From: Jeff Garzik <jgarzik@pobox.com>
Date: Fri, 20 May 2005 23:58:12 GMT
Subject: Re: checkout-cache -f: a better way?
Message-ID: <428E7994.1090402@pobox.com>
In-Reply-To: <7vvf5dqxfq.fsf@assigned-by-dhcp.cox.net>

Junio C Hamano wrote:
>>>>>>"JG" == Jeff Garzik <jgarzik@pobox.com> writes:
> 
> 
> JG> git-checkout-cache -f writes out every file, even if it exists, correct?
> 
> No, that's not correct.  To translate my prose, you would want
> this:

Thanks, I stand corrected :)


>     git-read-tree -m HEAD && git-checkout-cache -q -f -u -a
> 
> (notice that I do not have git-update-cache --refresh after
> that).

Yep, thanks.  Script does seem faster now.  Numbers for hot cache (first 
is pre-modification, post is your mod):

[jgarzik@pretzel libata-dev]$ time git-switch-tree adma-mwi

real    0m7.069s
user    0m4.183s
sys     0m2.817s
[jgarzik@pretzel libata-dev]$ time git-switch-tree adma

real    0m0.389s
user    0m0.294s
sys     0m0.094s