projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 4af5b23) | patch
[X86] Use zero vector to extend to 512-bits for strict_fp_to_uint v2i1->v2f64 on...
authorCraig Topper <craig.topper@gmail.com>
Wed, 25 Dec 2019 18:40:40 +0000 (10:40 -0800)
committerCraig Topper <craig.topper@gmail.com>
Wed, 25 Dec 2019 18:46:00 +0000 (10:46 -0800)
commitc5b4a2386b51a18daad7e42040c685c2e9708c47
tree10230c17996ca7724daf50716af03b2bd0780690tree | snapshot
parent4af5b23db308c89edeb9fdc6dfbe7e6457b22f1dcommit | diff
[X86] Use zero vector to extend to 512-bits for strict_fp_to_uint v2i1->v2f64 on targets with AVX512F, but not AVX512VL.

In the worst case, this requires a 128-bit move instruction to
implicitly zero the upper bits. In the common case, we should
recognize the producing instruction already zeroed the upper bits.
llvm/lib/Target/X86/X86ISelLowering.cpp diff | blob | history
llvm/test/CodeGen/X86/vec-strict-fptoint-128.ll diff | blob | history
Domain: System / Toolchain;