On Mon, 18 Dec 2006, Linus Torvalds wrote: > > > On Mon, 18 Dec 2006, Randal L. Schwartz wrote: > > > > In preparation for my git-talk tonight, I thought having a *big* > > archive would be nice, so I entered: > > > > git-clone git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git > > > > The network traffic took about five minutes, resulting in: > > No. That's just the preparatory work, not "the network traffic". > > > > > % git-clone git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git > > Initialized empty Git repository in /Volumes/UFS/MIRROR/linus-GIT/linux-2.6/.git/ > > remote: Generating pack... > > remote: Done counting 393554 objects. > > remote: Deltifying 393554 objects. > > remote: 100% (393554/393554) done > > Indexing 393554 objects. > > remote: Total 393554, written 393554 (delta 313037), reused 393392 (delta 312908) > > 100% (393554/393554) done > > This is the REMOTE telling you that it has indexed the objects in the > pack, but it hasn't actually _created_ the pack yet, it's just ready to > start feeding it to you. It's just created the list of objects and their > relationships. Not quite actually. When the message "remote: Total 393554, written 393554" appears, the remote is done with sending the pack over. When " 100% (393554/393554) done" appears this means the pack has been parsed locally. > > But then it took nearly an *hour* at the next phase: > > > > Resolving 313037 deltas. > > 100% (313037/313037) done > > And _this_ is the phase where it will actually create all the deltas, and > write them out, and you're receiving them. No. This is the phase where deltas are replayed locally in order to compute the resulting object's sha1 in order to reconstruct the index. I cannot imagine it would take _longer_ for that than downloading the full pack. Nicolas - To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html