diff options
| author | David Green <david.green@arm.com> | 2022-04-04 23:07:47 +0100 |
|---|---|---|
| committer | David Green <david.green@arm.com> | 2022-04-04 23:07:47 +0100 |
| commit | 3b9833597e810d4c485487d2f094a8e223af5548 (patch) | |
| tree | fa872b0f5eee6a9c4c8c77b5452c39925f122f09 /flang/lib/Frontend/CompilerInvocation.cpp | |
| parent | a70480dd13c3317f4f9e2f55dc9d67dfb61d8218 (diff) | |
| download | llvm-3b9833597e810d4c485487d2f094a8e223af5548.zip llvm-3b9833597e810d4c485487d2f094a8e223af5548.tar.gz llvm-3b9833597e810d4c485487d2f094a8e223af5548.tar.bz2 | |
[AArch64] Alter mull buildvectors(ext(..)) combine to work on shuffles
D120018 altered this combine to work on buildvectors as opposed to
shuffle dup's. This works well for dups and other things that are
expanded into buildvectors. Some shuffles are legal though, and stay as
vector_shuffle through lowering. This expands the transform to also
handle shuffles, so that we can turn mul(shuffle(sext into
mul(sext(shuffle and more readily make smull/umull instructions. This
can come up from the SLP vectorizer adding shuffles that are costed from
extends.
Differential Revision: https://reviews.llvm.org/D123012
Diffstat (limited to 'flang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions