replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

math: Use single precision fmax in sp path 

Fixes fdim piglit on Turks

v2: use CL fmax instead of __builtin

Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Tom Stellard <tom.stellard@amd.com>
llvm-svn: 269807
This commit is contained in:
Jan Vesely 2016-05-17 19:44:01 +00:00
parent 35b544adc5
commit 973c1fa5f5
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
libclc/generic/lib/math/fdim.inc
View File

@ -25,7 +25,7 @@
_CLC_OVERLOAD _CLC_DEF __CLC_GENTYPE fdim(__CLC_GENTYPE x, __CLC_GENTYPE y) {
if (__builtin_isnan(x) || __builtin_isnan(y))
return as_float(QNANBITPATT_SP32);
return __builtin_fmax(x - y, 0);
return fmax(x - y, 0.0f);
}
#define __CLC_FDIM_VEC(width) \
_CLC_OVERLOAD _CLC_DEF float##width fdim(float##width x, float##width y) { \