fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-01-30 20:10:24 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

nir/opt_algebraic: optimize open coded pack_32_2x16
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
Details
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Details

Browse source 
Foz-DB Navi48:
Totals from 4 (0.00% of 80287) affected shaders:
Instrs: 6231 -> 6101 (-2.09%)
CodeSize: 35916 -> 35156 (-2.12%)
Latency: 72190 -> 71317 (-1.21%)
InvThroughput: 20817 -> 19962 (-4.11%)
VALU: 3145 -> 3029 (-3.69%)
VOPD: 310 -> 312 (+0.65%)

Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37937>
This commit is contained in:
Georg Lehmann 2025-10-17 14:42:45 +02:00 • committed by Marge Bot
parent d9bed33c11
commit 9ef0c96f26
1 changed files with 2 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
src/compiler/nir/nir_opt_algebraic.py
View file

@ -2134,6 +2134,8 @@ optimizations.extend([
(('ior', ('ishl', ('u2u32', 'a@8'), 24), ('ior', ('ishl', ('u2u32', 'b@8'), 16), ('ior', ('ishl', ('u2u32', 'c@8'), 8), ('u2u32', 'd@8')))),
('pack_32_4x8', ('vec4', d, c, b, a)), 'options->has_pack_32_4x8'),
(('ior', ('ishl', a, 16), ('u2u32', 'b@16')), ('pack_32_2x16_split', b, ('u2u16', a)), '!options->lower_pack_32_2x16_split && !options->lower_pack_split'),
# Mixed 16-bit/8-bit loads vectorized to 8-bit vector load and then lowered to 32-bit
(('ior', ('u2u16', ('unpack_32_4x8', a)), ('ishl', ('u2u16', ('unpack_32_4x8.y', a)), 8)),
('unpack_32_2x16_split_x', a), '!options->lower_unpack_32_2x16_split'),