Eric Engestrom
d3e96aa02e
ci_run_n_monitor: dependency jobs must always be started
...
Fixes: 6b49b477ac
("ci/ci_run_n_monitor: simplify enable/cancel logic in monitor_pipeline()")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25715 >
2023-10-13 18:37:56 +00:00
Samuel Pitoiset
0cbaf6cc8d
zink/ci: remove expected failures that are skipped for RADV
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25711 >
2023-10-13 17:02:30 +00:00
Erik Faye-Lund
5fe5c3e223
meson: add wayland-protocols from meson wrapdb
...
Sometimes, users don't have a recent enough version of wayland-protocols
to build Mesa. But these days, Meson's WrapDB has a wrap that's more
than new enough for our needs.
If we add the wrap-file to the subprojects folder, we'll download a more
recent wrap if the installed version is too old (or doesn't exist at
all).
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25683 >
2023-10-13 16:44:31 +00:00
Gert Wollny
7953e14878
r600: drop egcm_load_index_reg
...
This is now handled in SFN.
v2: remove obsolte comments (Vitaliy Kuzmin)
Signed-off-by: Gert Wollny <gert.wollny@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25475 >
2023-10-13 16:28:11 +00:00
Gert Wollny
76993af858
r600/sfn: don't remove texture sources by using the enum value
...
We have to query the index first, otherwise we remove the wrong value.
Fixes: 02bb506c54
r600/sfn: Lower tex,txl,txb and txf to backend
Signed-off-by: Gert Wollny <gert.wollny@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25475 >
2023-10-13 16:28:11 +00:00
Martin Roukala (né Peres)
d6c6599a49
zink/ci: tighten the zink-radv-vangogh timeouts
...
The jobs should never take longer than 30 minutes, so let's enforce it!
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25704 >
2023-10-13 15:59:18 +00:00
Martin Roukala (né Peres)
f5cf90fbea
radv/ci: tighten the vkcts-navi21 timeouts
...
The jobs should never take longer than 30 minutes, so let's enforce it!
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25704 >
2023-10-13 15:59:18 +00:00
Erik Faye-Lund
a740a9f1dc
ci/etnaviv: move failure to flake
...
Turns out, this passes sometimes... So let's mark it as a flake
instead.
Fixes: 7368a89752
("ci/etnaviv: update ci expectation")
Reviewed-by: Christian Gmeiner <cgmeiner@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25699 >
2023-10-13 15:41:20 +00:00
Matt Coster
1f112abd7f
pvr: Use common physical device properties
...
Make use of the common vulkan properties code introduced in [1].
[1]: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24575
Signed-off-by: Matt Coster <matt.coster@imgtec.com >
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25326 >
2023-10-13 12:21:18 +00:00
Matt Coster
989e5e4c70
pvr: Minor refactor of pvr_device.c
...
Moving a few functions further up here to prepare for the next commit;
should make the diffs a lot nicer. No (intentional) functional changes.
Signed-off-by: Matt Coster <matt.coster@imgtec.com >
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25326 >
2023-10-13 12:21:18 +00:00
Matt Coster
6046f735b9
pvr: Don't pass pvr_physical_device when only device info is needed
...
Signed-off-by: Matt Coster <matt.coster@imgtec.com >
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25326 >
2023-10-13 12:21:18 +00:00
Erico Nunes
eb34a86b3b
Revert "ci/lima: farm is down, disable for now"
...
This reverts commit c7806daf43
.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25702 >
2023-10-13 11:52:42 +00:00
cheyang
cf22954973
isaspec : fix isaspec build error in aosp
...
in Android 12 build have error "ninja:
'external/mesa/src/compiler/isaspec/README.rst', needed by
'out/target/product/s/obj/MESON_MESA3D_GEN/.timestamp', missing and
no known rule to make it" because commit:
d48d8aefdf
(docs: Move isaspec out of
drivers/freedreno) modify isaspec.rst Location.
Signed-off-by: cheyang <cheyang@bytedance.com >
Reviewed-by: Christian Gmeiner <cgmeiner@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25697 >
2023-10-13 11:09:40 +00:00
Karol Herbst
65b663cc9d
rusticl/kernel: Fix creation from programs not built for every device
...
OpenCL does not require that a kernel is created for every device. So we
shouldn't assume there is a build for every device.
API validation around launching kernels already takes this possibility
into account.
I did not verify if the commit below is actually the culprit and whether
this bug existed before that, but a fix for older code also would have to
look differently anyway.
Fixes: 323dcbb4b5
("rusticl: Move NirKernelBuild to ProgramDevBuild")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9968
Signed-off-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25696 >
2023-10-13 08:51:30 +00:00
Danylo Piliaiev
734bbe33cf
freedreno/rddecompiler: Decompile repeated IBs
...
Otherwise we don't reconstruct the whole cmdstream.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25677 >
2023-10-13 08:34:41 +00:00
Danylo Piliaiev
0338b14657
freedreno/rddecompiler: Use fd_dev_gen to pass gpu_id to ir3 disasm
...
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25677 >
2023-10-13 08:34:41 +00:00
Eric Engestrom
e41bc40571
include/dri_interface.h: restore define mistakenly removed in !25587
...
This file is a public API used by Xserver; removing something from it
means Xserver can't build anymore, and even if we fix Xserver, old
versions are still around and we want to keep them working to allow for
bisecting issues.
Fixes: 7301914755
("dri: Remove __driDriverExtensions leftovers")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9976
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25692 >
2023-10-13 06:11:02 +00:00
Lionel Landwerlin
3f973a4f45
Revert "intel/fs: limit register flag interaction of FIND_*LIVE_CHANNEL"
...
This reverts commit c9739e8912
.
We don't have a full understanding of what is going on but reverting
definitely fixes a hang.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: c9739e8912
("intel/fs: limit register flag interaction of FIND_*LIVE_CHANNEL")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9868
Tested-By: Valentin Geyer <trayshar@t-online.de >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25563 >
2023-10-13 08:37:28 +03:00
José Roberto de Souza
07eede0970
intel: Prepare implementation of Wa_18019816803 and Wa_16013994831 for future platforms
...
Those workarounds are temporary for newer platforms so we can't use
INTEL_NEEDS_WA_*, luckly those already had runtime checks.
INTEL_NEEDS_WA_* was only used because it was accessing instructions
or fields of the instructions that only exists in gfx12 or gfx125.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25685 >
2023-10-13 03:54:50 +00:00
Alyssa Rosenzweig
be0ab37bac
nir/opt_algebraic: Optimize LLVM booleans
...
Helps OpenCL kernels.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25687 >
2023-10-13 02:55:48 +00:00
Jordan Justen
c74a578c54
intel/dev: Add 0x56ba-0x56bd DG2 PCI IDs
...
Cc: mesa-stable
Ref: https://lists.freedesktop.org/archives/intel-gfx/2023-October/337287.html
Ref: BSpec 44477
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25673 >
2023-10-13 02:16:11 +00:00
Jordan Justen
ee482ad660
anv/batch: Assert that extend_cb is non-NULL if the batch is out of space
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25672 >
2023-10-13 01:46:58 +00:00
Jordan Justen
ef8dcb0aa4
anv/batch: Check if batch already has an error in anv_queue_submit_simple_batch()
...
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25672 >
2023-10-13 01:46:58 +00:00
Emma Anholt
534511935d
ci/radeonsi: Drop an xfail for vangogh.
...
It's passed in the last 3 nightly runs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25693 >
2023-10-13 01:12:59 +00:00
Emma Anholt
01a1975fce
ci/zink: Add a TGL flake that's showed up in nightlies recently.
...
I don't know how recently, since the nightlies were timing out for a long
time.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25693 >
2023-10-13 01:12:59 +00:00
Emma Anholt
40f22f8ecd
nir/print: Decode system values in the variable declarations.
...
decl_var system INTERP_MODE_NONE none vec4 #0
decl_var system INTERP_MODE_FLAT none mediump uint #1
turns into:
decl_var system INTERP_MODE_NONE none vec4 #0 (SYSTEM_VALUE_FRAG_COORD)
decl_var system INTERP_MODE_FLAT none mediump uint #1 (SYSTEM_VALUE_SUBGROUP_INVOCATION)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25647 >
2023-10-12 22:52:42 +00:00
Mike Blumenkrantz
0d652c0c8d
zink: shrink vectors during optimization
...
this avoids a number of cases where a shader was reading more components
from an input than an output was providing. functionally there was never
any issue as these read components were subsequently rewritten to use
constant data, but the read itself is a spec violation
shrinking can't be done in finalize, however, as that enables the frontend
to optimize vertex states, which seems like a good thing but ends up being
a bad thing since it may or may not be consistent across frontends and I
don't wanna deal with having to reorder i/o locations in unintuitive ways
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25433 >
2023-10-12 22:01:44 +00:00
Alyssa Rosenzweig
8df8d1e2f2
nir/opt_algebraic: Reduce int64
...
If we just want the bottom 32-bits we don't need a full 64-bit operation.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25625 >
2023-10-12 21:03:31 +00:00
Alyssa Rosenzweig
8b5b362be6
nir/lower_io: Use load_global_constant for OpenCL
...
Map __constant with a 64-bit address format to load_global_constant instead of
load_global. This notably allows nir_opt_preamble to hoist the load.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25625 >
2023-10-12 21:03:31 +00:00
Alyssa Rosenzweig
569d44eff4
nir/print: Handle KERNEL
...
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25625 >
2023-10-12 21:03:31 +00:00
Alyssa Rosenzweig
6d0efa8701
nir/legalize_16bit_sampler_srcs: Use instr_pass
...
Fixes the pass with multiple functions.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25625 >
2023-10-12 21:03:31 +00:00
Alyssa Rosenzweig
b1b7616418
nir/opt_phi_precision: Work with libraries
...
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Reviewed-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25625 >
2023-10-12 21:03:31 +00:00
Alyssa Rosenzweig
cd1ba21958
r600/sfn: Handle load_global_constant
...
as an alias of load_global, for CL.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Suggested-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25625 >
2023-10-12 21:03:31 +00:00
Corentin Noël
aef37305a1
ci: Uprev virglrenderer
...
Include the latest virglrenderer version.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25561 >
2023-10-12 19:22:33 +00:00
Lionel Landwerlin
fe05e6610b
anv: fixup spirv cap for ImageReadWithoutFormat on Gfx12.5
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: 2e2491b76c
("anv: enable shaderStorageImageReadWithoutFormat on Gfx12.5+")
Reviewed-by: Tapani Pälli <tapani.palli@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25573 >
2023-10-12 18:57:55 +00:00
Mike Blumenkrantz
80d837cf40
ci: add a630 trace flakes
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25684 >
2023-10-12 17:59:28 +00:00
Sagar Ghuge
bed0542b2f
anv: Enable transfer queue only on ACM+ platforms
...
On older platforms, we have the blitter engine, but it lacks compression
handling and other features we need, unfortunately, so enable the
transfer queue only on ACM+ platforms.
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/25667 >
2023-10-12 17:24:06 +00:00
Samuel Pitoiset
1f77f52bbe
radv: skip GDS allocation for NGG streamout on GFX11
...
Only GDS OA is needed on GFX11.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25284 >
2023-10-12 16:58:22 +00:00
Samuel Pitoiset
2cc3288089
radv: mark GDS as needed for XFB queries with NGG streamout on GFX11
...
This doesn't fix anything because gds_needed should already be TRUE
because it's initialized at pipeline bind time, but this will be needed
for skipping GDS allocation on GFX11.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25284 >
2023-10-12 16:58:22 +00:00
Samuel Pitoiset
f0abdaea9f
amd/llvm,aco,radv: implement NGG streamout with GDS_STRMOUT registers on GFX11
...
According to RadeonSI, this is required for preemption, user queues,
and we only have to wait for VS after streamout which should be more
performant.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25284 >
2023-10-12 16:58:22 +00:00
Samuel Pitoiset
7a3981b9a7
radv: enable radv_disable_aniso_single_level=true for Zink too
...
This seems the default OpenGL behavior as well as DX.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25642 >
2023-10-12 12:54:16 +00:00
Karmjit Mahil
f1c9efc52e
pvr: Don't merge subpasses on framebuffer-global dependancy
...
When we have a framebuffer-local dependency we cannot merge
subpasses.
dEQP tests fixed:
dEQP-VK.renderpass.suballocation.attachment_allocation
.input_output.{15,23,35,61,80,93}
dEQP-VK.renderpass.suballocation.attachment_allocation
.roll.{13,31,42,47,55,59,61,98}
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/25612 >
2023-10-12 09:30:07 +00:00
Karmjit Mahil
1c35e495d9
tu: Use common vk_subpass_dependency_is_fb_local()
...
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com >
Reviewed-by: Connor Abbott <cwabbott0@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25612 >
2023-10-12 09:30:07 +00:00
Karmjit Mahil
9b7d437653
vulkan: Add vk_subpass_dependency_is_fb_local()
helper
...
Some tilers check for framebuffer local dependancy to determine if
they can rearrange or merge some sub-passes without breaking
their dependencies. Adding a helper for that.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com >
Reviewed-by: Connor Abbott <cwabbott0@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25612 >
2023-10-12 09:30:07 +00:00
Samuel Pitoiset
7397502a1f
radv: disable primitive restart for non-indexed draws on GFX11
...
Primitive restart is also applied to non-indexed draws on AMD GPUs. On
GFX11, DISABLE_FOR_AUTO_INDEX can be set but we will need a different
solution for older GPUs.
This fixes all line related flakes in CI (at least).
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25639 >
2023-10-12 06:33:40 +00:00
Tapani Pälli
f33d937608
mesa/st: ignore StencilSampling if stencil not part of the format
...
This avoids hitting assert in debug builds and incorrect rendering
on release in case GL_DEPTH_STENCIL_TEXTURE_MODE has been set to
GL_STENCIL_INDEX with a texture that does not have stencil.
Cc: mesa-stable
Signed-off-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/25610 >
2023-10-12 03:52:26 +00:00
Yiwei Zhang
ed2cac0f1e
venus: enable Vulkan 1.3 for Android 13 and above
...
Now that ndk translation support is ready for both Intel and AMD, we can
advertise Vulkan 1.3 on Android 13.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25666 >
2023-10-11 22:29:46 +00:00
Helen Koike
51521e1f9d
ci/ci_run_n_monitor: limit repetitions on --stress
...
--stress options now receives the number of repetitions or -1 to behave
as before.
Signed-off-by: Helen Koike <helen.koike@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25473 >
2023-10-11 21:50:58 +00:00
Helen Koike
57fa35f19c
ci/ci_run_n_monitor: allow <user>/<project> in --project
...
Allow to monitor pipelines in any <user>/<project> on gitlab.fdo.
If developers want to monitor MR pipelines, they can run with
--project mesa/mesa to force it to be picked from mesa username.
Signed-off-by: Helen Koike <helen.koike@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25473 >
2023-10-11 21:50:58 +00:00
Helen Koike
6b49b477ac
ci/ci_run_n_monitor: simplify enable/cancel logic in monitor_pipeline()
...
leave to enable_job() and cancel_job() to do the right thing according
to the job status.
let target canceled jobs to be re-runned.
Signed-off-by: Helen Koike <helen.koike@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25473 >
2023-10-11 21:50:58 +00:00