[PATCH] D48128: [ARM] Parallel DSP IR Pass

[Sjoerd Meijer via Phabricator via llvm-commits]

SjoerdMeijer updated this revision to Diff 152242.
SjoerdMeijer added a comment.

Addressed Sam's comments:

- set the alignment on the load instructions.
- added alias analysis checks.
- check that the accumulator is the right integer type.
- bail out early when the operand chains of the muls are integers constants or more complicated patterns we don't yet support.

Tests have been modified and added for all these cases.


https://reviews.llvm.org/D48128

Files:
  lib/Target/ARM/ARM.h
  lib/Target/ARM/ARMParallelDSP.cpp
  lib/Target/ARM/ARMTargetMachine.cpp
  lib/Target/ARM/CMakeLists.txt
  lib/Target/ARM/LLVMBuild.txt
  test/CodeGen/ARM/smlad0.ll
  test/CodeGen/ARM/smlad1.ll
  test/CodeGen/ARM/smlad10.ll
  test/CodeGen/ARM/smlad2.ll
  test/CodeGen/ARM/smlad3.ll
  test/CodeGen/ARM/smlad4.ll
  test/CodeGen/ARM/smlad5.ll
  test/CodeGen/ARM/smlad6.ll
  test/CodeGen/ARM/smlad7.ll
  test/CodeGen/ARM/smlad8.ll
  test/CodeGen/ARM/smlad9.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D48128.152242.patch
Type: text/x-patch
Size: 45825 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180621/7d2c0440/attachment-0001.bin>