diff options
| author | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2022-04-21 16:45:19 -0700 |
|---|---|---|
| committer | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2022-11-28 15:52:32 -0800 |
| commit | c46634554d507b6b764eeba881ed4eef3b28baac (patch) | |
| tree | b3e9ca8c8ae9b1ef4f159abd54ff10300bf53b8f /lldb/packages/Python/lldbsuite/test/lldbgdbclient.py | |
| parent | 55378ae87cce07c8897f9a0abb4a7edfed387696 (diff) | |
| download | llvm-c46634554d507b6b764eeba881ed4eef3b28baac.zip llvm-c46634554d507b6b764eeba881ed4eef3b28baac.tar.gz llvm-c46634554d507b6b764eeba881ed4eef3b28baac.tar.bz2 | |
[LoadStoreVectorizer] Consider if operation is faster than before
Compare a relative speed of misaligned accesses before and
after vectorization, not just check the new instruction is
not going to be slower.
Since no target now returns anything but 0 or 1 for Fast
argument of the allowsMisalignedMemoryAccesses this is still NFCI.
The subsequent patch will tune actual vaues of Fast on AMDGPU.
Differential Revision: https://reviews.llvm.org/D124218
Diffstat (limited to 'lldb/packages/Python/lldbsuite/test/lldbgdbclient.py')
0 files changed, 0 insertions, 0 deletions