[PATCH] D145163: Add support for vectorization of interleaved memory accesses for scalable VF

[Graham Hunter via Phabricator via llvm-commits]

huntergr updated this revision to Diff 511067.
huntergr added a comment.
Herald added subscribers: frasercrmck, luismarques, apazos, sameer.abuasal, s.egerton, Jim, jocewei, PkmX, the_o, brucehoult, MartinMosbeck, rogfer01, edward-jones, zzheng, MaskRay, jrtc27, niosHD, sabuasal, simoncook, johnrusso, rbar, asb, arichardson.

Separated out the bitcast fix and committed. Precommitted tests.

Changed IRBuilder interface to focus on the intrinsics (and fixed-length shuffle equivalents) instead of mixing in loads/stores.

There's a few unit tests which will fail with the new interface -- although we generate the same IR instructions, the order is different. Assuming the interface is suitable I'll update the tests before posting the next patch revision.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D145163/new/

https://reviews.llvm.org/D145163

Files:
  llvm/include/llvm/IR/IRBuilder.h
  llvm/lib/IR/IRBuilder.cpp
  llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
  llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
  llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
  llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
  llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll
  llvm/tools/llvm-profdata/CMakeLists.txt

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D145163.511067.patch
Type: text/x-patch
Size: 89138 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230405/336e97ce/attachment-0001.bin>