d57611fe25
intel/isl: Add scores for GEN12_RC_CCS and MTL_RC_CCS
...
Now that these CCS-enabled modifiers have non-zero scores, anv is
enabled to use them.
We found this to improve the performance of Borderlands 3 by 18.73%.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6701
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Tested-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
9e402e93d2
anv: Delete implicit CCS code
...
Stop allocating CCS at the end of some BOs. Anv no longer uses that
memory range.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
4cdd3178fb
anv: Meet CCS alignment reqs with dedicated allocs
...
At image bind time, we require BOs to meet aux-map alignment
requirements in order to enable CCS on images. This is a heuristic
controlled by anv_bo_allows_aux_map().
To improve the chances of getting a properly aligned BO, we make use of
the dedicated allocation extension. Firstly, we report to applications a
preference for dedicated memory if an image would like to use the aux
map. Secondly, we align the VMA for dedicated allocations to meet
aux-map requirements.
To make enabling modifiers much easier on integrated gfx12, report
dedicated allocations as a requirement for modifiers which specify CCS.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com > (v1)
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
2cbec81041
anv: Loosen anv_bo_allows_aux_map
...
Instead of requiring that a BO has the has_implicit_ccs flag set, simply
require that the BO is aligned according to aux-map requirements.
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
ee6e2bc4a3
anv: Place images into the aux-map when safe to do so
...
At image bind time, if an image's addresses can be placed into the
aux-map without causing conflicts with a pre-existing mapping, do so.
The code aux management code in the binding function operates on a
per-plane basis. So, use the per-plane CCS memory range from the image
rather than the CCS memory region for the entire BO.
Another way to avoid aux-map conflicts is to rely solely on having a
dedicated allocation for an image. Unfortunately, not all workloads
change their behavior when drivers report a preference for dedicated
allocations. In particular, 3DMark Wild Life Extreme does not make more
dedicated allocations and such a solution was measured to perform ~16%
worse than this solution. With this solution, I did not measure a loss
of CCS on that benchmark.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6304
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com > (v1)
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
207db22117
anv: Refactor CCS disabling at image bind time
...
Split out the discrete and integrated implicit CCS cases. We'll do more
work in the integrated case in a future commit.
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
d31c62f384
anv: Wrap aux surface image binding queries
...
Add and use anv_image_get_aux_memory_range.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
cd12eec496
anv: Allocate space for aux-map CCS in image bindings
...
This makes images a bit larger by reserving space to store the
compression control surface when the device uses an aux-map.
This space is not used currently because anv still maps main surface
addresses to space at the end of the anv_bo.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
5e07255148
anv: Move scope of CCS binding determination
...
Move the determination of the image binding for CCS to a larger scope,
so that it can be reused for other aux usages in
add_aux_surface_if_supported().
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
b1a14fe923
intel: Return a bool from intel_aux_map_add_mapping
...
Make intel_aux_map_add_mapping return false if a mapping is attempted
that would conflict with an existing one. If this function doesn't
return false, it will either fail to return or return true.
The Vulkan driver will make use of this feature to opportunistically
enable CCS if a BO's VMA range has not been already mapped.
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25003 >
2023-10-23 21:37:24 +00:00
16e1163651
radv/rmv: Recognize LPDDR memory
...
Fixes: 8d0e6c02https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25845 >
2023-10-23 18:08:02 +00:00
5e734a7cbf
radv/rra: Recognize LPDDR memory
...
Also move away from using unnamed constants for memory types.
Fixes: aef7ea868fhttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25845 >
2023-10-23 18:08:02 +00:00
7a3fb60ac8
docs/ci: Add some links in the CI docs to how to track job flakes
...
and also figuring out how many boards are available for sharding
management.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25806 >
2023-10-23 17:59:55 +00:00
553070f993
docs/ci: Drop old instructions for farm disabling
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25806 >
2023-10-23 17:59:55 +00:00
2ef0e6be17
util: Remove MESA_TRACE_BEGIN/END
...
There are no user of these macros anymore, simplify the cpu_trace header.
Signed-off-by: Corentin Noël <corentin.noel@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25308 >
2023-10-23 17:20:03 +00:00
d64765c200
util/perf: Remove the tracing categories
...
The slow category remains unused in Mesa currently, readd it once we have
a use for it.
Signed-off-by: Corentin Noël <corentin.noel@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25308 >
2023-10-23 17:20:03 +00:00
b7a4e78071
util: Avoid the use of MESA_TRACE_BEGIN/END
...
Switch the last occurrences of it to the MESA_TRACE_SCOPE macro which is
easier to use.
Signed-off-by: Corentin Noël <corentin.noel@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25308 >
2023-10-23 17:20:03 +00:00
bf1a68833b
venus: Change the only occurrence of VN_TRACE_BEGIN/END to VN_TRACE_SCOPE
...
This is the only part of venus using this macro, let's replace it with the scope
one.
Signed-off-by: Corentin Noël <corentin.noel@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25308 >
2023-10-23 17:20:03 +00:00
c637901392
aux/tc: Use MESA_TRACE_SCOPE instead of _BEGIN/_END
...
The MESA_TRACE_SCOPE macro is an exact replacement for this.
Signed-off-by: Corentin Noël <corentin.noel@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25308 >
2023-10-23 17:20:03 +00:00
57c6e79db5
tu: Use MESA_TRACE_SCOPE instead of _BEGIN/_END
...
The MESA_TRACE_SCOPE macro is an exact replacement for this.
Signed-off-by: Corentin Noël <corentin.noel@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25308 >
2023-10-23 17:20:03 +00:00
f60a9d89c1
freedreno/drm/virtio: Use MESA_TRACE_SCOPE instead of _BEGIN/_END
...
The MESA_TRACE_SCOPE macro is an exact replacement for this.
Signed-off-by: Corentin Noël <corentin.noel@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25308 >
2023-10-23 17:20:03 +00:00
e0301659f1
zink: fix zink_destroy_screen for early screen creation fails
...
Fixes crashes on at least my system with multiple GPUs.
Fixes: 0c2045553fkherbst@redhat.com >
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25844 >
2023-10-23 16:52:15 +00:00
90cd5f3799
etnaviv: Skip 'empty' cmd streams
...
Make use of etna_cmd_stream_mark_end_of_context_init(..) to mark the
end of context init states. If the cmd stream is untouched until
etna_cmd_stream_flush(..) gets called the drm layer will downgrade this
specific cmd stream to a noop.
This safes us from calling into the kernel with an 'empty' cmd stream.
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com >
Reviewed-by: Lucas Stach <l.stach@pengutronix.de >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25849 >
2023-10-23 14:56:48 +00:00
4c8fad4fd8
etnaviv: drm: Be able to mark end of context init
...
Add etna_cmd_stream_mark_end_of_context_init(..) which stores the current
offset of the cmd stream as offset_end_of_context_init. This information
will be used to downgrade a cmd stream flush to a noop.
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com >
Reviewed-by: Lucas Stach <l.stach@pengutronix.de >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25849 >
2023-10-23 14:56:48 +00:00
2c62fdc8a7
pvr: Only setup the bgobj to load if we have a load_op
...
We only need to setup the background object if we have a load_op,
to perform the loads/clears.
This fixes segfaults for cases where we don't have a load_op.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com >
Reviewed-by: Frank Binns <frank.binns@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25703 >
2023-10-23 08:18:33 +00:00
fc62df19b7
ci/b2c: change artifacts path to match baremetal and LAVA
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25815 >
2023-10-22 21:23:01 +00:00
9f4f52f4f3
ci/ci_run_n_monitor: keep monitoring if a job is still running
...
When running multiple targets, the script was returning if any of the
targets failed.
But sometimes some of the targets are still running and we want to
monitor it until it finishes, even if some of the targets failed.
So check if any targeted jobs are in pending or running before deciding
to finish.
Signed-off-by: Helen Koike <helen.koike@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25826 >
2023-10-22 09:30:56 +00:00
59ee87b606
ci/freedreno: disable Adreno 660 Vulkan pre-merge
...
No need to move it to manual/nightly - we already run full runs there.
Revert when Adreno 660 gets stable again!
Reported-by: Helen Koike <helen.koike@collabora.com >
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25839 >
2023-10-22 10:49:10 +02:00
db5fc29116
radeonsi: fixes compilaton error when perfetto is disabled
...
Signed-off-by: Saroj Kumar <saroj.kumar@amd.com >
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25827 >
2023-10-22 05:43:59 +00:00
5d87cd4e51
ci/freedreno: fix copy paste causing a618_gl being run only in manual pipeline
...
Fixes: 9d442b459aCloses : #9992
Acked-by: Guilherme Gallo <guilherme.gallo@collabora.com >
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25816 >
2023-10-21 15:36:16 +02:00
f4e7aefe07
svga: Unify gmr and mob surface pool managers
...
The svga driver contains several buffer managers with gmr* and mob* prefixes,
but they are of similar functionality. The underlying vmwgfx kernel driver
allocates memory of one type - mobs if supported by the virtual hardware,
and gmrs otherwise - and returns an opaque pointer to userspace, so several
managers are reduntant.
This patch reduces the number of buffer managers by unifying similar managers
and performing refactors where necessary.
Signed-off-by: Maaz Mombasawala <mombasawalam@vmware.com >
Reviewed-by: Charmaine Lee <charmainel@vmware.com >
Reviewed-by: Zack Rusin <zackr@vmware.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25832 >
2023-10-20 22:01:04 +00:00
354f0b5b02
svga: Make surfaces shareable at creation.
...
There is no need to have a separate shareable pool of surfaces, we can make
all surfaces shareable.
This change makes all surfaces shareable when creating them.
Signed-off-by: Maaz Mombasawala <mombasawalam@vmware.com >
Reviewed-by: Charmaine Lee <charmainel@vmware.com >
Reviewed-by: Martin Krastev <krastevm@vmware.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25832 >
2023-10-20 22:01:04 +00:00
b0100d23fb
ci/venus: mark more flaky tests after recent cts uprev
...
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25799 >
2023-10-20 21:33:22 +00:00
46fc9decad
ci/marge_queue: add missing python-dateutils to requirements.txt
...
marge_queue.py requires python-dateutils to run, so add it.
Signed-off-by: Helen Koike <helen.koike@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25828 >
2023-10-20 18:27:44 +00:00
f8753a542b
mesa: add GL_APPLE_sync support
...
This extension is a backport of GL_ARB_sync to GLES 1.1 and 2.0
and reuses token IDs and entry point prototypes from that extension,
so adding support is pretty trivial.
Signed-off-by: Lucas Stach <dev@lynxeye.de >
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Reviewed-by: Adam Jackson <ajax@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25789 >
2023-10-20 17:20:11 +00:00
c3516861e8
zink: properly alias shared memory
...
Signed-off-by: Karol Herbst <kherbst@redhat.com >
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25763 >
2023-10-20 18:49:47 +02:00
389b0ce604
zink: wrap shared memory blocks in a struct
...
Signed-off-by: Karol Herbst <kherbst@redhat.com >
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25763 >
2023-10-20 18:49:38 +02:00
2c1dff3851
frontends/va: Fix locking in vlVaBeginPicture
...
The assert in vlVaSetSurfaceContext would very rarely fail because
the mutex was already unlocked when calling this function from
vlVaBeginPicture.
Keep the lock until returning from vlVaBeginPicture, as that's what
other functions are already doing.
Reviewed-by: Leo Liu <leo.liu@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25823 >
2023-10-20 15:14:41 +00:00
c70687afe7
pvr: Implement VK_EXT_host_query_reset
...
Signed-off-by: Vlad Schiller <vlad-radu.schiller@imgtec.com >
Reviewed-by: Matt Coster <matt.coster@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25783 >
2023-10-20 14:58:12 +00:00
649ebbb0fb
pvr: Implement VK_EXT_texel_buffer_alignment
...
This commit will implement the VK_EXT_texel_buffer_alignment
extension and add the texture_baseaddress_byte_aligned feature.
Signed-off-by: Vlad Schiller <vlad-radu.schiller@imgtec.com >
Reviewed-by: Matt Coster <matt.coster@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25781 >
2023-10-20 14:42:05 +00:00
ea59d61294
pvr: Enable VK_KHR_bind_memory2 extension
...
The two functions 'vkBindBufferMemory2' and 'vkBindImageMemory2' are aleady
implemented, and it seems that the flag does not need to be enabled for the
tests to pass.
Signed-off-by: Vlad Schiller <vlad-radu.schiller@imgtec.com >
Reviewed-by: Matt Coster <matt.coster@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25779 >
2023-10-20 14:26:36 +00:00
1f056f229a
pvr: Implement VK_KHR_external_semaphore
...
In order for the tests to pass, this commit also enables
the VK_KHR_external_semaphore_fd extension.
Signed-off-by: Vlad Schiller <vlad-radu.schiller@imgtec.com >
Reviewed-by: Matt Coster <matt.coster@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25784 >
2023-10-20 13:58:07 +00:00
89f9724b1a
pvr: Implement VK_KHR_external_fence
...
In order for the tests to pass, this commit also enables
the VK_KHR_external_fence_fd extension.
Signed-off-by: Vlad Schiller <vlad-radu.schiller@imgtec.com >
Reviewed-by: Matt Coster <matt.coster@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25784 >
2023-10-20 13:58:06 +00:00
6f3dfba803
anv: document the draw indirect optimization ring mode
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Tested-by: Felix DeGrood <felix.j.degrood@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25361 >
2023-10-20 13:07:53 +00:00
454870dd5f
anv: merge gfx9/11 indirect draw generation shaders
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Tested-by: Felix DeGrood <felix.j.degrood@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25361 >
2023-10-20 13:07:53 +00:00
11b4c23d19
anv: add ring buffer mode to generated draw optimization
...
When the number of draw calls is very large, instead of allocating
large amounts of batch buffer space for the draws, use a ring buffer
and process the draw calls by batches.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8645
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Tested-by: Felix DeGrood <felix.j.degrood@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25361 >
2023-10-20 13:07:53 +00:00
718e77eee5
anv: index indirect data buffer with absolute offset
...
This will help for a follow up change where we will respawn the shader
multiple times in a loop and the base offset will be edited by the
shader itself.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Tested-by: Felix DeGrood <felix.j.degrood@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25361 >
2023-10-20 13:07:53 +00:00
db335d9b73
anv: factor out host/gpu internal shaders interfaces
...
This will prevent host/gpu structure definitions to go out of sync.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Tested-by: Felix DeGrood <felix.j.degrood@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25361 >
2023-10-20 13:07:53 +00:00
da1e39faa4
util/glsl2spirv: add ability to pass defines
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Tested-by: Felix DeGrood <felix.j.degrood@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25361 >
2023-10-20 13:07:53 +00:00
c700d47c56
anv: move generation batch fields to a sub-struct
...
Just tyding things a bit since we're about to add more.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ivan Briano <ivan.briano@intel.com >
Tested-by: Felix DeGrood <felix.j.degrood@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25361 >
2023-10-20 13:07:53 +00:00
Nanley Chery