Fri, 03 Apr 2015 23:12:21 -0400 test-subrepo-recursion: set progress.changedelay really high
Augie Fackler <augie@google.com> [Fri, 03 Apr 2015 23:12:21 -0400] rev 24612
test-subrepo-recursion: set progress.changedelay really high This avoids subtopics from showing up, which were introduced by my recent change that added progress bars to hardlink clones in some cases.
Fri, 03 Apr 2015 18:10:12 -0400 test-clone: fix test expectations on systems without hardlinks
Augie Fackler <augie@google.com> [Fri, 03 Apr 2015 18:10:12 -0400] rev 24611
test-clone: fix test expectations on systems without hardlinks
Wed, 01 Apr 2015 00:44:33 -0700 dirstate: use parsers.make_file_foldmap when available
Siddharth Agarwal <sid0@fb.com> [Wed, 01 Apr 2015 00:44:33 -0700] rev 24610
dirstate: use parsers.make_file_foldmap when available This is a significant performance win on large repositories. perffilefoldmap: On Linux/gcc, on a test repo with over 500,000 files: before: wall 0.605021 comb 0.600000 user 0.560000 sys 0.040000 (best of 17) after: wall 0.280530 comb 0.280000 user 0.250000 sys 0.030000 (best of 35) On Mac OS X/clang, on a real-world repo with over 200,000 files: before: wall 0.281103 comb 0.280000 user 0.260000 sys 0.020000 (best of 34) after: wall 0.133622 comb 0.140000 user 0.120000 sys 0.020000 (best of 65) This visibly impacts status times on case-insensitive file systems. On the Mac OS X repo, status goes from 3.64 seconds to 3.50. With the third-party hgwatchman extension [1], 'hg status' on the same repo goes from 0.80 seconds to 0.65. [1] https://bitbucket.org/facebook/hgwatchman
(0) -10000 -3000 -1000 -300 -100 -30 -10 -3 +3 +10 +30 +100 +300 +1000 +3000 +10000 tip