lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Tue, 3 Aug 2021 09:00:25 +0800
From:   Gang He <ghe@...e.com>
To:     Wengang Wang <wen.gang.wang@...cle.com>
CC:     "mark@...heh.com" <mark@...heh.com>,
        "jlbec@...lplan.org" <jlbec@...lplan.org>,
        "joseph.qi@...ux.alibaba.com" <joseph.qi@...ux.alibaba.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "ocfs2-devel@....oracle.com" <ocfs2-devel@....oracle.com>
Subject: Re: [Ocfs2-devel] [PATCH] ocfs2: reflink deadlock when clone file to
 the same directory simultaneously



On 2021/8/3 2:22, Wengang Wang wrote:
> 
> 
>> On Aug 1, 2021, at 9:35 PM, Gang He <ghe@...e.com> wrote:
>>
>> Hi Wengang,
>>
>> The ocfs2 file system is mounted at /mnt/shared,
>> The test file(e.g. 40G VM image) is located at /mnt/shared/test, the snapshot direcory is /mnt/shared/.snapshots, then the user runs a reflink command from each node to generate a clone file repeatedly like,
>> reflink "/mnt/shared/test" \
>>    "/mnt/shared/.snapshots/test.`date +%m%d%H%M%S`.`hostname`"
>>
>> after a while, the reflink process on each node is hung, the user
>> cannot list the file system again.
>>
>> The problematic reflink command process is blocked at one node, like
>> __schedule+0x2fd/0x750
>> schedule+0x2f/0xa0
>> schedule_timeout+0x1cc/0x310
>> ? ocfs2_control_cfu+0x50/0x50 [ocfs2_stack_user]
>> ? 0xffffffffc0e3e000
>> wait_for_completion+0xba/0x140
>> ? wake_up_q+0xa0/0xa0
>> __ocfs2_cluster_lock.isra.41+0x3b5/0x820 [ocfs2]
>> ? ocfs2_inode_lock_full_nested+0x1fc/0x960 [ocfs2]
>> ocfs2_inode_lock_full_nested+0x1fc/0x960 [ocfs2]
>> ocfs2_init_security_and_acl+0xbe/0x1d0 [ocfs2]
>> ocfs2_reflink+0x436/0x4c0 [ocfs2]
>> ? ocfs2_reflink_ioctl+0x2ca/0x360 [ocfs2]
>> ocfs2_reflink_ioctl+0x2ca/0x360 [ocfs2]
>> ocfs2_ioctl+0x25e/0x670 [ocfs2]
>> do_vfs_ioctl+0xa0/0x680
>> ksys_ioctl+0x70/0x80
>> __x64_sys_ioctl+0x16/0x20
>>
>> In fact, this node has acquired .snapshots directory inode dlm lock,
>> but the reflink process is blocked at ocfs2_init_security_and_acl+0xbe/0x1d0, the process is wait for getting
>> .snapshots directory inode dlm lock.
>> The reflink process on other nodes is also blocked at getting .snapshots
>> directory inode dlm lock, then the file system cannot be listed since "ls -l" command also need to get .snapshot directory's attributes.
>>
> 
> I still didn’t get which locks are involved in the dead lock party.
> If that is not clear, it’s hard to review your patch further.
.snapshot inode dlm lock, the node has acquired this lock, then
acquire it again, this is a deadlock key point.


> 
> 
>> Why the the reflink process is blocked at ocfs2_init_security_and_acl+0xbe/0x1d0? I feel it is caused by multiple acquiring the .snapshots directory inode dlm lock in ocfs2_reflink function. For the deeper reasons, it is probably related to dlmglue layer or there is any factor which blocked .snapshots inode dlm lock to downconvert.
> 
> I suggest you take tcpdump in your system and get clear what was going on.  So you can remove the words like “feel” or “probably” next time.
> If the problem is in downconvert, we should fix that instead.
I think we can fix it via using one time acquire the lock before trying 
to investigate further. In fact, this way also can improve the 
performance in case the user run the reflink to clone file to the same 
directory simultaneously.


Thanks
Gang

> 
> thanks,
> wengang
> 
>>
>> Thanks
>> Gang
>>
>>
>>
>>
>> On 2021/7/31 0:35, Wengang Wang wrote:
>>> Hi Gang,
>>>> On Jul 29, 2021, at 11:16 PM, Gang He <ghe@...e.com> wrote:
>>>>
>>>> Hello Wengang and all,
>>>>
>>>> This issue can be reproduced stably when you run the below reflink command line(maybe you also can follow a "rm this file" command line and  sleep some usecs) from each node repeatedly for a while.
>>>> Based on my observation, the reflink processes are always blocked at the below points.
>>>>  From dlm_tool output and crash analysis, the node1 has acquired .snapshots directory inode EX dlm lock, but the reflink process is blocked at ocfs2_init_security_and_acl+0xbe/0x1d0 to acqure it's inode dlm lock again.
>>> So the reflink path on node 1 have .snapshots inode lock granted and is blocking at the new created inode under orphan directory.  BTW, what’s .snapshots directory? What’s the call path to lock that .snapshots inode?
>>>> On the other two nodes, the reflink processes are blocked at acquire .snapshots directory inode dlm lock, then the whole file system is hung,
>>>> you can not list this file again.
>>> So there are reflink paths on the other two nodes blocking at .snapshots inode. But what lock they are granted already?
>>> For a typical ABBA deadlock,
>>> path 1 granted lock A and blocks at lock B
>>> path 2 granted lock B and blocks at lock A
>>> Per your description, I see this:
>>> reflink path on node1  granted .snapshots lock and blocks at new inode lock
>>> reflnk paths on onde2/3 block at .snapshots lock.
>>> I don't see how deadlock formed… the new inode lock is granted to any of the reflink path on node2/3? how?
>>> thanks,
>>> wengang
>>>>
>>>> The problem looks like acquiring the destination direcory multiple during ocfs2_reflink, dlm glue layer cannot downconvert lock in some case.
>>>> e.g.
>>>> kernel: (ocfs2dc-F50B203,1593,0):ocfs2_downconvert_lock:3674 ERROR: DLM error -16 while calling ocfs2_dlm_lock on resource M000000000000000004661c00000000
>>>> kernel: (ocfs2dc-F50B203,1593,0):ocfs2_unblock_lock:3918 ERROR: status = -16
>>>> kernel: (ocfs2dc-F50B203,1593,0):ocfs2_process_blocked_lock:4317 ERROR: status = -16
>>>>
>>>> Then, I change the code to acquire this destination direcory dlm lock, and hold the lock until the end of ocfs2_reflink function.
>>>> After this change, I did not encounter this hang problem again after lots of testing. Second, I find the code change also improve reflink performance, since the code avoids the previous ping-pong effect.
>>>>
>>>> Thanks
>>>> Gang
>>>>
>>>>
>>>> On 2021/7/30 6:07, Wengang Wang wrote:
>>>>> Hi Gang,
>>>>> I’d suggest you list the call paths on the related nodes, Say call path 1 on node one granted lock A and is requesting for lock B, at the same time, path2 on node two granted lock B and now is requesting for lock A.
>>>>> With that, the problem would be easier to understand.
>>>>> thanks,
>>>>> wengang
>>>>>> On Jul 29, 2021, at 4:02 AM, Gang He <ghe@...e.com> wrote:
>>>>>>
>>>>>> Running reflink from multiple nodes simultaneously to clone a file
>>>>>> to the same directory probably triggers a deadlock issue.
>>>>>> For example, there is a three node ocfs2 cluster, each node mounts
>>>>>> the ocfs2 file system to /mnt/shared, and run the reflink command
>>>>>> from each node repeatedly, like
>>>>>>   reflink "/mnt/shared/test" \
>>>>>>   "/mnt/shared/.snapshots/test.`date +%m%d%H%M%S`.`hostname`"
>>>>>> then, reflink command process will be hung on each node, and you
>>>>>> can't list this file system directory.
>>>>>> The problematic reflink command process is blocked at one node,
>>>>>> task:reflink         state:D stack:    0 pid: 1283 ppid:  4154
>>>>>> Call Trace:
>>>>>>   __schedule+0x2fd/0x750
>>>>>>   schedule+0x2f/0xa0
>>>>>>   schedule_timeout+0x1cc/0x310
>>>>>>   ? ocfs2_control_cfu+0x50/0x50 [ocfs2_stack_user]
>>>>>>   ? 0xffffffffc0e3e000
>>>>>>   wait_for_completion+0xba/0x140
>>>>>>   ? wake_up_q+0xa0/0xa0
>>>>>>   __ocfs2_cluster_lock.isra.41+0x3b5/0x820 [ocfs2]
>>>>>>   ? ocfs2_inode_lock_full_nested+0x1fc/0x960 [ocfs2]
>>>>>>   ocfs2_inode_lock_full_nested+0x1fc/0x960 [ocfs2]
>>>>>>   ocfs2_init_security_and_acl+0xbe/0x1d0 [ocfs2]
>>>>>>   ocfs2_reflink+0x436/0x4c0 [ocfs2]
>>>>>>   ? ocfs2_reflink_ioctl+0x2ca/0x360 [ocfs2]
>>>>>>   ocfs2_reflink_ioctl+0x2ca/0x360 [ocfs2]
>>>>>>   ocfs2_ioctl+0x25e/0x670 [ocfs2]
>>>>>>   do_vfs_ioctl+0xa0/0x680
>>>>>>   ksys_ioctl+0x70/0x80
>>>>>>   __x64_sys_ioctl+0x16/0x20
>>>>>>   do_syscall_64+0x5b/0x1e0
>>>>>> The other reflink command processes are blocked at other nodes,
>>>>>> task:reflink         state:D stack:    0 pid:29759 ppid:  4088
>>>>>> Call Trace:
>>>>>>   __schedule+0x2fd/0x750
>>>>>>   schedule+0x2f/0xa0
>>>>>>   schedule_timeout+0x1cc/0x310
>>>>>>   ? ocfs2_control_cfu+0x50/0x50 [ocfs2_stack_user]
>>>>>>   ? 0xffffffffc0b19000
>>>>>>   wait_for_completion+0xba/0x140
>>>>>>   ? wake_up_q+0xa0/0xa0
>>>>>>   __ocfs2_cluster_lock.isra.41+0x3b5/0x820 [ocfs2]
>>>>>>   ? ocfs2_inode_lock_full_nested+0x1fc/0x960 [ocfs2]
>>>>>>   ocfs2_inode_lock_full_nested+0x1fc/0x960 [ocfs2]
>>>>>>   ocfs2_mv_orphaned_inode_to_new+0x87/0x7e0 [ocfs2]
>>>>>>   ocfs2_reflink+0x335/0x4c0 [ocfs2]
>>>>>>   ? ocfs2_reflink_ioctl+0x2ca/0x360 [ocfs2]
>>>>>>   ocfs2_reflink_ioctl+0x2ca/0x360 [ocfs2]
>>>>>>   ocfs2_ioctl+0x25e/0x670 [ocfs2]
>>>>>>   do_vfs_ioctl+0xa0/0x680
>>>>>>   ksys_ioctl+0x70/0x80
>>>>>>   __x64_sys_ioctl+0x16/0x20
>>>>>>   do_syscall_64+0x5b/0x1e0
>>>>>> or
>>>>>> task:reflink         state:D stack:    0 pid:18465 ppid:  4156
>>>>>> Call Trace:
>>>>>>   __schedule+0x302/0x940
>>>>>>   ? usleep_range+0x80/0x80
>>>>>>   schedule+0x46/0xb0
>>>>>>   schedule_timeout+0xff/0x140
>>>>>>   ? ocfs2_control_cfu+0x50/0x50 [ocfs2_stack_user]
>>>>>>   ? 0xffffffffc0c3b000
>>>>>>   __wait_for_common+0xb9/0x170
>>>>>>   __ocfs2_cluster_lock.constprop.0+0x1d6/0x860 [ocfs2]
>>>>>>   ? ocfs2_wait_for_recovery+0x49/0xd0 [ocfs2]
>>>>>>   ? ocfs2_inode_lock_full_nested+0x30f/0xa50 [ocfs2]
>>>>>>   ocfs2_inode_lock_full_nested+0x30f/0xa50 [ocfs2]
>>>>>>   ocfs2_inode_lock_tracker+0xf2/0x2b0 [ocfs2]
>>>>>>   ? dput+0x32/0x2f0
>>>>>>   ocfs2_permission+0x45/0xe0 [ocfs2]
>>>>>>   inode_permission+0xcc/0x170
>>>>>>   link_path_walk.part.0.constprop.0+0x2a2/0x380
>>>>>>   ? path_init+0x2c1/0x3f0
>>>>>>   path_parentat+0x3c/0x90
>>>>>>   filename_parentat+0xc1/0x1d0
>>>>>>   ? filename_lookup+0x138/0x1c0
>>>>>>   filename_create+0x43/0x160
>>>>>>   ocfs2_reflink_ioctl+0xe6/0x380 [ocfs2]
>>>>>>   ocfs2_ioctl+0x1ea/0x2c0 [ocfs2]
>>>>>>   ? do_sys_openat2+0x81/0x150
>>>>>>   __x64_sys_ioctl+0x82/0xb0
>>>>>>   do_syscall_64+0x61/0xb0
>>>>>>
>>>>>> The deadlock is caused by multiple acquiring the destination directory
>>>>>> inode dlm lock in ocfs2_reflink function, we should acquire this
>>>>>> directory inode dlm lock at the beginning, and hold this dlm lock until
>>>>>> end of the function.
>>>>>>
>>>>>> Signed-off-by: Gang He <ghe@...e.com>
>>>>>> ---
>>>>>> fs/ocfs2/namei.c        | 32 +++++++++++++-------------------
>>>>>> fs/ocfs2/namei.h        |  2 ++
>>>>>> fs/ocfs2/refcounttree.c | 15 +++++++++++----
>>>>>> fs/ocfs2/xattr.c        | 12 +-----------
>>>>>> fs/ocfs2/xattr.h        |  1 +
>>>>>> 5 files changed, 28 insertions(+), 34 deletions(-)
>>>>>>
>>>>>> diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
>>>>>> index 2c46ff6ba4ea..f8bbb22cc60b 100644
>>>>>> --- a/fs/ocfs2/namei.c
>>>>>> +++ b/fs/ocfs2/namei.c
>>>>>> @@ -2489,6 +2489,7 @@ static int ocfs2_prep_new_orphaned_file(struct inode *dir,
>>>>>> }
>>>>>>
>>>>>> int ocfs2_create_inode_in_orphan(struct inode *dir,
>>>>>> +				 struct buffer_head **dir_bh,
>>>>>> 				 int mode,
>>>>>> 				 struct inode **new_inode)
>>>>>> {
>>>>>> @@ -2597,13 +2598,16 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
>>>>>>
>>>>>> 	brelse(new_di_bh);
>>>>>>
>>>>>> -	if (!status)
>>>>>> -		*new_inode = inode;
>>>>>> -
>>>>>> 	ocfs2_free_dir_lookup_result(&orphan_insert);
>>>>>>
>>>>>> -	ocfs2_inode_unlock(dir, 1);
>>>>>> -	brelse(parent_di_bh);
>>>>>> +	if (!status) {
>>>>>> +		*new_inode = inode;
>>>>>> +		*dir_bh = parent_di_bh;
>>>>>> +	} else {
>>>>>> +		ocfs2_inode_unlock(dir, 1);
>>>>>> +		brelse(parent_di_bh);
>>>>>> +	}
>>>>>> +
>>>>>> 	return status;
>>>>>> }
>>>>>>
>>>>>> @@ -2760,11 +2764,11 @@ int ocfs2_del_inode_from_orphan(struct ocfs2_super *osb,
>>>>>> }
>>>>>>
>>>>>> int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
>>>>>> +				   struct buffer_head *dir_bh,
>>>>>> 				   struct inode *inode,
>>>>>> 				   struct dentry *dentry)
>>>>>> {
>>>>>> 	int status = 0;
>>>>>> -	struct buffer_head *parent_di_bh = NULL;
>>>>>> 	handle_t *handle = NULL;
>>>>>> 	struct ocfs2_super *osb = OCFS2_SB(dir->i_sb);
>>>>>> 	struct ocfs2_dinode *dir_di, *di;
>>>>>> @@ -2778,14 +2782,7 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
>>>>>> 				(unsigned long long)OCFS2_I(dir)->ip_blkno,
>>>>>> 				(unsigned long long)OCFS2_I(inode)->ip_blkno);
>>>>>>
>>>>>> -	status = ocfs2_inode_lock(dir, &parent_di_bh, 1);
>>>>>> -	if (status < 0) {
>>>>>> -		if (status != -ENOENT)
>>>>>> -			mlog_errno(status);
>>>>>> -		return status;
>>>>>> -	}
>>>>>> -
>>>>>> -	dir_di = (struct ocfs2_dinode *) parent_di_bh->b_data;
>>>>>> +	dir_di = (struct ocfs2_dinode *) dir_bh->b_data;
>>>>>> 	if (!dir_di->i_links_count) {
>>>>>> 		/* can't make a file in a deleted directory. */
>>>>>> 		status = -ENOENT;
>>>>>> @@ -2798,7 +2795,7 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
>>>>>> 		goto leave;
>>>>>>
>>>>>> 	/* get a spot inside the dir. */
>>>>>> -	status = ocfs2_prepare_dir_for_insert(osb, dir, parent_di_bh,
>>>>>> +	status = ocfs2_prepare_dir_for_insert(osb, dir, dir_bh,
>>>>>> 					      dentry->d_name.name,
>>>>>> 					      dentry->d_name.len, &lookup);
>>>>>> 	if (status < 0) {
>>>>>> @@ -2862,7 +2859,7 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
>>>>>> 	ocfs2_journal_dirty(handle, di_bh);
>>>>>>
>>>>>> 	status = ocfs2_add_entry(handle, dentry, inode,
>>>>>> -				 OCFS2_I(inode)->ip_blkno, parent_di_bh,
>>>>>> +				 OCFS2_I(inode)->ip_blkno, dir_bh,
>>>>>> 				 &lookup);
>>>>>> 	if (status < 0) {
>>>>>> 		mlog_errno(status);
>>>>>> @@ -2886,10 +2883,7 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
>>>>>> 	iput(orphan_dir_inode);
>>>>>> leave:
>>>>>>
>>>>>> -	ocfs2_inode_unlock(dir, 1);
>>>>>> -
>>>>>> 	brelse(di_bh);
>>>>>> -	brelse(parent_di_bh);
>>>>>> 	brelse(orphan_dir_bh);
>>>>>>
>>>>>> 	ocfs2_free_dir_lookup_result(&lookup);
>>>>>> diff --git a/fs/ocfs2/namei.h b/fs/ocfs2/namei.h
>>>>>> index 9cc891eb874e..03a2c526e2c1 100644
>>>>>> --- a/fs/ocfs2/namei.h
>>>>>> +++ b/fs/ocfs2/namei.h
>>>>>> @@ -24,6 +24,7 @@ int ocfs2_orphan_del(struct ocfs2_super *osb,
>>>>>> 		     struct buffer_head *orphan_dir_bh,
>>>>>> 		     bool dio);
>>>>>> int ocfs2_create_inode_in_orphan(struct inode *dir,
>>>>>> +				 struct buffer_head **dir_bh,
>>>>>> 				 int mode,
>>>>>> 				 struct inode **new_inode);
>>>>>> int ocfs2_add_inode_to_orphan(struct ocfs2_super *osb,
>>>>>> @@ -32,6 +33,7 @@ int ocfs2_del_inode_from_orphan(struct ocfs2_super *osb,
>>>>>> 		struct inode *inode, struct buffer_head *di_bh,
>>>>>> 		int update_isize, loff_t end);
>>>>>> int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
>>>>>> +				   struct buffer_head *dir_bh,
>>>>>> 				   struct inode *new_inode,
>>>>>> 				   struct dentry *new_dentry);
>>>>>>
>>>>>> diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
>>>>>> index 7f6355cbb587..a9a0c7c37e8e 100644
>>>>>> --- a/fs/ocfs2/refcounttree.c
>>>>>> +++ b/fs/ocfs2/refcounttree.c
>>>>>> @@ -4250,7 +4250,7 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir,
>>>>>> {
>>>>>> 	int error, had_lock;
>>>>>> 	struct inode *inode = d_inode(old_dentry);
>>>>>> -	struct buffer_head *old_bh = NULL;
>>>>>> +	struct buffer_head *old_bh = NULL, *dir_bh = NULL;
>>>>>> 	struct inode *new_orphan_inode = NULL;
>>>>>> 	struct ocfs2_lock_holder oh;
>>>>>>
>>>>>> @@ -4258,7 +4258,7 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir,
>>>>>> 		return -EOPNOTSUPP;
>>>>>>
>>>>>>
>>>>>> -	error = ocfs2_create_inode_in_orphan(dir, inode->i_mode,
>>>>>> +	error = ocfs2_create_inode_in_orphan(dir, &dir_bh, inode->i_mode,
>>>>>> 					     &new_orphan_inode);
>>>>>> 	if (error) {
>>>>>> 		mlog_errno(error);
>>>>>> @@ -4304,13 +4304,15 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir,
>>>>>>
>>>>>> 	/* If the security isn't preserved, we need to re-initialize them. */
>>>>>> 	if (!preserve) {
>>>>>> -		error = ocfs2_init_security_and_acl(dir, new_orphan_inode,
>>>>>> +		error = ocfs2_init_security_and_acl(dir, dir_bh,
>>>>>> +						    new_orphan_inode,
>>>>>> 						    &new_dentry->d_name);
>>>>>> 		if (error)
>>>>>> 			mlog_errno(error);
>>>>>> 	}
>>>>>> 	if (!error) {
>>>>>> -		error = ocfs2_mv_orphaned_inode_to_new(dir, new_orphan_inode,
>>>>>> +		error = ocfs2_mv_orphaned_inode_to_new(dir, dir_bh,
>>>>>> +						       new_orphan_inode,
>>>>>> 						       new_dentry);
>>>>>> 		if (error)
>>>>>> 			mlog_errno(error);
>>>>>> @@ -4328,6 +4330,11 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir,
>>>>>> 			iput(new_orphan_inode);
>>>>>> 	}
>>>>>>
>>>>>> +	if (dir_bh) {
>>>>>> +		ocfs2_inode_unlock(dir, 1);
>>>>>> +		brelse(dir_bh);
>>>>>> +	}
>>>>>> +
>>>>>> 	return error;
>>>>>> }
>>>>>>
>>>>>> diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
>>>>>> index dd784eb0cd7c..3f23e3a5018c 100644
>>>>>> --- a/fs/ocfs2/xattr.c
>>>>>> +++ b/fs/ocfs2/xattr.c
>>>>>> @@ -7203,16 +7203,13 @@ int ocfs2_reflink_xattrs(struct inode *old_inode,
>>>>>> /*
>>>>>>   * Initialize security and acl for a already created inode.
>>>>>>   * Used for reflink a non-preserve-security file.
>>>>>> - *
>>>>>> - * It uses common api like ocfs2_xattr_set, so the caller
>>>>>> - * must not hold any lock expect i_mutex.
>>>>>>   */
>>>>>> int ocfs2_init_security_and_acl(struct inode *dir,
>>>>>> +				struct buffer_head *dir_bh,
>>>>>> 				struct inode *inode,
>>>>>> 				const struct qstr *qstr)
>>>>>> {
>>>>>> 	int ret = 0;
>>>>>> -	struct buffer_head *dir_bh = NULL;
>>>>>>
>>>>>> 	ret = ocfs2_init_security_get(inode, dir, qstr, NULL);
>>>>>> 	if (ret) {
>>>>>> @@ -7220,17 +7217,10 @@ int ocfs2_init_security_and_acl(struct inode *dir,
>>>>>> 		goto leave;
>>>>>> 	}
>>>>>>
>>>>>> -	ret = ocfs2_inode_lock(dir, &dir_bh, 0);
>>>>>> -	if (ret) {
>>>>>> -		mlog_errno(ret);
>>>>>> -		goto leave;
>>>>>> -	}
>>>>>> 	ret = ocfs2_init_acl(NULL, inode, dir, NULL, dir_bh, NULL, NULL);
>>>>>> 	if (ret)
>>>>>> 		mlog_errno(ret);
>>>>>>
>>>>>> -	ocfs2_inode_unlock(dir, 0);
>>>>>> -	brelse(dir_bh);
>>>>>> leave:
>>>>>> 	return ret;
>>>>>> }
>>>>>> diff --git a/fs/ocfs2/xattr.h b/fs/ocfs2/xattr.h
>>>>>> index 00308b57f64f..b27fd8ba0019 100644
>>>>>> --- a/fs/ocfs2/xattr.h
>>>>>> +++ b/fs/ocfs2/xattr.h
>>>>>> @@ -83,6 +83,7 @@ int ocfs2_reflink_xattrs(struct inode *old_inode,
>>>>>> 			 struct buffer_head *new_bh,
>>>>>> 			 bool preserve_security);
>>>>>> int ocfs2_init_security_and_acl(struct inode *dir,
>>>>>> +				struct buffer_head *dir_bh,
>>>>>> 				struct inode *inode,
>>>>>> 				const struct qstr *qstr);
>>>>>> #endif /* OCFS2_XATTR_H */
>>>>>> -- 
>>>>>> 2.21.0
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Ocfs2-devel mailing list
>>>>>> Ocfs2-devel@....oracle.com
>>>>>> https://oss.oracle.com/mailman/listinfo/ocfs2-devel
>>>>
>>
> 

Powered by blists - more mailing lists