[PATCH] [analyzer] Refactor conditional expression evaluating code

Pavel Labath labath at google.com
Thu Aug 22 02:18:43 PDT 2013


  Posting measurements on memory usage when analyzing sqlite3.c.

  After the patch:
  $/usr/bin/time -v bin/clang --analyze ~/Downloads/sqlite3.c
  ...
  42 warnings generated.
  	User time (seconds): 357.35
  	System time (seconds): 0.41
  	Percent of CPU this job got: 99%
  	Elapsed (wall clock) time (h:mm:ss or m:ss): 5:58.84
  	Maximum resident set size (kbytes): 478688
  	Average resident set size (kbytes): 0
  	Major (requiring I/O) page faults: 0
  	Minor (reclaiming a frame) page faults: 130624
  	Voluntary context switches: 6
  	Involuntary context switches: 37812
  	Swaps: 0
  	File system inputs: 16
  	File system outputs: 2720

  Before the patch:
  /usr/bin/time -v bin/clang --analyze ~/Downloads/sqlite3.c
  ...
  42 warnings generated.
  	User time (seconds): 354.07
  	System time (seconds): 1.12
  	Percent of CPU this job got: 99%
  	Elapsed (wall clock) time (h:mm:ss or m:ss): 5:56.19
  	Maximum resident set size (kbytes): 483232
  	Average resident set size (kbytes): 0
  	Major (requiring I/O) page faults: 0
  	Minor (reclaiming a frame) page faults: 607164
  	Voluntary context switches: 5
  	Involuntary context switches: 37310
  	Swaps: 0
  	File system inputs: 0
  	File system outputs: 2720

  The difference is not definitely not big, and I strongly suspect it is within experimental error, because the measurements vary across different runs.

http://llvm-reviews.chandlerc.com/D1340



More information about the cfe-commits mailing list