[PATCH] D27321: Fix LSR ImmCost calculation for profitable chains

[Evgeny Stupachenko via Phabricator via llvm-commits]

evstupac created this revision.
evstupac added a reviewer: qcolombet.
evstupac added subscribers: llvm-commits, anna, wmi, hfinkel.
evstupac set the repository for this revision to rL LLVM.
Herald added a subscriber: mzolotukhin.

For the following case:
for(j = 0; j < n; j++) {

  s += *(in++);
  s += *(in++);

}
LSR consider "in" as profitable chain and do count ImmCost only for first load.
However when we have a base set before the loop
in += 1024;
ImmCost could be a significant value.


Repository:
  rL LLVM

https://reviews.llvm.org/D27321

Files:
  lib/Transforms/Scalar/LoopStrengthReduce.cpp
  test/Transforms/LoopStrengthReduce/X86/imm-cost.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D27321.79976.patch
Type: text/x-patch
Size: 6367 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161201/e7f755ea/attachment.bin>