AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
141,606 Commits 1 Branch 0 Tags
f953dc2cedaa412ecf56ee804ac19d4dee3b47db
Commit Graph

141606 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Daniel Schürmann
f953dc2ced amd/ci: add hawaii-specific skip and fail lists 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10424>
 2021-06-23 15:17:22 +00:00
Icecream95
08495a948e pan/mdg: Add 16 bytes of padding to the end of shaders 
Fixes INSTR_INVALID_PC faults when a shader ends on a 16MB boundary.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11551>
 2021-06-23 14:42:55 +00:00
Tapani Pälli
55951ac28e anv: fix emitting dynamic primitive topology 
Initial implementation missed various fields that derive from the
primitive topology. This patch fixes 3DSTATE_RASTER/3DSTATE_SF,
3DSTATE_CLIP and 3DSTATE_WM (gen7.x) emission in the dynamic case.

Fixes: f6fa4a8000 ("anv: add support for dynamic primitive topology change")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4924
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11379>
 2021-06-23 12:01:45 +00:00
Daniel Stone
43b99e48b1 Revert "ci: disable panfrost t760 jobs" 
Both our T760 machines took a dive in beautiful synchronicity last
night, were recovered early this morning.

This reverts commit 854d93f73d.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11555>
 2021-06-23 12:30:07 +01:00
Tapani Pälli
2d87ea3166 iris: take a reference to memobj bo in iris_resource_from_memobj 
v2: and remove it from iris_memobj_create_from_handle ... (Nanley)

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4969
Fixes: 772dc50d16 ("iris: hook up resource creation from memory object")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11552>
 2021-06-23 10:46:09 +00:00
Pierre-Eric Pelloux-Prayer
52547cfbd0 radeonsi: disable ngg culling on llvm < 12 
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4874
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
CC: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11362>
 2021-06-23 08:42:46 +00:00
Lionel Landwerlin
9b7cba7724 anv: bound checks buffer memory binding in debug builds 
Validation layers should warn you about this
(VUID-VkBindBufferMemoryInfo-size-01037) but this would be useful for
zink debugging.

Requested by Zmike.

v2: Also check memoryOffset (Jason)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11542>
 2021-06-23 08:16:57 +00:00
Francisco Jerez
4dc4284342 intel/fs: Implement Wa_14013745556 on TGL+. 
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
 2021-06-23 07:34:22 +00:00
Francisco Jerez
c19cfa9dc2 intel/fs: Fix synchronization of accumulator-clearing W/A move on TGL+. 
Right now the accumulator-clearing move emitted by the generator for
Wa_14010017096 inherits the SWSB field from the previous instruction.
This can lead to redundant synchronization, or possibly more serious
issues if the previous instruction had a TGL_SBID_SET SWSB
synchronization mode.  Take the SWSB synchronization information from
the IR.

Fixes: a27542c5dd ("intel/compiler: Clear accumulator register before EOT")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
 2021-06-23 07:34:22 +00:00
Francisco Jerez
63abc083ce intel/fs: Teach IR about EOT instruction writing the accumulator implicitly on TGL+. 
This is unlikely to have had any negative side effect on the original
TGL, but will lead to issues on XeHP+ if the software scoreboard pass
isn't able to synchronize the accumulator writes.

Fixes: a27542c5dd ("intel/compiler: Clear accumulator register before EOT")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
 2021-06-23 07:34:22 +00:00
Francisco Jerez
5e7f443de0 intel/fs: Add SWSB dependency annotations for cross-pipeline WaR data hazards on XeHP+. 
In cases where an in-order instruction is overwriting a register
previously read by another in-order instruction, drop the dependency
iff the previous read is guaranteed to have occurred from the same
in-order pipeline.  This should only have an effect on XeHP+ since
previous Xe platforms only had one in-order FPU pipeline.

The previous workaround we were using for this treated all ordered
read dependencies as write dependencies to avoid noise from our
simulation environment.  Relative to our previous workaround this
improves performance of GFXBench5 gl_tess by ~7% on a DG2 system
among other single-digit percentual FPS improvements.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
 2021-06-23 07:34:22 +00:00
Francisco Jerez
d46bb14d14 intel/fs: Implement Wa_22012725308 for cross-pipe accumulator data hazard. 
The hardware fails to provide the expected data coherency guarantees
for accumulator registers when accessed from multiple FPU pipelines.
Fix this by tracking implicit accumulator accesses just like we do for
regular GRF registers, but instead of adding synchronization
annotations for any dependency we only do it for dependencies with a
pipeline mismatch, since the hardware should be able to guarantee
proper synchronization for matching pipelines.

Note that this workaround handles RaW and WaW dependencies in addition
to the WaR dependencies described in the hardware bug report even
though cross-pipeline RaW accumulator dependencies should be extremely
rare, since chances are the hardware will also hang if we ever hit
such a condition.  This only affects XeHP+, since all FPU instructions
are executed as a single in-order pipeline on earlier Xe platforms.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
 2021-06-23 07:34:22 +00:00
Francisco Jerez
385da1fe36 intel/fs: Track single accumulator in scoreboard lowering pass. 
This change reduces the precision of the scoreboard data structure for
accumulator registers, because the rules determining the aliasing of
accumulator registers are non-trivial and poorly documented (e.g. acc0
overlaps the storage of acc1 when the former is accessed with an
integer type).  We could implement those rules but it wouldn't have
any practical benefit since we currently only use acc0-1, and for the
most part we can rely on the hardware's accumulator dependency
tracking.  Instead make our lives easier by representing it as a
single register.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
 2021-06-23 07:34:22 +00:00
Francisco Jerez
231337a13a intel/fs/xehp: Assert that the compiler is sending all 3 coords for cubemaps. 
As required by HSDES:14013363432.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11433>
 2021-06-23 07:34:22 +00:00
Matt Turner
85315f5fb1 freedreno/ci: Use TU_IGNORE_CONFORMANCE_WARNING to reduce warnings 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11543>
 2021-06-23 07:07:42 +00:00
Matt Turner
205d6e582c tu: Provide a toggle to avoid warnings about unsupported devices 
In the CI, we have such devices, and this message is printed many
hundreds of times. This results in a useless spam which makes it
difficult to see real issues.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11543>
 2021-06-23 07:07:42 +00:00
Dave Airlie
17a1cc6163 meson/crocus: add prefer-crocus option. 
This just allows picking crocus without having to set the env var.

Acked-by: Alyssa Rosenzweig <alyssa@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11353>
 2021-06-23 06:42:24 +00:00
Marek Olšák
ec7e262349 mesa: unreference zombie buffers when creating buffers to lower memory usage 
This fixes an issue where one context only creates buffers while another
context only destroys buffers. Only the creating context can release its
buffers and the destroying context only turns them into zombie buffers.

This fix makes the creating context release its zombie buffers.
It's not a plot from an apocalyptic movie.

Fixes: e014e3b6be "mesa: don't count buffer references for the context that created them"
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4840

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11514>
 2021-06-23 06:06:04 +00:00
Dave Airlie
29d272f144 crocus: fix batch state bo leak 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11547>
 2021-06-23 05:31:59 +00:00
Dave Airlie
9027aae546 crocus: fix vertex buffer leak on screen end. 
this was stopping the screen from being cleaned up as well

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11547>
 2021-06-23 05:31:59 +00:00
Dave Airlie
6036121a55 crocus: free context state properly. 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11547>
 2021-06-23 05:31:59 +00:00
Mike Blumenkrantz
c9bc5e3a86 zink: ci updates 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:42 +00:00
Mike Blumenkrantz
a784713080 zink: remove primconvert 
this is no longer used

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:42 +00:00
Mike Blumenkrantz
c032c9e32d zink: export supported primitive restart types 
this is now handled by gallium, so the codepath can be dropped

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:42 +00:00
Mike Blumenkrantz
2953d70994 zink: export supported prim types 
this is now handled by gallium, so the related codepaths can be dropped

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:42 +00:00
Mike Blumenkrantz
8ac4be916e zink: export 8bit index buffer support based on extension presence 
this is now handled by gallium, so the codepath can be dropped

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:42 +00:00
Mike Blumenkrantz
09c28a35a5 zink: export PIPE_CAP_EMULATE_NONFIXED_PRIMITIVE_RESTART 
gallium will now handle unsupported restart index rewriting, so
this codepath can be removed

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
3fc651110c gallium: add a pipe cap for determining driver support for prim type in restarts 
this is another prim type bitmask which will trigger automatic draw rewriting
to a direct draw any time a prim-restart draw occurs with a prim type that is
not supported by the driver for prim restart, even if that prim type is supported
for normal drawing

the default is set to all prim types to preserve existing functionality, and PrimitiveRestartForPatches
is now explicitly set to false because no driver supports it

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
f8ff6eb78b gallium: add a pipe cap for performing automatic prim type conversion 
drivers can now export a bitmask of the primitive types they support,
and all others will be automatically be rewritten

the default value is set to all primitive types supported to preserve
existing behavior

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
c8e674d4e3 gallium: handle automatic 8bit -> 16bit index buffer rewrites 
this is based on PIPE_FORMAT_R8_UINT format support

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
c9a65e5f77 gallium: add a pipe cap to rewrite index buffers for draws using a non-fixed restart index 
for drivers that set it, this now automatically handles restart index rewriting
by running draws through primconvert when necessary

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
99a60ed378 util/primconvert: handle multidraws in primconvert 
once a draw reaches primconvert, it should never be able to reach the driver
until all draw operations have been converted as necessary

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
583070748c util/primconvert: handle rewriting of prim-restart draws with unsupported primtype 
this is a bit more work, as the primitive restart needs to be rewritten into a multidraw,
then the multidraw converted to the new primitive type and serialized back into a
single draw

detection is handled using a new primconvert config member, which is set to the full
primtype mask by default for compatibility with existing drivers

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
f3fe70f10e util/primconvert: map index buffer before getting index translator function 
also set other local values on init using new util functions

no functional changes

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
26e718fb37 util/primconvert: handle indirect draws 
this avoids ping-ponging to the driver, which would have to call back
to here anyway with the unwrapped indirect draw

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
4673febf91 util/prim_restart: break out draw rewriting into separate function 
this enables doing the draw rewrites without calling draw_vbo

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
3eb9932317 aux/draw: add a util function for reading back indirect draw params 
the return type of this is a bit clunky because instance values can change,
but it's simpler to just return the full draw info struct than to force the
caller to keep pulling from arrays or whatever

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10973>
 2021-06-23 03:13:41 +00:00
Mike Blumenkrantz
854d93f73d ci: disable panfrost t760 jobs 
these are timing out

Acked-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11548>
 2021-06-22 22:58:02 -04:00
Dave Airlie
fd09727a7d crocus: Explicitly cast value to uint64_t 
Ports 565a80450d from iris.

Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11546>
 2021-06-23 02:22:40 +00:00
Simon Ser
17f0d7ef83 amd/addrlib: remove Meson debug message() 
This message pops up in the Meson build logs, but has no context
and doesn't seem to be directed at end-users. Sounds like a leftover
from a debugging session.

Signed-off-by: Simon Ser <contact@emersion.fr>
Fixes: d0767fc045 ("amd/addrlib: use cpp.has_argument() to filter compiler arguments")
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11524>
 2021-06-22 23:47:19 +00:00
Thong Thai
d977e7fbe1 radeon/vcn/enc: Add missing line to HEVC SPS header code 
Add previously missing line in HEVC SPS header generation on VCN 1.0
devices, for when an image does not need to be cropped.

Fixes: e62c7e7c6c ("radeon: Add cropping to encoded H.265 when padding is used")
Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11535>
 2021-06-22 23:36:29 +00:00
Marek Olšák
d3fddf63ec ac/llvm: don't set skip-uniform-regions to fix atomic.cmpswap 
This fixes SSBO atomic cmpswap tests and transform feedback order piglit tests.

Fixes: 3bb65c0670 - ac: force enable -structurizecfg-skip-uniform-regions for LLVM 11
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3884

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11517>
 2021-06-22 22:13:07 +00:00
Paul Gofman
bdd8646306 util: add force_gl_names_reuse for SWKOTOR. 
SWKOTOR is expecting texture names generated by glGenTextures() to
be small enough and crashes otherwise.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11527>
 2021-06-22 21:43:33 +00:00
Emma Anholt
a9e3ddfc63 i915g: Fix GL_ARB_copy_buffer assertion fails. 
The i915_texture() asserts that the resource is not a buffer, so check for
the buffer fallback first.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11540>
 2021-06-22 21:32:36 +00:00
Lionel Landwerlin
7ed0aaced7 nir: use a more fitting index for btd_stack_push_intel 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8637>
 2021-06-22 21:09:25 +00:00
Lionel Landwerlin
423c47de99 nir: drop the btd_resume_intel intrinsic 
This is now 100% equivalent to the new rt_resume intrinsic.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8637>
 2021-06-22 21:09:25 +00:00
Lionel Landwerlin
4d9fcf2799 intel/rt: switch to common pass for shader calls lowering 
v2: rename for new indices

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8637>
 2021-06-22 21:09:25 +00:00
Lionel Landwerlin
1dacea10f3 anv: implement caching for ray tracing pipelines 
v2: Turn a bunch of pointer checks into checks against NULL (Jason)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8637>
 2021-06-22 21:09:25 +00:00
Lionel Landwerlin
fed7ac932f anv: move trivial return shader to device 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8637>
 2021-06-22 21:09:25 +00:00
Lionel Landwerlin
7c852f78c4 anv: store more RT shader data in pipeline_stage object 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8637>
 2021-06-22 21:09:25 +00:00