Ian Romanick
37ee462e03
nir/algebraic: Fix up extract_[iu]8 after loop unrolling
Skylake, Broadwell, and Haswell had similar results. (Skylake shown)
total instructions in shared programs: 15256840 -> 15256837 (<.01%)
instructions in affected programs: 4713 -> 4710 (-0.06%)
helped: 3
HURT: 0
helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
helped stats (rel) min: 0.06% max: 0.08% x̄: 0.06% x̃: 0.06%
total cycles in shared programs: 372286583 -> 372286583 (0.00%)
cycles in affected programs: 198516 -> 198516 (0.00%)
helped: 1
HURT: 1
helped stats (abs) min: 10 max: 10 x̄: 10.00 x̃: 10
helped stats (rel) min: <.01% max: <.01% x̄: <.01% x̃: <.01%
HURT stats (abs) min: 10 max: 10 x̄: 10.00 x̃: 10
HURT stats (rel) min: 0.01% max: 0.01% x̄: 0.01% x̃: 0.01%
No changes on any other Intel platform.
v2: Use a loop to generate patterns. Suggested by Jason.
Reviewed-by: Matt Turner <mattst88@gmail.com> [v1]
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
2019-03-08 22:24:19 -08:00
..
2019-03-08 15:00:50 +01:00
2019-03-06 14:13:45 -08:00
2019-03-08 22:24:19 -08:00
2019-03-05 13:04:06 +00:00
2019-03-08 18:44:57 -05:00
2019-03-08 18:44:57 -05:00
2019-02-14 11:20:00 +00:00
2019-03-08 18:00:19 +00:00
2018-10-31 18:20:25 +00:00
2019-03-08 18:00:19 +00:00
2019-02-26 12:49:07 +00:00
2019-03-08 22:03:34 -06:00
2019-02-26 11:07:23 +00:00
2019-03-08 15:13:36 +00:00
2019-03-08 10:35:12 -08:00
2019-03-05 12:27:35 +00:00
2019-03-07 13:45:14 +00:00
2019-02-21 18:06:05 +00:00
2019-03-05 10:02:47 +00:00
2018-10-19 16:50:26 +01:00