fscache: Fix the volume collision wait condition

The condition that the waits in fscache_wait_on_volume_collision() are
waiting until are inverted.  This suddenly started happening on the
upstream kernel with something like the following appearing in dmesg when
running xfstests:

	CacheFiles: cachefiles: Inode already in use: Iafs,example.com,100055

Fix them by inverting the conditions.

Fixes: 62ab63352350 ("fscache: Implement volume registration")
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
cc: linux-cachefs@redhat.com
Link: https://lore.kernel.org/r/164251398010.3435901.943876048104930939.stgit@warthog.procyon.org.uk/

diff --git a/fs/fscache/volume.c b/fs/fscache/volume.c
index a57c6cb..f2aa7db 100644 (file)
--- a/fs/fscache/volume.c
+++ b/fs/fscache/volume.c
@@ -142,12 +142,12 @@ static void fscache_wait_on_volume_collision(struct fscache_volume *candidate,
                                             unsigned int collidee_debug_id)
 {
        wait_var_event_timeout(&candidate->flags,
-                              fscache_is_acquire_pending(candidate), 20 * HZ);
+                              !fscache_is_acquire_pending(candidate), 20 * HZ);
        if (!fscache_is_acquire_pending(candidate)) {
                pr_notice("Potential volume collision new=%08x old=%08x",
                          candidate->debug_id, collidee_debug_id);
                fscache_stat(&fscache_n_volumes_collision);
-               wait_var_event(&candidate->flags, fscache_is_acquire_pending(candidate));
+               wait_var_event(&candidate->flags, !fscache_is_acquire_pending(candidate));
        }
 }