Skip to contentSkip to conversation

Possible-Patch: unregister bdi on error path.

PATCH Neil Brown 2 participants 4 messages Mar 10, 2010 – Mar 11, 2010 3 messages not in the archive

Subsystems (2)
lore
Share & tools
lore

Neil Brown

author
#0 lore raw

Hi,
I've had an internal report of the following error:

[145405.949506] WARNING: at /usr/src/packages/BUILD/kernel-default-2.6.32.8/linux-2.6.32/fs/sysfs/dir.c:491 sysfs_add_one+0xf4/0x120()
[145405.949516] sysfs: cannot create duplicate filename '/devices/virtual/bdi/0:26'
...
[145405.949598] Pid: 8745, comm: mount.nfs Not tainted 2.6.32.8-0.3-default #1
[145405.949601] Call Trace:
...
[145405.949644]  [<ffffffff811877c4>] sysfs_add_one+0xf4/0x120
[145405.949650]  [<ffffffff811883c0>] create_dir+0x60/0xb0
[145405.949656]  [<ffffffff81188441>] sysfs_create_dir+0x31/0x50
[145405.949663]  [<ffffffff8120a8ef>] kobject_add_internal+0xff/0x260
[145405.949669]  [<ffffffff8120ac56>] kobject_add+0x46/0x70
[145405.949675]  [<ffffffff812c06b6>] device_add+0xc6/0x4e0
[145405.949682]  [<ffffffff812c0c2f>] device_create_vargs+0x13f/0x150
[145405.949689]  [<ffffffff810eceea>] bdi_register+0x7a/0x180
[145405.949707]  [<ffffffffa04fb006>] nfs_get_sb+0x316/0x430 [nfs]
[145405.949735]  [<ffffffff8112195d>] vfs_kern_mount+0x7d/0x1b0
[145405.949741]  [<ffffffff81121af3>] do_kern_mount+0x53/0x120
[145405.949748]  [<ffffffff8113b7dc>] do_mount+0x21c/0x250
[145405.949754]  [<ffffffff8113b8d0>] sys_mount+0xc0/0xf0

i.e. it tried to bdi_register 0:26 while it was already registered.

Looking through the code, it seems that if nfs_get_sb calls nfs_bdi_register, but then doesn't set ->s_root, then ->put_super is not called (generic_shutdown_super only calls ->put_super if s_root is set), so the bdi is never unregistered. The next mount would then trigger the above error.

I haven't found a way to trigger this situation, but it seems like it should be possible.

Does this patch make sense?

Thanks,
NeilBrown

----------------------------
NFS: ensure bdi_unregister is called on mount failure.

bdi_unregister is called by nfs_put_super which is only called by generic_shutdown_super if ->s_root is not NULL. So if we error out in a circumstance where we called nfs_bdi_register (i.e. server != NULL) but have not set s_root, then we need to call bdi_unregister explicitly in nfs_get_sb.

fs/nfs/super.c +4 -1
index f1afee4..da11ec9 100644
@@ -2214,7 +2214,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
 	} else {
 		error = nfs_bdi_register(server);
 		if (error)
-			goto error_splat_super;
+			goto error_splat_bdi;
 	}
 
 	if (!s->s_root) {
@@ -2256,6 +2256,9 @@ out_err_nosb:
 error_splat_root:
 	dput(mntroot);
 error_splat_super:
+	if (server && !s->s_root)
+		bdi_unregister(&server->backing_dev_info);
+error_splat_bdi:
 	deactivate_locked_super(s);
 	goto out;
 }

A message that is not in the archive (sent off-list or not yet received): [email protected]

#2 lore raw
Quoted 68 lines · Neil Brown

Hi,
I've had an internal report of the following error:

[145405.949506] WARNING: at /usr/src/packages/BUILD/kernel-default-2.6.32.8/linux-2.6.32/fs/sysfs/dir.c:491 sysfs_add_one+0xf4/0x120()
[145405.949516] sysfs: cannot create duplicate filename '/devices/virtual/bdi/0:26'
...
[145405.949598] Pid: 8745, comm: mount.nfs Not tainted 2.6.32.8-0.3-default #1
[145405.949601] Call Trace:
...
[145405.949644]  [<ffffffff811877c4>] sysfs_add_one+0xf4/0x120
[145405.949650]  [<ffffffff811883c0>] create_dir+0x60/0xb0
[145405.949656]  [<ffffffff81188441>] sysfs_create_dir+0x31/0x50
[145405.949663]  [<ffffffff8120a8ef>] kobject_add_internal+0xff/0x260
[145405.949669]  [<ffffffff8120ac56>] kobject_add+0x46/0x70
[145405.949675]  [<ffffffff812c06b6>] device_add+0xc6/0x4e0
[145405.949682]  [<ffffffff812c0c2f>] device_create_vargs+0x13f/0x150
[145405.949689]  [<ffffffff810eceea>] bdi_register+0x7a/0x180
[145405.949707]  [<ffffffffa04fb006>] nfs_get_sb+0x316/0x430 [nfs]
[145405.949735]  [<ffffffff8112195d>] vfs_kern_mount+0x7d/0x1b0
[145405.949741]  [<ffffffff81121af3>] do_kern_mount+0x53/0x120
[145405.949748]  [<ffffffff8113b7dc>] do_mount+0x21c/0x250
[145405.949754]  [<ffffffff8113b8d0>] sys_mount+0xc0/0xf0

i.e. it tried to bdi_register 0:26 while it was already registered.

Looking through the code, it seems that if nfs_get_sb calls nfs_bdi_register, but then doesn't set ->s_root, then ->put_super is not called (generic_shutdown_super only calls ->put_super if s_root is set), so the bdi is never unregistered. The next mount would then trigger the above error.

I haven't found a way to trigger this situation, but it seems like it should be possible.

Does this patch make sense?

Thanks,
NeilBrown

----------------------------
NFS: ensure bdi_unregister is called on mount failure.

bdi_unregister is called by nfs_put_super which is only called by generic_shutdown_super if ->s_root is not NULL. So if we error out in a circumstance where we called nfs_bdi_register (i.e. server != NULL) but have not set s_root, then we need to call bdi_unregister explicitly in nfs_get_sb.

fs/nfs/super.c +1 -1
index f1afee4..da11ec9 100644
@@ -2214,7 +2214,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
 	} else {
 		error = nfs_bdi_register(server);
 		if (error)
-			goto error_splat_super;
+			goto error_splat_bdi;
 	}
 
 	if (!s->s_root) {

fs/nfs/super.c out_err_nosb: · line 2264

Quoted 9 lines
fs/nfs/super.c +3 -0
@@ -2256,6 +2256,9 @@ out_err_nosb:
 error_splat_root:
 	dput(mntroot);
 error_splat_super:
+	if (server && !s->s_root)
+		bdi_unregister(&server->backing_dev_info);
+error_splat_bdi:
 	deactivate_locked_super(s);
 	goto out;
 }

Yes. It looks as if you have indeed found a bug.

My main comment is that in order to be complete the patch should also address the cases of nfs_xdev_get_sb(), nfs4_remote_get_sb(), nfs4_xdev_get_sb(), and nfs4_remote_referral_get_sb(). They appear to suffer from the same problem...

Cheers
Trond

A message that is not in the archive (sent off-list or not yet received): 1268262107.3096.207.camel-bi+AKbBUZKY6gyzm1THtWbp2dZbC/[email protected]

author
#4 lore raw
Quoted 7 lines · On Wed, 10 Mar 2010 18:01:47 -0500 Trond Myklebust…

Yes. It looks as if you have indeed found a bug.

My main comment is that in order to be complete the patch should also address the cases of nfs_xdev_get_sb(), nfs4_remote_get_sb(), nfs4_xdev_get_sb(), and nfs4_remote_referral_get_sb(). They appear to suffer from the same problem...

Good point. Patch follows.
I have compile-tested this but nothing more. I'm not even sure how I can test it as it only affect error paths that are not easy to force.

Thanks,
NeilBrown

Quoted 1 line

From 8db673fbc28ea018ab0b80c3b1f53bad0d684251 Mon Sep 17 00:00:00 2001

From: NeilBrown <[email protected]>
Date: Thu, 11 Mar 2010 11:20:17 +1100
Subject: [PATCH] NFS: ensure bdi_unregister is called on mount failure.

bdi_unregister is called by nfs_put_super which is only called by generic_shutdown_super if ->s_root is not NULL. So if we error out in a circumstance where we called nfs_bdi_register (i.e. server != NULL) but have not set s_root, then we need to call bdi_unregister explicitly in nfs_get_sb and various other *_get_sb() functions.

fs/nfs/super.c +20 -5
index f1afee4..6baf9a3 100644
@@ -2214,7 +2214,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
 	} else {
 		error = nfs_bdi_register(server);
 		if (error)
-			goto error_splat_super;
+			goto error_splat_bdi;
 	}
 
 	if (!s->s_root) {
@@ -2256,6 +2256,9 @@ out_err_nosb:
 error_splat_root:
 	dput(mntroot);
 error_splat_super:
+	if (server && !s->s_root)
+		bdi_unregister(&server->backing_dev_info);
+error_splat_bdi:
 	deactivate_locked_super(s);
 	goto out;
 }
@@ -2326,7 +2329,7 @@ static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags,
 	} else {
 		error = nfs_bdi_register(server);
 		if (error)
-			goto error_splat_super;
+			goto error_splat_bdi;
 	}
 
 	if (!s->s_root) {
@@ -2363,6 +2366,9 @@ out_err_noserver:
 	return error;
 
 error_splat_super:
+	if (server && !s->s_root)
+		bdi_unregister(&server->backing_dev_info);
+error_splat_bdi:
 	deactivate_locked_super(s);
 	dprintk("<-- nfs_xdev_get_sb() = %d [splat]\n", error);
 	return error;
@@ -2578,7 +2584,7 @@ static int nfs4_remote_get_sb(struct file_system_type *fs_type,
 	} else {
 		error = nfs_bdi_register(server);
 		if (error)
-			goto error_splat_super;
+			goto error_splat_bdi;
 	}
 
 	if (!s->s_root) {
@@ -2616,6 +2622,9 @@ out_free:
 error_splat_root:
 	dput(mntroot);
 error_splat_super:
+	if (server && !s->s_root)
+		bdi_unregister(&server->backing_dev_info);
+error_splat_bdi:
 	deactivate_locked_super(s);
 	goto out;
 }
@@ -2811,7 +2820,7 @@ static int nfs4_xdev_get_sb(struct file_system_type *fs_type, int flags,
 	} else {
 		error = nfs_bdi_register(server);
 		if (error)
-			goto error_splat_super;
+			goto error_splat_bdi;
 	}
 
 	if (!s->s_root) {
@@ -2847,6 +2856,9 @@ out_err_noserver:
 	return error;
 
 error_splat_super:
+	if (server && !s->s_root)
+		bdi_unregister(&server->backing_dev_info);
+error_splat_bdi:
 	deactivate_locked_super(s);
 	dprintk("<-- nfs4_xdev_get_sb() = %d [splat]\n", error);
 	return error;
@@ -2893,7 +2905,7 @@ static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type,
 	} else {
 		error = nfs_bdi_register(server);
 		if (error)
-			goto error_splat_super;
+			goto error_splat_bdi;
 	}
 
 	if (!s->s_root) {
@@ -2929,6 +2941,9 @@ out_err_noserver:
 	return error;
 
 error_splat_super:
+	if (server && !s->s_root)
+		bdi_unregister(&server->backing_dev_info);
+error_splat_bdi:
 	deactivate_locked_super(s);
 	dprintk("<-- nfs4_referral_get_sb() = %d [splat]\n", error);
 	return error;

A message that is not in the archive (sent off-list or not yet received): [email protected]

#6 lore raw
Quoted 11 lines · Neil Brown
Quoted 7 lines · On Wed, 10 Mar 2010 18:01:47 -0500 Trond Myklebust…

Yes. It looks as if you have indeed found a bug.

My main comment is that in order to be complete the patch should also address the cases of nfs_xdev_get_sb(), nfs4_remote_get_sb(), nfs4_xdev_get_sb(), and nfs4_remote_referral_get_sb(). They appear to suffer from the same problem...

Quoted 3 lines

Good point. Patch follows.
I have compile-tested this but nothing more. I'm not even sure how I can test it as it only affect error paths that are not easy to force.

Agreed that it is faith based patching :-) but I'm willing to go with that unless someone can show me that your argument is incorrect.

Will apply...

Cheers
Trond