OpenHarmony/third_party_mesa3d
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

nir/algebraic: Allow fmin(a,a) optimization when flush denorm to zero is not set

Browse Source 
I was surprised this had any affect on Intel GPUs because we have been
unconditionally performing this optimization in the backend since June
2014.

Once that error is fixed (later in this MR), this change prevents a
couple dozen regressions in shader-db and around 90 regressions in
fossil-db. Many of the regressions in fossil-db were loss of SIMD32, and
that can be a big deal.

v2: Add 64-bit too. Suggested by Alyssa.

shader-db:

All Intel platforms had similar results. (Lunar Lake shown)
total instructions in shared programs: 16970141 -> 16970139 (<.01%)
instructions in affected programs: 40 -> 38 (-5.00%)
helped: 2 / HURT: 0

total cycles in shared programs: 914617580 -> 914617548 (<.01%)
cycles in affected programs: 3428 -> 3396 (-0.93%)
helped: 2 / HURT: 0

fossil-db:

All Intel platforms had similar results. (Lunar Lake shown)
Totals:
Cycle count: 30546028462 -> 30546025224 (-0.00%); split: -0.00%, +0.00%
Non SSA regs after NIR: 237017827 -> 237017731 (-0.00%)

Totals from 83 (0.01% of 706657) affected shaders:
Cycle count: 3042978 -> 3039740 (-0.11%); split: -0.13%, +0.02%
Non SSA regs after NIR: 78997 -> 78901 (-0.12%)

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34192>
This commit is contained in:
Ian Romanick
2025-03-06 16:16:53 -08:00
committed by Marge Bot
parent 44b01b55d5
commit 1d2ebeca17
1 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/compiler/nir/nir_opt_algebraic.py
View File

@@ -888,8 +888,12 @@ optimizations.extend([
(('bcsel', a, b, a), ('iand', a, b)),
(('bcsel', a, b, True), ('ior', ('inot', a), b)),
(('bcsel', a, False, b), ('iand', ('inot', a), b)),
(('~fmin', a, a), a),
(('~fmax', a, a), a),
(('fmin', 'a@64', a), a, '!nir_is_denorm_flush_to_zero(info->float_controls_execution_mode, 64)'),
(('fmin', 'a@32', a), a, '!nir_is_denorm_flush_to_zero(info->float_controls_execution_mode, 32)'),
(('fmin', 'a@16', a), a, '!nir_is_denorm_flush_to_zero(info->float_controls_execution_mode, 16)'),
(('fmax', 'a@64', a), a, '!nir_is_denorm_flush_to_zero(info->float_controls_execution_mode, 64)'),
(('fmax', 'a@32', a), a, '!nir_is_denorm_flush_to_zero(info->float_controls_execution_mode, 32)'),
(('fmax', 'a@16', a), a, '!nir_is_denorm_flush_to_zero(info->float_controls_execution_mode, 16)'),
(('imin', a, a), a),
(('imax', a, a), a),
(('umin', a, a), a),