diff options
| author | Gil Rapaport <gil.rapaport@intel.com> | 2018-10-24 07:08:38 +0000 |
|---|---|---|
| committer | Gil Rapaport <gil.rapaport@intel.com> | 2018-10-24 07:08:38 +0000 |
| commit | 5012e7f6acaf42d8bdd92394b9d82066f2fab94b (patch) | |
| tree | 28c6cef17347d5f947ed75ce6d482cabf4b6ce42 /llvm/lib/Transforms/Vectorize/LoopVectorize.cpp | |
| parent | 668ac94ba46d65cdedbe4194243c1d46c5f62ff1 (diff) | |
| download | bcm5719-llvm-5012e7f6acaf42d8bdd92394b9d82066f2fab94b.tar.gz bcm5719-llvm-5012e7f6acaf42d8bdd92394b9d82066f2fab94b.zip | |
[LSR] Combine unfolded offset into invariant register
LSR reassociates constants as unfolded offsets when the constants fit as
immediate add operands, which currently prevents such constants from being
combined later with loop invariant registers.
This patch modifies GenerateCombinations() to generate a second formula which
includes the unfolded offset in the combined loop-invariant register.
Differential Revision: https://reviews.llvm.org/D51861
llvm-svn: 345114
Diffstat (limited to 'llvm/lib/Transforms/Vectorize/LoopVectorize.cpp')
0 files changed, 0 insertions, 0 deletions
