projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 05d02e5) | patch
[x86] add tests for fmax/fmin experimental intrinsics with 'fast' FMF; NFC
authorSanjay Patel <spatel@rotateright.com>
Thu, 10 Sep 2020 12:45:36 +0000 (08:45 -0400)
committerSanjay Patel <spatel@rotateright.com>
Thu, 10 Sep 2020 12:55:40 +0000 (08:55 -0400)
commit1ebb31b14cd175b3f272e232958d342221eb875c
treee50ee14e25aadce14f55407f18b5dcbb3a25a83atree | snapshot
parent05d02e5a4e54a04f050b52ee30d1860073bd8b34commit | diff
[x86] add tests for fmax/fmin experimental intrinsics with 'fast' FMF; NFC

D87391 proposes to change the lowerings for 'nnan'-only FMF.
That's the minimal requirement to get good codegen for x86,
but currently we have bugs hindering that output unless the
full 'fast' FMF is applied. These tests provide coverage for
the ideal lowerings.
llvm/test/CodeGen/X86/vector-reduce-fmax-fmin-fast.ll [new file with mode: 0644] blob
Domain: System / Toolchain;