This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: XFS reports lchmod failure, but changes file system contents
- From: Christoph Hellwig <hch at infradead dot org>
- To: Florian Weimer <fw at deneb dot enyo dot de>
- Cc: Christoph Hellwig <hch at infradead dot org>, "Darrick J. Wong" <darrick dot wong at oracle dot com>, linux-xfs at vger dot kernel dot org, libc-alpha at sourceware dot org, linux-fsdevel at vger dot kernel dot org, viro at zeniv dot linux dot org dot uk
- Date: Wed, 12 Feb 2020 10:55:43 -0800
- Subject: Re: XFS reports lchmod failure, but changes file system contents
- References: <874kvwowke.fsf@mid.deneb.enyo.de> <20200212161604.GP6870@magnolia> <20200212181128.GA31394@infradead.org> <87pnejmyhy.fsf@mid.deneb.enyo.de>
On Wed, Feb 12, 2020 at 07:50:01PM +0100, Florian Weimer wrote:
> * Christoph Hellwig:
>
> > xfs doesn't seem all that different from the other file systems,
> > so I suspect you'll also see it with other on-disk file systems.
> > We probably need a check high up in the chmod and co code to reject
> > the operation early for O_PATH file descriptors pointing to symlinks.
>
> We will change the glibc emulation to avoid trying to lchmod symbolic
> links in this way. This will avoid triggering the kernel bug.
We'll still need to fix it.
> (We'd really like to get a proper fchmodat system call with a flags
> argument, though, for AT_EMPTY_PATH and AT_SYMLINK_NOFOLLOW.)
Send a patch (o find a minion to do so).