Rhys Perry
8829fc3bd6
amd/lower_mem_access_bit_sizes: improve subdword/unaligned SMEM lowering
Summary of changes:
- handle unaligned 16-bit scalar loads when supported_dword=true
- increases the size of 8/16/32/64-bit buffer loads which are not dword
aligned, which can create less SMEM loads.
- handles when "bytes" is less than "bit_size / 8"
fossil-db (gfx1201):
Totals from 26 (0.03% of 79839) affected shaders:
Instrs: 12676 -> 12710 (+0.27%); split: -0.30%, +0.57%
CodeSize: 67272 -> 67384 (+0.17%); split: -0.24%, +0.40%
Latency: 44399 -> 44375 (-0.05%); split: -0.09%, +0.04%
SClause: 352 -> 344 (-2.27%)
SALU: 3972 -> 3992 (+0.50%)
SMEM: 554 -> 528 (-4.69%)
fossil-db (navi21):
Totals from 6 (0.01% of 79825) affected shaders:
Instrs: 2192 -> 2186 (-0.27%)
CodeSize: 12188 -> 12140 (-0.39%)
Latency: 10037 -> 10033 (-0.04%); split: -0.12%, +0.08%
SMEM: 124 -> 118 (-4.84%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: fbf0399517 ("amd/lower_mem_access_bit_sizes: lower all SMEM instructions to supported sizes")
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37953>
2025-10-21 22:10:34 +00:00
..
2025-10-21 22:10:34 +00:00
2025-10-21 12:39:53 +02:00
2025-10-15 23:01:33 +00:00
2025-10-21 12:39:53 +02:00
2025-09-03 11:47:00 +00:00
2025-10-09 12:29:55 +00:00
2025-10-07 19:34:29 +00:00
2025-10-21 12:37:58 +02:00
2025-10-21 19:50:47 +02:00
2025-08-21 02:05:26 +00:00
2025-10-20 15:45:07 +00:00
2025-10-14 20:53:10 +00:00
2025-10-17 08:11:11 +00:00
2025-09-16 06:16:19 +00:00
2025-10-21 21:55:53 +00:00
2025-10-21 12:39:53 +02:00
2025-09-22 16:35:26 +00:00
2025-10-17 03:52:19 +00:00
2025-10-10 19:53:15 +00:00
2025-10-21 20:57:41 +00:00
2025-10-21 19:15:58 +00:00
2025-10-21 12:39:53 +02:00
2025-09-09 07:04:55 +00:00
2025-10-16 20:23:21 +00:00
2025-10-20 15:45:06 +00:00
2025-10-21 19:32:55 +00:00
2025-10-16 19:25:35 +00:00
2025-10-20 17:46:38 +00:00