Rob Clark [Thu, 8 Jul 2021 18:35:31 +0000 (11:35 -0700)]
freedreno+turnip: Get device name from device-info table
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 17:43:08 +0000 (10:43 -0700)]
freedreno+turnip: Add has_8bpp_ubwc
Newer a6xx devices seem to drop 8b/pixel UBWC support.
The turnip part was adapted from Jonathans patch on !10892
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 17:28:43 +0000 (10:28 -0700)]
freedreno+turnip: Add has_cp_reg_write
Newer a6xx devices drop this packet from the sqe firmware, and use
direct (pkt4) register writes instead for the few cases that previously
used CP_REG_WRITE.
The turnip part was adapted from Jonathans patch on !10892
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 16:44:45 +0000 (09:44 -0700)]
turnip: Get has_sample_locations from fd_dev_info
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 16:31:25 +0000 (09:31 -0700)]
turnip: Get has_tex_filter_cubic from fd_dev_info
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 16:25:20 +0000 (09:25 -0700)]
turnip: Get indirect_draw_wfm_quirk from fd_dev_info
At some point we might want to change this to minimum fw version, but
for now it can be a bool.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 16:01:12 +0000 (09:01 -0700)]
turnip: Get storage_16bit from fd_dev_info
Removing more gpu_id checks that will become bogus as we add more a6xx.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 15:49:18 +0000 (08:49 -0700)]
turnip: Drop unused vshs_workgroup param
Unused since
d968995c67f5e3a48e49015184f9ea1f143251ec, and this gets rid
of one more gpu_id check.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 17:02:19 +0000 (10:02 -0700)]
freedreno/ir3: Get reg_size_vec4 from fd_dev_info
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 15:37:55 +0000 (08:37 -0700)]
freedreno/ir3: Get tess_use_shared from fd_dev_info
A step towards getting rid of checks for gpu_id sprinkled around.
Checking major generation is ok, but checking for == or >= a specific
gpu_id is going to start getting messy as we add more a6xx.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 01:46:49 +0000 (18:46 -0700)]
turnip: Convert fd_dev_info to const pointer
Split out from earlier patch to reduce churn.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Thu, 8 Jul 2021 01:37:09 +0000 (18:37 -0700)]
freedreno: Convert fd_dev_info to const pointer
Split out from previous patch to reduce churn.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Wed, 7 Jul 2021 21:47:02 +0000 (14:47 -0700)]
freedreno: Generate device-info tables at build time
This way we can make the tables const. At the same time, for a6xx, this
introduces a "sub-generation template" to reduce the copy/paste for
parameters which are keyed to the sub-generation. It also explicitly
lists every supported GPU, to get rid of duplicate lists of supported
gpus between the device-info and drivers.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Rob Clark [Wed, 7 Jul 2021 20:48:21 +0000 (13:48 -0700)]
freedreno: Rename *_dev_info
Everywhere else symbols/types/etc are shortend to "fd_*", so lets do the
same here for consistency.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Jonathan Marek [Thu, 20 May 2021 02:46:16 +0000 (22:46 -0400)]
freedreno/common: unhardcode CCU color cache offset
Replace it with a calculation which works for all current GPUs.
Duplicated the calculation in both drivers because freedreno_dev_info isn't
meant for derived parameters (and drivers might want to just calculate on
the fly instead).
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Jonathan Marek [Thu, 20 May 2021 01:56:53 +0000 (21:56 -0400)]
freedreno/a6xx: larger gmem_page_align from tile align instead of gpu id
The larger page alignment is directly related to the 96 tile alignment,
so check for that instead of a specific gpu id.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Jonathan Marek [Wed, 19 May 2021 23:04:43 +0000 (19:04 -0400)]
tu: remove workaround for conditional rendering + hw binning
- It hurts users with newer firmware who don't need the workaround
- Kernel now rejects older firmware due to security issues, so this will
prevent users from using older firmware anyway.
- Only whitelisting 650 enables the workaround by default for any new GPUs
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11790>
Kenneth Graunke [Tue, 9 Jun 2020 19:05:07 +0000 (12:05 -0700)]
iris: Use simple_mtx in the bufmgr.
This avoids having to call out through the PLT just to lock/unlock.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11858>
Icecream95 [Wed, 14 Jul 2021 00:04:42 +0000 (12:04 +1200)]
panfrost: Initialise the blend equation in create_blend_state
Valgrind still complains about uninitialised values, but tests don't
flake anymore.
Fixes flakes in dEQP-GLES3.functional.fragment_out.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11862>
Dave Airlie [Tue, 13 Jul 2021 22:56:09 +0000 (08:56 +1000)]
crocus: use simple_mtx in the bufmgr
anything you can do I can do better^W^Wadapt for crocus
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11859>
Timur Kristóf [Mon, 5 Jul 2021 13:26:18 +0000 (15:26 +0200)]
radv, aco, ac/nir: Tweak position export scheduling for NGG culling.
The result is about +5-ish fps in Doom Eternal.
It turns out that the location of position exports matters more
than we thought, and it's actually better to keep them at the bottom
for culling shaders rather than schedule it up to the top.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Mon, 5 Jul 2021 13:18:28 +0000 (15:18 +0200)]
ac/nir: Reuse uniforms from top part of culling shaders.
Uniforms have the same value in all invocations, therefore they
can safely be reused by invocations even after repacking.
This saves several instructions from culling shaders, mainly
UBO loads and such.
We exclude uniform floats, because those would harm the VGPR
usage of the shaders too much.
Fossil DB results on Sienna Cichlid (with NGG culling on):
Totals from 55379 (43.05% of 128647) affected shaders:
VGPRs: 1926472 -> 1925360 (-0.06%); split: -0.07%, +0.01%
SpillSGPRs: 139 -> 330 (+137.41%)
CodeSize:
159472988 ->
157462856 (-1.26%); split: -1.27%, +0.00%
MaxWaves: 1571492 -> 1571412 (-0.01%)
Instrs:
30665685 ->
30302076 (-1.19%); split: -1.21%, +0.02%
Latency:
127385148 ->
126723891 (-0.52%); split: -0.55%, +0.03%
InvThroughput:
21096298 ->
20773069 (-1.53%); split: -1.53%, +0.00%
VClause: 514792 -> 511231 (-0.69%); split: -0.83%, +0.13%
SClause: 713959 -> 679556 (-4.82%); split: -4.84%, +0.02%
Copies: 2975106 -> 2828185 (-4.94%); split: -5.39%, +0.45%
Branches: 1201921 -> 1152766 (-4.09%)
PreSGPRs: 1753786 -> 1892848 (+7.93%); split: -0.00%, +7.93%
PreVGPRs: 1590522 -> 1583574 (-0.44%); split: -0.44%, +0.00%
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Mon, 5 Jul 2021 13:17:10 +0000 (15:17 +0200)]
ac/nir: Analyze culling shaders to remember which inputs are used when.
These will be useful for some optimizations.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Tue, 4 May 2021 09:36:02 +0000 (11:36 +0200)]
ac/nir: Reuse the repacked output positions of culling shaders.
The position outputs are stored into LDS and reloaded after
repacking, therefore the repacked position values can be
reused in the bottom part of the shader.
Fossil DB results on Sienna Cichlid (with NGG culling on):
Totals from 9016 (7.01% of 128647) affected shaders:
VGPRs: 372472 -> 347560 (-6.69%); split: -6.82%, +0.13%
SpillSGPRs: 437 -> 87 (-80.09%)
CodeSize:
32359340 ->
30441692 (-5.93%); split: -5.93%, +0.00%
MaxWaves: 222030 -> 238970 (+7.63%); split: +7.83%, -0.20%
Instrs: 6207833 -> 5834149 (-6.02%); split: -6.02%, +0.00%
Latency:
27626263 ->
27890632 (+0.96%); split: -5.34%, +6.29%
InvThroughput: 4792958 -> 4361336 (-9.01%); split: -9.01%, +0.00%
VClause: 144385 -> 139586 (-3.32%); split: -9.29%, +5.97%
SClause: 141350 -> 129875 (-8.12%); split: -8.57%, +0.45%
Copies: 580017 -> 568916 (-1.91%); split: -3.60%, +1.68%
Branches: 209067 -> 209154 (+0.04%); split: -0.24%, +0.28%
PreSGPRs: 281320 -> 277814 (-1.25%)
PreVGPRs: 290040 -> 273861 (-5.58%)
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Tue, 6 Jul 2021 06:45:33 +0000 (08:45 +0200)]
radv: Run algebraic optimizations before NGG lowering.
This makes culling shaders more efficient because they split the
shader in two parts. It is better to optimize before this split
happens.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Tue, 8 Jun 2021 10:32:35 +0000 (12:32 +0200)]
radv: Support NGG culling with new perftest environment variable.
Currently we don't enable it on any chip by default, but
we plan to enable it soon on GFX10.3 when we are comfortable
with its performance.
RADV_PERFTEST=nggc environment variable enables it on GFX10+ GPUs.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Thu, 29 Apr 2021 11:33:45 +0000 (13:33 +0200)]
aco: Implement NGG culling related intrinsics.
These are very straightforward as they just copy data from
the newly added shader arguments.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Mon, 7 Jun 2021 21:23:38 +0000 (23:23 +0200)]
radv: New shader args for NGG culling settings and viewport.
Add new shader arguments in RADV for:
- NGG culling settings
- Viewport transform
These will be used by NGG culling shaders.
Additionally, some tweaks are made to some config registers
in order to make culling shaders more efficient.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Mon, 7 Jun 2021 21:43:27 +0000 (23:43 +0200)]
radv: Expose radv_get_viewport_xform in radv_private.h
We need to emit viewport transform information for culling shaders.
This is used for small primitive culling.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Mon, 26 Apr 2021 14:56:11 +0000 (16:56 +0200)]
ac/nir: Implement NGG deferred attribute culling in NIR.
Culling is traditionally done by the rasterizer, but that
can be a bottleneck when an app creates a large number
of primitives. Eg. a lot of tiny triangles reduce the
rasterziation efficiency.
NGG makes it possible for the shader to check primitives
and delete those that it can prove are not needed.
After this is done, we have to repack the surviving invocations
so they remain compact. This also saves bandwidth, because
some memory loads are only executed by those vertices that
survived the culling.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Mon, 5 Jul 2021 09:19:14 +0000 (11:19 +0200)]
ac/nir: Use a ballot that matches the wave size during NGG lowering.
This generates slightly more efficient code in Wave32 mode.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Thu, 1 Apr 2021 10:43:31 +0000 (12:43 +0200)]
ac/nir: Add a NIR port of ac_llvm_cull.
The algorithms were originally implemented by Marek Olšák,
hence the copyright to AMD.
This commit just ports the LLVM based implementation to NIR,
using the new intrinsics added earlier.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Timur Kristóf [Tue, 4 May 2021 11:41:14 +0000 (13:41 +0200)]
nir: Add AMD specific intrinsics for NGG shader based culling.
The new intrinsics fall into the following categories:
1. New viewport intrinsics:
For missing components that we need.
RADV will emit new SGPR arguments which will contain the
viewport information for culling shaders. These are used to
compute the screen space coordinates for small primitive culling.
2. load_cull_xxx:
Load the culling settings in runtime.
These will be a new SGPR argument in RADV.
3. overwrite_xxx:
These are needed because system values such as vertex and
instance ID are not writeable, but we need to change them
after repacking shader invocations of VS and TES.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10525>
Emma Anholt [Tue, 13 Jul 2021 18:25:19 +0000 (11:25 -0700)]
ci: Enable testing of i915g in the the debian -Werror release build.
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11852>
Emma Anholt [Tue, 6 Jul 2021 23:03:52 +0000 (16:03 -0700)]
i915g: Fix release build compiler warnings.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11852>
Emma Anholt [Thu, 24 Jun 2021 23:11:59 +0000 (16:11 -0700)]
freedreno: Optimize duplicate obj-obj ring relocs.
No need to include the same BO multiple times in the long-lived ringbuffer
object's list of relocs to be added to the submit.
Improves non-TC drawoverhead -test 9 (8 tex updates) throughput by 1.4901%
+/- 0.8705% (n=20)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11697>
Emma Anholt [Fri, 2 Jul 2021 22:41:01 +0000 (15:41 -0700)]
freedreno/a6xx: Allocate just enough memory for SO state, only if we do SO.
Continuing to improve our suballocation packing.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11697>
Emma Anholt [Fri, 2 Jul 2021 22:21:56 +0000 (15:21 -0700)]
freedrneo/a6xx: Reduce the size of the long-lived texture stateobj.
It's just a few commands to upload the sampler/texconst data. Improves
the efficiency of suballocation.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11697>
Emma Anholt [Fri, 2 Jul 2021 22:21:28 +0000 (15:21 -0700)]
freedreno/a6xx: Reduce the size of the config stateobj allocation.
Improves the efficiency of suballocation.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11697>
Emma Anholt [Fri, 2 Jul 2021 17:51:16 +0000 (10:51 -0700)]
freedreno: Suballocate our long-lived ring objects.
On drawoverhead -test 9 (8 texture changes), this saves us 172kb of
memory. That's only ~1% of the GEM memory while the test is running, but
more importantly it saves us 29% of the gem BO allocations.
non-TC drawoverhead -test 9 (8 texture change) throughput 0.449019% +/-
0.336296% (n=100), but this gets better as we get better suballocation
density.
Note that this means that all fd_ringbuffer_new_object calls can now
return data aligned to 64 bytes, instead of 4k. We may find that we need
to increase it if some of our objects (tex consts, sampler consts, etc.)
require more alignment than that. But, this may help non-drawoverhead
perf if any of our RB objects have a cache in front of them (indirect
consts?) and we don't have most of our data in the same cache set any
more.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11697>
Paul Kocialkowski [Fri, 28 May 2021 21:06:49 +0000 (23:06 +0200)]
lima: Take offset in account when checking BO size
BO resources imported from a handle may have an offset provided, which
reduces the available size within the BO. Take this in account when
checking that the size is sufficient in lima.
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Andreas Baierl <ichgeh@imkreisrum.de>
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11076>
Joshua Ashton [Sun, 11 Jul 2021 15:46:21 +0000 (16:46 +0100)]
driconf: Add more workarounds for Teardown
Enable radeonsi_no_infinite_interp for Teardown to fix hangs.
Based on comments from #3714.
Tested-by: Joshua Ashton <joshua@froggi.es>
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Acked-by: Martin Peres <martin.peres@mupuf.org>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11814>
Simon Zeni [Tue, 6 Jul 2021 17:04:09 +0000 (13:04 -0400)]
anv: Implement VK_EXT_acquire_drm_display
Signed-off-by: Simon Zeni <simon@bl4ckb0ne.ca>
Reviewed-by: Simon Ser <contact@emersion.fr>
Tested-by: Simon Ser <contact@emersion.fr>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11735>
Tony Wasserka [Tue, 29 Jun 2021 17:32:00 +0000 (19:32 +0200)]
aco: Remove deprecated Operand constructors
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
Tony Wasserka [Tue, 29 Jun 2021 17:29:36 +0000 (19:29 +0200)]
aco: Clean up unneeded literal casts
These were only needed to select the appropriate Operand constructor before.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
Tony Wasserka [Tue, 13 Jul 2021 09:22:46 +0000 (11:22 +0200)]
aco: Remove use of deprecated Operand constructors
This migration was done with libclang-based automatic tooling, which
performed these replacements:
* Operand(uint8_t) -> Operand::c8
* Operand(uint16_t) -> Operand::c16
* Operand(uint32_t, false) -> Operand::c32
* Operand(uint32_t, bool) -> Operand::c32_or_c64
* Operand(uint64_t) -> Operand::c64
* Operand(0) -> Operand::zero(num_bytes)
Casts that were previously used for constructor selection have automatically
been removed (e.g. Operand((uint16_t)1) -> Operand::c16(1)).
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
Tony Wasserka [Tue, 29 Jun 2021 16:51:41 +0000 (18:51 +0200)]
aco: Remove use of deprecated Operand constructors in aco_builder.h
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
Tony Wasserka [Tue, 29 Jun 2021 16:53:36 +0000 (18:53 +0200)]
aco: Remove use of deprecated Operand constructors in test_to_hw_instr.cpp
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
Tony Wasserka [Tue, 29 Jun 2021 17:24:16 +0000 (19:24 +0200)]
aco: Replace Operand literal constructors with factory member functions
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
Emma Anholt [Mon, 12 Jul 2021 21:39:30 +0000 (14:39 -0700)]
ci: Make sure that we build the piglit dmabuf tests.
Force the option rather than relying on autodetection -- ARM runners were
apparently finding the necessary deps, but the x86 rootfs (radeonsi, iris)
and x86_test-gl container (i915g) were not.
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11834>
Mike Blumenkrantz [Tue, 13 Jul 2021 12:55:27 +0000 (08:55 -0400)]
zink: simplify modifier ifdefs
these are the only two defines referenced, so they can be defined to 0
for platforms that don't support modifiers in order to remove a ton of
ifdefs and make the code more readable
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11847>
Rob Clark [Mon, 25 Jan 2021 17:43:24 +0000 (09:43 -0800)]
freedreno/ir3: Add float immed "FLUT" support
We can encode a limited set of float immeds into cat2 instructions,
using hw's float lookup table (FLUT) feature.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/36
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8705>
Rob Clark [Sat, 10 Jul 2021 17:29:00 +0000 (10:29 -0700)]
freedreno/computerator: Add script to probe FLUT values
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8705>
Rob Clark [Mon, 12 Jul 2021 19:00:38 +0000 (12:00 -0700)]
turnip: Split tu6_emit_xs()
Emit all the state layout config (such as push-const CONSTLEN) first,
before emitting anything that depends on that state. This fixes an
issue that was showing up when FLUT is enabled in ir3 (which results
in higher probability of not having any immediats lowered to push-
consts).
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8705>
Rob Clark [Sun, 11 Jul 2021 19:40:05 +0000 (12:40 -0700)]
turnip: avoid some UB
Reduce a bit of extra noise that makes diffing cmdstream traces more
annoying.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8705>
Jason Ekstrand [Mon, 12 Jul 2021 16:50:04 +0000 (11:50 -0500)]
glsl: Delete lower_texture_projection
This is only used by i965 and we've been getting it through
nir_lower_tex since forever. Get rid of the GLSL IR pass.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11827>
Mike Blumenkrantz [Mon, 12 Jul 2021 19:07:01 +0000 (15:07 -0400)]
ci: only trigger gallium_core_file_list jobs from dri and glx frontend changes
these are the only frontends which may be used by gallium drivers in ci,
so stop triggering all driver jobs when other frontends are changed since
those changes can never affect ci
<MrCooper> Not that simple unfortunately. E.g. the llvmpipe-piglit-cl job hits
src/gallium/frontends/clover & possibly src/gallium/targets/opencl,
many jobs hit src/gallium/{frontends,targets}/dri and probably
src/gallium/targets/pipe-loader, lavapipe jobs hit src/gallium/{frontends,targets}/lavapipe.
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11832>
Mike Blumenkrantz [Mon, 12 Jul 2021 19:26:04 +0000 (15:26 -0400)]
ci: add vulkan files to lavapipe rules
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11833>
Icecream95 [Tue, 13 Jul 2021 08:01:07 +0000 (20:01 +1200)]
pan/bi: Create a nop clause when the shader starts with ATEST
Otherwise there would be no clause with the dependencies needed for
ATEST set, so the GPU would get stuck.
Not needed on v7, as there shader_wait_dependency in the RSD will wait
for the dependencies before the shader starts.
Explicitly create a NOP instruction, as it is assumed that clauses
have a non-zero count of instructions in various places.
Fixes GPU timeouts in many applications, such as SuperTuxKart and
GZDoom.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11842>
Icecream95 [Tue, 13 Jul 2021 06:29:13 +0000 (18:29 +1200)]
panfrost: Fix full_threads calculation on v6
Fixes:
8ba2f9f6985 ("panfrost: Create a blitter library to replace the existing preload helpers")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11842>
Heinrich Fink [Tue, 13 Jul 2021 09:12:56 +0000 (11:12 +0200)]
gbm/dri: Fix leaking bo memory on failure path
In gbm_dri_bo_create, when modifiers are requested but not supported, do
not return NULL immediately, but first go to cleanup section to free
already allocated buffer object.
Fixes:
cb9ae4273d68 ("dri: add loader_dri_create_image helper")
Signed-off-by: Heinrich Fink <hfink@snap.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11844>
Antonio Caggiano [Thu, 6 May 2021 11:20:01 +0000 (13:20 +0200)]
pps: Panfrost documentation
Add documentation for the Panfrost Perfetto datasource.
Signed-off-by: Antonio Caggiano <antonio.caggiano@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10215>
Antonio Caggiano [Thu, 18 Mar 2021 18:29:33 +0000 (19:29 +0100)]
pps: Panfrost pps driver
Add the Panfrost pps driver.
v2: Human readable names for counter blocks and use `unreachable`.
v3: Use libpanfrost_perf to collect counter values.
Signed-off-by: Antonio Caggiano <antonio.caggiano@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10215>
Pierre-Eric Pelloux-Prayer [Tue, 29 Jun 2021 08:10:17 +0000 (10:10 +0200)]
mesa: clear shader_info::is_lowered in prog_to_nir
This needs to be resetted each time prog_to_nir is called because it
turns st_nir_assign_vs_in_locations into a no-op when set.
Fixes:
81d106d6eca ("radeonsi: lower IO intrinsics - complete rewrite of input/output scanning")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5001
Reviewed-by: Isaac Bosompem <mrisaacb@google.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11636>
Iago Toral Quiroga [Tue, 13 Jul 2021 07:47:53 +0000 (09:47 +0200)]
v3dv: use defines for push constant offsets used by texel buffer copy shaders
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11843>
Iago Toral Quiroga [Tue, 13 Jul 2021 07:23:13 +0000 (09:23 +0200)]
v3dv: allow batching texel buffer copies for 3D images
For these we only need to check that the depth extent we are
copying is the same across regions in the batch, since we use
that to specify the number of layers in the framebuffer used
for the copy.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11843>
Iago Toral Quiroga [Mon, 12 Jul 2021 11:16:06 +0000 (13:16 +0200)]
v3dv: implement layered texel buffer copies using a geometry shader
Instead of specifying a separate framebuffer per layer which is expected
to be much slower.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11843>
Iago Toral Quiroga [Mon, 12 Jul 2021 10:38:05 +0000 (12:38 +0200)]
v3dv: fix push constant range for texel buffer copy pipelines
As per get_texel_buffer_copy_fs(), we load 24 bytes of data.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11843>
Erik Faye-Lund [Wed, 16 Jun 2021 11:00:57 +0000 (13:00 +0200)]
zink: remove duplicate format-mapping on little-endian
Doing *both* of thse ends up rewriting the previous mapping. Since this
doesn't seem to have lead to issues, it seems like the new mapping works
just as well.
Fixes:
a22a1c0324e ("zink: Fix VK_FORMAT_A8B8G8R8_SRGB_PACK32 mapping on big-endian")
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11417>
James Jones [Fri, 9 Jul 2021 15:40:19 +0000 (08:40 -0700)]
loader: Handle failure to load DRI driver library
I factored out the chunk of loader code that dlopen()s
libraries from the rest of the DRI driver loader function
in this commit:
commit
bc343154f816b637b200933e1c1a35db19f478ef
Author: James Jones <jajones@nvidia.com>
Date: Thu Apr 22 23:17:08 2021 -0700
loader: Factor out driver library loading code
However, I failed to adjust the DRI loader function that
now uses the new helper function to handle the case where
the requested DRI library is not found.
This change restores the prior behavior, and also ensures
loader_open_driver() consistently returns NULL in the
out_driver_handle parameter on failure.
Fixes:
bc343154f816 ("loader: Factor out driver library loading code")
Signed-off-by: James Jones <jajones@nvidia.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11807>
Daniel Schürmann [Thu, 8 Jul 2021 14:49:18 +0000 (16:49 +0200)]
aco: fix extract_vector optimization
If the allocated_vec map contains a different RegType
for the elements, ensure that the size matches exactly.
Otherwise, it could happen that extracting a dword
element matched with a subdword element.
No fossil-db changes.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11823>
Daniel Schürmann [Wed, 7 Jul 2021 15:41:56 +0000 (17:41 +0200)]
aco: fix self-intersecting register swaps
Splitting self-intersecting register swaps into
3 sections was unnecessary and only worked because
the middle section was always empty for full dword
swaps.
No fossil-db changes.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11823>
Dave Airlie [Tue, 13 Jul 2021 02:09:02 +0000 (12:09 +1000)]
lavapipe: add the separate depth/stencil layout enable.
This was missing.
Fixes:
ac07ad8414cd ("lavapipe: implement KHR_separate_depth_stencil_layouts")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11840>
Alyssa Rosenzweig [Tue, 13 Jul 2021 00:27:03 +0000 (20:27 -0400)]
panfrost: Fix format swizzles on G72
Careless typo. Cleaner fix suggested by Icecream95.
Reported-by: macc24
Tested-by: macc24
Fixes:
531d80de7c7 ("panfrost: Specialize blendable_formats for v6")
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11837>
Mike Blumenkrantz [Tue, 13 Jul 2021 00:38:17 +0000 (20:38 -0400)]
relnotes: add some missing zink/lavapipe updates
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11838>
Mike Blumenkrantz [Tue, 13 Jul 2021 00:34:46 +0000 (20:34 -0400)]
features: mark off line rasterization for lavapipe
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11838>
Mike Blumenkrantz [Fri, 9 Apr 2021 14:14:30 +0000 (10:14 -0400)]
zink: add a pipe_screen::resource_create_with_modifiers hook
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Tue, 22 Jun 2021 12:33:04 +0000 (08:33 -0400)]
zink: add fallback for linear modifier use
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Wed, 14 Apr 2021 14:07:32 +0000 (10:07 -0400)]
zink: don't pass modifier count to first image create
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Wed, 14 Apr 2021 14:06:20 +0000 (10:06 -0400)]
zink: explicitly disallow using the modifier image create for non-linear images
the first call is just for the base image, the second call is the one that needs
modifiers
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Fri, 9 Apr 2021 14:13:53 +0000 (10:13 -0400)]
zink: use VkImageDrmFormatModifierListCreateInfoEXT for creating from modifier array
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Fri, 9 Apr 2021 14:12:40 +0000 (10:12 -0400)]
zink: add a pipe_screen::resource_get_param hook
this is used for querying image properties
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Wed, 14 Apr 2021 12:29:16 +0000 (08:29 -0400)]
zink: store modifier aspect to resource
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Thu, 8 Apr 2021 20:24:51 +0000 (16:24 -0400)]
zink: start storing modifiers to the base resource struct
need these for rebinds
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Mike Blumenkrantz [Thu, 8 Apr 2021 20:10:24 +0000 (16:10 -0400)]
zink: populate modifier props onto screen object during init
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Erik Faye-Lund [Mon, 19 Oct 2020 12:17:27 +0000 (14:17 +0200)]
zink: implement support for non-planar DRM modifiers
Planar DRM modifiers are a bit more tricky, so let's kick that ball down
the road. For now, this should help a bit.
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10782>
Jason Ekstrand [Sun, 11 Jul 2021 14:44:24 +0000 (09:44 -0500)]
mailmap: Add two more lines for Alyssa Rosenzweig
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11788>
Jason Ekstrand [Sun, 11 Jul 2021 04:28:52 +0000 (23:28 -0500)]
Convert a few files to UTF-8
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11788>
Jason Ekstrand [Thu, 8 Jul 2021 20:35:43 +0000 (15:35 -0500)]
mailmap: Update for Emma's new e-mail address
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11788>
Alyssa Rosenzweig [Wed, 5 May 2021 21:38:02 +0000 (17:38 -0400)]
pan/bi: Add a bundling heuristic
Pick instructions locally that are likely to minimize nop's globally.
Note the obvious greedy solution is ineffective. Instead the principle
is to choose instructions which will increase _future_ scheduler
freedom, on the assumption that means fewer nop's overall will be
needed.
There is no concern about register pressure and little concern about
message timing, since this is post-RA and message ordering was fixed.
total tuples in shared programs: 125304 -> 123770 (-1.22%)
tuples in affected programs: 88301 -> 86767 (-1.74%)
helped: 548
HURT: 198
helped stats (abs) min: 1.0 max: 31.0 x̄: 3.43 x̃: 2
helped stats (rel) min: 0.33% max: 16.67% x̄: 3.09% x̃: 2.44%
HURT stats (abs) min: 1.0 max: 12.0 x̄: 1.73 x̃: 1
HURT stats (rel) min: 0.27% max: 11.11% x̄: 2.82% x̃: 2.35%
95% mean confidence interval for tuples value: -2.39 -1.72
95% mean confidence interval for tuples %-change: -1.77% -1.27%
Tuples are helped.
total clauses in shared programs: 26059 -> 25655 (-1.55%)
clauses in affected programs: 10821 -> 10417 (-3.73%)
helped: 311
HURT: 42
helped stats (abs) min: 1.0 max: 6.0 x̄: 1.44 x̃: 1
helped stats (rel) min: 0.48% max: 16.00% x̄: 5.26% x̃: 5.00%
HURT stats (abs) min: 1.0 max: 2.0 x̄: 1.05 x̃: 1
HURT stats (rel) min: 2.13% max: 33.33% x̄: 7.49% x̃: 6.25%
95% mean confidence interval for clauses value: -1.26 -1.02
95% mean confidence interval for clauses %-change: -4.28% -3.21%
Clauses are helped.
total cycles in shared programs: 12154.79 -> 12114.83 (-0.33%)
cycles in affected programs: 1907.63 -> 1867.67 (-2.09%)
helped: 227
HURT: 99
helped stats (abs) min: 0.
041665999999999315 max: 1.
2083360000000027 x̄: 0.22 x̃: 0
helped stats (rel) min: 0.36% max: 20.00% x̄: 4.14% x̃: 3.85%
HURT stats (abs) min: 0.
041665999999999315 max: 0.5 x̄: 0.09 x̃: 0
HURT stats (rel) min: 0.28% max: 9.09% x̄: 2.75% x̃: 2.63%
95% mean confidence interval for cycles value: -0.15 -0.09
95% mean confidence interval for cycles %-change: -2.51% -1.59%
Cycles are helped.
total arith in shared programs: 4658.13 -> 4603.42 (-1.17%)
arith in affected programs: 3449.83 -> 3395.12 (-1.59%)
helped: 509
HURT: 249
helped stats (abs) min: 0.
041665999999999315 max: 1.
2083360000000027 x̄: 0.14 x̃: 0
helped stats (rel) min: 0.36% max: 20.00% x̄: 3.56% x̃: 2.63%
HURT stats (abs) min: 0.
041665999999999315 max: 0.5 x̄: 0.07 x̃: 0
HURT stats (rel) min: 0.28% max: 25.00% x̄: 3.32% x̃: 2.27%
95% mean confidence interval for arith value: -0.09 -0.06
95% mean confidence interval for arith %-change: -1.64% -0.97%
Arith are helped.
total quadwords in shared programs: 111394 -> 110114 (-1.15%)
quadwords in affected programs: 78074 -> 76794 (-1.64%)
helped: 503
HURT: 204
helped stats (abs) min: 1.0 max: 43.0 x̄: 3.33 x̃: 2
helped stats (rel) min: 0.41% max: 13.79% x̄: 3.02% x̃: 2.44%
HURT stats (abs) min: 1.0 max: 26.0 x̄: 1.94 x̃: 1
HURT stats (rel) min: 0.48% max: 7.55% x̄: 2.44% x̃: 2.22%
95% mean confidence interval for quadwords value: -2.16 -1.46
95% mean confidence interval for quadwords %-change: -1.67% -1.21%
Quadwords are helped.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Thu, 21 Jan 2021 02:02:12 +0000 (21:02 -0500)]
pan/bi: Calculate dependency graph when bundling
Code is ported from Midgard, modified to be scalar, post-RA, and to put
the arrays on the worklist instead of the instruction to save memory.
This enables out-of-order scheduling.
total tuples in shared programs: 128691 -> 125304 (-2.63%)
tuples in affected programs: 114091 -> 110704 (-2.97%)
helped: 844
HURT: 377
helped stats (abs) min: 1.0 max: 150.0 x̄: 4.88 x̃: 3
helped stats (rel) min: 0.30% max: 26.42% x̄: 5.56% x̃: 4.35%
HURT stats (abs) min: 1.0 max: 8.0 x̄: 1.94 x̃: 1
HURT stats (rel) min: 0.20% max: 33.33% x̄: 6.84% x̃: 3.23%
95% mean confidence interval for tuples value: -3.16 -2.38
95% mean confidence interval for tuples %-change: -2.19% -1.27%
Tuples are helped.
total clauses in shared programs: 27579 -> 26059 (-5.51%)
clauses in affected programs: 20606 -> 19086 (-7.38%)
helped: 941
HURT: 39
helped stats (abs) min: 1.0 max: 21.0 x̄: 1.66 x̃: 1
helped stats (rel) min: 0.69% max: 44.44% x̄: 10.48% x̃: 9.09%
HURT stats (abs) min: 1.0 max: 2.0 x̄: 1.15 x̃: 1
HURT stats (rel) min: 1.89% max: 10.00% x̄: 4.73% x̃: 4.55%
95% mean confidence interval for clauses value: -1.63 -1.47
95% mean confidence interval for clauses %-change: -10.27% -9.48%
Clauses are helped.
total cycles in shared programs: 12262.54 -> 12154.79 (-0.88%)
cycles in affected programs: 2210.54 -> 2102.79 (-4.87%)
helped: 374
HURT: 56
helped stats (abs) min: 0.
041665999999999315 max: 6.25 x̄: 0.30 x̃: 0
helped stats (rel) min: 0.42% max: 26.00% x̄: 6.90% x̃: 7.14%
HURT stats (abs) min: 0.
041665999999999315 max: 0.
5833319999999986 x̄: 0.11 x̃: 0
HURT stats (rel) min: 0.16% max: 100.00% x̄: 55.17% x̃: 50.00%
95% mean confidence interval for cycles value: -0.29 -0.21
95% mean confidence interval for cycles %-change: -1.37% 3.73%
Inconclusive result (%-change mean confidence interval includes 0).
total arith in shared programs: 4852.29 -> 4658.13 (-4.00%)
arith in affected programs: 4525.17 -> 4331 (-4.29%)
helped: 1112
HURT: 166
helped stats (abs) min: 0.
041665999999999315 max: 6.25 x̄: 0.19 x̃: 0
helped stats (rel) min: 0.42% max: 33.33% x̄: 6.59% x̃: 5.36%
HURT stats (abs) min: 0.
041665999999999315 max: 0.
5833319999999986 x̄: 0.07 x̃: 0
HURT stats (rel) min: 0.16% max: 100.00% x̄: 25.05% x̃: 2.40%
95% mean confidence interval for arith value: -0.17 -0.14
95% mean confidence interval for arith %-change: -3.44% -1.51%
Arith are helped.
total quadwords in shared programs: 117141 -> 111394 (-4.91%)
quadwords in affected programs: 104390 -> 98643 (-5.51%)
helped: 1245
HURT: 76
helped stats (abs) min: 1.0 max: 69.0 x̄: 4.74 x̃: 4
helped stats (rel) min: 0.28% max: 35.00% x̄: 7.88% x̃: 6.45%
HURT stats (abs) min: 1.0 max: 8.0 x̄: 2.01 x̃: 1
HURT stats (rel) min: 0.20% max: 10.00% x̄: 3.52% x̃: 4.25%
95% mean confidence interval for quadwords value: -4.61 -4.09
95% mean confidence interval for quadwords %-change: -7.56% -6.88%
Quadwords are helped.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Thu, 8 Jul 2021 21:32:28 +0000 (17:32 -0400)]
pan/bi: Handle 4-src instructions in scheduler
Spill to a move. This allows us to emit SHADDX and general CSEL.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Thu, 8 Jul 2021 21:31:05 +0000 (17:31 -0400)]
pan/bi: Add bi_before_tuple convenience method
For constructing instructions during scheduling.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Thu, 8 Jul 2021 21:30:32 +0000 (17:30 -0400)]
pan/bi: Handle multiple destinations in scheduler
Needed for correct handling of pseudo ops like SHADDX and CUBEFACE
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Mon, 12 Jul 2021 15:11:27 +0000 (11:11 -0400)]
pan/bi: Do helper termination analysis on clauses
Unlike the dependency analysis for the skip bits which is a function of
the data flow graph, the thread termination analysis is dependent on the
actual sequence of instructions. As such, it must be done after
scheduling to be correct in the presence of out-of-order scheduling.
Furthermore it's specified in terms of clauses, not instructions.
Reflecting this in our code gets a nice simplification. As a side effect
this puts extra td flags on subsequent clauses, which matches the DDK's
behaviour. (Maybe td is just a hint?)
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Mon, 12 Jul 2021 16:05:28 +0000 (12:05 -0400)]
pan/bi: Update ins->link after scheduling
Otherwise foreach_instr and friends will be subtly wrong. None of our
current passes rely on this but it's a footgun.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Mon, 12 Jul 2021 16:00:44 +0000 (12:00 -0400)]
pan/bi: Make bi_foreach_instr_in_tuple safer
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Mon, 12 Jul 2021 14:26:05 +0000 (10:26 -0400)]
pan/bi: Refuse to CSE non-SSA sources
They might have changed in between.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10961>
Alyssa Rosenzweig [Mon, 12 Jul 2021 22:09:43 +0000 (18:09 -0400)]
panfrost: Only access blitter from per-gen
Now the blitter may be compiled per-gen legally.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11785>
Alyssa Rosenzweig [Mon, 12 Jul 2021 22:05:18 +0000 (18:05 -0400)]
panfrost: Init/destroy blitter from per-gen file
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11785>