forked from OSchip/llvm-project
d48c981697
SROA will drop the explicit alignment on allocas when the ABI guarantees enough alignment. Because the alignment on new load/store instructions are set based on the alloca's alignment, that means SROA would end up dropping the alignment from atomic loads and stores, which is not allowed (see bug). For those, make sure to always carry over the alignment from the previous instruction. Differential revision: https://reviews.llvm.org/D75266 |
||
|---|---|---|
| .. | ||
| address-spaces.ll | ||
| addrspacecast.ll | ||
| alignment.ll | ||
| alloca-address-space.ll | ||
| basictest.ll | ||
| big-endian.ll | ||
| dbg-addr-diamond.ll | ||
| dbg-single-piece.ll | ||
| dead-inst.ll | ||
| fca.ll | ||
| mem-par-metadata-sroa.ll | ||
| non-integral-pointers.ll | ||
| phi-and-select.ll | ||
| phi-with-duplicate-pred.ll | ||
| pointer-offset-size.ll | ||
| ppcf128-no-fold.ll | ||
| pr26972.ll | ||
| pr37267.ll | ||
| preserve-nonnull.ll | ||
| slice-order-independence.ll | ||
| slice-width.ll | ||
| tbaa-struct.ll | ||
| vector-conversion.ll | ||
| vector-lifetime-intrinsic.ll | ||
| vector-promotion-different-size.ll | ||
| vector-promotion.ll | ||
| vectors-of-pointers.ll | ||