fran/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
188,419 Commits 63 Branches 889 Tags
mesa-24.1.0-rc3
Commit Graph

188419 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Mike Blumenkrantz
744307289c frontends/dri: plumb an 'implicit' param through screen init 
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28139>
 2024-04-22 23:25:58 +00:00
Mike Blumenkrantz
4742d9bc1a gbm: plumb an 'implicit' param through device creation 
this is always true except in the software fallback

Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28139>
 2024-04-22 23:25:58 +00:00
Mike Blumenkrantz
14c44aacff dri: plumb a 'implicit' param through createNewScreen interfaces 
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28139>
 2024-04-22 23:25:58 +00:00
Mike Blumenkrantz
014bbae4bf glx: pass implicit load param through allocation 
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28139>
 2024-04-22 23:25:58 +00:00
Mike Blumenkrantz
91c757bda1 glx: add an 'implicit' param to createScreen 
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28139>
 2024-04-22 23:25:58 +00:00
Mike Blumenkrantz
1b9ee76369 glx: fix some indentation 
ifdefs are hard

Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28139>
 2024-04-22 23:25:58 +00:00
Mike Blumenkrantz
0e8202cc24 loader: delete unused param from pipe_loader_vk_probe_dri() 
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28139>
 2024-04-22 23:25:58 +00:00
Guilherme Gallo
4b81ee6418 ci/lava: Fix how exception entry in structured log 
Improves the error logging in the LAVA job submitter by capturing and
logging the exception message rather than just the exception type when a
job fails to run.

Additionally, introduces a clearer script interruption
message to aid in debugging and immediate understanding of job
submission failures.

Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28778>
 2024-04-22 21:20:07 +00:00
Guilherme Gallo
e96e25f323 ci/lava: Don't run jobs if the remaining execution time is too short 
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28778>
 2024-04-22 21:20:07 +00:00
Guilherme Gallo
3e33171471 ci/lava: Introduce unretriable exception handling 
This commit refactors the exception hierarchy to differentiate between
retriable and fatal errors in the CI pipeline, specifically within the
LAVA job submission process.

A new base class, `MesaCIRetriableException`, is introduced for
exceptions that should trigger a retry of the CI job, while
`MesaCIFatalException` is added for non-recoverable errors that halt the
process immediately.

Additionally, the logic for deciding whether a job should be retried or
not is updated to check for instances of `MesaCIRetriableException`,
improving the robustness and reliability of the CI job execution
strategy.

Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28778>
 2024-04-22 21:20:07 +00:00
Guilherme Gallo
5363874676 ci/lava: A few formatting cleanups 
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28778>
 2024-04-22 21:20:07 +00:00
Caio Oliveira
13093ceb3c intel/brw: Move validate out of fs_visitor 
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28534>
 2024-04-22 13:38:41 -07:00
Caio Oliveira
671d216f39 intel/brw: Remove two duplicated validate calls in optimizer 
The OPT macro will call validate() after each pass, so both cases
removed by this patch are just redundant calls.  Will only affect
Debug builds since in Release builds validation is a no-op.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28534>
 2024-04-22 13:38:41 -07:00
Caio Oliveira
8a6fe54409 intel/brw: Refactor FS validation macros 
Use `a` and `b` (already identified as that in the output message)
instead of `f` and `s` for the two values being compared, since in
a later patch `s` will be used to hold the fs_visitor shader.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28534>
 2024-04-22 13:38:41 -07:00
Echo J
d184808124 nvk: Don't advertise residencyAlignedMipSize on MaxwellB+ 
DXVK/vkd3d-proton require this feature to be advertised as VK_FALSE for FL12 support:
https://github.com/doitsujin/dxvk/blob/v2.3.1/src/d3d11/d3d11_features.cpp#L305
https://github.com/HansKristian-Work/vkd3d-proton/blob/v2.12/libs/vkd3d/device.c#L7426

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28850>
 2024-04-22 20:11:49 +00:00
Echo J
be940a7dc6 nvk: Use implicit pipeline cache 
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28851>
 2024-04-22 14:37:59 -05:00
Faith Ekstrand
59bba821ef nvk: Hash ycbcr conversions in the descriptor set layout hash 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28851>
 2024-04-22 14:37:59 -05:00
Echo J
0f46e279ba vulkan: Add implicit pipeline caching support 
This mirrors RADV's pipeline behavior (which is more performant
when programs like DXVK don't use the pipeline cache functionality)

Drivers need to set the implicit cache variable to use this though
(the next patch will enable this for NVK)

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28851>
 2024-04-22 14:37:59 -05:00
Eric R. Smith
dae6b6a23d panfrost: fix an incorrect stencil clear optimization 
We track stencil clears and writes to optimize them. Unfortunately, the
code for doing this tracks the whole resource, not individual layers or
levels within the resource, which can result in incorrect output when
different levels or layers are accessed. Modified to optimize only the first
layer/level; this will handle the common case of a single stencil texture
while allowing arrays or mipmaps to still work (albeit slightly slower).

The original optimization was introduced in a2463ec271 ("panfrost:
Constant stencil buffer tracking") but the code has been reformatted
since then, so this change won't apply as-is that far back (although it's
fairly obvious how to apply it by hand).

Fixes: a2463ec271 ("panfrost: Constant stencil value tracking")
Signed-off-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28832>
 2024-04-22 16:43:51 +00:00
Mike Blumenkrantz
e89123ec73 zink: prune some piglit cts fails 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28859>
 2024-04-22 16:16:59 +00:00
Yonggang Luo
bf2df78575 broadcom/common: Now "util/box.h" is under src, so remove the FIXME 
Remove the redundant inc_gallium_aux and inc_gallium

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28854>
 2024-04-22 15:01:34 +00:00
Tomeu Vizoso
ef111f5f07 etnaviv: Don't init the blitter in compute-only contexts 
Otherwise, we hit this assertion:

etna_vertex_elements_state_create: Assertion `buffer_idx < screen->specs.stream_count' failed.

As specs.stream_count can be zero in GPUs that are compute only.

Reviewed-by: Christian Gmeiner <cgmeiner@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28848>
 2024-04-22 14:28:46 +00:00
Samuel Pitoiset
095e3af2b0 radv: add RADV_DEBUG=psocachestats to report per-pipeline cache hits/misses 
This can be useful to make sure precompilation works as expected.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28829>
 2024-04-22 13:54:05 +00:00
Samuel Pitoiset
1f4ee45914 radv: rework pipeline cache search helpers 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28829>
 2024-04-22 13:54:05 +00:00
Samuel Pitoiset
bbe52934b6 radv: use radv_pipeline::sha1 for graphics/compute pipelines 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28829>
 2024-04-22 13:54:05 +00:00
Samuel Pitoiset
6e24da3ad4 radv/rt: move radv_ray_tracing_pipeline::sha1 to radv_pipeline 
To re-use it for graphics/compute pipelines.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28829>
 2024-04-22 13:54:05 +00:00
Samuel Pitoiset
862d53f00a radv/rt: remove unnecessary param to radv_ray_tracing_pipeline_cache_insert() 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28829>
 2024-04-22 13:54:05 +00:00
Robert Mader
6bbf8a08a4 panfrost: Use pipe resource helper 
Like we already do in panfrost_resource_get_param().
While on it, update the later to use pan_resource() again for
consistency. No functional changes intended.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28362>
 2024-04-22 13:28:41 +00:00
Rhys Perry
e0db4b458b radv: cache RT stage info 
If a RT pipeline misses the cache but it's imported stages hit the cache,
then can_inline might be false when it should be true.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28788>
 2024-04-22 11:55:26 +00:00
Joshua Ashton
122455b686 radv: Properly initialize imageCreateFlags in GetPhysicalDeviceVideoFormatPropertiesKHR 
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28841>
 2024-04-22 11:33:39 +00:00
Eric Engestrom
ba32ea6da3 llvmpipe/ci: update expectations after piglit uprev 
Fixes: ec45e8294c ("Uprev Piglit to f7ece74a107a2f99b2f494d978c84f8d51faa703")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28853>
 2024-04-22 10:56:45 +00:00
Eric Engestrom
dc387e000c radeonsi/ci: update vangogh expectations after piglit uprev 
Fixes: ec45e8294c ("Uprev Piglit to f7ece74a107a2f99b2f494d978c84f8d51faa703")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28852>
 2024-04-22 12:14:44 +02:00
M Henning
6b22fff658 nvk: Don't use a descriptor cbuf if it's too large 
This fixes a test on vkd3d-proton commit 836446ce25
VKD3D_TEST_FILTER=test_typed_buffers_many_objects_dxil build/tests/d3d12

Fixes: f1c909edd5 ("nvk/nir: Add cbuf analysis to nvi_nir_lower_descriptors()")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28844>
 2024-04-22 08:45:37 +00:00
M Henning
f6301a7727 nak: Count GLOBAL_SIZE_OFFSET in bytes, not words 
This matches what nvk_mme_dispatch_indirect expects

Fixes: 16f2249a ("nvk: Use the NAK helpers to fill QMDs")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28844>
 2024-04-22 08:45:37 +00:00
Boris Brezillon
34ffa4cd10 nir/lower_blend: Fix nir_blend_logicop() for 8/16-bit integer formats 
src and dst can be integer types, and doing an f2f on such types
messes up with the original value. Make sure we keep the original type
when {up,down}sizing the src, dst and out values.

Fixes: f3de2bd6c2 ("nir: Add blend lowering pass")
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28839>
 2024-04-22 09:58:41 +02:00
Marek Olšák
e5f133ccc4 util: import pipe_box and its helpers 
We'll use them. There are also new helpers that we'll use.

Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28733>
 2024-04-22 01:42:52 +00:00
Marek Olšák
959891e33f util: add new format helpers 
we'll use them

Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28733>
 2024-04-22 01:42:52 +00:00
Gert Wollny
2bb102f020 r600/sfn: Don't put b2f64 conversion into ALU group 
There is no need to pin the ops into channels because
these are 32 bit ops that can be executed independent
from each other.

Fixes: 79ca456b48
     r600/sfn: rewrite NIR backend

v2: grammar fixes (lorn10)

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28840>
 2024-04-21 19:06:02 +00:00
Gert Wollny
07995b98a8 r600/sfn: when emitting fp64 op2 groups pre-load values 
Since the group is created from the onset, we have to make
sure that four or eight src values don't have a readport
conflict, so force a pre-loading of the values to registers
evenly distributed over the channels and let copy-propagation
take care of cleaning up un-neccesary moves.

Fixes: 79ca456b48
   r600/sfn: rewrite NIR backend

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28840>
 2024-04-21 19:06:02 +00:00
Gert Wollny
36c81b5e88 r600/sfn: call nir_lower_doubles explicitely 
Fixes: 5218cff34b
    nir/algebraic: avoid double lowering of some fp64 operations

v2: grammar fixes (lorn10)

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28840>
 2024-04-21 19:06:02 +00:00
Gert Wollny
bf44ce61bb r600/sfn: Use dependecies to order barriers and LDS/RAT instructions 
This gives more freedom to schedule the group barrier and removes
the need to add blocks around a barrier to keep the scheduler in
check. This should avoid emitting some CF instructions.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11002

Fixes: fe881bf097
    r600/sfn: move kill handling fully to scheduling

v2: grammar fixes (lorn10)

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28840>
 2024-04-21 19:06:02 +00:00
Gert Wollny
a61b658d5f r600/sfn: Add array element parent also to array 
This is probably overdoing debendencies in many cases,
but it fixes a bug where scheduling goes wrong.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10984

Fixes: ddb167e81a
  r600/sfn: Handle indirect array load/store dependencies better

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28840>
 2024-04-21 19:06:02 +00:00
Manuel Stoeckl
04f232ed99 util/disk_cache: try getenv(HOME) before getpwuid->pw_dir 
getenv("HOME") is significantly faster than getpwuid_r(...)->pw_dir,
because the latter may require loading NSS libraries, reading
/etc/passwd, etc.

Furthermore, the Linux man pages for getpwuid_r recommend using
getenv("HOME") instead of getpwuid_r, because the user may wish to
change the value of HOME after logging in.

Signed-off-by: Manuel Stoeckl <code@mstoeckl.com>
Reviewed-by: Zoltán Böszörményi <zboszor@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13614>
 2024-04-20 13:43:50 +00:00
Konstantin
c2b1555a57 ac/parse_ib: Always print the value of the whole register 
ac_dump_reg is used by radv for dumping descriptors when writing a hang
report. Including the hex-value of the whole register help correlating
descriptors with SGPR values in the UMR wave dump.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28759>
 2024-04-20 08:29:23 +00:00
Konstantin
f4c6664d4b radv/debug: Dump descriptor binding information 
Can be useful for correlating offsets with descriptors.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28759>
 2024-04-20 08:29:23 +00:00
Konstantin
e27deed2ab radv/debug: Try to find unbound shaders 
Use radv_find_shader to correlate shaders to PCs.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28759>
 2024-04-20 08:29:23 +00:00
Konstantin
23c167f826 radv: Canonicalize addresses in radv_find_shader 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28759>
 2024-04-20 08:29:23 +00:00
Konstantin
48b79a18c5 radv/debug: Canonicalize shader addr 
The most significant bits of the PC are cleared.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28759>
 2024-04-20 08:29:23 +00:00
Konstantin
9778e1b9a0 ac/debug: Handle the output of recent umr versions 
umr was changed to always display the human readable format.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10127
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28759>
 2024-04-20 08:29:22 +00:00
Timur Kristóf
65f5f1559f ac/nir/ngg: Fix packing 16-bit MS outputs. 
Fixes: 9f36fba9be

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28765>
 2024-04-20 09:30:34 +02:00