diff options
-rw-r--r-- | .pick_status.json | 2 | ||||
-rw-r--r-- | src/compiler/nir/nir_opt_algebraic.py | 12 |
2 files changed, 10 insertions, 4 deletions
diff --git a/.pick_status.json b/.pick_status.json index 164afa4274d..55b04bb680a 100644 --- a/.pick_status.json +++ b/.pick_status.json @@ -814,7 +814,7 @@ "description": "nir/algebraic: mark some optimizations with fsat(NaN) as inexact", "nominated": true, "nomination_type": 0, - "resolution": 0, + "resolution": 1, "master_sha": null, "because_sha": null }, diff --git a/src/compiler/nir/nir_opt_algebraic.py b/src/compiler/nir/nir_opt_algebraic.py index 8f45b99060b..92bea7a4510 100644 --- a/src/compiler/nir/nir_opt_algebraic.py +++ b/src/compiler/nir/nir_opt_algebraic.py @@ -327,10 +327,14 @@ optimizations.extend([ (('fne', ('fneg', a), -1.0), ('fne', 1.0, a)), (('feq', -1.0, ('fneg', a)), ('feq', a, 1.0)), - (('flt', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('flt', a, b)), + # flt(fsat(a), b > 0 && b < 1) is inexact if a is NaN (fsat(NaN) is 0) + # because it returns True while flt(a, b) always returns False. + (('~flt', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('flt', a, b)), (('flt', '#b(is_gt_0_and_lt_1)', ('fsat(is_used_once)', a)), ('flt', b, a)), (('fge', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('fge', a, b)), - (('fge', '#b(is_gt_0_and_lt_1)', ('fsat(is_used_once)', a)), ('fge', b, a)), + # fge(b > 0 && b < 1, fsat(a)) is inexact if a is NaN (fsat(NaN) is 0) + # because it returns True while fge(b, a) always returns False. + (('~fge', '#b(is_gt_0_and_lt_1)', ('fsat(is_used_once)', a)), ('fge', b, a)), (('feq', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('feq', a, b)), (('fne', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('fne', a, b)), @@ -1767,7 +1771,9 @@ late_optimizations = [ (('fne', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('fne', a, b)), (('fge', ('fsat(is_used_once)', a), 1.0), ('fge', a, 1.0)), - (('flt', ('fsat(is_used_once)', a), 1.0), ('flt', a, 1.0)), + # flt(fsat(a), 1.0) is inexact because it returns True if a is NaN + # (fsat(NaN) is 0), while flt(a, 1.0) always returns FALSE. + (('~flt', ('fsat(is_used_once)', a), 1.0), ('flt', a, 1.0)), (('~fge', ('fmin(is_used_once)', ('fadd(is_used_once)', a, b), ('fadd', c, d)), 0.0), ('iand', ('fge', a, ('fneg', b)), ('fge', c, ('fneg', d)))), |