Eric Kow <[email protected]> added the comment: On Mon, Jul 26, 2010 at 00:17:08 +0000, Jason Dagit wrote: > The benchmark is still running, but it has produce partial results and > the results look quite promising. Whatsnew is improved quite a bit. > Unrevert might be slightly slower and everything else seems to be > unaffected:
That's very interesting. It may be helpful to target the parser specifically with some stress tests, ie. instead of benchmarking, build a suite of increasingly large/complex patches and also of patch bundles. Save time and learn more? [then again, the benchmarking also implicitly tells us how important (or not) parsing performance actually is... for example, we could learn that it actually doesn't matter, so we can afford to just go for the cleanest/nicest code we can get] -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> For a faster response, please try +44 (0)1273 64 2905. __________________________________ Darcs bug tracker <[email protected]> <http://bugs.darcs.net/patch318> __________________________________ _______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
