Tue, 13 Dec 2016 21:49:58 -0800 convert: don't use long list comprehensions
David Soria Parra <davidsp@fb.com> [Tue, 13 Dec 2016 21:49:58 -0800] rev 30602
convert: don't use long list comprehensions We are iterating over p4changes. Make the continue condition more clear and easier to add new conditions in future patches, by removing the list comprehension and move the condition into the existing for-loop.
Thu, 15 Dec 2016 11:00:18 -0800 changelog: keep track of file end in appender (issue5444)
Durham Goode <durham@fb.com> [Thu, 15 Dec 2016 11:00:18 -0800] rev 30601
changelog: keep track of file end in appender (issue5444) Previously, changelog.appender.end() would compute the end of the file by joining all the current appended data and checking the length. This is an O(n) operation. e240e914d226 introduced a seek call before every revlog write, which means we are hitting this O(n) behavior n times, which causes changelog writes during a pull to be n^2. In our large repo, this caused pulling 100k commits to go from 17s to 130s. With this fix, it's back to 17s.
Thu, 15 Dec 2016 11:14:00 -0500 tests: fix test-bdiff to handle variance between pure and c bdiff code
Augie Fackler <augie@google.com> [Thu, 15 Dec 2016 11:14:00 -0500] rev 30600
tests: fix test-bdiff to handle variance between pure and c bdiff code Obviously we'd rather patch pure to have the same algorithmic win as the C code, but this is a quick fix for the pure build since pure isn't wrong, just not as fast as it could be.
(0) -30000 -10000 -3000 -1000 -300 -100 -30 -10 -3 +3 +10 +30 +100 +300 +1000 +3000 +10000 tip