Re: git and larger trees, not so fast?

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Aug 10, 2007, 19:39 UTC
Message-ID: <alpine.LFD.0.999.0708101231580.30176@woody.linux-foundation.org>
In-Reply-To: <20070809163026.GD568@mbox.bz>

On Thu, 9 Aug 2007, moe wrote:

> 
> earlier today i imported one of my larger trees
> (~70k files) into git and was quite disappointed
> by the performance.

Ok, I said I wouldn't have time to fix it yesterday, but today it's all done.

With the first fix from Junio yesterday (the one that fixed "git status"), and the fixes I've sent out today, your cases should not all be basically instantaneous (ie we're talking low seconds, even on not-the-fastest- possible-machines).

So with the following patches that were posted over the last 24 hours, you should be ok:

  Junio:
	Fix performance problem in "git status"

  Me:
	Start moving unpack-trees to "struct tree_desc"
	Fix "git commit directory/" performance anomaly (+ one-liner fix)
	Move old index entry removal from "unpack_trees()" into the individual functions
	Optimize the common cases of git-read-tree
	Optimize the two-way merge of git-read-tree too

(that patch from Junio was sent in an email in this thread, with the subject line "Re: git and larger trees, not so fast?" and a message ID of "<7v7io4xwvp.fsf@assigned-by-dhcp.cox.net>": the patches from me should all have the appropriate Subject lines and be findable that way).

If you can test with your real load to make sure, that would be good.

			Linus