diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2021-10-06 07:42:06 -0400 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2021-10-06 08:12:05 -0400 |
| commit | db231ebdb07f38f8b5d92311a4dc89e63c5a9d05 (patch) | |
| tree | ff958f88dd005812787d1b22fcaf84f0ce7d6f36 /llvm/lib/Support/ErrorHandling.cpp | |
| parent | 02e690ba0b10c422cbec77105eb40835dab57092 (diff) | |
| download | llvm-db231ebdb07f38f8b5d92311a4dc89e63c5a9d05.zip llvm-db231ebdb07f38f8b5d92311a4dc89e63c5a9d05.tar.gz llvm-db231ebdb07f38f8b5d92311a4dc89e63c5a9d05.tar.bz2 | |
[InstCombine] fold fake vector extract to shift+trunc
We already handle more complicated cases like:
extelt (bitcast (inselt poison, X, 0)) --> trunc (lshr X)
But we missed this simpler pattern:
https://alive2.llvm.org/ce/z/D55h64 / https://alive2.llvm.org/ce/z/GKzzRq
This is part of solving:
https://llvm.org/PR52057
I made the transform depend on legal/desirable int type to avoid creating
a shift of an illegal type (for example i128). I'm not sure if that
restriction is actually necessary, but we can change that as a follow-up
if the backend can deal with integer ops on too-wide illegal types.
The pile of AVX512 test changes are all neutral AFAICT - the x86 backend
seems to know how to turn that into the expected "kmov" instructions.
Differential Revision: https://reviews.llvm.org/D111082
Diffstat (limited to 'llvm/lib/Support/ErrorHandling.cpp')
0 files changed, 0 insertions, 0 deletions