[Mesa-dev] [PATCH 11.1/22] nir: Mark bcsel-to-fmin (or fmax) transformations as inexact

[Ian Romanick]

From: Ian Romanick <ian.d.romanick at intel.com>

These transformations are inexact because section 4.7.1 (Range and
Precision) says:

    Operations and built-in functions that operate on a NaN are not
    required to return a NaN as the result.

The fmin or fmax might not return NaN in cases where the original
expression would be required to return NaN.

Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
---
 src/compiler/nir/nir_opt_algebraic.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/compiler/nir/nir_opt_algebraic.py b/src/compiler/nir/nir_opt_algebraic.py
index 17f4d9c..b643cda 100644
--- a/src/compiler/nir/nir_opt_algebraic.py
+++ b/src/compiler/nir/nir_opt_algebraic.py
@@ -168,8 +168,8 @@ optimizations = [
    (('flt(is_not_used_by_conditional)', ('fadd(is_used_once)', a, ('fneg', b)), 0.0), ('flt', a, b)),
 
    (('fge', ('fneg', ('fabs', a)), 0.0), ('feq', a, 0.0)),
-   (('bcsel', ('flt', b, a), b, a), ('fmin', a, b)),
-   (('bcsel', ('flt', a, b), b, a), ('fmax', a, b)),
+   (('~bcsel', ('flt', b, a), b, a), ('fmin', a, b)),
+   (('~bcsel', ('flt', a, b), b, a), ('fmax', a, b)),
    (('~bcsel', ('fge', a, b), b, a), ('fmin', a, b)),
    (('~bcsel', ('fge', b, a), b, a), ('fmax', a, b)),
    (('bcsel', ('inot', a), b, c), ('bcsel', a, c, b)),
-- 
2.9.5