Re: [RFH] bug in unpack_trees

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Mar 4, 2008, 21:31 UTC
Message-ID: <alpine.LFD.1.00.0803041325370.12253@woody.linux-foundation.org>
In-Reply-To: <20080304115940.GA5260@sigill.intra.peff.net>

On Tue, 4 Mar 2008, Jeff King wrote:

>
> I am tracking down a bug in unpack_trees, but I can't seem to find the
> exact problem; I'm hoping to get help from people who have touched this
> code a bit more than I have.

Ok, I haven't (the blame for that unpack_trees function lies mainly at Dscho, I think ;), and now that I'm looking at it more closely I really don't think unpack_trees() is salvageable.

I tried. I can't make it work.

The only really sane way to traverse trees in parallel is with the walk-tree.c functionality (ie using "traverse_trees()"), which is quite straightforward and rather simple, and which I can pretty much guarantee works.

In contrast, the things that unpack_trees() does to try to figure out how to mix in the index into the pot really doesn't work.

I'll take a good hard look at trying to convert users of unpack_trees() into traverse_trees(), or perhaps even convert "unpack_trees()" itself.

		Linus