[PATCH] D57857: [PowerPC] custom lower `v2f64 fpext v2f32`

[Lei Huang via Phabricator via llvm-commits]

lei updated this revision to Diff 194183.
lei marked 3 inline comments as done.
lei added a comment.

Update comments and renamed new ppc ISD nodes.

I tried the suggested approach to fpextend v2f32 loads to v2f64 and then adding a custom load of two float values into a vector of two double values.  However this generated extra instructions for each load and we no longer have any of the performance gains seen with the original approach.  With the new approach we see negligible performance changes for all 2017 spec benchmarks with the exception of omnetpp which showed a 1.7% performance degradation.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D57857/new/

https://reviews.llvm.org/D57857

Files:
  llvm/lib/Target/PowerPC/PPCISelLowering.cpp
  llvm/lib/Target/PowerPC/PPCISelLowering.h
  llvm/lib/Target/PowerPC/PPCInstrVSX.td
  llvm/test/CodeGen/PowerPC/reduce_scalarization.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D57857.194183.patch
Type: text/x-patch
Size: 9982 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190408/b607cbbd/attachment.bin>