[PATCH] D79641: [AMDGPU] Vectorize alloca thru bitcast

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec created this revision.
rampitec added a reviewer: arsenm.
Herald added subscribers: kerbowa, hiraditya, t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl.
Herald added a project: LLVM.
rampitec added a comment.

PSDB passed.


This is mostly useful if alloca element type is not integer
and then casted to an integer for load or store. We now can
vectorize an [i32] alloca but cannot do so for [float].

There also a separate patch needed to properly lower 64 bit
types after they vectorized. At the moment these are lowered
via scratch anyway.


https://reviews.llvm.org/D79641

Files:
  llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
  llvm/test/CodeGen/AMDGPU/vector-alloca-bitcast.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D79641.262918.patch
Type: text/x-patch
Size: 18611 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200508/453d7188/attachment.bin>