5.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jan Kara jack@suse.cz
commit fcaf3b2683b05a9684acdebda706a12025a6927a upstream.
Currently quota recovery is synchronized with unmount using sb->s_umount semaphore. That is however prone to deadlocks because flush_workqueue(osb->ocfs2_wq) called from umount code can wait for quota recovery to complete while ocfs2_finish_quota_recovery() waits for sb->s_umount semaphore.
Grabbing of sb->s_umount semaphore in ocfs2_finish_quota_recovery() is only needed to protect that function from disabling of quotas from ocfs2_dismount_volume(). Handle this problem by disabling quota recovery early during unmount in ocfs2_dismount_volume() instead so that we can drop acquisition of sb->s_umount from ocfs2_finish_quota_recovery().
Link: https://lkml.kernel.org/r/20250424134515.18933-6-jack@suse.cz Fixes: 5f530de63cfc ("ocfs2: Use s_umount for quota recovery protection") Signed-off-by: Jan Kara jack@suse.cz Reported-by: Shichangkuo shi.changkuo@h3c.com Reported-by: Murad Masimov m.masimov@mt-integration.ru Reviewed-by: Heming Zhao heming.zhao@suse.com Tested-by: Heming Zhao heming.zhao@suse.com Acked-by: Joseph Qi joseph.qi@linux.alibaba.com Cc: Changwei Ge gechangwei@live.cn Cc: Joel Becker jlbec@evilplan.org Cc: Jun Piao piaojun@huawei.com Cc: Junxiao Bi junxiao.bi@oracle.com Cc: Mark Fasheh mark@fasheh.com Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- fs/ocfs2/journal.c | 20 ++++++++++++++++++-- fs/ocfs2/journal.h | 1 + fs/ocfs2/ocfs2.h | 6 ++++++ fs/ocfs2/quota_local.c | 9 ++------- fs/ocfs2/super.c | 3 +++ 5 files changed, 30 insertions(+), 9 deletions(-)
--- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -229,6 +229,11 @@ out_lock: flush_workqueue(osb->ocfs2_wq); }
+void ocfs2_recovery_disable_quota(struct ocfs2_super *osb) +{ + ocfs2_recovery_disable(osb, OCFS2_REC_QUOTA_WANT_DISABLE); +} + void ocfs2_recovery_exit(struct ocfs2_super *osb) { struct ocfs2_recovery_map *rm; @@ -1429,6 +1434,18 @@ static int __ocfs2_recovery_thread(void } } restart: + if (quota_enabled) { + mutex_lock(&osb->recovery_lock); + /* Confirm that recovery thread will no longer recover quotas */ + if (osb->recovery_state == OCFS2_REC_QUOTA_WANT_DISABLE) { + osb->recovery_state = OCFS2_REC_QUOTA_DISABLED; + wake_up(&osb->recovery_event); + } + if (osb->recovery_state >= OCFS2_REC_QUOTA_DISABLED) + quota_enabled = 0; + mutex_unlock(&osb->recovery_lock); + } + status = ocfs2_super_lock(osb, 1); if (status < 0) { mlog_errno(status); @@ -1532,8 +1549,7 @@ bail:
mutex_unlock(&osb->recovery_lock);
- if (quota_enabled) - kfree(rm_quota); + kfree(rm_quota);
/* no one is callint kthread_stop() for us so the kthread() api * requires that we call do_exit(). And it isn't exported, but --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -150,6 +150,7 @@ void ocfs2_wait_for_recovery(struct ocfs
int ocfs2_recovery_init(struct ocfs2_super *osb); void ocfs2_recovery_exit(struct ocfs2_super *osb); +void ocfs2_recovery_disable_quota(struct ocfs2_super *osb);
int ocfs2_compute_replay_slots(struct ocfs2_super *osb); void ocfs2_free_replay_slots(struct ocfs2_super *osb); --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -288,6 +288,12 @@ enum ocfs2_mount_options
enum ocfs2_recovery_state { OCFS2_REC_ENABLED = 0, + OCFS2_REC_QUOTA_WANT_DISABLE, + /* + * Must be OCFS2_REC_QUOTA_WANT_DISABLE + 1 for + * ocfs2_recovery_disable_quota() to work. + */ + OCFS2_REC_QUOTA_DISABLED, OCFS2_REC_WANT_DISABLE, /* * Must be OCFS2_REC_WANT_DISABLE + 1 for ocfs2_recovery_exit() to work --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -453,8 +453,7 @@ out:
/* Sync changes in local quota file into global quota file and * reinitialize local quota file. - * The function expects local quota file to be already locked and - * s_umount locked in shared mode. */ + * The function expects local quota file to be already locked. */ static int ocfs2_recover_local_quota_file(struct inode *lqinode, int type, struct ocfs2_quota_recovery *rec) @@ -585,7 +584,6 @@ int ocfs2_finish_quota_recovery(struct o { unsigned int ino[OCFS2_MAXQUOTAS] = { LOCAL_USER_QUOTA_SYSTEM_INODE, LOCAL_GROUP_QUOTA_SYSTEM_INODE }; - struct super_block *sb = osb->sb; struct ocfs2_local_disk_dqinfo *ldinfo; struct buffer_head *bh; handle_t *handle; @@ -597,7 +595,6 @@ int ocfs2_finish_quota_recovery(struct o printk(KERN_NOTICE "ocfs2: Finishing quota recovery on device (%s) for " "slot %u\n", osb->dev_str, slot_num);
- down_read(&sb->s_umount); for (type = 0; type < OCFS2_MAXQUOTAS; type++) { if (list_empty(&(rec->r_list[type]))) continue; @@ -674,7 +671,6 @@ out_put: break; } out: - up_read(&sb->s_umount); kfree(rec); return status; } @@ -840,8 +836,7 @@ static int ocfs2_local_free_info(struct ocfs2_release_local_quota_bitmaps(&oinfo->dqi_chunk);
/* - * s_umount held in exclusive mode protects us against racing with - * recovery thread... + * ocfs2_dismount_volume() has already aborted quota recovery... */ if (oinfo->dqi_rec) { ocfs2_free_quota_recovery(oinfo->dqi_rec); --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1876,6 +1876,9 @@ static void ocfs2_dismount_volume(struct /* Orphan scan should be stopped as early as possible */ ocfs2_orphan_scan_stop(osb);
+ /* Stop quota recovery so that we can disable quotas */ + ocfs2_recovery_disable_quota(osb); + ocfs2_disable_quotas(osb);
/* All dquots should be freed by now */