[PATCH] xfs: validate sb_logsunit is a multiple of the fs blocksize

[PATCH] xfs: validate sb_logsunit is a multiple of the fs blocksize

[Darrick J. Wong]

Make sure the log stripe unit is sane before proceeding with mounting.
AFAICT this means that logsunit has to be 0, 1, or a multiple of the fs
block size.  Found this by setting the LSB of logsunit in xfs/350 and
watching the system crash as soon as we try to write to the log.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/xfs_log.c |    6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index d729e00..da4dcab 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -659,6 +659,12 @@ xfs_log_mount(
 			 XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks),
 			 XFS_MAX_LOG_BYTES);
 		error = -EINVAL;
+	} else if (mp->m_sb.sb_logsunit > 1 &&
+		   mp->m_sb.sb_logsunit % mp->m_sb.sb_blocksize) {
+		xfs_warn(mp,
+		"log stripe unit %u bytes must be a multiple of block size",
+			 mp->m_sb.sb_logsunit);
+		error = -EINVAL;
 	}
 	if (error) {
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {

Re: [PATCH] xfs: validate sb_logsunit is a multiple of the fs blocksize

[Brian Foster]

On Mon, Oct 23, 2017 at 05:28:44PM -0700, Darrick J. Wong wrote:
> Make sure the log stripe unit is sane before proceeding with mounting.
> AFAICT this means that logsunit has to be 0, 1, or a multiple of the fs
> block size.  Found this by setting the LSB of logsunit in xfs/350 and
> watching the system crash as soon as we try to write to the log.
> 
> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> ---
>  fs/xfs/xfs_log.c |    6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
> index d729e00..da4dcab 100644
> --- a/fs/xfs/xfs_log.c
> +++ b/fs/xfs/xfs_log.c
> @@ -659,6 +659,12 @@ xfs_log_mount(
>  			 XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks),
>  			 XFS_MAX_LOG_BYTES);
>  		error = -EINVAL;
> +	} else if (mp->m_sb.sb_logsunit > 1 &&
> +		   mp->m_sb.sb_logsunit % mp->m_sb.sb_blocksize) {
> +		xfs_warn(mp,
> +		"log stripe unit %u bytes must be a multiple of block size",
> +			 mp->m_sb.sb_logsunit);
> +		error = -EINVAL;

Looks fine, but I notice that the error handling just below only fails
the mount on v5 filesystems. Otherwise we warn and carry on. I'm not
sure why that is.. but is this a crash vector on v4 filesystems as well?

Brian

>  	}
>  	if (error) {
>  		if (xfs_sb_version_hascrc(&mp->m_sb)) {
> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] xfs: validate sb_logsunit is a multiple of the fs blocksize

[Darrick J. Wong]

On Tue, Oct 24, 2017 at 08:54:13AM -0400, Brian Foster wrote:
> On Mon, Oct 23, 2017 at 05:28:44PM -0700, Darrick J. Wong wrote:
> > Make sure the log stripe unit is sane before proceeding with mounting.
> > AFAICT this means that logsunit has to be 0, 1, or a multiple of the fs
> > block size.  Found this by setting the LSB of logsunit in xfs/350 and
> > watching the system crash as soon as we try to write to the log.
> > 
> > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > ---
> >  fs/xfs/xfs_log.c |    6 ++++++
> >  1 file changed, 6 insertions(+)
> > 
> > diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
> > index d729e00..da4dcab 100644
> > --- a/fs/xfs/xfs_log.c
> > +++ b/fs/xfs/xfs_log.c
> > @@ -659,6 +659,12 @@ xfs_log_mount(
> >  			 XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks),
> >  			 XFS_MAX_LOG_BYTES);
> >  		error = -EINVAL;
> > +	} else if (mp->m_sb.sb_logsunit > 1 &&
> > +		   mp->m_sb.sb_logsunit % mp->m_sb.sb_blocksize) {
> > +		xfs_warn(mp,
> > +		"log stripe unit %u bytes must be a multiple of block size",
> > +			 mp->m_sb.sb_logsunit);
> > +		error = -EINVAL;
> 
> Looks fine, but I notice that the error handling just below only fails
> the mount on v5 filesystems. Otherwise we warn and carry on. I'm not
> sure why that is.. but is this a crash vector on v4 filesystems as well?

It is a crash vector on v4 as well, so I'll be stricter about failing
the mount.  I was trying to be cautious about this since there haven't
been general complaints about v4 filesystems, but ... eh.   A crash is a
crash. :)

--D

> 
> Brian
> 
> >  	}
> >  	if (error) {
> >  		if (xfs_sb_version_hascrc(&mp->m_sb)) {
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html