diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2016-06-10 15:17:54 +0000 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2016-06-10 15:17:54 +0000 |
| commit | cccccd9ab5d63b8f65b71c302bc7960a0a78045e (patch) | |
| tree | 8b96eac8ecbb8172ed872b42ebdc8031d5dadf87 /llvm/test/CodeGen/X86/large-gep-scale.ll | |
| parent | 330a359fb308b3671f954f595f0927ebbeba579a (diff) | |
| download | bcm5719-llvm-cccccd9ab5d63b8f65b71c302bc7960a0a78045e.tar.gz bcm5719-llvm-cccccd9ab5d63b8f65b71c302bc7960a0a78045e.zip | |
[x86] add missing tests for fcmp ueq/one
Somehow, the codegen logic for these sequences has gone completely untested
until now (note the 2 compare instructions generated per test).
There's also an *Intel* AVX optimization opportunity exposed in these cases
and the existing tests. Intel's (but not AMD's) AVX spec shows that extra FP
predicates were added, so a single comparison should always be sufficient,
and operand commutation should never be necessary.
llvm-svn: 272397
Diffstat (limited to 'llvm/test/CodeGen/X86/large-gep-scale.ll')
0 files changed, 0 insertions, 0 deletions
