Jason Ekstrand [Fri, 15 Jan 2021 06:18:37 +0000 (00:18 -0600)]
anv/apply_pipeline_layout: Lower UBO loads in the early pass
We're about to enable bindless UBOs via A64 memory access like we do for
SSBOs. In order to prevent 100% of UBOs from hitting that path, we
enable them in the early lowering. This way we'll still get binding
table-based UBO access for any non-bindless ones. In particular, we
need this for UBO pushing to work.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8635>
Jason Ekstrand [Thu, 21 Jan 2021 01:42:16 +0000 (19:42 -0600)]
anv/apply_pipeline_layout: Rework the early pass index/offset helpers
Rewrite them all to work on an index/offset vec2 instead of some only
returning the index. This means SSBO size handling is a tiny bit more
complicated but it will also mean we can use them for descriptor buffers
properly.
This also fixes a bug where we weren't bounds-checking re-index
intrinsics because we applied the bounds check at the tail of the
recursion and not at the beginning.
Fixes:
3cf78ec2bd "anv: Lower some SSBO operations in..."
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8635>
Jason Ekstrand [Thu, 21 Jan 2021 00:49:26 +0000 (18:49 -0600)]
anv/apply_pipeline_layout: Refactor descriptor chasing code
This makes things a bit more generic for use in the next commit.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8635>
Jason Ekstrand [Fri, 12 Mar 2021 19:00:47 +0000 (13:00 -0600)]
anv: Use nir_shader_instructions_pass in apply_pipeline_layout
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8635>
Jason Ekstrand [Wed, 20 Jan 2021 22:21:38 +0000 (16:21 -0600)]
anv: Use load_global_constant for shader constants
NIR can do a bit better job optimizing this version.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8635>
Jason Ekstrand [Fri, 15 Jan 2021 00:00:00 +0000 (18:00 -0600)]
intel/fs,rt: Add a predicate to load_global_const_block
This allows us to do bounds checked A64 block load without the it being
counted as control-flow by NIR. This means that NIR optimizations like
CSE will be able to work on these the same as a regular load.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8635>
Eric Anholt [Wed, 17 Mar 2021 16:19:18 +0000 (09:19 -0700)]
ci/bare-metal: Restart a run on intermittent kernel lockups.
Since enabling SMP on db820c and cranking up how many tests we run, we've
been seeing lockups like this a couple of times a week.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9655>
Rob Clark [Tue, 16 Mar 2021 22:45:38 +0000 (15:45 -0700)]
freedreno/drm: Avoid unitialized timestamp in submit fail
Saw a flood of "waiting on invalid fence" with a completely bogus
looking fence # in a log of a rather strange low-memory crash. Not
sure if it is coming from memory corruption in userspace, but if a
submit ioctl is failing due to failed allocation (or other reason)
we would get left with random stack garbage as the fence #. Let's
not have that as a potential problem.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9638>
Rhys Perry [Mon, 15 Mar 2021 13:35:54 +0000 (13:35 +0000)]
aco: use a single instruction for uadd32_sat() on GFX8
fossil-db (GFX8):
Totals from 8 (0.01% of 147787) affected shaders:
SGPRs: 352 -> 368 (+4.55%)
CodeSize: 49576 -> 48788 (-1.59%)
Instrs: 9487 -> 9318 (-1.78%)
Latency: 49935 -> 49607 (-0.66%)
InvThroughput: 138493 -> 137443 (-0.76%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9598>
Rhys Perry [Mon, 15 Mar 2021 13:33:29 +0000 (13:33 +0000)]
aco: use uadd32_sat() helper for nir_op_uadd_sat
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9598>
Rhys Perry [Mon, 15 Mar 2021 13:28:17 +0000 (13:28 +0000)]
aco: implement 64-bit VGPR {u,i}find_msb
This can be created by subgroupBallotFindMSB().
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4458
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9598>
Marek Olšák [Fri, 12 Mar 2021 15:26:54 +0000 (10:26 -0500)]
ac/gpu_info: fix more non-coherent RB and GL2 combinations
It ignored non-harvested chips with a non-power-of-two memory bus.
Fixes:
abed921ce71 - amd: add support for Navy Flounder
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9568>
Erik Faye-Lund [Wed, 17 Mar 2021 09:32:15 +0000 (10:32 +0100)]
zink: fix free of ralloced pointer
When we alloc with ralloc, we also need to free with it.
But let's take a step back here; we don't just need to use ralloc, we
also need to destroy all other screen-resources. So let's call the
destructor here instead.
Fixes:
2643f9ed284 ("zink: ralloc screen objects")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9647>
Erik Faye-Lund [Wed, 17 Mar 2021 10:36:59 +0000 (11:36 +0100)]
zink: fix emulation of no mipfilter
This approach is taken from the Vulkan spec[1], where a robust maxLod
of 0.25 is proposed instead of 0.0.
This has the effect of allowing room for both minification and
magnification filters, yet still rounding down to the right miplevel in
the end.
[1]: https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkSamplerCreateInfo.html#_description
Fixes:
8d46e35d16e ("zink: introduce opengl over vulkan")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9649>
Timur Kristóf [Mon, 22 Feb 2021 14:23:49 +0000 (15:23 +0100)]
aco: Delete superfluous tess and ESGS I/O code.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 26 Feb 2021 17:49:27 +0000 (18:49 +0100)]
radv/llvm: Delete superfluous tess and ESGS I/O code.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 26 Feb 2021 17:49:12 +0000 (18:49 +0100)]
radv/llvm: Only store TCS outputs where they are really needed.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 11 Mar 2021 16:45:10 +0000 (17:45 +0100)]
radv: Use new, NIR-based I/O lowering.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 11 Mar 2021 16:41:55 +0000 (17:41 +0100)]
radv: Reorder some NIR optimizations in preparation for the I/O changes.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Wed, 3 Mar 2021 15:29:59 +0000 (16:29 +0100)]
radv: Fill some tess shader info earlier.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 18 Feb 2021 12:39:40 +0000 (13:39 +0100)]
radv: Determine tcs_in_out_eq in radv_pipeline instead of the compiler.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Wed, 17 Feb 2021 16:26:29 +0000 (17:26 +0100)]
radv: Calculate tess patches and LDS use outside the backend compilers.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 18 Feb 2021 10:02:30 +0000 (11:02 +0100)]
radv: Save I/O usage data to both shader infos for merged stages.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Wed, 3 Mar 2021 15:17:55 +0000 (16:17 +0100)]
radv: Lower IO and set driver locations earlier.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Tue, 2 Mar 2021 14:30:58 +0000 (15:30 +0100)]
ac: Add NIR passes to lower ES->GS I/O to memory accesses.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Tue, 9 Feb 2021 18:19:53 +0000 (19:19 +0100)]
ac: Add NIR passes to lower VS->TCS->TES I/O to memory accesses.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Mon, 22 Feb 2021 13:18:05 +0000 (14:18 +0100)]
ac/llvm: Emit more efficient code for load_shared.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Mon, 22 Feb 2021 13:13:37 +0000 (14:13 +0100)]
ac/llvm: Add constant offset to load/store_shared.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 19 Feb 2021 16:30:32 +0000 (17:30 +0100)]
ac/llvm: Make sure to always emit integer comparison for nir_op_ieq.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 19 Feb 2021 16:30:08 +0000 (17:30 +0100)]
ac/llvm: Make shared loads/stores work correctly for non-CS stages.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 11 Mar 2021 15:41:38 +0000 (16:41 +0100)]
ac/llvm: Implement new Geometry Shader intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 19 Feb 2021 16:29:00 +0000 (17:29 +0100)]
ac/llvm: Implement the new tessellation intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 19 Feb 2021 16:29:25 +0000 (17:29 +0100)]
ac/llvm: Implement AMD-specific buffer load/store intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Mon, 22 Feb 2021 19:18:08 +0000 (20:18 +0100)]
aco: Implement new Geometry Shader intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Wed, 17 Feb 2021 15:39:50 +0000 (16:39 +0100)]
aco: Implement the new tessellation I/O related NIR intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Mon, 1 Feb 2021 15:25:13 +0000 (16:25 +0100)]
aco: Implement new buffer load/store intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Mon, 22 Feb 2021 19:16:32 +0000 (20:16 +0100)]
nir: Add AMD-specific Geometry Shader related intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Mon, 15 Feb 2021 21:01:02 +0000 (22:01 +0100)]
nir: Add tessellation related AMD-specific intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 11 Mar 2021 11:45:31 +0000 (12:45 +0100)]
nir: Add nir_opt_offsets to fold const adds into load/store offsets.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Mon, 25 Jan 2021 18:12:18 +0000 (19:12 +0100)]
nir: Add AMD-specific buffer load/store intrinsics.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 11 Mar 2021 11:43:56 +0000 (12:43 +0100)]
nir: Add default unsigned upper bound configuration.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 26 Feb 2021 15:54:04 +0000 (16:54 +0100)]
nir: Add unsigned upper bound for TCS load_invocation_id.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Thu, 25 Feb 2021 09:08:18 +0000 (10:08 +0100)]
nir: Shrink vectors for load_shared.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 26 Feb 2021 13:56:06 +0000 (14:56 +0100)]
nir: Fix unsigned upper bound of local_invocation_index for non-CS stages.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Fri, 26 Feb 2021 18:05:25 +0000 (19:05 +0100)]
nir: Add a few more algebraic optimizations to help address calculation.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Tue, 2 Mar 2021 15:00:53 +0000 (16:00 +0100)]
nir: Add nir_builder helper for I/O address offset calculations.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Timur Kristóf [Tue, 16 Feb 2021 17:18:52 +0000 (18:18 +0100)]
nir: Add new nir_builder helpers for iadd with no_unsigned_wrap.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Rhys Perry [Thu, 4 Mar 2021 18:35:08 +0000 (18:35 +0000)]
nir: Don't update base in vectorize_loads()
The offset is already updated with consideration to the base above under
"/* update the offset */".
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9201>
Rhys Perry [Wed, 10 Mar 2021 17:51:31 +0000 (17:51 +0000)]
aco/tests: add test for waNsaCannotFollowWritelane
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9187>
Rhys Perry [Wed, 10 Mar 2021 17:42:52 +0000 (17:42 +0000)]
aco: fix NSA following writelane
No fossil-db changes on GFX10.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes:
c353895c922 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9187>
Rhys Perry [Tue, 23 Feb 2021 09:54:04 +0000 (09:54 +0000)]
aco/tests: add test for NSAToVMEMBug
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9187>
Rhys Perry [Mon, 22 Feb 2021 11:12:15 +0000 (11:12 +0000)]
aco: fix NSA MIMG followed by MUBUF/MTBUF
No fossil-db changes on GFX10.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes:
c353895c922 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9187>
Danylo Piliaiev [Tue, 16 Mar 2021 19:04:02 +0000 (21:04 +0200)]
freedreno/isa: assert if field's range is out of bitset's range
Also, update outdated comment along the way.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9628>
Danylo Piliaiev [Tue, 16 Mar 2021 15:23:45 +0000 (17:23 +0200)]
ir3: match mova1 mnemonic when writing to A1
For MOV to A1 blob uses "mova1" mnemonic, which is mov.u16u16;
change s16 to u16 when creating MOV to A1 in order to match the blob.
Before, couldn't be parsed back:
mov.s16s16 ha0.y, 0
After, could be parsed back and matches blob behaviour:
mova1 a1.x, 0
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9628>
Danylo Piliaiev [Tue, 16 Mar 2021 15:13:38 +0000 (17:13 +0200)]
ir3/isa,parser: fix encoding and parsing of bindless s2en SAM
Before, decoding showed that there is an error:
sam.base0 (f32)(xyzw)r0.x, r0.z, a1.x ; no field 'HAS_SAMP',
WARNING: unexpected bits[0:7] in #cat5-samp-s2en-bindless-a1: 0x1 vs 0x0
After:
sam.base0 (f32)(xyzw)r0.x, r0.z, s#1, a1.x
Fixes textures on the ground in TauCeti Vulkan Technology Benchmark
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9628>
Juan A. Suarez Romero [Tue, 9 Mar 2021 15:08:34 +0000 (16:08 +0100)]
ci/vc4/v3d: run piglit testsuite against Xorg
This increases the coverage adding tests that require an X server to
run.
Update also the expected results, and skip some flake tests.
Reviewed-by: Andres Gomez <agomez@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9562>
Juan A. Suarez Romero [Mon, 15 Mar 2021 13:15:22 +0000 (14:15 +0100)]
ci/v3dv: add flaky test in the skip list
Reviewed-by: Andres Gomez <agomez@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9562>
Juan A. Suarez Romero [Thu, 11 Mar 2021 08:49:51 +0000 (09:49 +0100)]
ci/armXX: add libgl1-mesa-dev dependency
Waffle building requires libgl1-mesa-dev to add support for GLX. This
package is pulled automatically in arm64 container as a suggested
package, but no in armhf. Which means we end up having support for GLX
in waffle in arm64 but not in armhf.
Thus let's install explicitly this package to have support in both
cases.
Reviewed-by: Andres Gomez <agomez@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9562>
Juan A. Suarez Romero [Fri, 12 Mar 2021 14:53:47 +0000 (15:53 +0100)]
vc4: destroy renderonly object if present
When destroying the VC4 screen, check if renderonly exists before
freeing it.
This fixes for instance a segmentation fault caused when invoking
wflinfo.
Fixes:
187218395d7 ("renderonly: remove layering violations")
Reviewed-by: Andres Gomez <agomez@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9562>
Alejandro Piñeiro [Tue, 16 Mar 2021 23:07:12 +0000 (23:07 +0000)]
v3dv: avoid some maybe-uninitialized warnings
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9640>
Alejandro Piñeiro [Tue, 16 Mar 2021 22:07:44 +0000 (22:07 +0000)]
v3dv/descriptor_set: don't free individual set if not allowed
If we have a host_memory_base pointer it means that we are handling
the pool memory as a whole, so each set is a sub-slice of the memory
pool. In this case we can't just free the individual set. In other
words, VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT is not set.
Note tha at that point we were able to sub-allocate an set, and we are
failing to sub-allocate the pool bo for the descripto bo. So
technically we could try to return that set to the pool (so undo the
change on host_memory_ptr before), that I assume was my intention when
(wrongly) calling vk_free there. But in practice, at that point we are
already on a out of descriptor pool situation, so in the end it
doesn't even matter.
This fixes a double free crash with the UE4 VehicleGame demo.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9640>
Iago Toral Quiroga [Tue, 16 Mar 2021 08:51:45 +0000 (09:51 +0100)]
broadcom/compiler: use nir_lower_wrmasks to simplify TMU general stores
This pass splits writemaks with non-consecutive bits into multiple
store operations ensuring that each store only has consecutive
writemask bits set.
We can use this to simplify writemask handling in our backend removing
a loop solely intended to handle this case.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9619>
Iago Toral Quiroga [Mon, 15 Mar 2021 12:39:50 +0000 (13:39 +0100)]
broadcom/compiler: use nir_opt_load_store_vectorize
This will make it so we pack consecutive scalar operations into a vector
operation, reducing the amount of load/store operations in the NIR program.
Our backend can handle vector load/stores, and doing so may be more efficient
since we don't need to setup individual load/stores all the time.
A pathological case is:
dEQP-VK.spirv_assembly.instruction.compute.opcopymemory.array
which goes from 862 instructions to only 573 by converting
all scalar SSBO load/store operations to vec4 operations.
total instructions in shared programs:
13752607 ->
13733627 (-0.14%)
instructions in affected programs: 367117 -> 348137 (-5.17%)
helped: 1168
HURT: 371
Instructions are helped.
total threads in shared programs: 412230 -> 412272 (0.01%)
threads in affected programs: 54 -> 96 (77.78%)
helped: 23
HURT: 2
Threads are helped.
total uniforms in shared programs: 3790248 -> 3784601 (-0.15%)
uniforms in affected programs: 57417 -> 51770 (-9.84%)
helped: 1420
HURT: 19
Uniforms are helped.
total max-temps in shared programs: 2322170 -> 2322714 (0.02%)
max-temps in affected programs: 14353 -> 14897 (3.79%)
helped: 185
HURT: 306
Max-temps are HURT.
total spills in shared programs: 5940 -> 6010 (1.18%)
spills in affected programs: 65 -> 135 (107.69%)
helped: 0
HURT: 11
total fills in shared programs: 13372 -> 13494 (0.91%)
fills in affected programs: 75 -> 197 (162.67%)
helped: 0
HURT: 11
total sfu-stalls in shared programs: 31505 -> 31521 (0.05%)
sfu-stalls in affected programs: 751 -> 767 (2.13%)
helped: 210
HURT: 246
Inconclusive result (value mean confidence interval includes 0).
total inst-and-stalls in shared programs:
13784112 ->
13765148 (-0.14%)
inst-and-stalls in affected programs: 360283 -> 341319 (-5.26%)
helped: 1125
HURT: 366
Inst-and-stalls are helped.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9619>
Iago Toral Quiroga [Tue, 16 Mar 2021 08:02:20 +0000 (09:02 +0100)]
broadcom/compiler: fix end of tmu sequence detection
TMUWT always terminates a TMU sequence.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9619>
Samuel Pitoiset [Mon, 15 Mar 2021 16:51:27 +0000 (17:51 +0100)]
radv: extend the dirty bits to 64-bit
For future work.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9603>
Samuel Iglesias Gonsálvez [Tue, 9 Mar 2021 12:32:48 +0000 (13:32 +0100)]
turnip: set depth plane control zmode to A6XX_LATE_Z when sample mask is written
Otherwise, gl_SampleMask[] writes are ignored and the stencil test works like
if all samples were enabled.
Fixes: dEQP-VK.renderpass.suballocation.multisample.*s8*
Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Hyunjun Ko <zzoon@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9478>
Samuel Pitoiset [Wed, 3 Mar 2021 16:24:34 +0000 (17:24 +0100)]
radv: add notccompatcmask debug option
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9415>
Samuel Pitoiset [Wed, 3 Mar 2021 14:18:53 +0000 (15:18 +0100)]
radv: enable TC-compat CMASK on GFX10+
Untested on older chips. Should help MSAA games by 1-2%.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9415>
Samuel Pitoiset [Thu, 4 Mar 2021 18:37:41 +0000 (19:37 +0100)]
radv: initialize TC-compat CMASK images with the DCC clear code
0xff is for fast-clears without MSAA.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9415>
Samuel Pitoiset [Wed, 3 Mar 2021 16:21:16 +0000 (17:21 +0100)]
radv: only configure the CMASK tiling for TC-compat on GFX8
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9415>
Samuel Pitoiset [Tue, 19 Jan 2021 13:16:38 +0000 (14:16 +0100)]
radv: make sure FMASK is enabled for TC-compat CMASK
Otherwise it makes no sense at all.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9415>
Samuel Pitoiset [Tue, 19 Jan 2021 13:13:42 +0000 (14:13 +0100)]
radv: remove redundant check when enabling TC-compat CMASK
We already check that the image has CMASK.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9415>
Samuel Pitoiset [Tue, 19 Jan 2021 13:00:15 +0000 (14:00 +0100)]
radv: do not enable TC-compat CMASK if the image isn't readable by a shader
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9415>
Samuel Pitoiset [Mon, 15 Mar 2021 14:10:27 +0000 (15:10 +0100)]
radv: use common entrypoints for VK_KHR_copy_commands2
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9600>
Iago Toral Quiroga [Mon, 8 Mar 2021 07:34:50 +0000 (08:34 +0100)]
compiler/glsl: call util_cpu_detect from glsl_type_singleton_init_or_ref
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Closes: #4393
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9457>
Iago Toral Quiroga [Mon, 8 Mar 2021 07:26:28 +0000 (08:26 +0100)]
vulkan/util: call glsl_type_singleton_init_or_ref from vk_instance_init
v2: link libvulkan_util with libglsl so it can find the glsl singleton symbols.
v3: link with libcompiler instead of libglsl (Jason)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Dylan Baker <dylan.c.baker@intel.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com> for the v3dv bits.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com> for the turnip bits.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> for the radv bits.
Acked-by: Dave Airlie <airlied@redhat.com> for the lvp bits.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9457>
Lukas Feller [Mon, 8 Mar 2021 06:54:40 +0000 (07:54 +0100)]
v3dv: fix stride in buffer copy
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9402>
Lukas Feller [Mon, 8 Mar 2021 06:46:28 +0000 (07:46 +0100)]
v3dv: fix assertion in job_compute_frame_tiling
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9402>
Timur Kristóf [Tue, 16 Mar 2021 20:25:16 +0000 (21:25 +0100)]
anv: Use ASSERTED for results that are only used in asserts.
This gets rid of unused variable warnings for these results.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9634>
Timur Kristóf [Tue, 16 Mar 2021 20:20:32 +0000 (21:20 +0100)]
anv: Use unreachable() in anv_genX.
This gets rid of unused variable warnings on genX_thing, because
now the compiler will think that the unknown hardware generation
case is unreachable.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9634>
Timur Kristóf [Tue, 16 Mar 2021 20:06:14 +0000 (21:06 +0100)]
intel/compiler: Make room for maximum dest size in nir_emit_texture.
The maximum dest_size is 5, not 4.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9634>
Timur Kristóf [Tue, 16 Mar 2021 20:02:42 +0000 (21:02 +0100)]
intel/compiler: Use assume() instead of assert() for array bounds.
This should make both GCC and clang happy and make them believe that
the array bounds are not exceeded.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9634>
Daniel Stone [Tue, 16 Mar 2021 20:31:08 +0000 (20:31 +0000)]
CI: Change LAVA job visibility
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9635>
Mike Blumenkrantz [Tue, 27 Oct 2020 18:34:16 +0000 (14:34 -0400)]
zink: enable PIPE_CAP_INVALIDATE_BUFFER
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9546>
Mike Blumenkrantz [Tue, 27 Oct 2020 18:31:29 +0000 (14:31 -0400)]
zink: invalidate resources on map when discarding range
we can dump the whole vulkan object here for this case, which ends up being
much neater
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9546>
Mike Blumenkrantz [Fri, 12 Mar 2021 22:42:28 +0000 (17:42 -0500)]
zink: handle streamout buffer rebinds
this really just means to nuke the counter buffer, and this can be done
by using a special bind_history bit that can be unset when the buffer has
been rebound
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9546>
Mike Blumenkrantz [Fri, 12 Mar 2021 22:34:53 +0000 (17:34 -0500)]
zink: set valid region for streamout buffers on bind
this catches rebinds and is more accurate about the state of the region
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9546>
Mike Blumenkrantz [Tue, 27 Oct 2020 18:30:20 +0000 (14:30 -0400)]
zink: add a pipe_context::invalidate_resource hook
this creates new backing objects for the invalidated resource and, if
needed, rebinds it to any descriptor sets it might be used in by invalidating
the descriptor state and creating new view objects
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9546>
Mike Blumenkrantz [Tue, 16 Mar 2021 17:54:21 +0000 (13:54 -0400)]
zink: remove direct samplerview batch-tracking
this moves tracking onto the surface/bufferview, which is more accurate
and allows the removal of a temporary hack in resource invalidation
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9546>
Mike Blumenkrantz [Tue, 27 Oct 2020 16:44:58 +0000 (12:44 -0400)]
zink: hook up resource bind history
seems like this should be a gallium thing with how many drivers copy/paste it
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9546>
Mike Blumenkrantz [Tue, 16 Mar 2021 22:47:59 +0000 (18:47 -0400)]
ci/panfrost: disable the rest of these jobs temporarily
runners having a hard time right now
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9639>
Mike Blumenkrantz [Tue, 16 Mar 2021 15:51:50 +0000 (11:51 -0400)]
zink: ci updates
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Mon, 22 Feb 2021 14:34:56 +0000 (09:34 -0500)]
zink: set ntv variable descriptor sets during compile phase
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Fri, 4 Dec 2020 20:43:19 +0000 (15:43 -0500)]
zink: apply Delete All The Code methodology to the ubo/ssbo variables
gallium rewrites all the ubo/ssbo instructions to use the buffer index as
the instruction src. the nir variables which are passed to zink after that
point are completely worthless, and it's impossible to accurately determine
which one corresponds to which buffer
thus, deleting all the variables and creating new ones based on the buffers
that are used is the best option, and it also enables simplifying a ton
of code as well as doing even better improvements in the future
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Fri, 1 Jan 2021 16:32:42 +0000 (11:32 -0500)]
zink: move zink_binding() to compiler.c
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Thu, 31 Dec 2020 17:56:04 +0000 (12:56 -0500)]
zink: directly set nir variable bindings for reuse during ntv
this lets us prepopulate the binding values without "fetching" them
twice
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Fri, 4 Dec 2020 20:42:29 +0000 (15:42 -0500)]
zink: flatten binding numbers a bit
now that descriptor types are split, we can just use the type-max values
per stage as the increment instead of all type max values
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Fri, 4 Dec 2020 17:17:13 +0000 (12:17 -0500)]
zink: break out sized uint array construction into util function
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Fri, 4 Dec 2020 17:14:10 +0000 (12:14 -0500)]
zink: add ntv util function for checking if a glsl type is an atomic counter
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
Mike Blumenkrantz [Fri, 4 Dec 2020 17:13:32 +0000 (12:13 -0500)]
zink: add debug info about missing atomic ops
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9624>
projects / platform / upstream / mesa.git / log