Commit Graph

202918 Commits

Author SHA1 Message Date
Guilherme Gallo a33c0e1867 ci/lava: Split boot action into deploy and boot
The boot action was wrapping the deploy action, which could cause
timeout misalignment. For example, the boot `GitlabSection` timeout was
shorter than the deploy timeout in LAVA, leading to cases where LAVA
jobs were canceled during their own retry mechanism.

By splitting these actions, we can align the timeouts properly,
preventing interference and unnecessary job cancellations.

Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33906>
2025-03-10 05:44:25 +00:00
Guilherme Gallo d85af615f9 ci/lava: Remove depthcharge-start timeout
It has no effect, as it is overridden by depthcharge-retry timeout /
failure_retry.

Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33906>
2025-03-10 05:44:25 +00:00
Lionel Landwerlin 1835bf3520 brw: avoid calling lower_indirect_derefs multiple times
Lowering the indirect derefs multiple times leads to very inefficient
shaders because of all the control flow inserted.

In particular on some DGC tests with mesh shaders, the tests can spin
for 1hour on an i7 and still not complete compilation.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33809>
2025-03-09 20:52:01 +00:00
Yiwei Zhang c7bc90eaec venus: fix to ignore dstSet for push descriptor
Per push descriptor spec:

Each element of pDescriptorWrites is interpreted as in
VkWriteDescriptorSet, except the dstSet member is ignored.

Cc: mesa-stable
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33948>
2025-03-09 20:25:21 +00:00
Yiwei Zhang 7b228ef877 venus: use common memory report implementation
Looks to be the same, and there's no regression in the cts group
dEQP-VK.memory.device_memory_report.*

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33955>
2025-03-09 18:51:41 +00:00
Eric Engestrom 551770ccf8 wsi/x11: drop misleading reference to anv in var names
Fixes: ec0bc14a70 ("anv/wsi: remove all anv references from WSI common code")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33950>
2025-03-09 12:00:44 +00:00
Pavel Ondračka 28057971d2 r300/ci: add dEQP job with RADEON_DEBUG=notcl
This forces software vertex processing wia the draw module and should
hopefully test the exact same codepaths that the r300 chipsets without
built-in vertex engines use.

Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28216>
2025-03-09 08:39:53 +00:00
Faith Ekstrand 76883e0b3c vulkan: Update XML and headers to 1.4.309
Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33930>
2025-03-08 19:54:19 +00:00
Faith Ekstrand 57818436f8 spirv: Update the JSON and headers
Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33930>
2025-03-08 19:54:19 +00:00
Faith Ekstrand 1cdc420065 spirv: OpAsmTargetINTEL is untyped
We need to handle this in order to bump the JSON in the next commit.

Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33930>
2025-03-08 19:54:19 +00:00
Leonard Göhrs 240585cc6d etnaviv/ci: add pengutronix LAVA lab with one i.MX8MP device
The device is a MBa8MPxL[1]. Once we have verified that the infrastructure
is set up correctly and is operating reliably more devices can be added.

[1]: https://www.tq-group.com/de/produkte/tq-embedded/arm-architektur/mba8mpxl/

Reviewed-by: Christian Gmeiner <cgmeiner@igalia.com>
Reviewed-by: Valentine Burley <valentine.burley@collabora.com>
Signed-off-by: Leonard Göhrs <l.goehrs@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33759>
2025-03-08 10:18:11 +00:00
Timothy Arceri 25e008c639 util/u_idalloc: fix util_idalloc_sparse_alloc_range()
If the allocation didn't fit within the segment the loop incorrectly
freed ids of a range of different segments due to the loop redeclaring
i.

Fixes: d4085aaf56 ("util: add util_idalloc_sparse, solving the excessive virtual memory usage")

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33934>
2025-03-08 09:17:07 +00:00
Alyssa Rosenzweig 0568e57d21 pan/lower_helper_invocation: clean up
drive-by.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33939>
2025-03-08 07:47:40 +00:00
Alyssa Rosenzweig bc6b527b52 nir/lower_helper_writes: fix stores after discard
We need to use nir_is_helper_invocation instead of
nir_load_helper_invocation, to correctly predicate stores after demote.

Identified in a Piglit on AGX a year ago but I forgot to upstream this.

Fixes: 586da7b329 ("nir: Add nir_lower_helper_writes pass")
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33939>
2025-03-08 07:47:40 +00:00
Alyssa Rosenzweig e90ccf91a3 pan/mdg: call nir_lower_is_helper_invocation
needed to avoid regression from the next patch.

backported because the next patch is too

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Cc: mesa-stable
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33939>
2025-03-08 07:47:40 +00:00
Alyssa Rosenzweig d92304852a agx: call nir_lower_is_helper_invocation
needed for next patch.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33939>
2025-03-08 07:47:40 +00:00
Guilherme Gallo 47659ddf70 ci: Simplify LAVA farm detection
Refactor the LAVA farm detection to use a simpler environment
variable-based approach:
- Remove the complex regex-based farm detection
- Replace LavaFarm enum with a simple string-based farm identification
- Update related tests and job definition logic
- Remove hypothesis testing dependency

Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33888>
2025-03-08 02:45:02 +00:00
Guilherme Gallo 4a2717c4bd ci: Specify the FARM variable for DUT jobs
There are some jobs that were missing the FARM variable, which is useful
to lava_job_submitter.py to classify how it should interact with each
LAVA server and how it should assemble the job definition.
Right now, we use a set of regexex with the RUNNER_TAG variable, but
that is error-prone.

Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33888>
2025-03-08 02:45:02 +00:00
Mel Henning 0dad7857d8 nvk: Add NVK_DEBUG=trash_memory
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33947>
2025-03-08 02:24:18 +00:00
Mel Henning cab80223fd nvk: Don't zero imported memory
This fixes eg.
dEQP-VK.drm_format_modifiers.export_import_fmt_features2.a8b8g8r8_uint_pack32
with NVK_DEBUG=zero_memory

Fixes: 0399999dec ("nvk: Support dma-buf import")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33947>
2025-03-08 02:24:18 +00:00
Faith Ekstrand 18d206d67c zink: Check queue families when binding image resources
We check for iamge layouts and feedback loops when we bind image
resources but not queue families.  If the resource isn't on the graphics
queue, we need to add it to need_barriers so we can transition it back
to our queue.

Fixes: d4f8ad27f2 ("zink: handle implicit sync for dmabufs")
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33952>
2025-03-08 01:38:00 +00:00
Faith Ekstrand 396ece1ad8 zink: Set needs_barrier after transitioning to QUEUE_FAMILY_FOREIGN
Otherwise, we'll transition to QUEUE_FAMILY_FOREIGN and then forget that
we left it on the foreign queue and never transition back the next time
we use the resource.  This was kind-of okay with Wayland compositors
because they always re-import the BO so it's always fresh and they pick
up on the queue transfer the first time.  X11, on the other hand, does
not re-import BOs so they get stuck in this weird QUEUE_FAMILY_FOREIGN
limbo until something happens to randomly trigger a layout transition
check and then we find it and do the transition.  We should mark them as
needing a barrier the moment we transition to QUEUE_FAMILY_FOREIGN.

Fixes: d4f8ad27f2 ("zink: handle implicit sync for dmabufs")
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33952>
2025-03-08 01:38:00 +00:00
Yiwei Zhang e538a38017 lavapipe: fix accel struct device query copy
This change:
1. use vulkan flags instead of pipe query flags
2. set the avail bit when requested

Fixes: a26f96ed3d ("lavapipe: Handle accel struct queries in handle_copy_query_pool_results")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33951>
2025-03-08 01:11:44 +00:00
Yiwei Zhang bc190cab2d lavapipe: set availability bit for accel struct host queries
Fixes: 897ccbd180 ("lavapipe: Implement VK_KHR_acceleration_structure")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33951>
2025-03-08 01:11:44 +00:00
Sagar Ghuge 1bfe2571f5 intel/compiler: Lower sample index into coord for MSRT messages
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/32690>
2025-03-07 23:06:14 +00:00
Sagar Ghuge bea9d79cb9 intel/compiler: Add support for MSAA typed load/store messages
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/32690>
2025-03-07 23:06:14 +00:00
Yiwei Zhang 2923945020 venus: support VK_KHR_ray_tracing_maintenance1
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 7902d9ee32 venus: support VK_KHR_ray_tracing_position_fetch
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 7cfdd9e8d4 venus: enable VK_KHR_ray_tracing_pipeline
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang e081621dca venus: implement VK_KHR_ray_tracing_pipeline commands
Implement all commands involved. No need to scrub anything in the RT
pipeline info since it has been well validated by the VUs related.

The nature of VkDeferredOperationKHR plays well with venus multi-ring
support. So later we can properly define our own concurrent limits for
RT pipeline creations.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 9547efa6b0 venus: prepare push template for ray tracing pipeline
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 1e526c555b venus: support VK_KHR_ray_query
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 0deef3d037 venus: enable VK_KHR_acceleration_structure
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 4cb0f99379 venus: implement VK_KHR_acceleration_structure - Part III
Add acceleration structure descriptor type support:
- async set alloc
- descriptor update with and w/o template
- push descriptor with and w/o template

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 1075d03228 venus: implement VK_KHR_acceleration_structure - Part II
Add query cmd integrations.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang c0f4d9861d venus: implement VK_KHR_acceleration_structure - Part I
Implement all cmds except query updates.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang eb0ad64e80 venus: add stubs for accel struct host commands
They won't be implemented, and the host cmds feature will always be
disabled in venus.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 3eb554a2b5 venus: support VK_KHR_deferred_host_operations
Details are at:
https://gitlab.freedesktop.org/virgl/venus-protocol/-/merge_requests/97

For venus and vkr, we'll use pure driver side impl. Start with (1) and
move towards (2) later as things evolve. We'd like renderer side to be
simple and passive.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 90f9a1cca5 venus: add a debug option for ray tracing support
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
Yiwei Zhang 6bac77b75c venus: sync protocol for ray tracing support
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33907>
2025-03-07 22:44:10 +00:00
sarbes a3f827319f lima: add genxml for texture descriptor
This commit adds genxml to Lima, by copying mostly from Asahi. The definition of the texture descriptor has been moved there.

v2:
- remove mipaddress handling, use "shr(6)" modifier
- indent TD parser 8 spaces

v3:
- added copyright
- renamed ushort to unorm16

Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33385>
2025-03-07 21:10:20 +00:00
Benjamin Lee 871804a494 panfrost: implement 16-bit pack/unpack intrinsics
This significantly improves codegen for the 16-bit ldexp2 lowering.

Signed-off-by: Benjamin Lee <benjamin.lee@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33781>
2025-03-07 18:56:33 +00:00
Juan A. Suarez Romero 2662b9b71d v3d/v3dv/vc4: review all expected timeouts
Some of these tests are not valid anymore, others can now be executed
under the time budget.

This commit updates all the expected timeouts.

Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33913>
2025-03-07 17:37:18 +00:00
Daniel Schürmann 795706f3d6 zink: enable CapabilityDemoteToHelperInvocation and extension only if required
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33492>
2025-03-07 15:44:49 +00:00
Daniel Schürmann efaaaf45d1 zink: clean up HelperInvocation code emission
With SPIR-V 1.6 OpIsHelperInvocationEXT was effectively replaced with
Volatile loads of the HelperInvocation built-in variable.

This patch also drops the distinction between is_helper_invocation and
load_helper_invocation.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33492>
2025-03-07 15:44:49 +00:00
Daniel Schürmann dbd41e3ddd nir: set SYSTEM_VALUE_HELPER_INVOCATION read for nir_intrinsic_is_helper_invocation
is_helper_invocation is the volatile access of load_helper_invocation.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33492>
2025-03-07 15:44:49 +00:00
Daniel Schürmann a4cffa91b8 nir: remove nir_lower_discard_if_to_cf option
Since removing nir_intrinsic_discard{_if} it has no purpose anymore.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33492>
2025-03-07 15:44:49 +00:00
Daniel Schürmann 276dc751ce zink: lower {demote|terminate}_if to CF after lowering point smooth
This must have slipped somehow.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33492>
2025-03-07 15:44:49 +00:00
Erik Faye-Lund fa91b41f54 docs/features: add VK_EXT_hdr_metadata
This was missed when it was recently added.

Fixes: 4b663d561b ("vulkan/wsi: implement support for VK_EXT_hdr_metadata on Wayland")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33918>
2025-03-07 15:34:02 +00:00
Erik Faye-Lund 1a1412e66e docs/features: add missing panvk feature
I forgot to document this feature when I added it, whoops!

Fixes: ac05c2a2b8 ("panvk: expose subgroup operations")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33918>
2025-03-07 15:34:02 +00:00