Hi, Jason Dagit <b...@darcs.net> writes: > This is my attempt to improve the performance now that my refactor was > accepted. > > The majority of these patches happened as a natural part of me staring > at the code thinking of how to optimize it. Which is to say, if the > code was ugly and I had to stare at it I fet compelled to tidy it up.
I am glad that you are working on this. However, there may be still some work left, since with this bundle applied, we are getting failures in the testsuite. You can see the output at eg. http://buildbot.darcs.net/builders/6.10.4%20Debian%20TRY/builds/8/steps/test/logs/stdio also, it seems (but I am not sure, I haven't studied the numbers closely enough) that the refactor without this bundle performs better than after this bundle. Well, actually I would say that the performance is quite varied -- it seems that some things got faster and other things got slower. Maybe depending on the shape of the patches that appear in the given benchmark. I wish there was a better way to measure this... Yours, Petr. _______________________________________________ darcs-users mailing list darcs-users@darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users
