Since the stack pointer may wrap around the stack size in overflow
cases, traversal logic calculates the real stack pointer with
nir_umod_imm(b, stack, args->stack_entries * args->stack_stride).
For ray queries, "stack" was initialized to
"stack_base + local_invocation_idx * 4". This was completely broken, as
the umod would later delete the stack base completely and overwrite the
start of LDS, which belongs to the apps' shared memory.
Instead, add the stack base as a constant offset in the load/store_stack
callback. (This should also save 1 VALU per ray query)
Also, delete radv_ray_traversal_args::stack_base since it's unused now.
Cc: mesa-stable
(cherry picked from commit b046eaf36d)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40488>
Vertex shaders shorter than four instructions can hard-lock R3xx GPUs.
This seems to happen in combination with a small vertex count. This was
seen before, most notably with dummy shaders, but the earlier fix only
removed those dummy shaders, so some occurrences could still slip
through the cracks. Pad all vertex shaders to four instructions on R3xx.
Reviewed-by: Filip Gawin <filip@gawin.net>
Fixes: c6aa639ba9 ("r300: skip draws instead of using a dummy vertex shader")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/337
(cherry picked from commit 9b12664b72)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40488>
GFX12 encoding added one bit to the stack offset, doubling the limit on
the stack base offset that is possible to encode. In practice, this
always allows using bvh_stack_push* instructions on GFX12 since LDS is
still 64kB.
Cc: mesa-stable
Fixes: 59a39779 (radv/rt: Only use ds_bvh_stack_rtn if the stack base is possible to encode)
(cherry picked from commit 867d0b33b3)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40488>
Needed for some FSR macro changes I want to test.
Signed-off-by: Mary Guillemard <mary@mary.zone>
Fixes: 7d6cc15ab8 ("nvk/mme: Add a unit test framework for driver macros")
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
(cherry picked from commit 32895657b4)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40488>
The correct dependence is cs_flush_caches.cs_defer.signal to
signal cs_sync32_set.cs_defer.wait in occulusion query path.
Fixes: 443ddac ("panvk/csf: merge v10 and v11 paths in
issue_fragment_jobs")
Fixed: many random fail cases in VK-GL-CTS 1.4.4.2, eg.
dEQP-VK.query_pool.occlusion_query.get_results_conservative
_size_64_wait_query_without_availability_draw_points_clear_color
Signed-off-by: Ryan Zhang <ryan.zhang@nxp.com>
Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
(cherry picked from commit 93b58064f7)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40488>
Normally Venus on Nvidia GPUs takes the prime blit path. The exception
is when KWin or any wlroots based compositors are used:
1. KWin and wlroots based compositors always add LINEAR to dmabuf
feedback tranches assuming LINEAR can be handled by GPU drivers.
2. Venus + Virgl only sees the compositor injected LINEAR mod since
Virgl doesn't support explicit modifiers on the driver side.
3. Nvidia GPUs doesn't support LINEAR color attachment, and it's too
late to reject LINEAR mod when the native image path has already
been taken instead of the prime image path.
Gamescope requires VK_EXT_physical_device_drm and its runtime doesn't
use standard WSI extensions, so venus can spoof without impacting it.
Cc: mesa-stable
(cherry picked from commit 1a302155ee)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40488>
"explicit sw" means llvmpipe, which cannot be a real drm device. this requires also
returning only a single device so as to avoid leaking non-sw drivers
should fix LIBGL_ALWAYS_SOFTWARE=1 eglinfo
Fixes: 8a339cdebc ("egl: fix sw fallback rejection in non-sw EGL_PLATFORM=device")
(cherry picked from commit c9b2986607)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
WA states that we need to allocate maximum number of stackIDs per DSS
from RT_DISPATCH_GLOBALS to 2048.
We can still throttle/control the CFE_STATE::StackID to be in range
specified by the field.
This does impact performance having CFE_STATE::stackIDs capped to 2K
by default. More the outstanding ray queries, larger the working set and
have more impact on cache hit rate.
This affect performance on Xe2+ onwards:
* Boundary Benchmark: 36.2%
* Solar Bay extreme: 9.8%
* Hitman world of assassination: 3.9%
Fixes: c1a44e8d43 ("anv: force StackIDControl value for Wa_14021821874")
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
(cherry picked from commit cb423ee636)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
Previously, we assumed that the selector for bcsel could be whatever,
regardless of the bit sizes of the data and we'd just fix it in the
back-end. This works okay for scalars but falls over the moment we
vectorize because all our vector handling assumes bit sizes match.
Since matching bit sizes is what the hardware wants anyway, it's better
to do the right thing in NIR and hope copy-propagation can fold in
conversions if needed.
Unfortunately, copy prop isn't that smart yet so this does hurt a bit:
Instrs: 1193679 -> 1198086 (+0.37%); split: -0.06%, +0.43%
CodeSize: 11915136 -> 11950592 (+0.30%); split: -0.05%, +0.34%
Full: 160985 -> 160941 (-0.03%); split: -0.04%, +0.01%
Estimated normalized CVT cycles: 4456.938557000181 -> 4480.876069000186 (+0.54%); split: -0.13%, +0.67%
Estimated normalized SFU cycles: 6350.9375 -> 6392.21875 (+0.65%)
Estimated normalized Load/Store cycles: 205773.0 -> 205795.0 (+0.01%)
Maximum number of threads: 12864 -> 12863 (-0.01%)
Number of spill instructions: 22487 -> 22489 (+0.01%)
Number of fill instructions: 52179 -> 52219 (+0.08%)
Hurt shaders:
google-meet-clvk/BgBlur
google-meet-clvk/Relight
parallel-rdp/small_subgroup
parallel-rdp/small_uber_subgroup
The proper solution here is to teach copy-prop about this stuff so that
it can propagate swizzles into ALU ops when they're supported:
https://gitlab.freedesktop.org/panfrost/mesa/-/issues/265
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14945
Cc: mesa-stable
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
(cherry picked from commit 3fd471dca5)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
It calls both for some reason but never handles any other booleans than
32-bit. This was probably a mistake.
Fixes: e63a7882a0 ("etnaviv: call nir_lower_bool_to_bitsize")
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
Reviewed-by: Christian Gmeiner <cgmeiner@igalia.com>
(cherry picked from commit 6fb3995659)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
This should be doing a or and not an assign.
This fixes issues on NVK with mesh stages on DGC.
Signed-off-by: Mary Guillemard <mary@mary.zone>
Fixes: 9308e8d90d ("vulkan: Add generic graphics and compute VkPipeline implementations")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
(cherry picked from commit 8f2eeee7ba)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
Zink uses the output primitive of the last vertex stage when deciding
the raster primitive. When we generate the gs the output primitive
depends on the raster primitive.
Not only does the generated gs output primitive have no value in chosing
the raster primitive, it can also get us stuck with the last raster
primitve which is of course incorrect.
Ignore it for generated shaders.
Cc: mesa-stable
(cherry picked from commit d526bbc29b)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
If multiview is enabled on the render pass, baseLayer and layerCount
will be 0 and 1 respectively and throw us off.
We can still fast clear if view_mask == 1, but anything else hits the
BLORP_BATCH_NO_EMIT_DEPTH_STENCIL restriction.
Fixes: e488773b29 ("anv: Fast clear depth/stencil surface in vkCmdClearAttachments")
Signed-off-by: Iván Briano <ivan.briano@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
(cherry picked from commit 5d22f307d5)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
Commit 0d4aa5f55f introduced the watermark to optimize the guardband
state changes and always computed new_distance as MAX2(distance,
watermark).
That is correct for point/line paths where distance > 0, but it keeps a
non-zero discard distance alive when the next draw sets distance = 0
(triangles). This leaks wide point/line clip-discard state into later
triangle draws and can clip away large parts of geometry (as observed in
Sauerbraten). Only apply the watermark when distance > 0 and reset it to
zero otherwise so triangle draws disable clip-discard as intended.
Fixes: 0d4aa5f55f ("r300: pop-free clipping")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14959
(cherry picked from commit ce33f82f83)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
It must be lowered.
This fixes
dEQP-VK.spirv_assembly.instruction.compute.compute_shader_derivatives.{mesh,task}.*.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
(cherry picked from commit 3c4cb16159)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
If the secondary changes the fragment output state and if the same
PS epilog used before ExecuteCommands() is re-bind immediately after
that call, the PS epilog state wouldn't be re-emitted.
Apply the same change for VS prologs, although the logic is slightly
different and the bug shouldn't occur. The whole logic of secondaries
should be completely rewritten because it's definitely not robust.
This fixes a GPU hang in Where Winds Meet, see
https://github.com/doitsujin/dxvk/issues/5436.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
(cherry picked from commit 1a00587c44)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
Lavapipe relies on true udmabuf support for dmabuf export allocation.
This changes aligns the behavior with both llvmpipe_allocate_memory_fd
and llvmpipe_import_memory_fd.
Fixes: 7d0a631f20 ("llvmpipe: export dmabuf caps for kms_swrast")
Reviewed-by: Lucas Fryzek <lfryzek@igalia.com>
(cherry picked from commit 5ab8c8a439)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
This is not NaN correct.
And also make the pattern 32bit only because the constant is hard coded
FLT_MAX.
Fixes: 780b5c1037 ("nir/algebraic: Simplify some Inf and NaN avoidance code")
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
(cherry picked from commit ab773fc5d4)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
There was duplicated code to set unscaled_input_fragcoord and a read
from VK_ATTACHMENT_UNUSED attachment, which incorrectly updated
builder->unscaled_input_fragcoord.
ubsan:
tu_pipeline.cc:4734:44: runtime error: load of value 127, which is not a valid value for type 'bool'
Seen in:
dEQP-VK.renderpasses.renderpass1.custom_resolve.monolithic.stencil_only_s8
Fixes: 97da0a7734 ("tu: Rewrite to use common Vulkan dynamic state")
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
(cherry picked from commit 81a76be861)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
For dynamic renderpass we created a fake second subpass,
which would is used by CmdBeginCustomResolveEXT, however
CmdBeginCustomResolveEXT doesn't trigger tile stores, but
attachments didn't know they should be stored after fake
custom resolve subpass.
Fixes: 520e3f3a47 ("tu: Implement VK_EXT_custom_resolve")
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
(cherry picked from commit 67c54c4465)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
Use INT_MIN instead of INT_MAX for underflow.
Fixes: cc4b50b023 ("nir/opcodes: use u_overflow to fix incorrect checks")
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pelloux@gmail.com>
(cherry picked from commit da57fbfb07)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40359>
