[PATCH V3] xfs: Do not allow norecovery mount with quotacheck

[PATCH V3] xfs: Do not allow norecovery mount with quotacheck

[cem]

From: Carlos Maiolino <cem@kernel.org>

Mounting a filesystem that requires quota state changing will generate a
transaction.

We already check for a read-only device; we should do that for
norecovery too.

A quotacheck on a norecovery mount, and with the right log size, will cause
the mount process to hang on:

[<0>] xlog_grant_head_wait+0x5d/0x2a0 [xfs]
[<0>] xlog_grant_head_check+0x112/0x180 [xfs]
[<0>] xfs_log_reserve+0xe3/0x260 [xfs]
[<0>] xfs_trans_reserve+0x179/0x250 [xfs]
[<0>] xfs_trans_alloc+0x101/0x260 [xfs]
[<0>] xfs_sync_sb+0x3f/0x80 [xfs]
[<0>] xfs_qm_mount_quotas+0xe3/0x2f0 [xfs]
[<0>] xfs_mountfs+0x7ad/0xc20 [xfs]
[<0>] xfs_fs_fill_super+0x762/0xa50 [xfs]
[<0>] get_tree_bdev_flags+0x131/0x1d0
[<0>] vfs_get_tree+0x26/0xd0
[<0>] vfs_cmd_create+0x59/0xe0
[<0>] __do_sys_fsconfig+0x4e3/0x6b0
[<0>] do_syscall_64+0x82/0x160
[<0>] entry_SYSCALL_64_after_hwframe+0x76/0x7e

This is caused by a transaction running with bogus initialized head/tail

I initially hit this while running generic/050, with random log
sizes, but I managed to reproduce it reliably here with the steps
below:

mkfs.xfs -f -lsize=1025M -f -b size=4096 -m crc=1,reflink=1,rmapbt=1, -i
sparse=1 /dev/vdb2 > /dev/null
mount -o usrquota,grpquota,prjquota /dev/vdb2 /mnt
xfs_io -x -c 'shutdown -f' /mnt
umount /mnt
mount -o ro,norecovery,usrquota,grpquota,prjquota  /dev/vdb2 /mnt

Last mount hangs up

As we add yet another validation if quota state is changing, this also
add a new helper named xfs_qm_validate_state_change(), factoring the
quota state changes out of xfs_qm_newmount() to reduce cluttering
within it.

Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
---

Changelog V2 -> V3:
	- Update helper name
	- Update metadir warn message
	- Don't use typedef for xfs_mount

 fs/xfs/xfs_qm_bhv.c | 55 ++++++++++++++++++++++++++++++++-------------
 1 file changed, 39 insertions(+), 16 deletions(-)

diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
index 37f1230e7584..245d754f382a 100644
--- a/fs/xfs/xfs_qm_bhv.c
+++ b/fs/xfs/xfs_qm_bhv.c
@@ -78,6 +78,28 @@ xfs_qm_statvfs(
 	}
 }
 
+STATIC int
+xfs_qm_validate_state_change(
+	struct xfs_mount	*mp,
+	uint			uqd,
+	uint			gqd,
+	uint			pqd)
+{
+	int state;
+
+	/* Is quota state changing? */
+	state = ((uqd && !XFS_IS_UQUOTA_ON(mp)) ||
+		(!uqd &&  XFS_IS_UQUOTA_ON(mp)) ||
+		 (gqd && !XFS_IS_GQUOTA_ON(mp)) ||
+		(!gqd &&  XFS_IS_GQUOTA_ON(mp)) ||
+		 (pqd && !XFS_IS_PQUOTA_ON(mp)) ||
+		(!pqd &&  XFS_IS_PQUOTA_ON(mp)));
+
+	return  state &&
+		(xfs_dev_is_read_only(mp, "changing quota state") ||
+		xfs_has_norecovery(mp));
+}
+
 int
 xfs_qm_newmount(
 	xfs_mount_t	*mp,
@@ -97,24 +119,25 @@ xfs_qm_newmount(
 	}
 
 	/*
-	 * If the device itself is read-only, we can't allow
-	 * the user to change the state of quota on the mount -
-	 * this would generate a transaction on the ro device,
-	 * which would lead to an I/O error and shutdown
+	 * If the device itself is read-only and/or in norecovery
+	 * mode, we can't allow the user to change the state of
+	 * quota on the mount - this would generate a transaction
+	 * on the ro device, which would lead to an I/O error and
+	 * shutdown.
 	 */
 
-	if (((uquotaondisk && !XFS_IS_UQUOTA_ON(mp)) ||
-	    (!uquotaondisk &&  XFS_IS_UQUOTA_ON(mp)) ||
-	     (gquotaondisk && !XFS_IS_GQUOTA_ON(mp)) ||
-	    (!gquotaondisk &&  XFS_IS_GQUOTA_ON(mp)) ||
-	     (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) ||
-	    (!pquotaondisk &&  XFS_IS_PQUOTA_ON(mp)))  &&
-	    xfs_dev_is_read_only(mp, "changing quota state")) {
-		xfs_warn(mp, "please mount with%s%s%s%s.",
-			(!quotaondisk ? "out quota" : ""),
-			(uquotaondisk ? " usrquota" : ""),
-			(gquotaondisk ? " grpquota" : ""),
-			(pquotaondisk ? " prjquota" : ""));
+	if (xfs_qm_validate_state_change(mp, uquotaondisk,
+			    gquotaondisk, pquotaondisk)) {
+
+		if (xfs_has_metadir(mp))
+			xfs_warn(mp,
+		"metadir enabled, please mount without any quota mount options");
+		else
+			xfs_warn(mp, "please mount with%s%s%s%s.",
+				(!quotaondisk ? "out quota" : ""),
+				(uquotaondisk ? " usrquota" : ""),
+				(gquotaondisk ? " grpquota" : ""),
+				(pquotaondisk ? " prjquota" : ""));
 		return -EPERM;
 	}
 
-- 
2.48.1

Re: [PATCH V3] xfs: Do not allow norecovery mount with quotacheck

[Darrick J. Wong]

On Mon, Feb 03, 2025 at 02:04:57PM +0100, cem@kernel.org wrote:
> From: Carlos Maiolino <cem@kernel.org>
> 
> Mounting a filesystem that requires quota state changing will generate a
> transaction.
> 
> We already check for a read-only device; we should do that for
> norecovery too.
> 
> A quotacheck on a norecovery mount, and with the right log size, will cause
> the mount process to hang on:
> 
> [<0>] xlog_grant_head_wait+0x5d/0x2a0 [xfs]
> [<0>] xlog_grant_head_check+0x112/0x180 [xfs]
> [<0>] xfs_log_reserve+0xe3/0x260 [xfs]
> [<0>] xfs_trans_reserve+0x179/0x250 [xfs]
> [<0>] xfs_trans_alloc+0x101/0x260 [xfs]
> [<0>] xfs_sync_sb+0x3f/0x80 [xfs]
> [<0>] xfs_qm_mount_quotas+0xe3/0x2f0 [xfs]
> [<0>] xfs_mountfs+0x7ad/0xc20 [xfs]
> [<0>] xfs_fs_fill_super+0x762/0xa50 [xfs]
> [<0>] get_tree_bdev_flags+0x131/0x1d0
> [<0>] vfs_get_tree+0x26/0xd0
> [<0>] vfs_cmd_create+0x59/0xe0
> [<0>] __do_sys_fsconfig+0x4e3/0x6b0
> [<0>] do_syscall_64+0x82/0x160
> [<0>] entry_SYSCALL_64_after_hwframe+0x76/0x7e
> 
> This is caused by a transaction running with bogus initialized head/tail
> 
> I initially hit this while running generic/050, with random log
> sizes, but I managed to reproduce it reliably here with the steps
> below:
> 
> mkfs.xfs -f -lsize=1025M -f -b size=4096 -m crc=1,reflink=1,rmapbt=1, -i
> sparse=1 /dev/vdb2 > /dev/null
> mount -o usrquota,grpquota,prjquota /dev/vdb2 /mnt
> xfs_io -x -c 'shutdown -f' /mnt
> umount /mnt
> mount -o ro,norecovery,usrquota,grpquota,prjquota  /dev/vdb2 /mnt
> 
> Last mount hangs up
> 
> As we add yet another validation if quota state is changing, this also
> add a new helper named xfs_qm_validate_state_change(), factoring the
> quota state changes out of xfs_qm_newmount() to reduce cluttering
> within it.
> 
> Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>

Looks good to me now,
Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>

--D

> ---
> 
> Changelog V2 -> V3:
> 	- Update helper name
> 	- Update metadir warn message
> 	- Don't use typedef for xfs_mount
> 
>  fs/xfs/xfs_qm_bhv.c | 55 ++++++++++++++++++++++++++++++++-------------
>  1 file changed, 39 insertions(+), 16 deletions(-)
> 
> diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
> index 37f1230e7584..245d754f382a 100644
> --- a/fs/xfs/xfs_qm_bhv.c
> +++ b/fs/xfs/xfs_qm_bhv.c
> @@ -78,6 +78,28 @@ xfs_qm_statvfs(
>  	}
>  }
>  
> +STATIC int
> +xfs_qm_validate_state_change(
> +	struct xfs_mount	*mp,
> +	uint			uqd,
> +	uint			gqd,
> +	uint			pqd)
> +{
> +	int state;
> +
> +	/* Is quota state changing? */
> +	state = ((uqd && !XFS_IS_UQUOTA_ON(mp)) ||
> +		(!uqd &&  XFS_IS_UQUOTA_ON(mp)) ||
> +		 (gqd && !XFS_IS_GQUOTA_ON(mp)) ||
> +		(!gqd &&  XFS_IS_GQUOTA_ON(mp)) ||
> +		 (pqd && !XFS_IS_PQUOTA_ON(mp)) ||
> +		(!pqd &&  XFS_IS_PQUOTA_ON(mp)));
> +
> +	return  state &&
> +		(xfs_dev_is_read_only(mp, "changing quota state") ||
> +		xfs_has_norecovery(mp));
> +}
> +
>  int
>  xfs_qm_newmount(
>  	xfs_mount_t	*mp,
> @@ -97,24 +119,25 @@ xfs_qm_newmount(
>  	}
>  
>  	/*
> -	 * If the device itself is read-only, we can't allow
> -	 * the user to change the state of quota on the mount -
> -	 * this would generate a transaction on the ro device,
> -	 * which would lead to an I/O error and shutdown
> +	 * If the device itself is read-only and/or in norecovery
> +	 * mode, we can't allow the user to change the state of
> +	 * quota on the mount - this would generate a transaction
> +	 * on the ro device, which would lead to an I/O error and
> +	 * shutdown.
>  	 */
>  
> -	if (((uquotaondisk && !XFS_IS_UQUOTA_ON(mp)) ||
> -	    (!uquotaondisk &&  XFS_IS_UQUOTA_ON(mp)) ||
> -	     (gquotaondisk && !XFS_IS_GQUOTA_ON(mp)) ||
> -	    (!gquotaondisk &&  XFS_IS_GQUOTA_ON(mp)) ||
> -	     (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) ||
> -	    (!pquotaondisk &&  XFS_IS_PQUOTA_ON(mp)))  &&
> -	    xfs_dev_is_read_only(mp, "changing quota state")) {
> -		xfs_warn(mp, "please mount with%s%s%s%s.",
> -			(!quotaondisk ? "out quota" : ""),
> -			(uquotaondisk ? " usrquota" : ""),
> -			(gquotaondisk ? " grpquota" : ""),
> -			(pquotaondisk ? " prjquota" : ""));
> +	if (xfs_qm_validate_state_change(mp, uquotaondisk,
> +			    gquotaondisk, pquotaondisk)) {
> +
> +		if (xfs_has_metadir(mp))
> +			xfs_warn(mp,
> +		"metadir enabled, please mount without any quota mount options");
> +		else
> +			xfs_warn(mp, "please mount with%s%s%s%s.",
> +				(!quotaondisk ? "out quota" : ""),
> +				(uquotaondisk ? " usrquota" : ""),
> +				(gquotaondisk ? " grpquota" : ""),
> +				(pquotaondisk ? " prjquota" : ""));
>  		return -EPERM;
>  	}
>  
> -- 
> 2.48.1
> 
> 

Re: [PATCH V3] xfs: Do not allow norecovery mount with quotacheck

[Carlos Maiolino]

On Mon, 03 Feb 2025 14:04:57 +0100, cem@kernel.org wrote:
> Mounting a filesystem that requires quota state changing will generate a
> transaction.
> 
> We already check for a read-only device; we should do that for
> norecovery too.
> 
> A quotacheck on a norecovery mount, and with the right log size, will cause
> the mount process to hang on:
> 
> [...]

Applied to for-next, thanks!

[1/1] xfs: Do not allow norecovery mount with quotacheck
      commit: 18df3ca14c34e9fd6e12189a24238a12b064fbac

Best regards,
-- 
Carlos Maiolino <cem@kernel.org>