diff options
| author | Easwaran Raman <eraman@google.com> | 2017-11-14 19:31:51 +0000 |
|---|---|---|
| committer | Easwaran Raman <eraman@google.com> | 2017-11-14 19:31:51 +0000 |
| commit | 0d55b55bb616ce28b5570d488edfa294ea039879 (patch) | |
| tree | 7c4579c94f2853960a8a47e6302589290704457c /llvm/tools/llvm-strings/llvm-strings.cpp | |
| parent | 55b8590e03cf211563e6a7f86a48cf4c2c2ab10f (diff) | |
| download | bcm5719-llvm-0d55b55bb616ce28b5570d488edfa294ea039879.tar.gz bcm5719-llvm-0d55b55bb616ce28b5570d488edfa294ea039879.zip | |
[CodeGenPrepare] Disable div bypass when working set size is huge.
Summary:
Bypass of slow divs based on operand values is currently disabled for
-Os. Do the same when profile summary is available and the working set
size of the application is huge. This is similar to how loop peeling is
guarded by hasHugeWorkingSetSize. In the div bypass case, the generated
extra code (and the extra branch) tendss to outweigh the benefits of the
bypass. This results in noticeable performance improvement on an
internal application.
Reviewers: davidxl
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D39992
llvm-svn: 318179
Diffstat (limited to 'llvm/tools/llvm-strings/llvm-strings.cpp')
0 files changed, 0 insertions, 0 deletions
