Thanks for doing this. On Thu, Aug 07, 2008 at 02:33:09 +0200, Petr Rockai wrote: > The profiling-enabled version in consistently almost 2 seconds faster than the > non-profiling-enabled one. Graph colouring doesn't make the difference (it'd > be > cool if it did, though). All non-profiling darcs versions seem to perform > about > same. Both 2.0.2 and 2.0.2 (+ 29 patches) with profiling seem to perform about > same (and consistently better than non-profiling ones).
Could be worth asking #haskell I suppose this is not an free/open-source repository that could go into Standard Darcs Benchmark profiling zoo? I'm hoping we can set up a directory of real world repo tarballs used for the standard benchmarks. -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> PGP Key ID: 08AC04F9
pgpkNDmftXBpt.pgp
Description: PGP signature
_______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
