From: Steve French <stfrench@microsoft.com>
Date: Fri, 12 Nov 2021 09:55:03 -0600
Subject: [PATCH] cifs: release lock earlier in dequeue_mid error case
Git-commit: 9e7ffa77b26a8a00a12b6631aa13f161d9900b0c
References: bsc#1193629
Patch-mainline: v5.16-rc1
In dequeue_mid we can log an error while holding a spinlock,
GlobalMid_Lock. Coverity notes that the error logging
also grabs a lock so it is cleaner (and a bit safer) to
release the GlobalMid_Lock before logging the warning.
Addresses-Coverity: 1507573 ("Thread deadlock")
Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
Signed-off-by: Steve French <stfrench@microsoft.com>
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
---
fs/cifs/connect.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index df3365958603..ae21dff02f30 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -709,13 +709,14 @@ dequeue_mid(struct mid_q_entry *mid, bool malformed)
* Trying to handle/dequeue a mid after the send_recv()
* function has finished processing it is a bug.
*/
- if (mid->mid_flags & MID_DELETED)
+ if (mid->mid_flags & MID_DELETED) {
+ spin_unlock(&GlobalMid_Lock);
pr_warn_once("trying to dequeue a deleted mid\n");
- else {
+ } else {
list_del_init(&mid->qhead);
mid->mid_flags |= MID_DELETED;
+ spin_unlock(&GlobalMid_Lock);
}
- spin_unlock(&GlobalMid_Lock);
}
static unsigned int
--
2.34.1