From: Nicolas Pitre <nico@fluxnic.net>
Date: Tue, 05 Oct 2010 19:12:29 GMT
Subject: Re: large files and low memory
Message-ID: <alpine.LFD.2.00.1010051510160.3107@xanadu.home>
In-Reply-To: <AANLkTimHKyQhAzuh4Hsh7=bEet1T0mNRSMgiZkhWQwSN@mail.gmail.com>

On Tue, 5 Oct 2010, Nguyen Thai Ngoc Duy wrote:

> On Tue, Oct 5, 2010 at 2:41 PM, Enrico Weigelt <weigelt@metux.de> wrote:
> > * Enrico Weigelt <weigelt@metux.de> wrote:
> >
> > <snip>
> >
> > Found another possible bottleneck: git-commit seems to scan through
> > a lot of files. Shouldnt it just create a commit object from the
> > current index and update the head ?
> 
> You mean a lot of stat()? There is no way to avoid that unless you set
> assume-unchanged bits. Or you could use
> write-tree/commit-tree/update-ref directly.

Avoiding memory exhaustion is also going to help a lot as the stat() 
information will remain cached instead of requiring disk access.  Just a 
guess given $subject.


Nicolas