[PATCH v7 03/11] NFS/localio: nfs_uuid_put() fix the wake up after unlinking the file
From: Mike Snitzer <snitzer@kernel.org>
Date: 2025-08-05 23:21:12
Subsystem:
filesystems (vfs and infrastructure), kernel nfsd, sunrpc, and lockd servers, nfs, sunrpc, and lockd clients, the rest · Maintainers:
Alexander Viro, Christian Brauner, Chuck Lever, Jeff Layton, Trond Myklebust, Anna Schumaker, Linus Torvalds
From: Trond Myklebust <redacted> Use store_release_wake_up() instead of wake_up_var_locked(), because the waiter cannot retake the nfs_uuid->lock. Acked-by: Mike Snitzer <snitzer@kernel.org> Tested-by: Mike Snitzer <snitzer@kernel.org> Suggested-by: NeilBrown <neil@brown.name> Link: https://lore.kernel.org/all/175262948827.2234665.1891349021754495573@noble.neil.brown.name/ (local) Fixes: 21fb44034695 ("nfs_localio: protect race between nfs_uuid_put() and nfs_close_local_fh()") Signed-off-by: Trond Myklebust <redacted> --- fs/nfs_common/nfslocalio.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/fs/nfs_common/nfslocalio.c b/fs/nfs_common/nfslocalio.c
index f1f1592ac1348..dd715cdb6c043 100644
--- a/fs/nfs_common/nfslocalio.c
+++ b/fs/nfs_common/nfslocalio.c@@ -198,8 +198,7 @@ static bool nfs_uuid_put(nfs_uuid_t *nfs_uuid) /* Now we can allow racing nfs_close_local_fh() to * skip the locking. */ - RCU_INIT_POINTER(nfl->nfs_uuid, NULL); - wake_up_var_locked(&nfl->nfs_uuid, &nfs_uuid->lock); + store_release_wake_up(&nfl->nfs_uuid, RCU_INITIALIZER(NULL)); } /* Remove client from nn->local_clients */
--
2.44.0