[darcs-users] Re: fptools in darcs now available

David Roundy droundy at abridgegame.org
Wed May 18 17:41:22 UTC 2005


On Wed, May 18, 2005 at 05:37:59PM +0200, Juliusz Chroboczek wrote:
> > Actually, now that I think about it, optimize --checkpoint *also* is
> > worse than an "initial record", which is one of the things that darcs
> > has trouble with--but Ian has largely fixed in darcs-unstable.
> 
> IMHO, we should use a different format for checkpoints -- hunk patches
> from the empty tree are not a good idea.
> 
> Changing the on-disk format of patches will help a lot.

Indeed, that should help.  But even as things are now, a darcs-unstable
initial record of the linux kernel requires only 10 times the CPU time that
tar czf does, and only 7.5 times the wallclock time.  So if we assume that
tar is pretty much optimal, we only have one order of magnitude improvement
left to be made.  I expect that changing the hunk format (as we've
discussed) should pretty much get us that order of magnitude in
improvement in CPU time.

The memory usage is way worse than that of tar, but I'm optimistic that
we can improve things a bit in that realm.  Perhaps (for example) by
storing PackedString file paths, or by making the directory-reading portion
of slurp lazy.  In any case, 450M isn't such bad maximum memory consumption
for a project the size of the kernel.
-- 
David Roundy




More information about the darcs-users mailing list