xfs
[Top] [All Lists]

Re: [PATCH] - filesystem corruption on soft RAID5 in 2.4.0+ (fwd)

To: Russell Cattelan <cattelan@xxxxxxxxxxx>
Subject: Re: [PATCH] - filesystem corruption on soft RAID5 in 2.4.0+ (fwd)
From: Steve Lord <lord@xxxxxxx>
Date: Mon, 22 Jan 2001 13:41:49 -0600
Cc: Andi Kleen <ak@xxxxxxx>, Steve Lord <lord@xxxxxxx>, Scott Smyth <SSmyth@xxxxxxxxxx>, "'linux-xfs@xxxxxxxxxxx '" <linux-xfs@xxxxxxxxxxx>
In-reply-to: Message from Russell Cattelan <cattelan@xxxxxxxxxxx> of "Mon, 22 Jan 2001 13:37:53 CST." <3A6C8C11.FB35C453@xxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
> Andi Kleen wrote:
> 
> > On Mon, Jan 22, 2001 at 12:53:33PM -0600, Russell Cattelan wrote:
> > > Steve Lord wrote:
> > >
> > >
> > > Also Note: all of XFS's meta data writes when using LVM or MD devices
> > > should be in 512byte chunks.
> > > The switching sizes messeages is a bit weird as no size should be
> > > anything
> > > but 512 (meta data) or 4096 (file data)
> >
> > Just a random side comment... nothing really important.
> >
> > This means that XFS won't work on Linux/s390 because it has devices
> > that do not physically support 512 byte blocks. We had fun fixing LVM there
> > which used to rely on 512byte writes too.
> 
> Ouch that one would be really hard to fix. This would require a bunch of
> XFS work. :-(

There is always read/modify/write at the driver level - you just have to
make it atomic.

Steve



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