diff options
| author | Charlie Turner <charlie.turner@arm.com> | 2015-10-27 17:49:11 +0000 |
|---|---|---|
| committer | Charlie Turner <charlie.turner@arm.com> | 2015-10-27 17:49:11 +0000 |
| commit | 74c387feb7023dae60c760b5e98dcf617466a434 (patch) | |
| tree | e163188f5d341b9d78d1d33ef9afae2fab3765ba /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | 4a51e5ddaed42e0b43c92170649eca2443e7208b (diff) | |
| download | llvm-74c387feb7023dae60c760b5e98dcf617466a434.zip llvm-74c387feb7023dae60c760b5e98dcf617466a434.tar.gz llvm-74c387feb7023dae60c760b5e98dcf617466a434.tar.bz2 | |
[SLP] Treat SelectInsts as reduction values.
Summary:
Certain workloads, in particular sum-of-absdiff loops, can be vectorized using SLP if it can treat select instructions as reduction values.
The test case is a bit awkward. The AArch64 cost model needs some tuning to not be so pessimistic about selects. I've had to tweak the SLP threshold here.
Reviewers: jmolloy, mzolotukhin, spatel, nadav
Subscribers: nadav, mssimpso, aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D13949
llvm-svn: 251424
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions