From: Matt Turner <mattst88@gmail.com>
Date: Wed, 18 Mar 2015 06:30:43 +0000 (-0700)
Subject: nir: Recognize sat(add(b2f(a), b2f(b))) as a logical OR.
X-Git-Tag: upstream/17.1.0~19825
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=3fb56805f0dbfe929cd733e15566197bc0733290;p=platform%2Fupstream%2Fmesa.git

nir: Recognize sat(add(b2f(a), b2f(b))) as a logical OR.

Transform this into b2f(or(a, b)).

instructions in affected programs:     432 -> 430 (-0.46%)
helped:                                2

Acked-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
---

diff --git a/src/glsl/nir/nir_opt_algebraic.py b/src/glsl/nir/nir_opt_algebraic.py
index f956edf..1ee51a0 100644
--- a/src/glsl/nir/nir_opt_algebraic.py
+++ b/src/glsl/nir/nir_opt_algebraic.py
@@ -97,6 +97,7 @@ optimizations = [
    (('fmin', ('fmax', ('fmin', ('fmax', a, 0.0), 1.0), 0.0), 1.0), ('fmin', ('fmax', a, 0.0), 1.0)),
    # Emulating booleans
    (('fmul', ('b2f', a), ('b2f', b)), ('b2f', ('iand', a, b))),
+   (('fsat', ('fadd', ('b2f', a), ('b2f', b))), ('b2f', ('ior', a, b))),
    # Comparison with the same args.  Note that these are not done for
    # the float versions because NaN always returns false on float
    # inequalities.