AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
193,706 Commits 1 Branch 0 Tags
9e7091352c19af6133bfb6f33915c17d85fbe64d
Commit Graph

3117 Commits

Author SHA1 Message Date
Connor Abbott de1d36d054 ci: Uprev VK-CTS to 1.3.9.0 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29766>
 2024-08-15 09:01:26 +00:00
Connor Abbott bc1521e601 ci: Move two failing loader-related tests to all-skips.txt 
There's no value testing these tests in CI until the loader is upgraded,
so don't force every driver to add them to their fails list.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29766>
 2024-08-15 09:01:26 +00:00
Iago Toral Quiroga ad9ff707ce broadcom: drop backend implementation of nir_op_ufind_msb 
We can have NIR do this for us now that we have uclz.

Suggested by Georg Lehmann.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30614>
 2024-08-13 13:16:18 +02:00
Iago Toral Quiroga 35a10f5d5a broadcom: implement nir_op_uclz 
This enables some algebraic optimizations.

No changes in shader-db, but it does cause some CTS tests to
produce less instructions.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30614>
 2024-08-13 13:16:11 +02:00
Alyssa Rosenzweig c3d999dec9 broadcom: switch to derivative intrinsics 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30570>
 2024-08-09 13:54:11 +00:00
Zan Dobersek 7fd5f76393 nir/lower_vars_to_scratch: calculate threshold-limited variable size separately 
ir3's lowering of variables to scratch memory has to treat 8-bit values as
16-bit ones when comparing such value's size against the given threshold
since those values are handled through 16-bit half-registers. But those
values can still use natural 8-bit size and alignment for storing inside
scratch memory.

nir_lower_vars_to_scratch now accepts two size-and-alignment functions,
one used for calculating the variable size and the other for calculating
the size and alignment needed for storing inside scratch memory. Non-ir3
uses of this pass can just duplicate the currently-used function. ir3
provides a separate variable-size function that special-cases 8-bit types.

Signed-off-by: Zan Dobersek <zdobersek@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29875>
 2024-08-07 14:32:28 +00:00
Iago Toral Quiroga 086ed1e54b broadcom/compiler: emit instructions producing flags earlier 
We usually emit flags right before consuming them but this is
suboptimal from the point of view of register pressure: if an
instruction is only used to generate flags then waiting to emit
it right before reading the flags extends the liveness of the
sources used to generate the flags for no gain. This pass will
check for such instructions and try to move them as early as
possible.

Shader-db results below show this is effective to reduce register
pressure, allowing a few shaders to increase thread counts and/or
reduce spilling:

total instructions in shared programs: 11057173 -> 11057076 (<.01%)
instructions in affected programs: 1955543 -> 1955446 (<.01%)
helped: 4214
HURT: 3905
Inconclusive result (value mean confidence interval includes 0).

total threads in shared programs: 425096 -> 425170 (0.02%)
threads in affected programs: 74 -> 148 (100.00%)
helped: 37
HURT: 0
Threads are helped.

total uniforms in shared programs: 3846275 -> 3845674 (-0.02%)
uniforms in affected programs: 23574 -> 22973 (-2.55%)
helped: 217
HURT: 30
Uniforms are helped.

total max-temps in shared programs: 2222910 -> 2220488 (-0.11%)
max-temps in affected programs: 61904 -> 59482 (-3.91%)
helped: 2145
HURT: 113
Max-temps are helped.

total spills in shared programs: 4294 -> 4280 (-0.33%)
spills in affected programs: 148 -> 134 (-9.46%)
helped: 8
HURT: 0

total fills in shared programs: 6497 -> 6468 (-0.45%)
fills in affected programs: 291 -> 262 (-9.97%)
helped: 8
HURT: 0

total sfu-stalls in shared programs: 14344 -> 14611 (1.86%)
sfu-stalls in affected programs: 1308 -> 1575 (20.41%)
helped: 217
HURT: 335
Inconclusive result (%-change mean confidence interval includes 0).

total inst-and-stalls in shared programs: 11071517 -> 11071687 (<.01%)
inst-and-stalls in affected programs: 1946767 -> 1946937 (<.01%)
helped: 4191
HURT: 3909
Inconclusive result (value mean confidence interval includes 0).

total nops in shared programs: 270628 -> 269829 (-0.30%)
nops in affected programs: 22032 -> 21233 (-3.63%)
helped: 1213
HURT: 571
Inconclusive result (%-change mean confidence interval includes 0).

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30511>
 2024-08-07 09:28:39 +02:00
Alejandro Piñeiro 28df772935 v3dv: expose Vulkan 1.3 
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10896

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29476>
 2024-08-05 15:10:24 +00:00
Alejandro Piñeiro 3ae4da4ea7 v3dv: update conformanceVersion to 1.3.8.3 
As we used as reference to check if we fully pass Vulkan 1.3

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29476>
 2024-08-05 15:10:24 +00:00
Alejandro Piñeiro 781622ba5b v3dv/device: set value for maxInlineUniformTotalSize 
Fixes the following test, as far as you enable Vulkan 1.3 (if not it
is skipped):

dEQP-VK.api.info.vulkan1p3_limits_validation.max_inline_uniform_total_size

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29476>
 2024-08-05 15:10:24 +00:00
Eric Engestrom f427c9fe23 rpi3/ci: remove {version,render}_check from [piglit.env] 
v1 was moving it into the [[piglit]] section, but piglit-runner actually
doesn't support checking for this, so let's just remove it entirely.

Fixes: 3990463c48 ("v3d/vc4/ci: set full renderer version check")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30436>
 2024-08-01 21:05:27 +00:00
Juan A. Suarez Romero 14c82207a4 broadcom/ci: use kernel8 for rpi5 full jobs 
Test if using kernel8 instead of specific kernel_2712 fixes the issues
with rpi5.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30457>
 2024-08-01 07:12:24 +02:00
Juan A. Suarez Romero 52e7f2ca08 v3d/ci: update number of rpi5 available for testing 
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30438>
 2024-07-31 11:50:47 +00:00
Juan A. Suarez Romero fa0a8b8d2c v3d/ci: add new flakes in rpi5 
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30438>
 2024-07-31 11:50:47 +00:00
Juan A. Suarez Romero 07cd860b52 broadcom/ci: annotate some of the failures 
We already know the reason of some of the failures. Let's include the
reason.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30438>
 2024-07-31 11:50:47 +00:00
Eric Engestrom 38ede4334f rpi/ci: shorten the boot timeout to 90 seconds 
5 minutes is too long and doesn't leave enough time after the retry to
still complete the job, especially if two boots fail in a row, removing
10 out of the worst-case 20 minutes allowed.

90 seconds is more than enough, the boot output normally starts after
~40-45 seconds.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30409>
 2024-07-31 09:01:32 +02:00
Eric Engestrom 1e5fb15afa ci/poe-powered: rename --test-timeout to --test-timeout-minutes to be more explicit 
Due to how python's argparse works, anyone anywhere else that would
continue passing `--test-timeout` would not see any change, it will
transparently work for them too.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30409>
 2024-07-31 08:58:20 +02:00
Juan A. Suarez Romero 9b1302f80a v3d/ci: enable full run jobs for rpi5 
Now that there are more rpi5 devices in the CI, run all the GL/GLES
tests and CL tests nightly.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30349>
 2024-07-30 15:50:30 +00:00
Juan A. Suarez Romero 3c61489a3e v3d/vc4/ci: add address sanitizer jobs 
This runs part of GL and Vulkan tests in vc4/v3d/v3dv with the address
sanitizer enabled to detect memory issues and leaks.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30349>
 2024-07-30 15:50:30 +00:00
Eric Engestrom ff58f0afbe rpi4/ci: mark fs variant of arb_texture_buffer_object as flaky too 
See eg.:
https://gitlab.freedesktop.org/mesa/mesa/-/jobs/61482695
https://gitlab.freedesktop.org/mesa/mesa/-/jobs/61594844

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30404>
 2024-07-29 10:06:10 +02:00
Eric Engestrom 79389b9f6f v3d/ci: mark glx@glx-visuals-depth as flaky on rpi4, just like the -stencil variant 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30387>
 2024-07-26 20:18:12 +00:00
Eric Engestrom 1337e00a89 v3d/ci: mark spec@amd_performance_monitor@api as flaky on rpi4 
It's been failing a lot in the last few days, blocking a bunch of MRs.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30385>
 2024-07-26 19:31:56 +02:00
Eric Engestrom f1f8c465d5 vc4,v3d,v3dv: avoid compiling in unused sim_file field in the {vc4,v3d}_screen/v3dv_physical_device struct 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30287>
 2024-07-24 19:11:24 +00:00
Eric Engestrom eae740f2e4 v3d,v3dv: simplify USE_V3D_SIMULATOR/using_v3d_simulator logic 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30287>
 2024-07-24 19:11:24 +00:00
Eric Engestrom 5577078ac1 v3d,v3dv: figure out whether we're using the simulator only once 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30287>
 2024-07-24 19:11:23 +00:00
Eric Engestrom f90d1182e0 v3d,v3dv: reuse dep_v3d_hw from simulator/meson.build 
broadcom/simulator/ gets parsed before broadcom/vulkan/ or
gallium/drivers/v3d/ so this is safe.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30287>
 2024-07-24 19:11:23 +00:00
Juan A. Suarez Romero c30e5d44b1 v3d/ci: add new flakes 
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30337>
 2024-07-24 09:19:05 +02:00
Juan A. Suarez Romero 559eaaf0e6 v3d/ci: add failures from piglit all profile 
While we don't run this profile in the CI, our expected lists already
have some tests from old days when we run the all profile.

So let's update the list with new failures from this all profile,
allowing us to have a more comprehensive list of failing tests for our
drivers.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30318>
 2024-07-23 16:52:13 +00:00
Karol Herbst bad67ee77c spirv: handle function parameters passed by value 
Cc: mesa-stable
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29896>
 2024-07-22 21:16:58 +00:00
Eric Engestrom 547de1e928 v3d/ci: mark spec@amd_performance_monitor@vc4 tests as flaky 
Turns out it was not fixed, it just happened to pass a bunch of times in
a row, but it actually fails randomly, so mark it as such.

Fixes: 4696e9c49b ("v3d/ci: mark spec@amd_performance_monitor@vc4 tests as fixed")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30290>
 2024-07-22 10:56:09 +02:00
Daniel Stone e05415a82e format: Generate endian-independent format aliases 
Instead of having a hardcoded list of endian-independent format aliases
in the header, generate them from the format definitions.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29649>
 2024-07-19 13:50:42 +00:00
Eric Engestrom 4696e9c49b v3d/ci: mark spec@amd_performance_monitor@vc4 tests as fixed 
Fixed by a commit in the range 452fed52..decc040a, but no obvious
candidate stands out.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30259>
 2024-07-19 09:11:16 +00:00
Juan A. Suarez Romero 433a0422d5 v3dv: don't leak cache key 
The hashtable stores a hash generated from the key, so there is no need
to duplicate the key, as otherwise it is a leak.

Found through address sanitizer.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30204>
 2024-07-18 11:49:07 +00:00
Juan A. Suarez Romero 2d54a605fe v3dv: free temp image created when copying with blit 
This fixes a leak that happens when copying a image using blit, and the
image is a compressed one.

In this case a new image view is created that can be re-interpreted to
perform the copy, but was not properly free.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30161>
 2024-07-15 11:14:29 +00:00
Juan A. Suarez Romero f8623ea7da v3dv: adversise VK_EXT_depth_clamp_zero_one 
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30148>
 2024-07-15 09:17:40 +00:00
Eric Engestrom ce7e1e0f7a v3d/ci: add disabled job for CL testing on the RPi5 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30076>
 2024-07-15 08:29:36 +00:00
Eric Engestrom 7c40c35b97 v3d/ci: rename "rusticl on v3d" suite to v3d-rusticl 
There's nothing rpi4-specific in there, it just tests rusticl on v3d,
and the next commit will do exactly that on the rpi5.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30076>
 2024-07-15 08:29:36 +00:00
Eric Engestrom beb05d14ae v3d/ci: add disabled job for GL testing on the RPi5 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30076>
 2024-07-15 08:29:36 +00:00
Eric Engestrom 29c4961b53 v3d/ci: include results of CL run in expectations 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30134>
 2024-07-12 10:12:38 +00:00
Eric Engestrom 10af395f72 v3d/ci: include results of GL full run in expectations 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30134>
 2024-07-12 10:12:38 +00:00
Juan A. Suarez Romero 795b3f83ff v3d/ci: update expected list 
Add new failures, and update comments.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30117>
 2024-07-11 09:29:39 +02:00
Juan A. Suarez Romero 2c74872bbc broadcom/ci: update traces for rpi4 
Mainly mark some traces as flake.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30087>
 2024-07-10 10:31:24 +00:00
Juan A. Suarez Romero 4bb564f40d broadcom/ci: add more jobs to test with rpi5 
Now that there are more rpi5 in the CI, let's add pre-merge jobs.

This also restrict the nightly job to be executed by devices allocated
to run full runs.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30087>
 2024-07-10 10:31:24 +00:00
Iago Toral Quiroga bb63b7b802 v3dv: don't lower fsat on V3D 7.x 
This requires that our nir options are different across V3D versions
so we can't use a static global any more.

Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
 2024-07-10 08:30:21 +02:00
Iago Toral Quiroga 5b1e88760a v3dv: make nir helpers receive nir compiler options from caller 
We are about to make a change that will make compiler options
depend on v3d version, so helpers would usually need additional
parameters to retrieve them. Isntead of doing this, we will make
the callers get the options instead and provide them to the
helpers.

Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
 2024-07-10 08:30:15 +02:00
Iago Toral Quiroga 33187012ab broadcom/compiler: implement nir_op_fsat 
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
 2024-07-10 08:29:59 +02:00
Iago Toral Quiroga d62082a131 broadcom/compiler: disallow copy propagation of FMOV exclusive modifiers 
Since .sat, .nsat and .max0 are only supported with FMOV we can't copy
propagate an FMOV with any of these unpack modifiers into a different
opcode.

Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
 2024-07-10 08:29:50 +02:00
Iago Toral Quiroga fa959c2993 broadcom/compiler: add new float32 unpack modifiers in V3D 7.x 
These are only supported with FMOV.

Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30086>
 2024-07-10 08:29:40 +02:00
Juan A. Suarez Romero 4581bf595b broadcom: follow version naming convention 
We usually name the functions that depend on hardware version as
v3d<version>_foo.

Keep the same convention in QPU and lower_image_load_store, so it makes
easier when searching for versioned functions.

Acked-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30000>
 2024-07-08 11:19:31 +00:00
Juan A. Suarez Romero a10957adb6 broadcom/qpu: clean all versions not supported 
Right now we only support V3D 4.2 and V3D 7.1, so clean older versions
that were left in the QPU.

Acked-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30000>
 2024-07-08 11:19:30 +00:00