xfs
[Top] [All Lists]

Re: [REVIEW #4] bad_features2 support in userspace

To: Barry Naujok <bnaujok@xxxxxxx>
Subject: Re: [REVIEW #4] bad_features2 support in userspace
From: "Josef 'Jeff' Sipek" <jeffpc@xxxxxxxxxxxxxx>
Date: Tue, 4 Mar 2008 23:45:39 -0500
Cc: "xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>
In-reply-to: <op.t7i0f5123jf8g2@pc-bnaujok.melbourne.sgi.com>
References: <op.t7i0f5123jf8g2@pc-bnaujok.melbourne.sgi.com>
Sender: xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.5.16 (2007-06-11)
On Wed, Mar 05, 2008 at 03:37:07PM +1100, Barry Naujok wrote:
> Due to the issue of mounting filesystem with older kernels and
> potentially reading sb_features2 from the wrong location. It
> seems the best course of action is to always make sb_features2
> and sb_bad_features2 the same. This is pretty important as
> new bits in this are supposed to stop older kernels from
> mounting filesystems with unsupported features.
>
> If sb_bad_features2 is zero, and the old kernel tries to read
> sb_features2 from this location during mount, it will succeed
> as it will read zero.
>
> So, this patch changes mkfs.xfs to set sb_bad_features2 to
> the same as sb_features2, xfs_check and xfs_repair now also
> makes sure they are the same.

Idea: good
Implementation: I didn't see anything wrong.

Josef 'Jeff' Sipek.

P.S. Any reason why you inline the patch _and_ attach?

-- 
I think there is a world market for maybe five computers.
                - Thomas Watson, chairman of IBM, 1943.


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