diff options
| author | Arthur Eubanks <aeubanks@google.com> | 2021-11-03 15:45:30 -0700 |
|---|---|---|
| committer | Arthur Eubanks <aeubanks@google.com> | 2021-11-04 17:16:11 -0700 |
| commit | 7175886a0f612aded1430ae240ca7ffd53d260dd (patch) | |
| tree | 0b22dfdce98dc25ed2bfebc7f5a4bdf61dc79484 /llvm/lib/IR/Module.cpp | |
| parent | 41860e602aaa94556ba996fae7b8baa268e25096 (diff) | |
| download | llvm-7175886a0f612aded1430ae240ca7ffd53d260dd.zip llvm-7175886a0f612aded1430ae240ca7ffd53d260dd.tar.gz llvm-7175886a0f612aded1430ae240ca7ffd53d260dd.tar.bz2 | |
[NewPM] Make eager analysis invalidation per-adaptor
Follow-up change to D111575.
We don't need eager invalidation on every adaptor. Most notably,
adaptors running passes that use very few analyses, or passes that
purely invalidate specific analyses.
Also allow testing of this via a pipeline string
"function<eager-inv>()".
The compile time/memory impact of this is very comparable to D111575.
https://llvm-compile-time-tracker.com/compare.php?from=9a2eec512a29df45c90c2fcb741e9d5c693b1383&to=b9f20bcdea138060967d95a98eab87ce725b22bb&stat=instructions
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D113196
Diffstat (limited to 'llvm/lib/IR/Module.cpp')
0 files changed, 0 insertions, 0 deletions