Mesa (master): compiler/nir: add lowering option for 16-bit fmod

[GitLab Mirror]

Module: Mesa
Branch: master
Commit: ca31df6f1fbc2c502644dd7be7424000dc02541c
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=ca31df6f1fbc2c502644dd7be7424000dc02541c

Author: Iago Toral Quiroga <itoral at igalia.com>
Date:   Wed Apr 18 10:50:35 2018 +0200

compiler/nir: add lowering option for 16-bit fmod

And enable it on Intel.

v2:
 - Squash the change to enable this lowering on Intel (Jason)

Reviewed-by: Jason Ekstrand <jason at jlekstrand.net>

---

 src/compiler/nir/nir.h                | 1 +
 src/compiler/nir/nir_opt_algebraic.py | 1 +
 src/intel/compiler/brw_compiler.c     | 1 +
 3 files changed, 3 insertions(+)

diff --git a/src/compiler/nir/nir.h b/src/compiler/nir/nir.h
index b6a2ba7ec8c..ee766d5238a 100644
--- a/src/compiler/nir/nir.h
+++ b/src/compiler/nir/nir.h
@@ -2169,6 +2169,7 @@ typedef struct nir_shader_compiler_options {
    bool lower_fpow;
    bool lower_fsat;
    bool lower_fsqrt;
+   bool lower_fmod16;
    bool lower_fmod32;
    bool lower_fmod64;
    /** Lowers ibitfield_extract/ubitfield_extract to ibfe/ubfe. */
diff --git a/src/compiler/nir/nir_opt_algebraic.py b/src/compiler/nir/nir_opt_algebraic.py
index 3326bbef684..ddd8d40676f 100644
--- a/src/compiler/nir/nir_opt_algebraic.py
+++ b/src/compiler/nir/nir_opt_algebraic.py
@@ -674,6 +674,7 @@ optimizations = [
    (('bcsel', ('ine', a, -1), ('ifind_msb', a), -1), ('ifind_msb', a)),
 
    # Misc. lowering
+   (('fmod at 16', a, b), ('fsub', a, ('fmul', b, ('ffloor', ('fdiv', a, b)))), 'options->lower_fmod16'),
    (('fmod at 32', a, b), ('fsub', a, ('fmul', b, ('ffloor', ('fdiv', a, b)))), 'options->lower_fmod32'),
    (('fmod at 64', a, b), ('fsub', a, ('fmul', b, ('ffloor', ('fdiv', a, b)))), 'options->lower_fmod64'),
    (('frem', a, b), ('fsub', a, ('fmul', b, ('ftrunc', ('fdiv', a, b)))), 'options->lower_fmod32'),
diff --git a/src/intel/compiler/brw_compiler.c b/src/intel/compiler/brw_compiler.c
index 4fa02ca4d10..b588d5be97b 100644
--- a/src/intel/compiler/brw_compiler.c
+++ b/src/intel/compiler/brw_compiler.c
@@ -33,6 +33,7 @@
    .lower_sub = true,                                                         \
    .lower_fdiv = true,                                                        \
    .lower_scmp = true,                                                        \
+   .lower_fmod16 = true,                                                      \
    .lower_fmod32 = true,                                                      \
    .lower_fmod64 = false,                                                     \
    .lower_bitfield_extract = true,                                            \