diff options
| author | Craig Topper <craig.topper@intel.com> | 2017-11-12 02:28:21 +0000 |
|---|---|---|
| committer | Craig Topper <craig.topper@intel.com> | 2017-11-12 02:28:21 +0000 |
| commit | d3e5781e53aaeded7d6ce85aa1118275fcf82bdc (patch) | |
| tree | 9c6972978523731097a8739e74c08d8788a275e2 /llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp | |
| parent | bfb990b78325a0af160178cdf3803686ec7ba695 (diff) | |
| download | bcm5719-llvm-d3e5781e53aaeded7d6ce85aa1118275fcf82bdc.tar.gz bcm5719-llvm-d3e5781e53aaeded7d6ce85aa1118275fcf82bdc.zip | |
[InstCombine] Teach visitICmpInst to not break integer absolute value idioms
Summary:
This patch adds an early out to visitICmpInst if we are looking at a compare as part of an integer absolute value idiom. Similar is already done for min/max.
In the particular case I observed in a benchmark we had an absolute value of a load from an indexed global. We simplified the compare using foldCmpLoadFromIndexedGlobal into a magic bit vector, a shift, and an and. But the load result was still used for the select and the negate part of the absolute valute idiom. So we overcomplicated the code and lost the ability to recognize it as an absolute value.
I've chosen a simpler case for the test here.
Reviewers: spatel, davide, majnemer
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39766
llvm-svn: 317994
Diffstat (limited to 'llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp')
0 files changed, 0 insertions, 0 deletions
