fa4c1a191
compiler/rust: Copy NirInstrPrinter from NAK
...
Switch NAK to it.
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com >
Reviewed-by: M Henning <drawoc@darkrefraction.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31706 >
2024-10-18 12:43:52 +00:00
d52a9d832e
nvk: Put a sample map in the descriptor for MSAA storage images
...
This tells us exactly where each sample instead of assuming they're laid
out row-major. NIL_SAMPLE_LAYOUT_4X2_D3D is not row-major.
Fixes: 8f1697b12dhttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31727 >
2024-10-18 04:56:26 +00:00
9724028c15
nil: Add a nil_sample_offset() helper
...
Fixes: 8f1697b12dhttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31727 >
2024-10-18 04:56:26 +00:00
6e30ab8b16
nvk: Fix host copies for array images
...
Our offset handling was hitting asserts inside of NIL because we're
passing in offsets and extents with non-trivial array dimensions.
Fixes: 6c5420cd30https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31727 >
2024-10-18 04:56:26 +00:00
aee0ce980b
nak/bindings: Use an enum for IOCTL numbers
...
This should be more robust against weird bindgen shenanigans because it
forces LLVM to constant-fold and give us an actual thing we can generate
bindings for.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12008
Fixes: cd7128c2e3https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31727 >
2024-10-18 04:56:26 +00:00
f49a3e0a68
nvk: Use nvk_queue_submit_simple() for nvk_queue_state_update()
...
This change has two main advantages. First is that the previous code
had a subtle bug where state updates would drop the previous state
pushbuf without first guaranteeing that the queue was idle, potentially
leading to a fault. This worked when I first wrote the code because we
were using the old nouveau UAPI which took a list of BOs on each submit
and the kernel would interanlly reference count them. With the modern
UAPI, there is no such safety net.
The second effect of this commit is that we're now only submitting the
pushbuf when one of the pointers or ranges actually changes. This
should hopefully reduce GPU stalling as setting those pointers is a
pretty heavy-weight operation.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30596 >
2024-10-18 02:58:00 +00:00
5fa5c5a300
nvk: move nvk_queue_state_update to nvk_queue_submit_exec
...
The only reason we had it pulled out was because that made more sense
back when we had drm/nouveau baked into our queue code. With the new
NVKMD interface, there's no point.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30596 >
2024-10-18 02:58:00 +00:00
86a6b894a8
nvk: Only submit the state pushbuf if there are commands
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30596 >
2024-10-18 02:58:00 +00:00
296746d4b9
nvk: Only wait on the upload queue if there are commands
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30596 >
2024-10-18 02:58:00 +00:00
21beb7a6bd
nvk: implement VK_EXT_depth_clamp_zero_one
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31488 >
2024-10-18 00:56:15 +00:00
ade4512803
nvk: handle driconf for zeroing vram
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29892 >
2024-10-17 22:19:33 +00:00
cba575f4df
nir: always emit ddx intrinsics
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Rob Clark <robdclark@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31014 >
2024-10-17 09:50:19 +00:00
279f38918f
nak: memstream: move into common code
...
Move the memstream code into common code. Other Rust code interfacing
with FILE pointers will find the memstream abstraction useful.
Most notably, pinning is actually enforced this time with PhantomPinned.
Add a .clang-format from a sibling dir (i.e.: compiler/nir) while we're
at it to keep things tidy.
Signed-off-by: Daniel Almeida <daniel.almeida@collabora.com >
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30594 >
2024-10-17 02:50:21 +00:00
9f58d0b0e7
nvk/ci: document more flakes seen
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31691 >
2024-10-16 20:40:18 +00:00
a33538cb9f
nvk: Remap 10 and 12 bit formats to 16 bit formats
...
Preserves the previous behavior while handling the new formats.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Signed-off-by: Valentine Burley <valentine.burley@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30821 >
2024-10-16 14:30:15 +00:00
537ada2308
nak: Phi coalescing via biased register coloring
...
Reduces code size by -29.08% on shaderdb + nvk-fossils-foss
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31498 >
2024-10-15 22:29:11 +00:00
03a393d6ca
nak: Handle annotations in legalization
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31665 >
2024-10-15 17:13:27 +00:00
36d9d11882
nak: Remove annotations before calc_instr_deps()
...
Otherwise the annotations might throw off latency information which
needs exact instruction counts.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31665 >
2024-10-15 17:13:27 +00:00
02923e237d
nir: add hole_size parameter into the vectorize callback
...
It will be used to allow merging loads with a hole between them.
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29398 >
2024-10-15 05:50:24 +00:00
c2684968de
nvk: Advertise 64-bit atomics on buffer views
...
We also add an nvk_format_supports_atomics() helper. This helper lives
in NVK for now because it's not just about the format and hardware but
also about whether or not we have compiler support in NAK.
Fixes: 1d10de539chttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31633 >
2024-10-15 05:21:03 +00:00
d3d8271620
nvk: Re-sort the features table
...
There were a couple of KHR extensions that got mixed in with the EXTs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31633 >
2024-10-15 05:21:03 +00:00
681f807747
nvk: Only set texture/sampler tables and SLM for enabled engines
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31633 >
2024-10-15 05:21:02 +00:00
7ae2cc7f0a
nvk: Add an nvkmd_engines bitfield to nvk_queue
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31633 >
2024-10-15 05:21:02 +00:00
2fb4aed9d8
nvk: Advertise VK_KHR_fragment_shading_rate
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
e45effe555
nvk: Set VARIABLE_PIXEL_RATE_SHADING_TABLE_SELECT
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
99107ef8d4
nak: Add support for gl_PrimitiveShadingRateEXT
...
This aliases gl_ViewportIndex in the NVIDIA attribute space.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
ab8e6bf4c3
nak: Pass a nak_compiler into sysval and attrib helpers
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
58252622f1
nak: Add support for gl_ShadingRateEXT
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
957a1add26
nak: Rename SV_VERTEX_COUNT to PRIM_TYPE
...
This is what nvdisasm calls it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
1e7c3ddc33
nak: Get rid of some dead code warnings
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
75bcb656d9
nvk: Add support for binding fragment shading rate images
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
16bd3f0f50
nvk: Emit FSR state
...
This is mostly a matter of filling out the
VARIABLE_PIXEL_SHADING_INDEX_TO_RATE tables.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
55854de584
nvk: Rework setting of min_sample_shading
...
Instead of just storing a pre-computed min_sample_shading, store both
sample_shading_enable and min_sample_shading and then compute the final
min_sample_shading value when we emit state.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31585 >
2024-10-10 23:16:57 +00:00
Marek Olšák