From: Linus Torvalds <torvalds@osdl.org>
Date: Wed, 18 Oct 2006 21:48:43 GMT
Subject: Re: heads-up: git-index-pack in "next" is broken
Message-ID: <Pine.LNX.4.64.0610181434510.3962@g5.osdl.org>
In-Reply-To: <Pine.LNX.4.64.0610181407040.18885@alien.or.mcafeemobile.com>



On Wed, 18 Oct 2006, Davide Libenzi wrote:
> 
> The hash value (hence the hash bucket index) simply directs you to the 
> bucket where a real record-compare loop is performed.

As far as I can tell not all loops do a real "record-compare" thing.

Some of the hash loops _only_ look at the hash, and as such a bad hash 
will do more than just cause bad performance, it will actually degrade the 
diff itself. Isn't that what XDL_MAX_EQLIMIT effectively does?

Btw, the binary delta generator doesn't seem to have this issue at all: it 
uses "unsigned int" for the hash values, so the xdiff delta generation 
will give the same exact results on 32-bit and 64-bit architectures.

Or was that one of the changes by Nico? (I only looked at the git version 
of that code)

		Linus