xfs
[Top] [All Lists]

Re: xfs on raid questions

To: Chris Wedgwood <cw@xxxxxxxx>
Subject: Re: xfs on raid questions
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Fri, 18 Jun 2004 22:55:30 +0100
Cc: Joshua Baker-LePain <jlb17@xxxxxxxx>, Seth Mos <seth.mos@xxxxxxxxx>, campbell@xxxxxxxxxxxx, linux-xfs@xxxxxxxxxxx
In-reply-to: <20040618215335.GA16407@taniwha.stupidest.org>
References: <20040617152353.GB2511@helium.inexs.com> <40D1D87D.9030208@xfs.org> <20040617184510.GB4309@helium.inexs.com> <40D2BC7E.3040402@xs4all.nl> <Pine.LNX.4.58.0406180851210.10588@chaos.egr.duke.edu> <20040618215034.GA16343@taniwha.stupidest.org> <20040618215206.GA28367@infradead.org> <20040618215335.GA16407@taniwha.stupidest.org>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.4.1i
On Fri, Jun 18, 2004 at 02:53:35PM -0700, Chris Wedgwood wrote:
> On Fri, Jun 18, 2004 at 10:52:06PM +0100, Christoph Hellwig wrote:
> 
> > It's always 16TB as that's a pagecache limit (page->index is a 32bit
> > scalar in PAGE_CACHE_SIZE = 4k units).
> 
> Yeah, I was more worried about signedness cleanliness as some people
> has trouble over 8TB apparently.

well, the driver doesn't see those indices at all, they always see flat
64bit block numbers.  So if someone manages to get in a signedness issue
it'd have to be the fs or pagecache code..


<Prev in Thread] Current Thread [Next in Thread>