nir: Mark the 0.0 < abs(a) transformation as imprecise
authorIan Romanick <ian.d.romanick@intel.com>
Wed, 1 Aug 2018 17:13:55 +0000 (10:13 -0700)
committerIan Romanick <ian.d.romanick@intel.com>
Sat, 4 Aug 2018 08:12:03 +0000 (01:12 -0700)
commitea6c276436851d27394abd132b239e1c21f0e923
tree2c2faf69f0d3725b08cb7bb87585f637a70f7867
parent4bad50ded99229c5c4f2f7e6c213b226cfd00a2c
nir: Mark the 0.0 < abs(a) transformation as imprecise

Unlike the much older -abs(a) >= 0.0 transformation, this is not
precise.  The behavior changes if the source is NaN.

No shader-db changes on any platform.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Thomas Helland <thomashelland90@gmail.com>
src/compiler/nir/nir_opt_algebraic.py