linux-kernel - Re: [PATCH 01/24] GFS2: Support for FIEMAP ioctl

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Thu, 18 Dec 2008 11:04:14 -0800
From:	Andrew Morton <akpm@...ux-foundation.org>
To:	Steven Whitehouse <swhiteho@...hat.com>
Cc:	linux-kernel@...r.kernel.org, cluster-devel@...hat.com,
	tytso@....edu, sandeen@...hat.com
Subject: Re: [PATCH 01/24] GFS2: Support for FIEMAP ioctl

On Thu, 18 Dec 2008 10:29:04 +0000
Steven Whitehouse <swhiteho@...hat.com> wrote:

> Hi,
> 
> On Wed, 2008-12-17 at 17:22 -0800, Andrew Morton wrote:
> > On Wed, 17 Dec 2008 11:30:00 +0000
> > swhiteho@...hat.com wrote:
> > 
> > > +static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
> > > +		       u64 start, u64 len)
> > > +{
> > > +	struct gfs2_inode *ip = GFS2_I(inode);
> > > +	struct gfs2_holder gh;
> > > +	int ret;
> > > +
> > > +	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC);
> > > +	if (ret)
> > > +		return ret;
> > > +
> > > +	mutex_lock(&inode->i_mutex);
> > > +
> > > +	ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
> > > +	if (ret)
> > > +		goto out;
> > > +
> > > +	if (gfs2_is_stuffed(ip)) {
> > > +		u64 phys = ip->i_no_addr << inode->i_blkbits;
> > > +		u64 size = i_size_read(inode);
> > 
> > It's actually safe to directly access i_size inside i_mutex.  Although
> > not terribly maintainable.
> > 
> > 
> I did wonder about that at the time, but I'm not quite sure if you are
> suggesting that I should change this now, or leave it as it is?

No strong opinion, really.  If it's a performance-sensitive path (it
isn't) then bypassing i_size_read() might be advantageous.  Leaving the
i_size_read() there provides some future-safety and sets a good
example.

It would be an easier decision if the SMP, 32-bit version of
i_size_read() wasn't inlined, and tremendously huge :(

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/