Skip to contentSkip to conversation

afs: Miscellaneous fixes

PATCH Superseded David Howells 3 participants 11 messages Jun 9

Subsystems (3)

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore

David Howells

author maintainer
#0 lore raw

cover afs: Miscellaneous fixes

Hi Christian,

Can you pick up these miscellaneous AFS fixes, please?

(1) Fix the CB.InitCallBackState3 service handler to handle an unknown server (server pointer is NULL).

(2) Fix the clobbering of the default error code in

     afs_extract_vl_addrs().

(3) Fix a NULL pointer in a trace point in afs_get_tree().

(4) Fix setting of AS_RELEASE_ALWAYS for symlinks (and mountpoints) as there's no release_folio function provided.

(5) Fix the release of op->more_files to Use kvfree().

(6) Fix erroneous seq |= 1 in volume lookup loop.

(7) Drop for duplicate server records when parsing DNS reply into the VL server list (this is not strictly a bug fix, so could be punted to the merge window).

The patches can be found here:

https://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git/log/?h=afs-fixes

Thanks,
David

Dan Carpenter (1):
afs: Fix error code in afs_extract_vl_addrs()

David Howells (1):
afs: Remove setting of AS_RELEASE_ALWAYS for symlinks and mountpoints

Li RongQing (1):
afs: Remove erroneous seq |= 1 in volume lookup loop

Matvey Kovalev (1):
afs: fix NULL pointer dereference in afs_get_tree()

Nan Li (1):
afs: handle CB.InitCallBackState3 requests without a server record

Yuto Ohnuki (1):
afs: check for duplicate servers in VL server list

Zilin Guan (1):
afs: use kvfree() to free memory allocated by kvcalloc()

 fs/afs/callback.c     |  1 -
 fs/afs/cmservice.c    |  5 +++++
 fs/afs/fs_operation.c |  2 +-
 fs/afs/inode.c        |  1 -
 fs/afs/super.c        |  3 ++-
 fs/afs/vl_list.c      | 18 ++++++++++++++++--
 6 files changed, 24 insertions(+), 6 deletions(-)
author maintainer
#1 lore raw

PATCH 1/7 afs: handle CB.InitCallBackState3 requests without a server record

From: Nan Li <[email protected]>

The cache manager callback path now attaches the server record to an incoming call through the rxrpc peer's app data. That association is not guaranteed to exist for every callback request, and most callback handlers already tolerate that case.

Make CB.InitCallBackState3 follow the same pattern by checking whether a server record was attached before using it. If the peer is not mapped to a server record, trace the request and ignore it, matching the existing behaviour for other unmatched callback requests.

This keeps the callback handler consistent with the rest of the cache manager service and avoids depending on peer state that may not be available for a given request.

Fixes 40e8b52fe8c8 ("afs: Use the per-peer app data provided by rxrpc")
cc Marc Dionne <[email protected]>
---
 fs/afs/cmservice.c | 5 +++++
 1 file changed, 5 insertions(+)
fs/afs/cmservice.c +5 -0
index 5540ae1cad59..263c60c811a5 100644
@@ -364,6 +364,11 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call)
 	if (!afs_check_call_state(call, AFS_CALL_SV_REPLYING))
 		return afs_io_error(call, afs_io_error_cm_reply);
 
+	if (!call->server) {
+		trace_afs_cm_no_server_u(call, call->request);
+		return 0;
+	}
+
 	if (memcmp(call->request, &call->server->_uuid, sizeof(call->server->_uuid)) != 0) {
 		pr_notice("Callback UUID does not match fileserver UUID\n");
 		trace_afs_cm_no_server_u(call, call->request);
author maintainer
#2 lore raw

PATCH 2/7 afs: Fix error code in afs_extract_vl_addrs()

In reply to David Howells

From: Dan Carpenter <[email protected]>

The error codes on these paths are only set on the first iteration through the loop. Set the correct error code on every iteration.

Fixes 0a5143f2f89c ("afs: Implement VL server rotation")
cc Marc Dionne <[email protected]>
---
 fs/afs/vl_list.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
fs/afs/vl_list.c +3 -1
index 3e4966915ea4..003889cf0f18 100644
@@ -92,7 +92,7 @@ static struct afs_addr_list *afs_extract_vl_addrs(struct afs_net *net,
 {
 	struct afs_addr_list *alist;
 	const u8 *b = *_b;
-	int ret = -EINVAL;
+	int ret;
 
 	alist = afs_alloc_addrlist(nr_addrs);
 	if (!alist)
@@ -110,6 +110,7 @@ static struct afs_addr_list *afs_extract_vl_addrs(struct afs_net *net,
 		case DNS_ADDRESS_IS_IPV4:
 			if (end - b < 4) {
 				_leave(" = -EINVAL [short inet]");
+				ret = -EINVAL;
 				goto error;
 			}
 			memcpy(x, b, 4);
@@ -122,6 +123,7 @@ static struct afs_addr_list *afs_extract_vl_addrs(struct afs_net *net,
 		case DNS_ADDRESS_IS_IPV6:
 			if (end - b < 16) {
 				_leave(" = -EINVAL [short inet6]");
+				ret = -EINVAL;
 				goto error;
 			}
 			memcpy(x, b, 16);
author maintainer
#3 lore raw

PATCH 3/7 afs: fix NULL pointer dereference in afs_get_tree()

In reply to David Howells

From: Matvey Kovalev <[email protected]>

afs_alloc_sbi() uses kzalloc for memory allocation. And, if ctx->dyn_root is not null, as->cell and as->volume are null. In trace_afs_get_tree() they are dereferenced.

KASAN error message:

KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
CPU: 2 PID: 18478 Comm: syz-executor.7 Not tainted 5.10.246-syzkaller #0
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1
04/01/2014
RIP: 0010:perf_trace_afs_get_tree+0x1d9/0x550

include/trace/events/afs.h:1365

Call Trace:
trace_afs_get_tree include/trace/events/afs.h:1365 [inline]
afs_get_tree+0x922/0x1350 fs/afs/super.c:599
vfs_get_tree+0x8e/0x300 fs/super.c:1572
do_new_mount fs/namespace.c:3011 [inline]
path_mount+0x14a5/0x2220 fs/namespace.c:3341
do_mount fs/namespace.c:3354 [inline]
__do_sys_mount fs/namespace.c:3562 [inline]
__se_sys_mount fs/namespace.c:3539 [inline]
__x64_sys_mount+0x283/0x300 fs/namespace.c:3539
 do_syscall_64+0x33/0x50 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x67/0xd1

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Fixes 80548b03991f5 ("afs: Add more tracepoints")
cc Marc Dionne <[email protected]>
---
 fs/afs/super.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/afs/super.c +2 -1
index 942f3e9800d7..dec091e569c4 100644
@@ -587,7 +587,8 @@ static int afs_get_tree(struct fs_context *fc)
 	}
 
 	fc->root = dget(sb->s_root);
-	trace_afs_get_tree(as->cell, as->volume);
+	if (!ctx->dyn_root)
+		trace_afs_get_tree(as->cell, as->volume);
 	_leave(" = 0 [%p]", sb);
 	return 0;
 
author maintainer
#4 lore raw

PATCH 4/7 afs: Remove setting of AS_RELEASE_ALWAYS for symlinks and mountpoints

In reply to David Howells

Regular AFS files correctly use afs_file_aops which have release_folio set as netfs_release_folio, so AS_RELEASE_ALWAYS is valid for them when fscache is enabled (set via afs_vnode_set_cache()). Symlinks and mountpoints in AFS use afs_dir_aops, which does not provide a release_folio callback. However, afs_apply_status() unconditionally calls mapping_set_release_always() for these.

In such case when memory management code attempts to release folios, filemap_release_folio() checks folio_needs_release() which returns true due to AS_RELEASE_ALWAYS being set. Since there is no release_folio callback, it falls through to try_to_free_buffers(), which at present expects buffer_heads to be not null. For symlinks and mountpoints without buffer_heads, this causes pointer dereference.

Fixes eae9e78951bb ("afs: Use netfslib for symlinks, allowing them to be cached")
cc Marc Dionne <[email protected]>
---
 fs/afs/inode.c | 1 -
 1 file changed, 1 deletion(-)
fs/afs/inode.c +0 -1
index 3f48458694ba..21ac098d03ef 100644
@@ -126,7 +126,6 @@ static int afs_inode_init_from_status(struct afs_operation *op,
 		}
 		inode->i_mapping->a_ops	= &afs_symlink_aops;
 		inode_nohighmem(inode);
-		mapping_set_release_always(inode->i_mapping);
 		break;
 	default:
 		dump_vnode(vnode, op->file[0].vnode != vnode ? op->file[0].vnode : NULL);
author maintainer
#5 lore raw

PATCH 5/7 afs: use kvfree() to free memory allocated by kvcalloc()

In reply to David Howells

From: Zilin Guan <[email protected]>

op->more_files is allocated with kvcalloc() but released via afs_put_operation(), which uses kfree() internally. This mismach prevents the resource from being released properly and may lead to undefined behavior.

Fix this by using kvfree() to free op->more_files to match its allocation method.

Fixes e49c7b2f6de7 ("afs: Build an abstraction around an "operation" concept")
cc Marc Dionne <[email protected]>
---
 fs/afs/fs_operation.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/afs/fs_operation.c +1 -1
index c0dbbc6d3716..20801b29521d 100644
@@ -348,7 +348,7 @@ int afs_put_operation(struct afs_operation *op)
 		for (i = 0; i < op->nr_files - 2; i++)
 			if (op->more_files[i].put_vnode)
 				iput(&op->more_files[i].vnode->netfs.inode);
-		kfree(op->more_files);
+		kvfree(op->more_files);
 	}
 
 	if (op->estate) {
author maintainer
#6 lore raw

PATCH 6/7 afs: Remove erroneous seq |= 1 in volume lookup loop

In reply to David Howells

From: Li RongQing <[email protected]>

The seq |= 1 operation in the volume lookup loop is incorrect because:
seq is already incremented at start, making it odd in next iteration which triggers lock, but The |= 1 operation causes seq to be even and unintended lockless operation

Remove this erroneous operation to maintain proper lock sequencing.

Fixes 32222f09782f ("afs: Apply server breaks to mmap'd files in the call processor")
cc Marc Dionne <[email protected]>
---
 fs/afs/callback.c | 1 -
 1 file changed, 1 deletion(-)
fs/afs/callback.c +0 -1
index 894d2bad6b6c..833ac3178ddc 100644
@@ -140,7 +140,6 @@ static struct afs_volume *afs_lookup_volume_rcu(struct afs_cell *cell,
 			break;
 		if (!need_seqretry(&cell->volume_lock, seq))
 			break;
-		seq |= 1; /* Want a lock next time */
 	}
 
 	done_seqretry(&cell->volume_lock, seq);
#7 lore raw

Re: [PATCH 6/7] afs: Remove erroneous seq |= 1 in volume lookup loop

Quoted 25 lines · David Howells

From: Li RongQing <[email protected]>

The seq |= 1 operation in the volume lookup loop is incorrect because:
seq is already incremented at start, making it odd in next iteration which triggers lock, but The |= 1 operation causes seq to be even and unintended lockless operation

Remove this erroneous operation to maintain proper lock sequencing.

Fixes 32222f09782f ("afs: Apply server breaks to mmap'd files in the call processor")
cc Marc Dionne <[email protected]>
---
 fs/afs/callback.c | 1 -
 1 file changed, 1 deletion(-)
fs/afs/callback.c +0 -0
index 894d2bad6b6c..833ac3178ddc 100644

fs/afs/callback.c static struct afs_volume *afs_lookup_volume_rcu(struct afs_cell *cell, · line 143

Quoted 4 lines
fs/afs/callback.c +0 -1
@@ -140,7 +140,6 @@ static struct afs_volume *afs_lookup_volume_rcu(struct afs_cell *cell,
 			break;
 		if (!need_seqretry(&cell->volume_lock, seq))
 			break;
-		seq |= 1; /* Want a lock next time */

Yes, but now that we have scoped_seqlock_read() we can make another change on top of this fix. afs_lookup_volume_rcu() can just do

	static struct afs_volume *afs_lookup_volume_rcu(struct afs_cell *cell,
							afs_volid_t vid)
	{
		struct afs_volume *volume = NULL;
		struct rb_node *p;

		scoped_seqlock_read (&cell->volume_lock, ss_lock) {
			/* Unfortunately, rbtree walking doesn't give reliable results
			 * under just the RCU read lock, so we have to check for
			 * changes.
			 */
			p = rcu_dereference_raw(cell->volumes.rb_node);
			while (p) {
				volume = rb_entry(p, struct afs_volume, cell_node);

				if (volume->vid < vid)
					p = rcu_dereference_raw(p->rb_left);
				else if (volume->vid > vid)
					p = rcu_dereference_raw(p->rb_right);
				else
					break;
				volume = NULL;
			}

			if (volume && afs_try_get_volume(volume, afs_volume_trace_get_callback))
				break;
		}

		return volume;
	}

Oleg.

author maintainer
#8 lore raw

Re: [PATCH 6/7] afs: Remove erroneous seq |= 1 in volume lookup loop

Quoted 2 lines · Oleg Nesterov

Yes, but now that we have scoped_seqlock_read() we can make another change on top of this fix. afs_lookup_volume_rcu() can just do

I wonder if that would be better as a separate patch rather than merged into this one.

David

#9 lore raw

Re: [PATCH 6/7] afs: Remove erroneous seq |= 1 in volume lookup loop

Quoted 8 lines · David Howells
Quoted 2 lines · Oleg Nesterov

Yes, but now that we have scoped_seqlock_read() we can make another change on top of this fix. afs_lookup_volume_rcu() can just do

I wonder if that would be better as a separate patch rather than merged into this one.

David

IMO, a separate change would be preferred so that this bug fix patch can be back-ported to stable where scoped_seqlock_read() is not present.

Jeffrey

author maintainer
#10 lore raw

PATCH 7/7 afs: check for duplicate servers in VL server list

In reply to David Howells

From: Yuto Ohnuki <[email protected]>

The DNS response may contain the same server more than once. Check for duplicates by name and port before inserting into the list to avoid duplicate entries.

Addresses the TODO comment in afs_extract_vlserver_list().

---
 fs/afs/vl_list.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
fs/afs/vl_list.c +13 -1
index 003889cf0f18..8e1cf6cdcf71 100644
@@ -289,8 +289,20 @@ struct afs_vlserver_list *afs_extract_vlserver_list(struct afs_cell *cell,
 			afs_put_addrlist(old, afs_alist_trace_put_vlserver_old);
 		}
 
+		/* Check for duplicates in the server list */
+		for (j = 0; j < vllist->nr_servers; j++) {
+			struct afs_vlserver *s = vllist->servers[j].server;
 
-		/* TODO: Might want to check for duplicates */
+			if (s->name_len == server->name_len &&
+			    s->port == server->port &&
+			    strncasecmp(s->name, server->name, server->name_len) == 0) {
+				afs_put_vlserver(cell->net, server);
+				server = NULL;
+				break;
+			}
+		}
+		if (!server)
+			continue;
 
 		/* Insertion-sort by priority and weight */
 		for (j = 0; j < vllist->nr_servers; j++) {