NFSv4: When recovering state fails with EAGAIN, retry the same recovery

If the server returns with EAGAIN when we're trying to recover from
a server reboot, we currently delay for 1 second, but then mark the
stateid as needing recovery after the grace period has expired.

Instead, we should just retry the same recovery process immediately
after the 1 second delay. Break out of the loop after 10 retries.

Fixes: 35a61606a612 ("NFS: Reduce indentation of the switch statement...")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>

1 files changed, 8 insertions, 1 deletions

diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index a71a61e5fe2c..d03b9cf42bd0 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1607,6 +1607,7 @@ static int __nfs4_reclaim_open_state(struct nfs4_state_owner *sp, struct nfs4_st
 static int nfs4_reclaim_open_state(struct nfs4_state_owner *sp, const struct nfs4_state_recovery_ops *ops)
 {
         struct nfs4_state *state;
+        unsigned int loop = 0;
         int status = 0;
 
         /* Note: we rely on the sp->so_states list being ordered 
@@ -1633,8 +1634,10 @@ restart:
 
                 switch (status) {
                 default:
-                        if (status >= 0)
+                        if (status >= 0) {
+                                loop = 0;
                                 break;
+                        }
                         printk(KERN_ERR "NFS: %s: unhandled error %d\n", __func__, status);
                         /* Fall through */
                 case -ENOENT:
@@ -1648,6 +1651,10 @@ restart:
                         break;
                 case -EAGAIN:
                         ssleep(1);
+                        if (loop++ < 10) {
+                                set_bit(ops->state_flag_bit, &state->flags);
+                                break;
+                        }
                         /* Fall through */
                 case -NFS4ERR_ADMIN_REVOKED:
                 case -NFS4ERR_STALE_STATEID: