diff options
| author | Sam McCall <sam.mccall@gmail.com> | 2022-06-21 22:19:06 +0200 |
|---|---|---|
| committer | Sam McCall <sam.mccall@gmail.com> | 2022-06-23 19:27:47 +0200 |
| commit | 3e610f2cdc1e156c82f5b59c287165e769b24f40 (patch) | |
| tree | 1e0154258027d222d2b15ba9cb00ce33db8e2946 /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | 1cc97922811a27c2cd8fd88a1367f29f3d64487a (diff) | |
| download | llvm-3e610f2cdc1e156c82f5b59c287165e769b24f40.zip llvm-3e610f2cdc1e156c82f5b59c287165e769b24f40.tar.gz llvm-3e610f2cdc1e156c82f5b59c287165e769b24f40.tar.bz2 | |
[pseudo] Turn glrReduce into a class, reuse storage across calls.
This is a ~5% speedup, we no longer have to allocate the priority queues and
other collections for each reduction step where we use them.
It's also IMO easier to understand the structure of a class with methods vs a
function with nested lambdas.
Differential Revision: https://reviews.llvm.org/D128301
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions