diff options
| author | Ian Romanick <[email protected]> | 2016-03-07 10:55:21 -0800 |
|---|---|---|
| committer | Ian Romanick <[email protected]> | 2016-03-22 14:42:42 -0700 |
| commit | 9442db4f89156713bcb241803f6de7c0777262c5 (patch) | |
| tree | 5c1f865b90f12115a1c0623bb647efd1cdf244a9 /src/compiler/nir/nir_opt_algebraic.py | |
| parent | 18c5fa1122eddc2da0912749760356e570631163 (diff) | |
i965: Have NIR lower flrp on pre-GEN6 vec4 backend
Previously we were doing the lowering by hand in vec4_visitor::emit_lrp.
By doing it in NIR, we have the opportunity for NIR to do additional
optimization of the expanded code.
This also enables optimizations added by the next commit.
shader-db results:
G4X / Ironlake
total instructions in shared programs: 4024401 -> 4016538 (-0.20%)
instructions in affected programs: 447686 -> 439823 (-1.76%)
helped: 2623
HURT: 0
total cycles in shared programs: 84375846 -> 84328296 (-0.06%)
cycles in affected programs: 16964960 -> 16917410 (-0.28%)
helped: 2556
HURT: 41
Unsurprisingly, no changes on later platforms.
v2: Formatting and comment changes suggested by Matt.
Signed-off-by: Ian Romanick <[email protected]>
Reviewed-by: Matt Turner <[email protected]>
Diffstat (limited to 'src/compiler/nir/nir_opt_algebraic.py')
0 files changed, 0 insertions, 0 deletions