[PATCH RFC] cg1packer: use fastpath when changesets are not skipped or reordered

Mike Hommey mh at glandium.org
Wed May 13 23:21:04 CDT 2015


On Tue, May 12, 2015 at 01:10:25PM -0700, Pierre-Yves David wrote:
> >Because I wasn't sure the approach was right, and sure enough, 2
> >messages later I figured it introduced a regression. And maybe we'd be
> >better off with a more aggressive approach, trying to minimize the slow
> >path even more.
> >
> >Mike
> >
> >PS: the funny thing is that the slow path is based on data that may not
> >be accurate (as the mozilla-central history demonstrates), but
> >fortunately, /shouldn't/ indicate new files that aren't new. That's
> >funny because if it was happening, then the slow path would need to be
> >even slower.
> 
> I think martin did some work in this area recently. And your patch does not
> apply anymore.
> 
> Can you have a look at what martin did and wrote about it and come back with
> a V2 and new data?

AIUI, the changes he did made the slow path be used in more cases than
before, so there is no change wrt the original issue. Running the same
commands as in the original patch message still lead to the same
timings. OTOH the original patch was also buggy, so /that/ would need to
be addressed. At this point, I'd rather someone who actually know this
code took a look.

Mike


More information about the Mercurial-devel mailing list