diff options
author | Aaron Watry <awatry@gmail.com> | 2016-05-06 03:34:45 +0000 |
---|---|---|
committer | Aaron Watry <awatry@gmail.com> | 2016-05-06 03:34:45 +0000 |
commit | 55a8e0fd6debffb14bb6a40ee461d276460ec18a (patch) | |
tree | 08db19a50d6e93ace76536aa8991d920edf8bfad /llvm/test/Transforms/ForcedFunctionAttrs | |
parent | 832e7fa5794bb2259cb39006b58000ee5d56fa07 (diff) | |
download | bcm5719-llvm-55a8e0fd6debffb14bb6a40ee461d276460ec18a.tar.gz bcm5719-llvm-55a8e0fd6debffb14bb6a40ee461d276460ec18a.zip |
math: Add fdim implementation
Based on the amd-builtin, but explicitly vectorized for all sizes (not just
float4), and includes a vectorized double implementation.
Passes piglit (float) tests on pitcairn.
Signed-off-by: Aaron Watry <awatry@gmail.com>
Reviewed-by: Jan Vesely <jan.vesely@rutgers.edu>
llvm-svn: 268708
Diffstat (limited to 'llvm/test/Transforms/ForcedFunctionAttrs')
0 files changed, 0 insertions, 0 deletions