On Wed, 11 Dec 2013, Jeff Liu wrote:
> Date: Wed, 11 Dec 2013 21:10:48 +0800
> From: Jeff Liu <jeff.liu@xxxxxxxxxx>
> To: Lukas Czerner <lczerner@xxxxxxxxxx>, xfs@xxxxxxxxxxx
> Subject: Re: [PATCH] xfs: prevent range.len underflow in xfs_ioc_trim()
> Hi Lukas,
> Thanks for the fix. We have corrected this problem with the blow patch
> which got merged intwo weeks ago. :)
> [ commit f9fd013561 ]
Not sure where it got merged, I do not see this upstream in linus
> xfs: don't perform discard if the given range length is less than block size
> That is also discovered by generic/288 you contributed.
Right, I mentioned that in the description.
> On 12/11 2013 20:05 PM, Lukas Czerner wrote:
> > Currently when range.len is set to 0 it will underflow. Fix it by
> > checking for this scenario and return EINVAL in case range.len is
> > smaller than block size.
> > This was discovered by the xfstests generic/288 and with this patch
> > the problem goes away.
> > Signed-off-by: Lukas Czerner <lczerner@xxxxxxxxxx>
> > ---
> > fs/xfs/xfs_discard.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> > diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
> > index 8367d6d..9029082 100644
> > --- a/fs/xfs/xfs_discard.c
> > +++ b/fs/xfs/xfs_discard.c
> > @@ -180,7 +180,8 @@ xfs_ioc_trim(
> > * matter as trimming blocks is an advisory interface.
> > */
> > if (range.start >= XFS_FSB_TO_B(mp, mp->m_sb.sb_dblocks) ||
> > - range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp)))
> > + range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp)) ||
> > + range.len < XFS_FSB_TO_B(mp, 1))
> > return -XFS_ERROR(EINVAL);
> > start = BTOBB(range.start);