diff options
| author | David Sherwood <david.sherwood@arm.com> | 2021-12-06 16:02:30 +0000 |
|---|---|---|
| committer | David Sherwood <david.sherwood@arm.com> | 2022-01-18 11:59:30 +0000 |
| commit | e781620dee4415018bdf3769a97c2ec31a85ff68 (patch) | |
| tree | 52d589acebc3450add905744c5d68b2f69823acc /llvm/lib/Object/ArchiveWriter.cpp | |
| parent | 59e031ff9057b103c73f22bebc32304ee79fa139 (diff) | |
| download | llvm-e781620dee4415018bdf3769a97c2ec31a85ff68.zip llvm-e781620dee4415018bdf3769a97c2ec31a85ff68.tar.gz llvm-e781620dee4415018bdf3769a97c2ec31a85ff68.tar.bz2 | |
[LoopVectorize][AArch64] Use get.active.lane.mask intrinsic when SVE is enabled
When SVE is enabled for AArch64 targets it makes more sense to use the
get.active.lane.mask intrinsic, because SVE has an exact 1-1 mapping
from the intrinsic to the 'whilelo' instruction for legal vector types.
This instruction neatly takes overflow into account as well. This patch
fixes an issue in VPInstruction::generateInstruction that assumed we are
only dealing with fixed-width vectors.
Differential Revision: https://reviews.llvm.org/D117109
Diffstat (limited to 'llvm/lib/Object/ArchiveWriter.cpp')
0 files changed, 0 insertions, 0 deletions