[PATCH] Do not model scalar accesses in non-affine subregions

[Johannes Doerfert]

Hi grosser, sebpop, simbuerg, zinob,

  If a scalar was defined and used only in a non-affine subregion we do
  not need to model the accesses. However, if the scalar was defined
  inside the region and escapes the region we have to model the access.
  The same is true if the scalar was defined outside and used inside the
  region.

http://reviews.llvm.org/D7985

Files:
  include/polly/TempScopInfo.h
  lib/Analysis/TempScopInfo.cpp
  test/ScopInfo/no-scalar-deps-in-non-affine-subregion.ll
  test/ScopInfo/non-affine-region-phi.ll
  test/ScopInfo/non_affine_region_1.ll
  test/ScopInfo/non_affine_region_2.ll
  test/ScopInfo/non_affine_region_3.ll
  test/ScopInfo/non_affine_region_4.ll

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D7985.20960.patch
Type: text/x-patch
Size: 27039 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150301/ab4d246f/attachment.bin>