[PATCH] D80202: [ASTMatchers] Performance optimization for memoization

[Nathan James via Phabricator via cfe-commits]

njames93 added a comment.

Running most of the clang tidy checks on the clang-tidy folder yields these results

  =================================BeforePatch===================================
  
  RUN1:
  4045.17user 83.93system 11:28.80elapsed 599%CPU (0avgtext+0avgdata 534024maxresident)k
  0inputs+0outputs (0major+27584683minor)pagefaults 0swaps
  
  RUN2:
  4078.06user 84.99system 11:35.99elapsed 598%CPU (0avgtext+0avgdata 506912maxresident)k
  55312inputs+0outputs (663major+27661947minor)pagefaults 0swaps
  
  RUN3:
  4040.77user 86.02system 11:28.85elapsed 599%CPU (0avgtext+0avgdata 547096maxresident)k
  0inputs+0outputs (0major+27698937minor)pagefaults 0swaps



  ==================================AfterPatch===================================
  
  RUN1: 
  4025.33user 83.32system 11:27.00elapsed 598%CPU (0avgtext+0avgdata 530568maxresident)k
  0inputs+0outputs (0major+27689512minor)pagefaults 0swaps
  
  RUN2:
  4056.93user 83.36system 11:32.31elapsed 598%CPU (0avgtext+0avgdata 529120maxresident)k
  3752inputs+0outputs (19major+27794845minor)pagefaults 0swaps
  
  RUN3:
  4029.05user 85.45system 11:26.31elapsed 599%CPU (0avgtext+0avgdata 533508maxresident)k
  0inputs+0outputs (0major+27730918minor)pagefaults 0swaps

Shows a consistent improvement but there tests are very noisy and dont focus on just the matching, they also include all the other boilderplate when running clang-tidy over a database. not to mention a small sample size


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D80202/new/

https://reviews.llvm.org/D80202