On Thu, Aug 19, 2010 at 04:14:28PM +0100, Eric Kow wrote: > How about ensuring that for users who are still deliberately using > hashed repositories, hashed repositories are no longer a significant > performance regression? > > How about answering concerns Ian may have about hashed repository > performance in practice? > > - Garbage collection? > - Large _darcs/pristine.hashed due to insufficient bucketing?
I think you might have meant 'deliberately using old fashioned repositories'. I still deliberately use old fashioned repositories, with darcs-1.0.9. As soon as I can show that we will have no performance regressions, I'm all for moving to darcs-2 and hashed repositories. I was originally thinking we would migrate all the way to darcs 2 format repositories, but the migration path is sufficiently difficult that I do not know if we will ever attempt it. I'd like to see 'darcs add' added to darcs-benchmark. I attempted it many months ago, but did not quite get it working before being overwhelmed with other stuff. I'm not sure if 'darcs diff' needs benchmarking, or if 'darcs annotate' takes care of that. I'm also interested in what has come out of 'camp', with regards to repository format, patch theory, and such, or if it is still too early. -kolibrie
signature.asc
Description: Digital signature
_______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
