Folks, On Fri, Oct 31, 2008 at 09:50:08 -0400, David Roundy wrote: > Just run timings yourself. I've wasted far too much time on this > stupid patch set already.
I would like everybody please keep in mind that we are *all* volunteers here. Reviewing patches (especially the nice and thorough approach that David takes) takes time; time pressure can can sometimes lead to stress; and stress can sometimes lead to the occasional undiplomatic remark. So please keep cool, don't take things personally and stay friendly :-) That said, here are the latest performance runs, best of 3 trials on an eight 3.16Ghz CPU machine (X86_64) with 32 GiB of RAM; again on the GHC hashed repo using the scripts in http://code.haskell.org/darcs/darcs-benchmarks This table is comparing the pre-salvo-8 darcs with --disable-bytestring against the post-salvo-8 darcs. || old | new ================++========+====== get-complete || 7.6 | 5.5 check || 580.5 | 416.9 init || 0.0 | 0.0 add-r-ghc || 0.4 | 0.4 get-lazy || 0.5 | 0.5 unpull-10 || 9.2 | 7.4 pull-10 || 3.5 | 2.6 unpull-100 || 62.4 | 15.3 pull-100 || 72.5 | 27.3 unpull-1000 || 7.3 | 5.5 pull-1000 || 73.1 | 27.3 whatsnew-empty || 0.1 | 0.1 whatsnew-id-ego || 2.0 | 1.8 revert || 7.5 | 5.7 record || 5.7 | 4.6 unrecord || 12.1 | 5.7 record-l || 5.5 | 4.5 annotate || 979.6 | 858.7 -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> PGP Key ID: 08AC04F9
signature.asc
Description: Digital signature
_______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
