[PATCH] D67921: [ARM][MVE] Change VCTP operand

Mon Sep 23 08:53:37 PDT 2019

samparker created this revision.
samparker added reviewers: dmgreen, SjoerdMeijer.
Herald added a subscriber: kristof.beyls.

The VCTP instruction will calculate the predicate masked based upon the number of elements that need to be processed. I had inserted the sub before the vctp intrinsic and supplied it as the operand, but this is incorrect as the sub is calculating the value for the //next// iteration.

https://reviews.llvm.org/D67921

Files:
  lib/Target/ARM/MVETailPredication.cpp
  test/CodeGen/Thumb2/LowOverheadLoops/basic-tail-pred.ll
  test/CodeGen/Thumb2/LowOverheadLoops/cond-vector-reduce-mve-codegen.ll
  test/CodeGen/Thumb2/LowOverheadLoops/nested.ll
  test/CodeGen/Thumb2/LowOverheadLoops/tail-pred-widen.ll
  test/CodeGen/Thumb2/LowOverheadLoops/tail-reduce.ll
  test/CodeGen/Thumb2/LowOverheadLoops/vector-arith-codegen.ll
  test/CodeGen/Thumb2/LowOverheadLoops/vector-reduce-mve-tail.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D67921.221349.patch
Type: text/x-patch
Size: 16924 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190923/b55edb6b/attachment.bin>