It is safe to allocate command buffer memory as MEM_LOCAL (VRAM or GART)
when we can map GPU memory (we trust NVKMD to force GART).
This reduces latency on pyrowave for compute dispatch (~1.90ms -> ~1.48ms),
likely as QMD and the root tables should not have been in GART in the first place.
Signed-off-by: Mary Guillemard <mary@mary.zone>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36970>
This increases rate of depth fast clear rate on BMG G31
per HSD 22020044224.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35966>
Field that we currently utilize does not change place, however
there are some new fields so let's update contents to match spec.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35966>
It turns out that the `occlusion_query.syncobj` is used to set
state that later code relies on, and setting it to NULL causes
some Vulkan CTS tests to fail. Instead, we should explicitly check
for the mode being `MALI_OCCLUSION_MODE_DISABLED` to avoid using
an invalid `ptr` field.
Fixes: 24c692c981 ("panvk: fix a NULL pointer dereference in occlusion queries")
Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36794>
For now only add the fallback variants with multiple sampler views in
order to keep things simple.
Signed-off-by: Robert Mader <robert.mader@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36890>
Which are essentially variants of NV12. All of them have been tested
with the Weston client-buffer test with llvmpipe and radeonsi.
Signed-off-by: Robert Mader <robert.mader@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36890>
`clRetainDevice` and `clReleaseDevice` ignore the device argument, it
should be checked for validity and return `CL_INVALID_DEVICE` if it
fails.
`check_cl_device_type` fails if entrypoint is passed `0` as the device
type. Zero as an input for a bitwise and operation will always result
in a zero, so the utility function fails to return
`CL_INVALID_DEVICE_TYPE`.
Signed-off-by: Ahmed Hesham <ahmed.hesham@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37008>
Strictly speaking, for an empty batch, there is no inherit/defined
ordering between the wait and signal operations within the batch. We
are relying on MESA impl defined behavior to transfer payloads of the
wait sempahores to the signal sempahore here.
Ideally we need a MESA internal hint to ensure transitivity, but since
this is on the legacy fallback path, it's not worth the effort. So here
we follow common wsi to use TRANSFER_BIT which avoids most unnecessary
stalls on the queue (an improvement over the previous ALL_COMMANDS_BIT).
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36855>
This follows common wsi side as suggested, and simplifies the later
adoption of vk_device_copy_semaphore_payloads.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36855>
This way we report the *actual* tilebuffer sizes rather than reporting
twice the amount and then correcting for it later on. The distinction
really only matters on V4 and V6 GPUs, but let's get it correct.
This shouln't have any functional changes, it just makes it easier to
reason about what each GPU supports.
While we're at it, correct the typo in the comment about what gens
supports disabling pipelining. It's *not* supported on V6.
Reviewed-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35755>
When using formats with less than 32-bits per pixel, we pad the
tile-buffer to a multiple of 32-bits so we can store additional bits
used by dithering.
Account for this when computing the max MSAA setting.
Fixes: 329568b5eb ("panfrost: add color-attachment and msaa helpers")
Reviewed-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35755>
Make sure that depth and color buffers have the same number of
samples. This *should* be guaranteed by the higher levels, but
in the case of a bug we want to notice quickly (and in fact
such a bug has bitten us, see panfrost/mesa/issue#200).
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36876>
The actual number of samples chosen is allowed to equal the number
requested, but currently we just check for sample counts greater
than the request.
Fixes: 894b37e060 ("mesa: fix sample count handling for MSRTT")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36878>
Those are for push constants, no point in doing that because :
- there is no HW constant offsets in push constants (payload
delivery), it's just register offset calculation
- if we have an dynamic value it's already using MOV_INDIRECT
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: e103afe7be ("brw: run the nir_opt_offsets pass and set the maximum offset size")
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36958>
I'm not sure why exactly it didn't work because
TPL1_A2D_SRC_TEXTURE_SIZE seemingly has (1 << 15) width
limit. However tests have shown that it doesn't work out.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36939>
The bottom right corner of the copy exceeded the maximum allowed
value in GRAS_A2D_DEST_BR.x
In order to fix this, we have to do a second copy per line of
the last texels.
Fixes asserts in:
dEQP-GLES31.functional.copy_image.*
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36939>
Turnip crashes under drm-shim when enabling VM_BIND. We don't care about
VM_BIND for shader compilation so just disable it.
Signed-off-by: Job Noorman <jnoorman@igalia.com>
Fixes: 4efbfa1441 ("tu/drm: Enable VM_BIND")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37000>
It's possible with GPL.
This fixes a NULL pointer dereference with updated pipeline binaries
tests in VKCTS.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36999>
"scale" was a bad name as it meant nothing others could comprehend.
However that value corresponded to the tile size of the Matrix layouts.
For int8 we have a tile size of 4, for 16 and 32 bit values we have a size
of 2.
For the future the same value need to be 32 for booleans, 8 for int4, 1
for fp64 and tf32 and 4 for all em types.
The "scale = 1 << scale;" assignment can simply be removed because it was
4 for 2 and 2 for 1 simply being the expected value after this change.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36391>
Without this, mesa build on some distros may silently produce a version
of radv that silently refuses to save RGP traces
Signed-off-by: Arseny Kapoulkine <arseny.kapoulkine@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36960>
FdMemoryDataSource was being registered as a Perfetto data source
unconditionally which led to anything calling fd_device_new(...)
attempting to do this even when they might not have Perfetto
initialized which is done as a part of util_perfetto_init, without
which trying to register the event causes a SEGFAULT.
Fixes: c7045e3e63 ("perfetto: unify init")
Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36993>
We are still in the process of moving our kernels to gfx-ci/linux, but
we got the request to uprev the kernel a month ago when I started my
holiday, so let's not delay it more. Anyway, it is better to change
only one variable at a time so no harm done.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32533>
