diff options
| author | Arnold Schwaighofer <aschwaighofer@apple.com> | 2014-01-10 18:20:32 +0000 |
|---|---|---|
| committer | Arnold Schwaighofer <aschwaighofer@apple.com> | 2014-01-10 18:20:32 +0000 |
| commit | c2e9d759f29602021daa26453452928c81adffbb (patch) | |
| tree | 0679457ea94f719d42bc91aaf748a01ebbf0959c /llvm/lib/Support/FileOutputBuffer.cpp | |
| parent | cebfcceec129af3cf0e8b43ac867b669d389c0d1 (diff) | |
| download | llvm-c2e9d759f29602021daa26453452928c81adffbb.zip llvm-c2e9d759f29602021daa26453452928c81adffbb.tar.gz llvm-c2e9d759f29602021daa26453452928c81adffbb.tar.bz2 | |
LoopVectorizer: Handle strided memory accesses by versioning
for (i = 0; i < N; ++i)
A[i * Stride1] += B[i * Stride2];
We take loops like this and check that the symbolic strides 'Strided1/2' are one
and drop to the scalar loop if they are not.
This is currently disabled by default and hidden behind the flag
'enable-mem-access-versioning'.
radar://13075509
llvm-svn: 198950
Diffstat (limited to 'llvm/lib/Support/FileOutputBuffer.cpp')
0 files changed, 0 insertions, 0 deletions