OSDN Git Service

nir: Add a bool to float32 lowering pass
authorJason Ekstrand <jason.ekstrand@intel.com>
Mon, 22 Oct 2018 21:44:59 +0000 (16:44 -0500)
committerJason Ekstrand <jason@jlekstrand.net>
Mon, 14 Jan 2019 19:27:06 +0000 (19:27 +0000)
commit2d2737dcfe60f1e1b601c2c75a2695a544c9a3dc
treeb167db1e8e395512504798f60d56075114358cd4
parent09c3ff01dfd72aa84334fe82cc3319594fc99ac6
nir: Add a bool to float32 lowering pass

From @jekstrand's nir-1-bit-bool branch, with improved ior/inot lowering.

ior: fmax instead of fadd allows removing the fsat.

inot: seq(x, 0) can be better than fsub(1, x). On a2xx, it works better
with the scalar instruction set.

Reviewed-by: Jonathan Marek <jonathan@marek.ca>
src/compiler/Makefile.sources
src/compiler/nir/meson.build
src/compiler/nir/nir.h
src/compiler/nir/nir_lower_bool_to_float.c [new file with mode: 0644]