diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2025-01-14 13:26:38 +0000 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-01-14 13:26:38 +0000 |
| commit | 0bd098b1cce62ad63eb63f7e57c3a55ccfaf7e0c (patch) | |
| tree | 8e4e4cdfd0f78dd83d1c4639d160522ec59156b1 /flang/lib/Frontend/CompilerInvocation.cpp | |
| parent | 6c7a53b78fd695ee58ea42d21362f6fa861eb37e (diff) | |
| download | llvm-0bd098b1cce62ad63eb63f7e57c3a55ccfaf7e0c.zip llvm-0bd098b1cce62ad63eb63f7e57c3a55ccfaf7e0c.tar.gz llvm-0bd098b1cce62ad63eb63f7e57c3a55ccfaf7e0c.tar.bz2 | |
[X86] Fold VPERMV3(WIDEN(X),M,WIDEN(Y)) -> VPERMV(CONCAT(X,Y),M') iff the CONCAT is free (#122750)
Minor followup to #122485 - if the source operands were widened half-size subvectors, then attempt to concatenate the subvectors directly, and then adjust the shuffle mask so references to the second operand now refer to the upper half of the concat result.
Diffstat (limited to 'flang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions