[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

aeubanks reopened this revision. aeubanks added a comment. This revision is now accepted and ready to land. Herald added a subscriber: ormris. Using valgrind's massif, I profiled using clang build build PassBuilder.cpp (it's one of the longest LLVM files to compile), and at peak memory usage: wi

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

aeubanks updated this revision to Diff 339782. aeubanks added a comment. Herald added subscribers: cfe-commits, steven_wu. Herald added a project: clang. add test use CFGAnalyses uncovered another place in updateCGAndAnalysisManager() to change depends on D101017

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

aeubanks updated this revision to Diff 342602. aeubanks added a comment. Herald added a subscriber: nikic. rebase Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D100917/new/ https://reviews.llvm.org/D100917 Files: clang/test/CodeGen/thinlto-distri

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGd14d84af2f5e: [NewPM] Only invalidate modified functions' analyses in CGSCC passes (authored by aeubanks). Repository: rG LLVM Github Monorepo CH

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

nikic added a comment. An unfortunate side-effect of this change is that NewPM uses even more memory. tramp3d-v4 is up 20% in max-rss (https://llvm-compile-time-tracker.com/compare.php?from=4ef1f90e4d564b872e3598ccef45adb740eb0f0d&to=d14d84af2f5ebb8ae2188ce6884a29a586dc0a40&stat=max-rss) Repos

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

aeubanks added a comment. In D100917#2735651 , @nikic wrote: > An unfortunate side-effect of this change is that NewPM uses even more > memory. tramp3d-v4 is up 20% in max-rss > (https://llvm-compile-time-tracker.com/compare.php?from=4ef1f90e4d564b872e3

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

mtrofin added a comment. In D100917#2736702 , @aeubanks wrote: > In D100917#2735651 , @nikic wrote: > >> An unfortunate side-effect of this change is that NewPM uses even more >> memory. tramp3d-v4 is up 20% in m

[PATCH] D100917: [NewPM] Only invalidate modified functions' analyses in CGSCC passes

aeubanks added a comment. Herald added a reviewer: ctetreau. I rebased this and there are still major memory regressions for tramp3d-v4: https://llvm-compile-time-tracker.com/compare.php?from=f63405f6e3d30f33e715ef5ad09136127535a3fb&to=aec1375e4afe90727f1c30dae1659d11a20d&stat=max-rss Perhaps