From: Junio C Hamano Date: Sun, 22 Jul 2012 18:52:35 GMT Subject: Re: [GSoC] Designing a faster index format - Progress report week 13 Message-ID: <7vfw8jsk5o.fsf@alter.siamese.dyndns.org> In-Reply-To: <500C1AA9.4000306@dewire.com> Robin Rosenberg writes: > A note on how JGit would work here. Java has none of the fields > that constitute statcrc. I guess we would write zero here when > creating new entries. Git could recognize that when checking status > and simply assume "clean" unless mtime or st_size says otherwise. Even though it may not be the end of the world, that is certainly bad. Recording the constituent fields separately without the statcrc microoptimization, thereby not shaving a handful of bytes per the index entry, is not the end of the world either in the same sense, which leads us to question the benefit we would be getting from such a change.