Friedrich Vock
f1742d36f3
radv/rt: Fix memory leak when compiling libraries
...
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29579 >
2024-06-06 21:56:11 +00:00
Daniel Schürmann
c452a4d1cc
aco/ra: use round robin register allocation
...
Totals from 74681 (94.06% of 79395) affected shaders: (GFX11)
MaxWaves: 2265668 -> 2263546 (-0.09%); split: +0.01%, -0.10%
Instrs: 44941647 -> 44412809 (-1.18%); split: -1.23%, +0.05%
CodeSize: 234173852 -> 232009132 (-0.92%); split: -0.97%, +0.05%
VGPRs: 3033208 -> 3403000 (+12.19%); split: -0.02%, +12.22%
Latency: 305575738 -> 301100302 (-1.46%); split: -1.70%, +0.23%
InvThroughput: 49366070 -> 49020000 (-0.70%); split: -0.91%, +0.21%
VClause: 875748 -> 854930 (-2.38%); split: -2.65%, +0.27%
SClause: 1369614 -> 1327212 (-3.10%); split: -3.43%, +0.33%
Copies: 2887932 -> 2883061 (-0.17%); split: -1.93%, +1.76%
Branches: 885041 -> 885101 (+0.01%); split: -0.01%, +0.02%
VALU: 25218078 -> 25215170 (-0.01%); split: -0.20%, +0.19%
SALU: 4328640 -> 4326052 (-0.06%); split: -0.20%, +0.14%
VOPD: 9129 -> 9611 (+5.28%); split: +7.48%, -2.20%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29235 >
2024-06-06 21:02:15 +00:00
Daniel Schürmann
197943ae27
aco/ra: change heuristic to first fit
...
Totals from 73175 (92.17% of 79395) affected shaders: (GFX11)
MaxWaves: 2217690 -> 2217930 (+0.01%); split: +0.02%, -0.01%
Instrs: 44780731 -> 44784895 (+0.01%); split: -0.14%, +0.15%
CodeSize: 233238960 -> 233255604 (+0.01%); split: -0.11%, +0.12%
VGPRs: 3009116 -> 3007684 (-0.05%); split: -0.29%, +0.24%
Latency: 304320163 -> 304286592 (-0.01%); split: -0.31%, +0.30%
InvThroughput: 49121992 -> 49145025 (+0.05%); split: -0.20%, +0.25%
VClause: 872566 -> 873242 (+0.08%); split: -0.25%, +0.33%
SClause: 1359666 -> 1361640 (+0.15%); split: -0.11%, +0.26%
Copies: 2879649 -> 2881646 (+0.07%); split: -1.13%, +1.20%
Branches: 887102 -> 887093 (-0.00%); split: -0.01%, +0.01%
VALU: 25128240 -> 25128572 (+0.00%); split: -0.12%, +0.12%
SALU: 4328852 -> 4330559 (+0.04%); split: -0.07%, +0.11%
VOPD: 8861 -> 8992 (+1.48%); split: +2.63%, -1.15%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29235 >
2024-06-06 21:02:15 +00:00
Daniel Schürmann
d76fc005b6
aco/ra: re-use registers from killed operands
...
Totals from 77283 (97.34% of 79395) affected shaders: (GFX11)
MaxWaves: 2348498 -> 2348250 (-0.01%); split: +0.01%, -0.02%
Instrs: 45304558 -> 45097367 (-0.46%); split: -0.57%, +0.11%
CodeSize: 235719656 -> 234957768 (-0.32%); split: -0.43%, +0.11%
VGPRs: 3065984 -> 3073244 (+0.24%); split: -0.41%, +0.65%
Latency: 308010576 -> 307008565 (-0.33%); split: -0.85%, +0.52%
InvThroughput: 49560307 -> 49464214 (-0.19%); split: -0.54%, +0.34%
VClause: 881895 -> 879739 (-0.24%); split: -0.78%, +0.53%
SClause: 1388139 -> 1374634 (-0.97%); split: -1.12%, +0.14%
Copies: 2918583 -> 2910434 (-0.28%); split: -1.92%, +1.64%
Branches: 893947 -> 893712 (-0.03%); split: -0.06%, +0.03%
VALU: 25260728 -> 25256766 (-0.02%); split: -0.20%, +0.19%
SALU: 4377750 -> 4373595 (-0.09%); split: -0.17%, +0.07%
VOPD: 8603 -> 9163 (+6.51%); split: +8.54%, -2.03%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29235 >
2024-06-06 21:02:15 +00:00
Daniel Schürmann
b054cfe704
aco/ra: move can_write_m0() check into get_reg_specified()
...
This way, affinities are also covered.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29235 >
2024-06-06 21:02:15 +00:00
Daniel Schürmann
8e817cf52b
aco/ra: refactor get_reg_simple() with increased stride.
...
This should avoid some redundant calls.
Totals from 153 (0.19% of 79395) affected shaders: (GFX11)
Instrs: 301717 -> 301687 (-0.01%); split: -0.06%, +0.05%
CodeSize: 1583080 -> 1582988 (-0.01%); split: -0.06%, +0.05%
VGPRs: 10068 -> 10348 (+2.78%)
Latency: 6685446 -> 6685475 (+0.00%); split: -0.11%, +0.11%
InvThroughput: 999241 -> 999316 (+0.01%); split: -0.01%, +0.02%
VClause: 3868 -> 3870 (+0.05%)
Copies: 23752 -> 23769 (+0.07%); split: -0.27%, +0.34%
Branches: 6479 -> 6480 (+0.02%)
VALU: 179290 -> 179307 (+0.01%); split: -0.04%, +0.04%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29235 >
2024-06-06 21:02:15 +00:00
Daniel Schürmann
1b0edf3f33
aco/ra: Fix array access when finding register for subdword variables
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29235 >
2024-06-06 21:02:15 +00:00
Daniel Schürmann
5326e033ff
aco/ra: fix handling of killed operands in compact_relocate_vars()
...
Found by inspection.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29235 >
2024-06-06 21:02:14 +00:00
Samuel Pitoiset
afa2070c99
radv: initialize compute preambles with the common helper
...
The PM4 mechanism can emit paired packets on GFX11+ when possible.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29452 >
2024-06-06 20:26:47 +00:00
Samuel Pitoiset
3c8b48e310
ac,radeonsi: add a function to initialize compute preambles
...
Preambles are very similar between RADV and RadeonSI.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29452 >
2024-06-06 20:26:47 +00:00
Samuel Pitoiset
428601095c
ac,radeonsi import PM4 state from RadeonSI
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29452 >
2024-06-06 20:26:47 +00:00
Samuel Pitoiset
15fe733703
radv: add a helper to get image VA
...
Similar to buffer, and less error prone.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29428 >
2024-06-06 18:21:33 +00:00
Rhys Perry
4cfb7a0c17
aco: remove support for sub-dword push constants
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29480 >
2024-06-06 17:52:05 +00:00
Rhys Perry
e21312018e
ac/llvm: remove support for sub-dword push constants
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29480 >
2024-06-06 17:52:05 +00:00
Rhys Perry
41c5f71343
radv: lower sub-dword push constants
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29480 >
2024-06-06 17:52:05 +00:00
Rhys Perry
69b7fcd775
ac/nir: support lowering of sub-dword push constants
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29480 >
2024-06-06 17:52:04 +00:00
Georg Lehmann
3fb1a64918
aco: move s_add_u32 -> s_addk_i32 optimization fully to ra
...
Having this in one place is better.
When I wrote the old I wasn't aware that checking the kill flag on definitions
is the same as checking zero uses.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29512 >
2024-06-06 16:28:23 +00:00
Georg Lehmann
60f3f0fdbb
aco/ra: use a switch to check vop2acc instruction support
...
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29512 >
2024-06-06 16:28:23 +00:00
Georg Lehmann
fdc2fb6835
aco: move literal unswizzle opt to RA
...
Much simpler.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29512 >
2024-06-06 16:28:23 +00:00
Georg Lehmann
c63c750380
aco/gfx11+: fix inline constants for v_pk_fmac_f16
...
On newer hardware, the hi operation reads the lo half of the inline constant.
On older hardware, it reads the hi half (zero).
I tested this on Navi31 for gfx11 and Raphael for gfx10.
Foz-DB Navi31:
Totals from 4 (0.01% of 79395) affected shaders:
CodeSize: 36832 -> 36448 (-1.04%)
Latency: 20362 -> 20334 (-0.14%)
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29512 >
2024-06-06 16:28:23 +00:00
Georg Lehmann
39380d475a
aco: add affinities for possible sopk optimizations
...
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29512 >
2024-06-06 16:28:23 +00:00
Georg Lehmann
fac475bc25
aco: rework how affinities for acc operands are determined
...
Improve accuracy by adding a helper that's also used by
the optimization function.
Foz-DB Navi31:
Totals from 50 (0.06% of 79206) affected shaders:
CodeSize: 126148 -> 126128 (-0.02%); split: -0.05%, +0.04%
Latency: 334049 -> 334060 (+0.00%); split: -0.00%, +0.00%
InvThroughput: 59203 -> 59205 (+0.00%)
Copies: 2011 -> 1998 (-0.65%); split: -0.75%, +0.10%
VALU: 14221 -> 14208 (-0.09%); split: -0.11%, +0.01%
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29512 >
2024-06-06 16:28:23 +00:00
Samuel Pitoiset
5b6207b282
radv: only set valid bitfields for CB/DS surfaces address on GFX12
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29566 >
2024-06-06 15:42:35 +00:00
Samuel Pitoiset
fe78ad2690
radv: fix emitting VGT_PRIMITIVEID_RESET in the GFX preamble on GFX12
...
It's a uconfig register.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29566 >
2024-06-06 15:42:35 +00:00
Samuel Pitoiset
be3c837c04
radv: update configuring COVERAGE_TO_SHADER_SELECT on GFX12
...
This bit has been moved to SPI_PS_INPUT_ENA.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29566 >
2024-06-06 15:42:35 +00:00
Samuel Pitoiset
27496928e4
radv: update configuring depth clamp enable on GFX12
...
DISABLE_VIEWPORT_CLAMP has been moved to a new register.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29566 >
2024-06-06 15:42:35 +00:00
Rhys Perry
8e475bba61
aco: implement nir_intrinsic_nop_amd and nir_intrinsic_sleep_amd
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29466 >
2024-06-06 14:26:52 +00:00
Rhys Perry
1ad05d4ca8
aco: implement nir_atomic_op_ordered_add_gfx12_amd
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29466 >
2024-06-06 14:26:52 +00:00
Rhys Perry
0dee5fdd3c
aco: don't combine vgpr into writelane src0
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29466 >
2024-06-06 14:26:52 +00:00
Rhys Perry
2a4424425a
aco/gfx12: fix s_wait_event immediate
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29466 >
2024-06-06 14:26:52 +00:00
Rhys Perry
26c981b2b8
ac/nir: skip subgroup_id/local_invocation_index lowering for gfx12
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29466 >
2024-06-06 14:26:52 +00:00
Rhys Perry
c651eed1d8
aco/gfx12: implement load_subgroup_id
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29466 >
2024-06-06 14:26:52 +00:00
Rhys Perry
61531b19cd
ac/llvm: implement load_subgroup_id
...
Usually this is lowered in NIR, but GFX12 needs to use an intrinsic.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29466 >
2024-06-06 14:26:51 +00:00
Eric Engestrom
ea5b3bfcd1
radv/ci: move radv manual rules into their own group
...
Makes it easier to re-use.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29550 >
2024-06-06 13:00:04 +00:00
Eric Engestrom
47bd1cff4b
radv/ci: fix manual rules
...
It was set to "always run" for amd common files changes when I obviously
meant for it to be manual and messed up my copy/paste when I wrote that.
Fixes: ebaede788e ("amd/ci: limit radv jobs to radv + aco files changes")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29550 >
2024-06-06 13:00:04 +00:00
Samuel Pitoiset
4400ecafa7
Revert "radv/ci: Bring back vkcts-navi21-llvm-valve"
...
This job is consuming CI resources for nothing, it's still broken and
it's completely useless.
This reverts commit 8dc364806e .
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29568 >
2024-06-06 12:22:33 +00:00
Samuel Pitoiset
e2db42298d
amd/common: add MIN_LOD for texture descriptors on GFX12
...
RADV will need that.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29321 >
2024-06-06 10:15:10 +00:00
Samuel Pitoiset
8cb2cad434
ac,radv,radeonsi: add a function to build texture descriptors
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29321 >
2024-06-06 10:15:10 +00:00
Samuel Pitoiset
4bb308d403
radv: use pipe_format when building image view descriptors
...
This simplifies things before adding a common helper for building
texture descriptors.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29321 >
2024-06-06 10:15:10 +00:00
Samuel Pitoiset
57d0d63d01
radv: only emit CB_COLOR0_DCC_CONTROL on GFX8
...
This register doesn't exist on GFX6-7 (no DCC at all).
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29545 >
2024-06-06 07:58:19 +02:00
Samuel Pitoiset
07b0096011
radv: only emit SPI_SHADER_PGM_SRC3_GS on GFX7+
...
This register doesn't exist on GFX6.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29545 >
2024-06-06 07:58:00 +02:00
Samuel Pitoiset
4a7150b469
radv: do not set VGT_SHADER_STAGES_EN.DYNAMIC_HS on GFX9
...
This bit doesn't exist.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29545 >
2024-06-06 07:58:00 +02:00
Samuel Pitoiset
4a75b50eb8
radv: only emit SQ_PERFCOUNTER_MASK on GFX7-9
...
This register doesn't exist on GFX10-10.3.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29545 >
2024-06-06 07:58:00 +02:00
Samuel Pitoiset
96e7ac027c
radv: only emit VGT_GS_MAX_PRIMS_PER_SUBGROUP on GFX9
...
This register doesn't exist on GFX10+.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29545 >
2024-06-06 07:58:00 +02:00
Samuel Pitoiset
f62a8f888f
radv: only set valid bitfields for CB/DS surfaces address
...
This isn't a problem in practice but better to mask them out.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29545 >
2024-06-06 07:57:59 +02:00
Marek Olšák
fbc237037c
radeonsi/gfx12: fix GPU deadlocks due to query result incoherency
...
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29510 >
2024-06-06 01:01:46 +00:00
Marek Olšák
060d5dacfd
ac: add gfx12 DCC shared code
...
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29510 >
2024-06-06 01:01:46 +00:00
Marek Olšák
1ea96a47cd
ac/nir/lower_ngg: use voffset in global_atomic_add for xfb
...
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29510 >
2024-06-06 01:01:46 +00:00
Marek Olšák
b002564633
ac/nir/lower_ngg: use global_atomic_amd to fix gfx12 streamout
...
The intrinsics are lowered before this is called.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29510 >
2024-06-06 01:01:46 +00:00
Marek Olšák
5db194b1d1
ac/descriptors: fix gfx12 regressions
...
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29510 >
2024-06-06 01:01:46 +00:00