diff options
author | Caroline Concatto <caroline.concatto@arm.com> | 2022-01-13 16:52:41 +0000 |
---|---|---|
committer | Caroline Concatto <caroline.concatto@arm.com> | 2022-02-03 18:58:37 +0000 |
commit | 019f0221d52dc14829b6011b5bccd9ba4c3849d8 (patch) | |
tree | 189199aa7a2252bfb83b09a6c38e9ede8c1b30e8 /llvm/tools/llvm-libtool-darwin/llvm-libtool-darwin.cpp | |
parent | 14124c30253600043eec74651faabccb06ac5627 (diff) | |
download | llvm-019f0221d52dc14829b6011b5bccd9ba4c3849d8.zip llvm-019f0221d52dc14829b6011b5bccd9ba4c3849d8.tar.gz llvm-019f0221d52dc14829b6011b5bccd9ba4c3849d8.tar.bz2 |
[AArch64][SVE] Fold gather/scatter with 32bits when possible
In AArch64ISelLowering.cpp this patch implements this fold:
GEP (%ptr, (splat(%offset) + stepvector(A)))
into GEP ((%ptr + %offset), stepvector(A))
The above transform simplifies the index operand so that it can be expressed
as i32 elements.
This allows using only one gather/scatter assembly instruction instead of two.
Patch by Paul Walker (@paulwalker-arm).
Depends on D118459
Differential Revision: https://reviews.llvm.org/D117900
Diffstat (limited to 'llvm/tools/llvm-libtool-darwin/llvm-libtool-darwin.cpp')
0 files changed, 0 insertions, 0 deletions