xfs: fix uflags detection at xfs_fs_rm_xquota
PATCH 1/1 xfs: fix uflags detection at xfs_fs_rm_xquota
From: Jie Liu <[email protected]>
We are intended to check up uflags against FS_PROJ_QUOTA rather than FS_USER_UQUOTA once more, it looks to me like a typo, but might cause the project quota metadata space can not be removed.
--- fs/xfs/xfs_quotaops.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/xfs/xfs_quotaops.c +1 -1
@@ -123,7 +123,7 @@ xfs_fs_rm_xquota( flags |=XFS_DQ_USER ;if (uflags &FS_GROUP_QUOTA ) flags |=XFS_DQ_GROUP ; -if (uflags &FS_USER_QUOTA ) +if (uflags &FS_PROJ_QUOTA ) flags |=XFS_DQ_PROJ ;return xfs_qm_scall_trunc_qfiles (mp, flags);
Signature · 6 lines
1.8.3.2 _______________________________________________ xfs mailing list [email protected] http://oss.sgi.com/mailman/listinfo/xfs
Quoted 5 lines · Jeff Liu
From: Jie Liu <[email protected]>
We are intended to check up uflags against
FS_PROJ_QUOTArather thanFS_USER_UQUOTAonce more, it looks to me like a typo, but might cause the project quota metadata space can not be removed.
This code was introduced by commit 9da93f9b7cdf8ab28da6b364cdc1fafc8670b4dc which attempts to fix the broken quotarm code. Seems like we need a testcase for this functionality or just remove it. Physically removing the quota file never seems that useful to me to start with..
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 7 lines · Christoph Hellwig
Quoted 5 lines · Jeff Liu
From: Jie Liu <[email protected]>
We are intended to check up uflags against
FS_PROJ_QUOTArather thanFS_USER_UQUOTAonce more, it looks to me like a typo, but might cause the project quota metadata space can not be removed.
Quoted 5 lines
This code was introduced by commit
9da93f9b7cdf8ab28da6b364cdc1fafc8670b4dcwhich attempts to fix the broken quotarm code. Seems like we need a testcase for this functionality or just remove it. Physically removing the quota file never seems that useful to me to start with..
Is there an approach to verify that a particular quota file inode has been successfully removed? It seems we have to do this in a test case but am not yet figure out a way, or any thoughts for the test case?
Cheers,
-Jeff
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 9 lines · Jeff Liu
Quoted 5 lines
This code was introduced by commit
9da93f9b7cdf8ab28da6b364cdc1fafc8670b4dcwhich attempts to fix the broken quotarm code. Seems like we need a testcase for this functionality or just remove it. Physically removing the quota file never seems that useful to me to start with..Is there an approach to verify that a particular quota file inode has been successfully removed? It seems we have to do this in a test case but am not yet figure out a way, or any thoughts for the test case?
umount the filesystem and check that the field in the superblock pointing to that kind of quota inode is zeroed out would be my suggestion.
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 7 lines · Christoph Hellwig
Quoted 6 lines · Jeff Liu
Quoted 5 lines
This code was introduced by commit
9da93f9b7cdf8ab28da6b364cdc1fafc8670b4dcwhich attempts to fix the broken quotarm code. Seems like we need a testcase for this functionality or just remove it. Physically removing the quota file never seems that useful to me to start with..
Quoted 7 lines
Quoted 3 lines
Is there an approach to verify that a particular quota file inode has been successfully removed? It seems we have to do this in a test case but am not yet figure out a way, or any thoughts for the test case?
umount the filesystem and check that the field in the superblock pointing to that kind of quota inode is zeroed out would be my suggestion.
Oh yes, after quota remove, the 'u' field of the quota inode will changed from u.bmx[0] = [startoff,startblock,blockcount,extentflag] 0:[0,16,1,0] to u = (empty) which can be used for the verification, thanks for your suggestion. :)
Cheers,
-Jeff
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 5 lines · Jeff Liu
From: Jie Liu <[email protected]>
We are intended to check up uflags against
FS_PROJ_QUOTArather thanFS_USER_UQUOTAonce more, it looks to me like a typo, but might cause the project quota metadata space can not be removed.
Ugh, mea culpa. Thanks for spotting. :(
Thankfully I don't think anyone uses this interface ;)
Quoted 11 lines
Signed-off-by Jie Liu <[email protected]>--- fs/xfs/xfs_quotaops.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)fs/xfs/xfs_quotaops.c +0
-0
fs/xfs/xfs_quotaops.c xfs_fs_rm_xquota( · line 129
Quoted 8 lines
fs/xfs/xfs_quotaops.c +1
-1@@ -123,7 +123,7 @@ xfs_fs_rm_xquota( flags |=XFS_DQ_USER ;if (uflags &FS_GROUP_QUOTA ) flags |=XFS_DQ_GROUP ; -if (uflags &FS_USER_QUOTA ) +if (uflags &FS_PROJ_QUOTA ) flags |=XFS_DQ_PROJ ;return xfs_qm_scall_trunc_qfiles (mp, flags);
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs