[PATCH V1] fuse: Abort the requests under processing queue with a spin_lock

[Pradeep P V K <quic_pragalla@quicinc.com>]

There is a potential race/timing issue while aborting the
requests on processing list between fuse_dev_release() and
fuse_abort_conn(). This is resulting into below warnings
and can even result into UAF issues.

[22809.190255][T31644] refcount_t: underflow; use-after-free.
[22809.190266][T31644] WARNING: CPU: 2 PID: 31644 at lib/refcount.c:28
refcount_warn_saturate+0x110/0x158
...
[22809.190567][T31644] Call trace:
[22809.190567][T31644]  refcount_warn_saturate+0x110/0x158
[22809.190569][T31644]  fuse_file_put+0xfc/0x104
[22809.190575][T31644]  fuse_readpages_end+0x210/0x29c
[22809.190579][T31644]  fuse_request_end+0x17c/0x200
[22809.190580][T31644]  fuse_dev_release+0xe0/0x1e4
[22809.190582][T31644]  __fput+0xfc/0x294
[22809.190588][T31644]  ____fput+0x18/0x2c
[22809.190590][T31644]  task_work_run+0xd8/0x104
[22809.190599][T31644]  do_exit+0x2a8/0xa5c
[22809.190605][T31644]  do_group_exit+0x78/0xa4
[22809.190608][T31644]  get_signal+0x778/0x8a8
[22809.190614][T31644]  do_notify_resume+0x134/0x340
[22809.190617][T31644]  el0_svc+0x68/0xc4
[22809.190623][T31644]  el0t_64_sync_handler+0x8c/0xfc
[22809.190626][T31644]  el0t_64_sync+0x1a0/0x1a4

Fix this by aborting the requests in fuse_dev_release()
under fpq spin lock.

Signed-off-by: Pradeep P V K <quic_pragalla@quicinc.com>
---
 fs/fuse/dev.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 1a8f82f478cb..bbc33a97ab7c 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -2208,9 +2208,8 @@ int fuse_dev_release(struct inode *inode, struct file *file)
 		WARN_ON(!list_empty(&fpq->io));
 		for (i = 0; i < FUSE_PQ_HASH_SIZE; i++)
 			list_splice_init(&fpq->processing[i], &to_end);
-		spin_unlock(&fpq->lock);
-
 		end_requests(&to_end);
+		spin_unlock(&fpq->lock);
 
 		/* Are we the last open device? */
 		if (atomic_dec_and_test(&fc->dev_count)) {
-- 
2.17.1