From ace7ade4f56cab0ea19d3665d7c14d713788c13d Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Mon, 13 Apr 2020 21:37:19 -0400
Subject: nfsd4: common stateid-printing code

There's a problem with how I'm formatting stateids.  Before I fix it,
I'd like to move the stateid formatting into a common helper.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfs4state.c | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index c107caa56525..ded08aeae497 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2422,6 +2422,11 @@ static void nfs4_show_owner(struct seq_file *s, struct nfs4_stateowner *oo)
 	seq_quote_mem(s, oo->so_owner.data, oo->so_owner.len);
 }
 
+static void nfs4_show_stateid(struct seq_file *s, stateid_t *stid)
+{
+	seq_printf(s, "0x%16phN", stid);
+}
+
 static int nfs4_show_open(struct seq_file *s, struct nfs4_stid *st)
 {
 	struct nfs4_ol_stateid *ols;
@@ -2437,7 +2442,9 @@ static int nfs4_show_open(struct seq_file *s, struct nfs4_stid *st)
 	nf = st->sc_file;
 	file = find_any_file(nf);
 
-	seq_printf(s, "- 0x%16phN: { type: open, ", &st->sc_stateid);
+	seq_printf(s, "- ");
+	nfs4_show_stateid(s, &st->sc_stateid);
+	seq_printf(s, ": { type: open, ");
 
 	access = bmap_to_share_mode(ols->st_access_bmap);
 	deny   = bmap_to_share_mode(ols->st_deny_bmap);
@@ -2470,7 +2477,9 @@ static int nfs4_show_lock(struct seq_file *s, struct nfs4_stid *st)
 	nf = st->sc_file;
 	file = find_any_file(nf);
 
-	seq_printf(s, "- 0x%16phN: { type: lock, ", &st->sc_stateid);
+	seq_printf(s, "- ");
+	nfs4_show_stateid(s, &st->sc_stateid);
+	seq_printf(s, ": { type: lock, ");
 
 	/*
 	 * Note: a lock stateid isn't really the same thing as a lock,
@@ -2499,7 +2508,9 @@ static int nfs4_show_deleg(struct seq_file *s, struct nfs4_stid *st)
 	nf = st->sc_file;
 	file = nf->fi_deleg_file;
 
-	seq_printf(s, "- 0x%16phN: { type: deleg, ", &st->sc_stateid);
+	seq_printf(s, "- ");
+	nfs4_show_stateid(s, &st->sc_stateid);
+	seq_printf(s, ": { type: deleg, ");
 
 	/* Kinda dead code as long as we only support read delegs: */
 	seq_printf(s, "access: %s, ",
@@ -2521,7 +2532,9 @@ static int nfs4_show_layout(struct seq_file *s, struct nfs4_stid *st)
 	ls = container_of(st, struct nfs4_layout_stateid, ls_stid);
 	file = ls->ls_file;
 
-	seq_printf(s, "- 0x%16phN: { type: layout, ", &st->sc_stateid);
+	seq_printf(s, "- ");
+	nfs4_show_stateid(s, &st->sc_stateid);
+	seq_printf(s, ": { type: layout, ");
 
 	/* XXX: What else would be useful? */
 
-- 
cgit v1.2.3


From ee590d259784806c32ad0523aa6a9c321dbf96a4 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Mon, 13 Apr 2020 22:03:06 -0400
Subject: nfsd4: stid display should preserve on-the-wire byte order

When we decode the stateid we byte-swap si_generation.

But for simplicity's sake and ease of comparison with network traces,
it's better to display the whole thing in network order.

Reported-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfs4state.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index ded08aeae497..6fed3bf00ca7 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2424,7 +2424,8 @@ static void nfs4_show_owner(struct seq_file *s, struct nfs4_stateowner *oo)
 
 static void nfs4_show_stateid(struct seq_file *s, stateid_t *stid)
 {
-	seq_printf(s, "0x%16phN", stid);
+	seq_printf(s, "0x%.8x", stid->si_generation);
+	seq_printf(s, "%12phN", &stid->si_opaque);
 }
 
 static int nfs4_show_open(struct seq_file *s, struct nfs4_stid *st)
-- 
cgit v1.2.3


From 580da465a0328d597df0087800717814a62da638 Mon Sep 17 00:00:00 2001
From: Achilles Gaikwad <agaikwad@redhat.com>
Date: Mon, 20 Apr 2020 18:20:31 +0530
Subject: nfsd4: add filename to states output

Add filename to states output for ease of debugging.

Signed-off-by: Achilles Gaikwad <agaikwad@redhat.com>
Signed-off-by: Kenneth Dsouza <kdsouza@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfs4state.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 6fed3bf00ca7..2c8567b32990 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2406,6 +2406,11 @@ static void states_stop(struct seq_file *s, void *v)
 	spin_unlock(&clp->cl_lock);
 }
 
+static void nfs4_show_fname(struct seq_file *s, struct nfsd_file *f)
+{
+         seq_printf(s, "filename: \"%pD2\"", f->nf_file);
+}
+
 static void nfs4_show_superblock(struct seq_file *s, struct nfsd_file *f)
 {
 	struct inode *inode = f->nf_inode;
@@ -2459,6 +2464,8 @@ static int nfs4_show_open(struct seq_file *s, struct nfs4_stid *st)
 
 	nfs4_show_superblock(s, file);
 	seq_printf(s, ", ");
+	nfs4_show_fname(s, file);
+	seq_printf(s, ", ");
 	nfs4_show_owner(s, oo);
 	seq_printf(s, " }\n");
 	nfsd_file_put(file);
@@ -2492,6 +2499,8 @@ static int nfs4_show_lock(struct seq_file *s, struct nfs4_stid *st)
 	nfs4_show_superblock(s, file);
 	/* XXX: open stateid? */
 	seq_printf(s, ", ");
+	nfs4_show_fname(s, file);
+	seq_printf(s, ", ");
 	nfs4_show_owner(s, oo);
 	seq_printf(s, " }\n");
 	nfsd_file_put(file);
@@ -2520,6 +2529,8 @@ static int nfs4_show_deleg(struct seq_file *s, struct nfs4_stid *st)
 	/* XXX: lease time, whether it's being recalled. */
 
 	nfs4_show_superblock(s, file);
+	seq_printf(s, ", ");
+	nfs4_show_fname(s, file);
 	seq_printf(s, " }\n");
 
 	return 0;
@@ -2540,6 +2551,8 @@ static int nfs4_show_layout(struct seq_file *s, struct nfs4_stid *st)
 	/* XXX: What else would be useful? */
 
 	nfs4_show_superblock(s, file);
+	seq_printf(s, ", ");
+	nfs4_show_fname(s, file);
 	seq_printf(s, " }\n");
 
 	return 0;
-- 
cgit v1.2.3


From c2d715a1af75443631b05a49edbd152aef7e8b93 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Mon, 27 Apr 2020 17:59:01 -0400
Subject: nfsd: handle repeated BIND_CONN_TO_SESSION

If the client attempts BIND_CONN_TO_SESSION on an already bound
connection, it should be either a no-op or an error.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfs4state.c | 54 +++++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 42 insertions(+), 12 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 2c8567b32990..fe88f5f23cf4 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -3485,6 +3485,45 @@ __be32 nfsd4_backchannel_ctl(struct svc_rqst *rqstp,
 	return nfs_ok;
 }
 
+static struct nfsd4_conn *__nfsd4_find_conn(struct svc_xprt *xpt, struct nfsd4_session *s)
+{
+	struct nfsd4_conn *c;
+
+	list_for_each_entry(c, &s->se_conns, cn_persession) {
+		if (c->cn_xprt == xpt) {
+			return c;
+		}
+	}
+	return NULL;
+}
+
+static __be32 nfsd4_match_existing_connection(struct svc_rqst *rqst,
+				struct nfsd4_session *session, u32 req)
+{
+	struct nfs4_client *clp = session->se_client;
+	struct svc_xprt *xpt = rqst->rq_xprt;
+	struct nfsd4_conn *c;
+	__be32 status;
+
+	/* Following the last paragraph of RFC 5661 Section 18.34.3: */
+	spin_lock(&clp->cl_lock);
+	c = __nfsd4_find_conn(xpt, session);
+	if (!c)
+		status = nfserr_noent;
+	else if (req == c->cn_flags)
+		status = nfs_ok;
+	else if (req == NFS4_CDFC4_FORE_OR_BOTH &&
+				c->cn_flags != NFS4_CDFC4_BACK)
+		status = nfs_ok;
+	else if (req == NFS4_CDFC4_BACK_OR_BOTH &&
+				c->cn_flags != NFS4_CDFC4_FORE)
+		status = nfs_ok;
+	else
+		status = nfserr_inval;
+	spin_unlock(&clp->cl_lock);
+	return status;
+}
+
 __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp,
 		     struct nfsd4_compound_state *cstate,
 		     union nfsd4_op_u *u)
@@ -3506,6 +3545,9 @@ __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp,
 	status = nfserr_wrong_cred;
 	if (!nfsd4_mach_creds_match(session->se_client, rqstp))
 		goto out;
+	status = nfsd4_match_existing_connection(rqstp, session, bcts->dir);
+	if (status == nfs_ok || status == nfserr_inval)
+		goto out;
 	status = nfsd4_map_bcts_dir(&bcts->dir);
 	if (status)
 		goto out;
@@ -3571,18 +3613,6 @@ out:
 	return status;
 }
 
-static struct nfsd4_conn *__nfsd4_find_conn(struct svc_xprt *xpt, struct nfsd4_session *s)
-{
-	struct nfsd4_conn *c;
-
-	list_for_each_entry(c, &s->se_conns, cn_persession) {
-		if (c->cn_xprt == xpt) {
-			return c;
-		}
-	}
-	return NULL;
-}
-
 static __be32 nfsd4_sequence_check_conn(struct nfsd4_conn *new, struct nfsd4_session *ses)
 {
 	struct nfs4_client *clp = ses->se_client;
-- 
cgit v1.2.3


From 28df3d1539de5090f7916f6fff03891b67f366f4 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Fri, 28 Jul 2017 16:35:15 -0400
Subject: nfsd: clients don't need to break their own delegations

We currently revoke read delegations on any write open or any operation
that modifies file data or metadata (including rename, link, and
unlink).  But if the delegation in question is the only read delegation
and is held by the client performing the operation, that's not really
necessary.

It's not always possible to prevent this in the NFSv4.0 case, because
there's not always a way to determine which client an NFSv4.0 delegation
came from.  (In theory we could try to guess this from the transport
layer, e.g., by assuming all traffic on a given TCP connection comes
from the same client.  But that's not really correct.)

In the NFSv4.1 case the session layer always tells us the client.

This patch should remove such self-conflicts in all cases where we can
reliably determine the client from the compound.

To do that we need to track "who" is performing a given (possibly
lease-breaking) file operation.  We're doing that by storing the
information in the svc_rqst and using kthread_data() to map the current
task back to a svc_rqst.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 Documentation/filesystems/locking.rst |  2 ++
 fs/locks.c                            |  3 +++
 fs/nfsd/nfs4proc.c                    |  2 ++
 fs/nfsd/nfs4state.c                   | 14 ++++++++++++++
 fs/nfsd/nfsd.h                        |  2 ++
 fs/nfsd/nfssvc.c                      |  6 ++++++
 include/linux/fs.h                    |  1 +
 include/linux/sunrpc/svc.h            |  1 +
 8 files changed, 31 insertions(+)

(limited to 'fs/nfsd')

diff --git a/Documentation/filesystems/locking.rst b/Documentation/filesystems/locking.rst
index 5057e4d9dcd1..9fdcec416614 100644
--- a/Documentation/filesystems/locking.rst
+++ b/Documentation/filesystems/locking.rst
@@ -425,6 +425,7 @@ prototypes::
 	int (*lm_grant)(struct file_lock *, struct file_lock *, int);
 	void (*lm_break)(struct file_lock *); /* break_lease callback */
 	int (*lm_change)(struct file_lock **, int);
+	bool (*lm_breaker_owns_lease)(struct file_lock *);
 
 locking rules:
 
@@ -435,6 +436,7 @@ lm_notify:		yes		yes			no
 lm_grant:		no		no			no
 lm_break:		yes		no			no
 lm_change		yes		no			no
+lm_breaker_owns_lease:	no		no			no
 ==========		=============	=================	=========
 
 buffer_head
diff --git a/fs/locks.c b/fs/locks.c
index b8a31c1c4fff..a3f186846e93 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1557,6 +1557,9 @@ static bool leases_conflict(struct file_lock *lease, struct file_lock *breaker)
 {
 	bool rc;
 
+	if (lease->fl_lmops->lm_breaker_owns_lease
+			&& lease->fl_lmops->lm_breaker_owns_lease(lease))
+		return false;
 	if ((breaker->fl_flags & FL_LAYOUT) != (lease->fl_flags & FL_LAYOUT)) {
 		rc = false;
 		goto trace;
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 0e75f7fb5fec..a6d73aa51ce4 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -2302,6 +2302,8 @@ nfsd4_proc_compound(struct svc_rqst *rqstp)
 	}
 	check_if_stalefh_allowed(args);
 
+	rqstp->rq_lease_breaker = (void **)&cstate->clp;
+
 	trace_nfsd_compound(rqstp, args->opcnt);
 	while (!status && resp->opcnt < args->opcnt) {
 		op = &args->ops[resp->opcnt++];
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index fe88f5f23cf4..31388046c6fe 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -4579,6 +4579,19 @@ nfsd_break_deleg_cb(struct file_lock *fl)
 	return ret;
 }
 
+static bool nfsd_breaker_owns_lease(struct file_lock *fl)
+{
+	struct nfs4_delegation *dl = fl->fl_owner;
+	struct svc_rqst *rqst;
+	struct nfs4_client *clp;
+
+	if (!i_am_nfsd())
+		return NULL;
+	rqst = kthread_data(current);
+	clp = *(rqst->rq_lease_breaker);
+	return dl->dl_stid.sc_client == clp;
+}
+
 static int
 nfsd_change_deleg_cb(struct file_lock *onlist, int arg,
 		     struct list_head *dispose)
@@ -4590,6 +4603,7 @@ nfsd_change_deleg_cb(struct file_lock *onlist, int arg,
 }
 
 static const struct lock_manager_operations nfsd_lease_mng_ops = {
+	.lm_breaker_owns_lease = nfsd_breaker_owns_lease,
 	.lm_break = nfsd_break_deleg_cb,
 	.lm_change = nfsd_change_deleg_cb,
 };
diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
index 2ab5569126b8..36cdd81b6688 100644
--- a/fs/nfsd/nfsd.h
+++ b/fs/nfsd/nfsd.h
@@ -88,6 +88,8 @@ int		nfsd_pool_stats_release(struct inode *, struct file *);
 
 void		nfsd_destroy(struct net *net);
 
+bool		i_am_nfsd(void);
+
 struct nfsdfs_client {
 	struct kref cl_ref;
 	void (*cl_release)(struct kref *kref);
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index ca9fd348548b..4f588c0eaaf4 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -601,6 +601,11 @@ static const struct svc_serv_ops nfsd_thread_sv_ops = {
 	.svo_module		= THIS_MODULE,
 };
 
+bool i_am_nfsd()
+{
+	return kthread_func(current) == nfsd;
+}
+
 int nfsd_create_serv(struct net *net)
 {
 	int error;
@@ -1011,6 +1016,7 @@ nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp)
 		*statp = rpc_garbage_args;
 		return 1;
 	}
+	rqstp->rq_lease_breaker = NULL;
 	/*
 	 * Give the xdr decoder a chance to change this if it wants
 	 * (necessary in the NFSv4.0 compound case)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4f6f59b4f22a..4b784560ffaf 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1045,6 +1045,7 @@ struct lock_manager_operations {
 	bool (*lm_break)(struct file_lock *);
 	int (*lm_change)(struct file_lock *, int, struct list_head *);
 	void (*lm_setup)(struct file_lock *, void **);
+	bool (*lm_breaker_owns_lease)(struct file_lock *);
 };
 
 struct lock_manager {
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index fd390894a584..abf4a57ce4a7 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -299,6 +299,7 @@ struct svc_rqst {
 	struct net		*rq_bc_net;	/* pointer to backchannel's
 						 * net namespace
 						 */
+	void **			rq_lease_breaker; /* The v4 client breaking a lease */
 };
 
 #define SVC_NET(rqst) (rqst->rq_xprt ? rqst->rq_xprt->xpt_net : rqst->rq_bc_net)
-- 
cgit v1.2.3


From 44fb26c6b4c5ad63f733df7936deffeae2950855 Mon Sep 17 00:00:00 2001
From: Ma Feng <mafeng.ma@huawei.com>
Date: Mon, 11 May 2020 20:07:08 +0800
Subject: nfsd: Fix old-style function definition

Fix warning:

fs/nfsd/nfssvc.c:604:6: warning: old-style function definition [-Wold-style-definition]
 bool i_am_nfsd()
      ^~~~~~~~~

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Ma Feng <mafeng.ma@huawei.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfssvc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 4f588c0eaaf4..b603dfcdd361 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -601,7 +601,7 @@ static const struct svc_serv_ops nfsd_thread_sv_ops = {
 	.svo_module		= THIS_MODULE,
 };
 
-bool i_am_nfsd()
+bool i_am_nfsd(void)
 {
 	return kthread_func(current) == nfsd;
 }
-- 
cgit v1.2.3


From 0b175b18648ebedfe255b11a7792f1d76848a8f7 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sat, 2 May 2020 11:34:40 -0400
Subject: NFSD: Add tracepoints to NFSD's duplicate reply cache

Try to capture DRC failures.

Two additional clean-ups:
- Introduce Doxygen-style comments for the main entry points
- Remove a dprintk that fires for an allocation failure. This was
  the only dprintk in the REPCACHE class.

Reported-by: kbuild test robot <lkp@intel.com>
[ cel: force typecast for display of checksum values ]
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfscache.c | 57 ++++++++++++++++++++++++++++++++--------------------
 fs/nfsd/trace.h    | 59 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 94 insertions(+), 22 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index 96352ab7bd81..945d2f5e760e 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -20,8 +20,7 @@
 
 #include "nfsd.h"
 #include "cache.h"
-
-#define NFSDDBG_FACILITY	NFSDDBG_REPCACHE
+#include "trace.h"
 
 /*
  * We use this value to determine the number of hash buckets from the max
@@ -323,8 +322,10 @@ nfsd_cache_key_cmp(const struct svc_cacherep *key,
 			const struct svc_cacherep *rp, struct nfsd_net *nn)
 {
 	if (key->c_key.k_xid == rp->c_key.k_xid &&
-	    key->c_key.k_csum != rp->c_key.k_csum)
+	    key->c_key.k_csum != rp->c_key.k_csum) {
 		++nn->payload_misses;
+		trace_nfsd_drc_mismatch(nn, key, rp);
+	}
 
 	return memcmp(&key->c_key, &rp->c_key, sizeof(key->c_key));
 }
@@ -377,15 +378,22 @@ out:
 	return ret;
 }
 
-/*
+/**
+ * nfsd_cache_lookup - Find an entry in the duplicate reply cache
+ * @rqstp: Incoming Call to find
+ *
  * Try to find an entry matching the current call in the cache. When none
  * is found, we try to grab the oldest expired entry off the LRU list. If
  * a suitable one isn't there, then drop the cache_lock and allocate a
  * new one, then search again in case one got inserted while this thread
  * didn't hold the lock.
+ *
+ * Return values:
+ *   %RC_DOIT: Process the request normally
+ *   %RC_REPLY: Reply from cache
+ *   %RC_DROPIT: Do not process the request further
  */
-int
-nfsd_cache_lookup(struct svc_rqst *rqstp)
+int nfsd_cache_lookup(struct svc_rqst *rqstp)
 {
 	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
 	struct svc_cacherep	*rp, *found;
@@ -399,7 +407,7 @@ nfsd_cache_lookup(struct svc_rqst *rqstp)
 	rqstp->rq_cacherep = NULL;
 	if (type == RC_NOCACHE) {
 		nfsdstats.rcnocache++;
-		return rtn;
+		goto out;
 	}
 
 	csum = nfsd_cache_csum(rqstp);
@@ -409,10 +417,8 @@ nfsd_cache_lookup(struct svc_rqst *rqstp)
 	 * preallocate an entry.
 	 */
 	rp = nfsd_reply_cache_alloc(rqstp, csum, nn);
-	if (!rp) {
-		dprintk("nfsd: unable to allocate DRC entry!\n");
-		return rtn;
-	}
+	if (!rp)
+		goto out;
 
 	spin_lock(&b->cache_lock);
 	found = nfsd_cache_insert(b, rp, nn);
@@ -431,8 +437,10 @@ nfsd_cache_lookup(struct svc_rqst *rqstp)
 
 	/* go ahead and prune the cache */
 	prune_bucket(b, nn);
- out:
+
+out_unlock:
 	spin_unlock(&b->cache_lock);
+out:
 	return rtn;
 
 found_entry:
@@ -442,13 +450,13 @@ found_entry:
 
 	/* Request being processed */
 	if (rp->c_state == RC_INPROG)
-		goto out;
+		goto out_trace;
 
 	/* From the hall of fame of impractical attacks:
 	 * Is this a user who tries to snoop on the cache? */
 	rtn = RC_DOIT;
 	if (!test_bit(RQ_SECURE, &rqstp->rq_flags) && rp->c_secure)
-		goto out;
+		goto out_trace;
 
 	/* Compose RPC reply header */
 	switch (rp->c_type) {
@@ -460,7 +468,7 @@ found_entry:
 		break;
 	case RC_REPLBUFF:
 		if (!nfsd_cache_append(rqstp, &rp->c_replvec))
-			goto out;	/* should not happen */
+			goto out_unlock; /* should not happen */
 		rtn = RC_REPLY;
 		break;
 	default:
@@ -468,13 +476,19 @@ found_entry:
 		nfsd_reply_cache_free_locked(b, rp, nn);
 	}
 
-	goto out;
+out_trace:
+	trace_nfsd_drc_found(nn, rqstp, rtn);
+	goto out_unlock;
 }
 
-/*
- * Update a cache entry. This is called from nfsd_dispatch when
- * the procedure has been executed and the complete reply is in
- * rqstp->rq_res.
+/**
+ * nfsd_cache_update - Update an entry in the duplicate reply cache.
+ * @rqstp: svc_rqst with a finished Reply
+ * @cachetype: which cache to update
+ * @statp: Reply's status code
+ *
+ * This is called from nfsd_dispatch when the procedure has been
+ * executed and the complete reply is in rqstp->rq_res.
  *
  * We're copying around data here rather than swapping buffers because
  * the toplevel loop requires max-sized buffers, which would be a waste
@@ -487,8 +501,7 @@ found_entry:
  * nfsd failed to encode a reply that otherwise would have been cached.
  * In this case, nfsd_cache_update is called with statp == NULL.
  */
-void
-nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp)
+void nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp)
 {
 	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
 	struct svc_cacherep *rp = rqstp->rq_cacherep;
diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
index 78c574251c60..371dc198d28e 100644
--- a/fs/nfsd/trace.h
+++ b/fs/nfsd/trace.h
@@ -432,6 +432,65 @@ TRACE_EVENT(nfsd_file_fsnotify_handle_event,
 			__entry->nlink, __entry->mode, __entry->mask)
 );
 
+#include "cache.h"
+
+TRACE_DEFINE_ENUM(RC_DROPIT);
+TRACE_DEFINE_ENUM(RC_REPLY);
+TRACE_DEFINE_ENUM(RC_DOIT);
+
+#define show_drc_retval(x)						\
+	__print_symbolic(x,						\
+		{ RC_DROPIT, "DROPIT" },				\
+		{ RC_REPLY, "REPLY" },					\
+		{ RC_DOIT, "DOIT" })
+
+TRACE_EVENT(nfsd_drc_found,
+	TP_PROTO(
+		const struct nfsd_net *nn,
+		const struct svc_rqst *rqstp,
+		int result
+	),
+	TP_ARGS(nn, rqstp, result),
+	TP_STRUCT__entry(
+		__field(unsigned long long, boot_time)
+		__field(unsigned long, result)
+		__field(u32, xid)
+	),
+	TP_fast_assign(
+		__entry->boot_time = nn->boot_time;
+		__entry->result = result;
+		__entry->xid = be32_to_cpu(rqstp->rq_xid);
+	),
+	TP_printk("boot_time=%16llx xid=0x%08x result=%s",
+		__entry->boot_time, __entry->xid,
+		show_drc_retval(__entry->result))
+
+);
+
+TRACE_EVENT(nfsd_drc_mismatch,
+	TP_PROTO(
+		const struct nfsd_net *nn,
+		const struct svc_cacherep *key,
+		const struct svc_cacherep *rp
+	),
+	TP_ARGS(nn, key, rp),
+	TP_STRUCT__entry(
+		__field(unsigned long long, boot_time)
+		__field(u32, xid)
+		__field(u32, cached)
+		__field(u32, ingress)
+	),
+	TP_fast_assign(
+		__entry->boot_time = nn->boot_time;
+		__entry->xid = be32_to_cpu(key->c_key.k_xid);
+		__entry->cached = (__force u32)key->c_key.k_csum;
+		__entry->ingress = (__force u32)rp->c_key.k_csum;
+	),
+	TP_printk("boot_time=%16llx xid=0x%08x cached-csum=0x%08x ingress-csum=0x%08x",
+		__entry->boot_time, __entry->xid, __entry->cached,
+		__entry->ingress)
+);
+
 #endif /* _NFSD_TRACE_H */
 
 #undef TRACE_INCLUDE_PATH
-- 
cgit v1.2.3


From dd5e3fbc1f472951afd7e7643ace5d1672b31358 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sun, 5 Apr 2020 11:42:29 -0400
Subject: NFSD: Add tracepoints to the NFSD state management code

Capture obvious events and replace dprintk() call sites. Introduce
infrastructure so that adding more tracepoints in this code later
is simplified.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfs4state.c |  52 ++++++++------------
 fs/nfsd/state.h     |   7 ---
 fs/nfsd/trace.h     | 133 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 152 insertions(+), 40 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index c107caa56525..04d80f9e2f91 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -51,6 +51,7 @@
 #include "netns.h"
 #include "pnfs.h"
 #include "filecache.h"
+#include "trace.h"
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
@@ -167,9 +168,6 @@ renew_client_locked(struct nfs4_client *clp)
 		return;
 	}
 
-	dprintk("renewing client (clientid %08x/%08x)\n",
-			clp->cl_clientid.cl_boot,
-			clp->cl_clientid.cl_id);
 	list_move_tail(&clp->cl_lru, &nn->client_lru);
 	clp->cl_time = ktime_get_boottime_seconds();
 }
@@ -1922,8 +1920,7 @@ STALE_CLIENTID(clientid_t *clid, struct nfsd_net *nn)
 	 */
 	if (clid->cl_boot == (u32)nn->boot_time)
 		return 0;
-	dprintk("NFSD stale clientid (%08x/%08x) boot_time %08llx\n",
-		clid->cl_boot, clid->cl_id, nn->boot_time);
+	trace_nfsd_clid_stale(clid);
 	return 1;
 }
 
@@ -3879,11 +3876,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		if (clp_used_exchangeid(conf))
 			goto out;
 		if (!same_creds(&conf->cl_cred, &rqstp->rq_cred)) {
-			char addr_str[INET6_ADDRSTRLEN];
-			rpc_ntop((struct sockaddr *) &conf->cl_addr, addr_str,
-				 sizeof(addr_str));
-			dprintk("NFSD: setclientid: string in use by client "
-				"at %s\n", addr_str);
+			trace_nfsd_clid_inuse_err(conf);
 			goto out;
 		}
 	}
@@ -4076,7 +4069,6 @@ out_free_openowner_slab:
 out_free_client_slab:
 	kmem_cache_destroy(client_slab);
 out:
-	dprintk("nfsd4: out of memory while initializing nfsv4\n");
 	return -ENOMEM;
 }
 
@@ -4508,6 +4500,8 @@ nfsd_break_deleg_cb(struct file_lock *fl)
 	struct nfs4_delegation *dp = (struct nfs4_delegation *)fl->fl_owner;
 	struct nfs4_file *fp = dp->dl_stid.sc_file;
 
+	trace_nfsd_deleg_break(&dp->dl_stid.sc_stateid);
+
 	/*
 	 * We don't want the locks code to timeout the lease for us;
 	 * we'll remove it ourself if a delegation isn't returned
@@ -5018,8 +5012,7 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open,
 
 	memcpy(&open->op_delegate_stateid, &dp->dl_stid.sc_stateid, sizeof(dp->dl_stid.sc_stateid));
 
-	dprintk("NFSD: delegation stateid=" STATEID_FMT "\n",
-		STATEID_VAL(&dp->dl_stid.sc_stateid));
+	trace_nfsd_deleg_open(&dp->dl_stid.sc_stateid);
 	open->op_delegate_type = NFS4_OPEN_DELEGATE_READ;
 	nfs4_put_stid(&dp->dl_stid);
 	return;
@@ -5136,9 +5129,7 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 	nfs4_open_delegation(current_fh, open, stp);
 nodeleg:
 	status = nfs_ok;
-
-	dprintk("%s: stateid=" STATEID_FMT "\n", __func__,
-		STATEID_VAL(&stp->st_stid.sc_stateid));
+	trace_nfsd_deleg_none(&stp->st_stid.sc_stateid);
 out:
 	/* 4.1 client trying to upgrade/downgrade delegation? */
 	if (open->op_delegate_type == NFS4_OPEN_DELEGATE_NONE && dp &&
@@ -5192,8 +5183,7 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	__be32 status;
 	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
 
-	dprintk("process_renew(%08x/%08x): starting\n", 
-			clid->cl_boot, clid->cl_id);
+	trace_nfsd_clid_renew(clid);
 	status = lookup_clientid(clid, cstate, nn, false);
 	if (status)
 		goto out;
@@ -5214,6 +5204,7 @@ nfsd4_end_grace(struct nfsd_net *nn)
 	if (nn->grace_ended)
 		return;
 
+	trace_nfsd_grace_complete(nn);
 	nn->grace_ended = true;
 	/*
 	 * If the server goes down again right now, an NFSv4
@@ -5279,13 +5270,10 @@ nfs4_laundromat(struct nfsd_net *nn)
 	copy_stateid_t *cps_t;
 	int i;
 
-	dprintk("NFSD: laundromat service - starting\n");
-
 	if (clients_still_reclaiming(nn)) {
 		new_timeo = 0;
 		goto out;
 	}
-	dprintk("NFSD: end of grace period\n");
 	nfsd4_end_grace(nn);
 	INIT_LIST_HEAD(&reaplist);
 
@@ -5307,8 +5295,7 @@ nfs4_laundromat(struct nfsd_net *nn)
 			break;
 		}
 		if (mark_client_expired_locked(clp)) {
-			dprintk("NFSD: client in use (clientid %08x)\n",
-				clp->cl_clientid.cl_id);
+			trace_nfsd_clid_expired(&clp->cl_clientid);
 			continue;
 		}
 		list_add(&clp->cl_lru, &reaplist);
@@ -5316,8 +5303,7 @@ nfs4_laundromat(struct nfsd_net *nn)
 	spin_unlock(&nn->client_lock);
 	list_for_each_safe(pos, next, &reaplist) {
 		clp = list_entry(pos, struct nfs4_client, cl_lru);
-		dprintk("NFSD: purging unused client (clientid %08x)\n",
-			clp->cl_clientid.cl_id);
+		trace_nfsd_clid_purged(&clp->cl_clientid);
 		list_del_init(&clp->cl_lru);
 		expire_client(clp);
 	}
@@ -5407,7 +5393,6 @@ laundromat_main(struct work_struct *laundry)
 					   laundromat_work);
 
 	t = nfs4_laundromat(nn);
-	dprintk("NFSD: laundromat_main - sleeping for %lld seconds\n", t);
 	queue_delayed_work(laundry_wq, &nn->laundromat_work, t*HZ);
 }
 
@@ -5948,8 +5933,7 @@ nfs4_preprocess_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid,
 	struct nfs4_stid *s;
 	struct nfs4_ol_stateid *stp = NULL;
 
-	dprintk("NFSD: %s: seqid=%d stateid = " STATEID_FMT "\n", __func__,
-		seqid, STATEID_VAL(stateid));
+	trace_nfsd_preprocess(seqid, stateid);
 
 	*stpp = NULL;
 	status = nfsd4_lookup_stateid(cstate, stateid, typemask, &s, nn);
@@ -6018,9 +6002,7 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	oo->oo_flags |= NFS4_OO_CONFIRMED;
 	nfs4_inc_and_copy_stateid(&oc->oc_resp_stateid, &stp->st_stid);
 	mutex_unlock(&stp->st_mutex);
-	dprintk("NFSD: %s: success, seqid=%d stateid=" STATEID_FMT "\n",
-		__func__, oc->oc_seqid, STATEID_VAL(&stp->st_stid.sc_stateid));
-
+	trace_nfsd_open_confirm(oc->oc_seqid, &stp->st_stid.sc_stateid);
 	nfsd4_client_record_create(oo->oo_owner.so_client);
 	status = nfs_ok;
 put_stateid:
@@ -7072,7 +7054,7 @@ nfs4_client_to_reclaim(struct xdr_netobj name, struct xdr_netobj princhash,
 	unsigned int strhashval;
 	struct nfs4_client_reclaim *crp;
 
-	dprintk("NFSD nfs4_client_to_reclaim NAME: %.*s\n", name.len, name.data);
+	trace_nfsd_clid_reclaim(nn, name.len, name.data);
 	crp = alloc_reclaim();
 	if (crp) {
 		strhashval = clientstr_hashval(name);
@@ -7122,7 +7104,7 @@ nfsd4_find_reclaim_client(struct xdr_netobj name, struct nfsd_net *nn)
 	unsigned int strhashval;
 	struct nfs4_client_reclaim *crp = NULL;
 
-	dprintk("NFSD: nfs4_find_reclaim_client for name %.*s\n", name.len, name.data);
+	trace_nfsd_clid_find(nn, name.len, name.data);
 
 	strhashval = clientstr_hashval(name);
 	list_for_each_entry(crp, &nn->reclaim_str_hashtbl[strhashval], cr_strhash) {
@@ -7686,6 +7668,9 @@ nfsd_recall_delegations(struct list_head *reaplist)
 	list_for_each_entry_safe(dp, next, reaplist, dl_recall_lru) {
 		list_del_init(&dp->dl_recall_lru);
 		clp = dp->dl_stid.sc_client;
+
+		trace_nfsd_deleg_recall(&dp->dl_stid.sc_stateid);
+
 		/*
 		 * We skipped all entries that had a zero dl_time before,
 		 * so we can now reset the dl_time back to 0. If a delegation
@@ -7868,6 +7853,7 @@ nfs4_state_start_net(struct net *net)
 		goto skip_grace;
 	printk(KERN_INFO "NFSD: starting %lld-second grace period (net %x)\n",
 	       nn->nfsd4_grace, net->ns.inum);
+	trace_nfsd_grace_start(nn);
 	queue_delayed_work(laundry_wq, &nn->laundromat_work, nn->nfsd4_grace * HZ);
 	return 0;
 
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
index 68d3f30ee760..3b408532a5dc 100644
--- a/fs/nfsd/state.h
+++ b/fs/nfsd/state.h
@@ -64,13 +64,6 @@ typedef struct {
 	refcount_t		sc_count;
 } copy_stateid_t;
 
-#define STATEID_FMT	"(%08x/%08x/%08x/%08x)"
-#define STATEID_VAL(s) \
-	(s)->si_opaque.so_clid.cl_boot, \
-	(s)->si_opaque.so_clid.cl_id, \
-	(s)->si_opaque.so_id, \
-	(s)->si_generation
-
 struct nfsd4_callback {
 	struct nfs4_client *cb_clp;
 	struct rpc_message cb_msg;
diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
index 371dc198d28e..7237fe2f3de9 100644
--- a/fs/nfsd/trace.h
+++ b/fs/nfsd/trace.h
@@ -277,6 +277,7 @@ DECLARE_EVENT_CLASS(nfsd_stateid_class,
 DEFINE_EVENT(nfsd_stateid_class, nfsd_##name, \
 	TP_PROTO(stateid_t *stp), \
 	TP_ARGS(stp))
+
 DEFINE_STATEID_EVENT(layoutstate_alloc);
 DEFINE_STATEID_EVENT(layoutstate_unhash);
 DEFINE_STATEID_EVENT(layoutstate_free);
@@ -288,6 +289,138 @@ DEFINE_STATEID_EVENT(layout_recall_done);
 DEFINE_STATEID_EVENT(layout_recall_fail);
 DEFINE_STATEID_EVENT(layout_recall_release);
 
+DEFINE_STATEID_EVENT(deleg_open);
+DEFINE_STATEID_EVENT(deleg_none);
+DEFINE_STATEID_EVENT(deleg_break);
+DEFINE_STATEID_EVENT(deleg_recall);
+
+DECLARE_EVENT_CLASS(nfsd_stateseqid_class,
+	TP_PROTO(u32 seqid, const stateid_t *stp),
+	TP_ARGS(seqid, stp),
+	TP_STRUCT__entry(
+		__field(u32, seqid)
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+		__field(u32, si_id)
+		__field(u32, si_generation)
+	),
+	TP_fast_assign(
+		__entry->seqid = seqid;
+		__entry->cl_boot = stp->si_opaque.so_clid.cl_boot;
+		__entry->cl_id = stp->si_opaque.so_clid.cl_id;
+		__entry->si_id = stp->si_opaque.so_id;
+		__entry->si_generation = stp->si_generation;
+	),
+	TP_printk("seqid=%u client %08x:%08x stateid %08x:%08x",
+		__entry->seqid, __entry->cl_boot, __entry->cl_id,
+		__entry->si_id, __entry->si_generation)
+)
+
+#define DEFINE_STATESEQID_EVENT(name) \
+DEFINE_EVENT(nfsd_stateseqid_class, nfsd_##name, \
+	TP_PROTO(u32 seqid, const stateid_t *stp), \
+	TP_ARGS(seqid, stp))
+
+DEFINE_STATESEQID_EVENT(preprocess);
+DEFINE_STATESEQID_EVENT(open_confirm);
+
+DECLARE_EVENT_CLASS(nfsd_clientid_class,
+	TP_PROTO(const clientid_t *clid),
+	TP_ARGS(clid),
+	TP_STRUCT__entry(
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+	),
+	TP_fast_assign(
+		__entry->cl_boot = clid->cl_boot;
+		__entry->cl_id = clid->cl_id;
+	),
+	TP_printk("client %08x:%08x", __entry->cl_boot, __entry->cl_id)
+)
+
+#define DEFINE_CLIENTID_EVENT(name) \
+DEFINE_EVENT(nfsd_clientid_class, nfsd_clid_##name, \
+	TP_PROTO(const clientid_t *clid), \
+	TP_ARGS(clid))
+
+DEFINE_CLIENTID_EVENT(expired);
+DEFINE_CLIENTID_EVENT(purged);
+DEFINE_CLIENTID_EVENT(renew);
+DEFINE_CLIENTID_EVENT(stale);
+
+DECLARE_EVENT_CLASS(nfsd_net_class,
+	TP_PROTO(const struct nfsd_net *nn),
+	TP_ARGS(nn),
+	TP_STRUCT__entry(
+		__field(unsigned long long, boot_time)
+	),
+	TP_fast_assign(
+		__entry->boot_time = nn->boot_time;
+	),
+	TP_printk("boot_time=%16llx", __entry->boot_time)
+)
+
+#define DEFINE_NET_EVENT(name) \
+DEFINE_EVENT(nfsd_net_class, nfsd_##name, \
+	TP_PROTO(const struct nfsd_net *nn), \
+	TP_ARGS(nn))
+
+DEFINE_NET_EVENT(grace_start);
+DEFINE_NET_EVENT(grace_complete);
+
+DECLARE_EVENT_CLASS(nfsd_clid_class,
+	TP_PROTO(const struct nfsd_net *nn,
+		 unsigned int namelen,
+		 const unsigned char *namedata),
+	TP_ARGS(nn, namelen, namedata),
+	TP_STRUCT__entry(
+		__field(unsigned long long, boot_time)
+		__field(unsigned int, namelen)
+		__dynamic_array(unsigned char,  name, namelen)
+	),
+	TP_fast_assign(
+		__entry->boot_time = nn->boot_time;
+		__entry->namelen = namelen;
+		memcpy(__get_dynamic_array(name), namedata, namelen);
+	),
+	TP_printk("boot_time=%16llx nfs4_clientid=%.*s",
+		__entry->boot_time, __entry->namelen, __get_str(name))
+)
+
+#define DEFINE_CLID_EVENT(name) \
+DEFINE_EVENT(nfsd_clid_class, nfsd_clid_##name, \
+	TP_PROTO(const struct nfsd_net *nn, \
+		 unsigned int namelen, \
+		 const unsigned char *namedata), \
+	TP_ARGS(nn, namelen, namedata))
+
+DEFINE_CLID_EVENT(find);
+DEFINE_CLID_EVENT(reclaim);
+
+TRACE_EVENT(nfsd_clid_inuse_err,
+	TP_PROTO(const struct nfs4_client *clp),
+	TP_ARGS(clp),
+	TP_STRUCT__entry(
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+		__array(unsigned char, addr, sizeof(struct sockaddr_in6))
+		__field(unsigned int, namelen)
+		__dynamic_array(unsigned char, name, clp->cl_name.len)
+	),
+	TP_fast_assign(
+		__entry->cl_boot = clp->cl_clientid.cl_boot;
+		__entry->cl_id = clp->cl_clientid.cl_id;
+		memcpy(__entry->addr, &clp->cl_addr,
+			sizeof(struct sockaddr_in6));
+		__entry->namelen = clp->cl_name.len;
+		memcpy(__get_dynamic_array(name), clp->cl_name.data,
+			clp->cl_name.len);
+	),
+	TP_printk("nfs4_clientid %.*s already in use by %pISpc, client %08x:%08x",
+		__entry->namelen, __get_str(name), __entry->addr,
+		__entry->cl_boot, __entry->cl_id)
+)
+
 TRACE_DEFINE_ENUM(NFSD_FILE_HASHED);
 TRACE_DEFINE_ENUM(NFSD_FILE_PENDING);
 TRACE_DEFINE_ENUM(NFSD_FILE_BREAK_READ);
-- 
cgit v1.2.3


From 1eace0d1e9d0db393be87572f79f6a07405461e9 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sun, 5 Apr 2020 14:15:29 -0400
Subject: NFSD: Add tracepoints for monitoring NFSD callbacks

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfs4callback.c |  37 +++++++-----
 fs/nfsd/nfs4state.c    |   6 +-
 fs/nfsd/trace.h        | 153 +++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 177 insertions(+), 19 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 5cf91322de0f..966ca75418c8 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -38,6 +38,7 @@
 #include "nfsd.h"
 #include "state.h"
 #include "netns.h"
+#include "trace.h"
 #include "xdr4cb.h"
 #include "xdr4.h"
 
@@ -904,16 +905,20 @@ static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *c
 
 	if (clp->cl_minorversion == 0) {
 		if (!clp->cl_cred.cr_principal &&
-				(clp->cl_cred.cr_flavor >= RPC_AUTH_GSS_KRB5))
+		    (clp->cl_cred.cr_flavor >= RPC_AUTH_GSS_KRB5)) {
+			trace_nfsd_cb_setup_err(clp, -EINVAL);
 			return -EINVAL;
+		}
 		args.client_name = clp->cl_cred.cr_principal;
 		args.prognumber	= conn->cb_prog;
 		args.protocol = XPRT_TRANSPORT_TCP;
 		args.authflavor = clp->cl_cred.cr_flavor;
 		clp->cl_cb_ident = conn->cb_ident;
 	} else {
-		if (!conn->cb_xprt)
+		if (!conn->cb_xprt) {
+			trace_nfsd_cb_setup_err(clp, -EINVAL);
 			return -EINVAL;
+		}
 		clp->cl_cb_conn.cb_xprt = conn->cb_xprt;
 		clp->cl_cb_session = ses;
 		args.bc_xprt = conn->cb_xprt;
@@ -925,32 +930,27 @@ static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *c
 	/* Create RPC client */
 	client = rpc_create(&args);
 	if (IS_ERR(client)) {
-		dprintk("NFSD: couldn't create callback client: %ld\n",
-			PTR_ERR(client));
+		trace_nfsd_cb_setup_err(clp, PTR_ERR(client));
 		return PTR_ERR(client);
 	}
 	cred = get_backchannel_cred(clp, client, ses);
 	if (!cred) {
+		trace_nfsd_cb_setup_err(clp, -ENOMEM);
 		rpc_shutdown_client(client);
 		return -ENOMEM;
 	}
 	clp->cl_cb_client = client;
 	clp->cl_cb_cred = cred;
+	trace_nfsd_cb_setup(clp);
 	return 0;
 }
 
-static void warn_no_callback_path(struct nfs4_client *clp, int reason)
-{
-	dprintk("NFSD: warning: no callback path to client %.*s: error %d\n",
-		(int)clp->cl_name.len, clp->cl_name.data, reason);
-}
-
 static void nfsd4_mark_cb_down(struct nfs4_client *clp, int reason)
 {
 	if (test_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_flags))
 		return;
 	clp->cl_cb_state = NFSD4_CB_DOWN;
-	warn_no_callback_path(clp, reason);
+	trace_nfsd_cb_state(clp);
 }
 
 static void nfsd4_mark_cb_fault(struct nfs4_client *clp, int reason)
@@ -958,17 +958,20 @@ static void nfsd4_mark_cb_fault(struct nfs4_client *clp, int reason)
 	if (test_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_flags))
 		return;
 	clp->cl_cb_state = NFSD4_CB_FAULT;
-	warn_no_callback_path(clp, reason);
+	trace_nfsd_cb_state(clp);
 }
 
 static void nfsd4_cb_probe_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_client *clp = container_of(calldata, struct nfs4_client, cl_cb_null);
 
+	trace_nfsd_cb_done(clp, task->tk_status);
 	if (task->tk_status)
 		nfsd4_mark_cb_down(clp, task->tk_status);
-	else
+	else {
 		clp->cl_cb_state = NFSD4_CB_UP;
+		trace_nfsd_cb_state(clp);
+	}
 }
 
 static void nfsd4_cb_probe_release(void *calldata)
@@ -993,6 +996,7 @@ static const struct rpc_call_ops nfsd4_cb_probe_ops = {
 void nfsd4_probe_callback(struct nfs4_client *clp)
 {
 	clp->cl_cb_state = NFSD4_CB_UNKNOWN;
+	trace_nfsd_cb_state(clp);
 	set_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_flags);
 	nfsd4_run_cb(&clp->cl_cb_null);
 }
@@ -1009,6 +1013,7 @@ void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn *conn)
 	spin_lock(&clp->cl_lock);
 	memcpy(&clp->cl_cb_conn, conn, sizeof(struct nfs4_cb_conn));
 	spin_unlock(&clp->cl_lock);
+	trace_nfsd_cb_state(clp);
 }
 
 /*
@@ -1165,8 +1170,7 @@ static void nfsd4_cb_done(struct rpc_task *task, void *calldata)
 	struct nfsd4_callback *cb = calldata;
 	struct nfs4_client *clp = cb->cb_clp;
 
-	dprintk("%s: minorversion=%d\n", __func__,
-		clp->cl_minorversion);
+	trace_nfsd_cb_done(clp, task->tk_status);
 
 	if (!nfsd4_cb_sequence_done(task, cb))
 		return;
@@ -1271,6 +1275,7 @@ static void nfsd4_process_cb_update(struct nfsd4_callback *cb)
 	 * kill the old client:
 	 */
 	if (clp->cl_cb_client) {
+		trace_nfsd_cb_shutdown(clp);
 		rpc_shutdown_client(clp->cl_cb_client);
 		clp->cl_cb_client = NULL;
 		put_cred(clp->cl_cb_cred);
@@ -1314,6 +1319,8 @@ nfsd4_run_cb_work(struct work_struct *work)
 	struct rpc_clnt *clnt;
 	int flags;
 
+	trace_nfsd_cb_work(clp, cb->cb_msg.rpc_proc->p_name);
+
 	if (cb->cb_need_restart) {
 		cb->cb_need_restart = false;
 	} else {
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 04d80f9e2f91..88433be551b7 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2842,14 +2842,12 @@ gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se, struct svc_r
 	conn->cb_prog = se->se_callback_prog;
 	conn->cb_ident = se->se_callback_ident;
 	memcpy(&conn->cb_saddr, &rqstp->rq_daddr, rqstp->rq_daddrlen);
+	trace_nfsd_cb_args(clp, conn);
 	return;
 out_err:
 	conn->cb_addr.ss_family = AF_UNSPEC;
 	conn->cb_addrlen = 0;
-	dprintk("NFSD: this client (clientid %08x/%08x) "
-		"will not receive delegations\n",
-		clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id);
-
+	trace_nfsd_cb_nodelegs(clp);
 	return;
 }
 
diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
index 7237fe2f3de9..1861db1bdc67 100644
--- a/fs/nfsd/trace.h
+++ b/fs/nfsd/trace.h
@@ -624,6 +624,159 @@ TRACE_EVENT(nfsd_drc_mismatch,
 		__entry->ingress)
 );
 
+TRACE_EVENT(nfsd_cb_args,
+	TP_PROTO(
+		const struct nfs4_client *clp,
+		const struct nfs4_cb_conn *conn
+	),
+	TP_ARGS(clp, conn),
+	TP_STRUCT__entry(
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+		__field(u32, prog)
+		__field(u32, ident)
+		__array(unsigned char, addr, sizeof(struct sockaddr_in6))
+	),
+	TP_fast_assign(
+		__entry->cl_boot = clp->cl_clientid.cl_boot;
+		__entry->cl_id = clp->cl_clientid.cl_id;
+		__entry->prog = conn->cb_prog;
+		__entry->ident = conn->cb_ident;
+		memcpy(__entry->addr, &conn->cb_addr,
+			sizeof(struct sockaddr_in6));
+	),
+	TP_printk("client %08x:%08x callback addr=%pISpc prog=%u ident=%u",
+		__entry->cl_boot, __entry->cl_id,
+		__entry->addr, __entry->prog, __entry->ident)
+);
+
+TRACE_EVENT(nfsd_cb_nodelegs,
+	TP_PROTO(const struct nfs4_client *clp),
+	TP_ARGS(clp),
+	TP_STRUCT__entry(
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+	),
+	TP_fast_assign(
+		__entry->cl_boot = clp->cl_clientid.cl_boot;
+		__entry->cl_id = clp->cl_clientid.cl_id;
+	),
+	TP_printk("client %08x:%08x", __entry->cl_boot, __entry->cl_id)
+)
+
+TRACE_DEFINE_ENUM(NFSD4_CB_UP);
+TRACE_DEFINE_ENUM(NFSD4_CB_UNKNOWN);
+TRACE_DEFINE_ENUM(NFSD4_CB_DOWN);
+TRACE_DEFINE_ENUM(NFSD4_CB_FAULT);
+
+#define show_cb_state(val)						\
+	__print_symbolic(val,						\
+		{ NFSD4_CB_UP,		"UP" },				\
+		{ NFSD4_CB_UNKNOWN,	"UNKNOWN" },			\
+		{ NFSD4_CB_DOWN,	"DOWN" },			\
+		{ NFSD4_CB_FAULT,	"FAULT"})
+
+DECLARE_EVENT_CLASS(nfsd_cb_class,
+	TP_PROTO(const struct nfs4_client *clp),
+	TP_ARGS(clp),
+	TP_STRUCT__entry(
+		__field(unsigned long, state)
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+		__array(unsigned char, addr, sizeof(struct sockaddr_in6))
+	),
+	TP_fast_assign(
+		__entry->state = clp->cl_cb_state;
+		__entry->cl_boot = clp->cl_clientid.cl_boot;
+		__entry->cl_id = clp->cl_clientid.cl_id;
+		memcpy(__entry->addr, &clp->cl_cb_conn.cb_addr,
+			sizeof(struct sockaddr_in6));
+	),
+	TP_printk("addr=%pISpc client %08x:%08x state=%s",
+		__entry->addr, __entry->cl_boot, __entry->cl_id,
+		show_cb_state(__entry->state))
+);
+
+#define DEFINE_NFSD_CB_EVENT(name)			\
+DEFINE_EVENT(nfsd_cb_class, nfsd_cb_##name,		\
+	TP_PROTO(const struct nfs4_client *clp),	\
+	TP_ARGS(clp))
+
+DEFINE_NFSD_CB_EVENT(setup);
+DEFINE_NFSD_CB_EVENT(state);
+DEFINE_NFSD_CB_EVENT(shutdown);
+
+TRACE_EVENT(nfsd_cb_setup_err,
+	TP_PROTO(
+		const struct nfs4_client *clp,
+		long error
+	),
+	TP_ARGS(clp, error),
+	TP_STRUCT__entry(
+		__field(long, error)
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+		__array(unsigned char, addr, sizeof(struct sockaddr_in6))
+	),
+	TP_fast_assign(
+		__entry->error = error;
+		__entry->cl_boot = clp->cl_clientid.cl_boot;
+		__entry->cl_id = clp->cl_clientid.cl_id;
+		memcpy(__entry->addr, &clp->cl_cb_conn.cb_addr,
+			sizeof(struct sockaddr_in6));
+	),
+	TP_printk("addr=%pISpc client %08x:%08x error=%ld",
+		__entry->addr, __entry->cl_boot, __entry->cl_id, __entry->error)
+);
+
+TRACE_EVENT(nfsd_cb_work,
+	TP_PROTO(
+		const struct nfs4_client *clp,
+		const char *procedure
+	),
+	TP_ARGS(clp, procedure),
+	TP_STRUCT__entry(
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+		__string(procedure, procedure)
+		__array(unsigned char, addr, sizeof(struct sockaddr_in6))
+	),
+	TP_fast_assign(
+		__entry->cl_boot = clp->cl_clientid.cl_boot;
+		__entry->cl_id = clp->cl_clientid.cl_id;
+		__assign_str(procedure, procedure)
+		memcpy(__entry->addr, &clp->cl_cb_conn.cb_addr,
+			sizeof(struct sockaddr_in6));
+	),
+	TP_printk("addr=%pISpc client %08x:%08x procedure=%s",
+		__entry->addr, __entry->cl_boot, __entry->cl_id,
+		__get_str(procedure))
+);
+
+TRACE_EVENT(nfsd_cb_done,
+	TP_PROTO(
+		const struct nfs4_client *clp,
+		int status
+	),
+	TP_ARGS(clp, status),
+	TP_STRUCT__entry(
+		__field(u32, cl_boot)
+		__field(u32, cl_id)
+		__field(int, status)
+		__array(unsigned char, addr, sizeof(struct sockaddr_in6))
+	),
+	TP_fast_assign(
+		__entry->cl_boot = clp->cl_clientid.cl_boot;
+		__entry->cl_id = clp->cl_clientid.cl_id;
+		__entry->status = status;
+		memcpy(__entry->addr, &clp->cl_cb_conn.cb_addr,
+			sizeof(struct sockaddr_in6));
+	),
+	TP_printk("addr=%pISpc client %08x:%08x status=%d",
+		__entry->addr, __entry->cl_boot, __entry->cl_id,
+		__entry->status)
+);
+
 #endif /* _NFSD_TRACE_H */
 
 #undef TRACE_INCLUDE_PATH
-- 
cgit v1.2.3


From 45f56da8224e6bc472320b154ca8821e50fc1819 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Sun, 5 Apr 2020 14:40:25 -0400
Subject: NFSD: Squash an annoying compiler warning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Clean up: Fix gcc empty-body warning when -Wextra is used.

../fs/nfsd/nfs4state.c:3898:3: warning: suggest braces around empty body in an ‘else’ statement [-Wempty-body]

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfs4state.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 88433be551b7..8b83341cae6b 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -3881,12 +3881,11 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	unconf = find_unconfirmed_client_by_name(&clname, nn);
 	if (unconf)
 		unhash_client_locked(unconf);
+	/* We need to handle only case 1: probable callback update */
 	if (conf && same_verf(&conf->cl_verifier, &clverifier)) {
-		/* case 1: probable callback update */
 		copy_clid(new, conf);
 		gen_confirm(new, nn);
-	} else /* case 4 (new client) or cases 2, 3 (client reboot): */
-		;
+	}
 	new->cl_minorversion = 0;
 	gen_callback(new, setclid, rqstp);
 	add_to_unconfirmed(new);
-- 
cgit v1.2.3


From f2453978a4f2ddb1938fa80e9bf0c9d6252bd5f8 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 6 Apr 2020 16:01:42 -0400
Subject: NFSD: Fix improperly-formatted Doxygen comments

fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'file' not described in 'write_unlock_ip'
fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'buf' not described in 'write_unlock_ip'
fs/nfsd/nfsctl.c:256: warning: Function parameter or member 'size' not described in 'write_unlock_ip'
fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'file' not described in 'write_unlock_fs'
fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'buf' not described in 'write_unlock_fs'
fs/nfsd/nfsctl.c:295: warning: Function parameter or member 'size' not described in 'write_unlock_fs'
fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'file' not described in 'write_filehandle'
fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'buf' not described in 'write_filehandle'
fs/nfsd/nfsctl.c:352: warning: Function parameter or member 'size' not described in 'write_filehandle'
fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'file' not described in 'write_threads'
fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'buf' not described in 'write_threads'
fs/nfsd/nfsctl.c:434: warning: Function parameter or member 'size' not described in 'write_threads'
fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'file' not described in 'write_pool_threads'
fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'buf' not described in 'write_pool_threads'
fs/nfsd/nfsctl.c:478: warning: Function parameter or member 'size' not described in 'write_pool_threads'
fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'file' not described in 'write_versions'
fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'buf' not described in 'write_versions'
fs/nfsd/nfsctl.c:697: warning: Function parameter or member 'size' not described in 'write_versions'
fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'file' not described in 'write_ports'
fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'buf' not described in 'write_ports'
fs/nfsd/nfsctl.c:858: warning: Function parameter or member 'size' not described in 'write_ports'
fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'file' not described in 'write_maxblksize'
fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'buf' not described in 'write_maxblksize'
fs/nfsd/nfsctl.c:892: warning: Function parameter or member 'size' not described in 'write_maxblksize'
fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'file' not described in 'write_maxconn'
fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'buf' not described in 'write_maxconn'
fs/nfsd/nfsctl.c:941: warning: Function parameter or member 'size' not described in 'write_maxconn'
fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'file' not described in 'write_leasetime'
fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'buf' not described in 'write_leasetime'
fs/nfsd/nfsctl.c:1023: warning: Function parameter or member 'size' not described in 'write_leasetime'
fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'file' not described in 'write_gracetime'
fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'buf' not described in 'write_gracetime'
fs/nfsd/nfsctl.c:1039: warning: Function parameter or member 'size' not described in 'write_gracetime'
fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'file' not described in 'write_recoverydir'
fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'buf' not described in 'write_recoverydir'
fs/nfsd/nfsctl.c:1094: warning: Function parameter or member 'size' not described in 'write_recoverydir'
fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'file' not described in 'write_v4_end_grace'
fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'buf' not described in 'write_v4_end_grace'
fs/nfsd/nfsctl.c:1125: warning: Function parameter or member 'size' not described in 'write_v4_end_grace'

fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'nss' not described in 'nfsd4_interssc_connect'
fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'rqstp' not described in 'nfsd4_interssc_connect'
fs/nfsd/nfs4proc.c:1164: warning: Function parameter or member 'mount' not described in 'nfsd4_interssc_connect'
fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'rqstp' not described in 'nfsd4_setup_inter_ssc'
fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'cstate' not described in 'nfsd4_setup_inter_ssc'
fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'copy' not described in 'nfsd4_setup_inter_ssc'
fs/nfsd/nfs4proc.c:1262: warning: Function parameter or member 'mount' not described in 'nfsd4_setup_inter_ssc'

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 fs/nfsd/nfs4proc.c |  7 +++----
 fs/nfsd/nfsctl.c   | 26 +++++++++++++-------------
 2 files changed, 16 insertions(+), 17 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 0e75f7fb5fec..ec1b28d08c1c 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -1155,7 +1155,7 @@ extern void nfs_sb_deactive(struct super_block *sb);
 
 #define NFSD42_INTERSSC_MOUNTOPS "vers=4.2,addr=%s,sec=sys"
 
-/**
+/*
  * Support one copy source server for now.
  */
 static __be32
@@ -1245,10 +1245,9 @@ nfsd4_interssc_disconnect(struct vfsmount *ss_mnt)
 	mntput(ss_mnt);
 }
 
-/**
- * nfsd4_setup_inter_ssc
- *
+/*
  * Verify COPY destination stateid.
+ *
  * Connect to the source server with NFSv4.1.
  * Create the source struct file for nfsd_copy_range.
  * Called with COPY cstate:
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 3bb2db947d29..b48eac3bb72b 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -238,7 +238,7 @@ static inline struct net *netns(struct file *file)
 	return file_inode(file)->i_sb->s_fs_info;
 }
 
-/**
+/*
  * write_unlock_ip - Release all locks used by a client
  *
  * Experimental.
@@ -277,7 +277,7 @@ static ssize_t write_unlock_ip(struct file *file, char *buf, size_t size)
 	return nlmsvc_unlock_all_by_ip(sap);
 }
 
-/**
+/*
  * write_unlock_fs - Release all locks on a local file system
  *
  * Experimental.
@@ -327,7 +327,7 @@ static ssize_t write_unlock_fs(struct file *file, char *buf, size_t size)
 	return error;
 }
 
-/**
+/*
  * write_filehandle - Get a variable-length NFS file handle by path
  *
  * On input, the buffer contains a '\n'-terminated C string comprised of
@@ -402,7 +402,7 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size)
 	return mesg - buf;	
 }
 
-/**
+/*
  * write_threads - Start NFSD, or report the current number of running threads
  *
  * Input:
@@ -452,7 +452,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size)
 	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%d\n", rv);
 }
 
-/**
+/*
  * write_pool_threads - Set or report the current number of threads per pool
  *
  * Input:
@@ -661,7 +661,7 @@ out:
 	return tlen + len;
 }
 
-/**
+/*
  * write_versions - Set or report the available NFS protocol versions
  *
  * Input:
@@ -811,7 +811,7 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size,
 	return -EINVAL;
 }
 
-/**
+/*
  * write_ports - Pass a socket file descriptor or transport name to listen on
  *
  * Input:
@@ -867,7 +867,7 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size)
 
 int nfsd_max_blksize;
 
-/**
+/*
  * write_maxblksize - Set or report the current NFS blksize
  *
  * Input:
@@ -917,7 +917,7 @@ static ssize_t write_maxblksize(struct file *file, char *buf, size_t size)
 							nfsd_max_blksize);
 }
 
-/**
+/*
  * write_maxconn - Set or report the current max number of connections
  *
  * Input:
@@ -998,7 +998,7 @@ static ssize_t nfsd4_write_time(struct file *file, char *buf, size_t size,
 	return rv;
 }
 
-/**
+/*
  * write_leasetime - Set or report the current NFSv4 lease time
  *
  * Input:
@@ -1025,7 +1025,7 @@ static ssize_t write_leasetime(struct file *file, char *buf, size_t size)
 	return nfsd4_write_time(file, buf, size, &nn->nfsd4_lease, nn);
 }
 
-/**
+/*
  * write_gracetime - Set or report current NFSv4 grace period time
  *
  * As above, but sets the time of the NFSv4 grace period.
@@ -1069,7 +1069,7 @@ static ssize_t __write_recoverydir(struct file *file, char *buf, size_t size,
 							nfs4_recoverydir());
 }
 
-/**
+/*
  * write_recoverydir - Set or report the pathname of the recovery directory
  *
  * Input:
@@ -1101,7 +1101,7 @@ static ssize_t write_recoverydir(struct file *file, char *buf, size_t size)
 	return rv;
 }
 
-/**
+/*
  * write_v4_end_grace - release grace period for nfsd's v4.x lock manager
  *
  * Input:
-- 
cgit v1.2.3


From a4abc6b12eb1f7a533c2e7484cfa555454ff0977 Mon Sep 17 00:00:00 2001
From: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Date: Mon, 25 May 2020 22:15:41 +0800
Subject: nfsd: Fix svc_xprt refcnt leak when setup callback client failed

nfsd4_process_cb_update() invokes svc_xprt_get(), which increases the
refcount of the "c->cn_xprt".

The reference counting issue happens in one exception handling path of
nfsd4_process_cb_update(). When setup callback client failed, the
function forgets to decrease the refcnt increased by svc_xprt_get(),
causing a refcnt leak.

Fix this issue by calling svc_xprt_put() when setup callback client
failed.

Signed-off-by: Xiyu Yang <xiyuyang19@fudan.edu.cn>
Signed-off-by: Xin Tan <tanxin.ctf@gmail.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfs4callback.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 966ca75418c8..7fbe9840a03e 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -1306,6 +1306,8 @@ static void nfsd4_process_cb_update(struct nfsd4_callback *cb)
 	err = setup_callback_client(clp, &conn, ses);
 	if (err) {
 		nfsd4_mark_cb_down(clp, err);
+		if (c)
+			svc_xprt_put(c->cn_xprt);
 		return;
 	}
 }
-- 
cgit v1.2.3


From 027690c75e8fd91b60a634d31c4891a6e39d45bd Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Mon, 1 Jun 2020 17:44:45 -0400
Subject: nfsd4: make drc_slab global, not per-net

I made every global per-network-namespace instead.  But perhaps doing
that to this slab was a step too far.

The kmem_cache_create call in our net init method also seems to be
responsible for this lockdep warning:

[   45.163710] Unable to find swap-space signature
[   45.375718] trinity-c1 (855): attempted to duplicate a private mapping with mremap.  This is not supported.
[   46.055744] futex_wake_op: trinity-c1 tries to shift op by -209; fix this program
[   51.011723]
[   51.013378] ======================================================
[   51.013875] WARNING: possible circular locking dependency detected
[   51.014378] 5.2.0-rc2 #1 Not tainted
[   51.014672] ------------------------------------------------------
[   51.015182] trinity-c2/886 is trying to acquire lock:
[   51.015593] 000000005405f099 (slab_mutex){+.+.}, at: slab_attr_store+0xa2/0x130
[   51.016190]
[   51.016190] but task is already holding lock:
[   51.016652] 00000000ac662005 (kn->count#43){++++}, at: kernfs_fop_write+0x286/0x500
[   51.017266]
[   51.017266] which lock already depends on the new lock.
[   51.017266]
[   51.017909]
[   51.017909] the existing dependency chain (in reverse order) is:
[   51.018497]
[   51.018497] -> #1 (kn->count#43){++++}:
[   51.018956]        __lock_acquire+0x7cf/0x1a20
[   51.019317]        lock_acquire+0x17d/0x390
[   51.019658]        __kernfs_remove+0x892/0xae0
[   51.020020]        kernfs_remove_by_name_ns+0x78/0x110
[   51.020435]        sysfs_remove_link+0x55/0xb0
[   51.020832]        sysfs_slab_add+0xc1/0x3e0
[   51.021332]        __kmem_cache_create+0x155/0x200
[   51.021720]        create_cache+0xf5/0x320
[   51.022054]        kmem_cache_create_usercopy+0x179/0x320
[   51.022486]        kmem_cache_create+0x1a/0x30
[   51.022867]        nfsd_reply_cache_init+0x278/0x560
[   51.023266]        nfsd_init_net+0x20f/0x5e0
[   51.023623]        ops_init+0xcb/0x4b0
[   51.023928]        setup_net+0x2fe/0x670
[   51.024315]        copy_net_ns+0x30a/0x3f0
[   51.024653]        create_new_namespaces+0x3c5/0x820
[   51.025257]        unshare_nsproxy_namespaces+0xd1/0x240
[   51.025881]        ksys_unshare+0x506/0x9c0
[   51.026381]        __x64_sys_unshare+0x3a/0x50
[   51.026937]        do_syscall_64+0x110/0x10b0
[   51.027509]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   51.028175]
[   51.028175] -> #0 (slab_mutex){+.+.}:
[   51.028817]        validate_chain+0x1c51/0x2cc0
[   51.029422]        __lock_acquire+0x7cf/0x1a20
[   51.029947]        lock_acquire+0x17d/0x390
[   51.030438]        __mutex_lock+0x100/0xfa0
[   51.030995]        mutex_lock_nested+0x27/0x30
[   51.031516]        slab_attr_store+0xa2/0x130
[   51.032020]        sysfs_kf_write+0x11d/0x180
[   51.032529]        kernfs_fop_write+0x32a/0x500
[   51.033056]        do_loop_readv_writev+0x21d/0x310
[   51.033627]        do_iter_write+0x2e5/0x380
[   51.034148]        vfs_writev+0x170/0x310
[   51.034616]        do_pwritev+0x13e/0x160
[   51.035100]        __x64_sys_pwritev+0xa3/0x110
[   51.035633]        do_syscall_64+0x110/0x10b0
[   51.036200]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   51.036924]
[   51.036924] other info that might help us debug this:
[   51.036924]
[   51.037876]  Possible unsafe locking scenario:
[   51.037876]
[   51.038556]        CPU0                    CPU1
[   51.039130]        ----                    ----
[   51.039676]   lock(kn->count#43);
[   51.040084]                                lock(slab_mutex);
[   51.040597]                                lock(kn->count#43);
[   51.041062]   lock(slab_mutex);
[   51.041320]
[   51.041320]  *** DEADLOCK ***
[   51.041320]
[   51.041793] 3 locks held by trinity-c2/886:
[   51.042128]  #0: 000000001f55e152 (sb_writers#5){.+.+}, at: vfs_writev+0x2b9/0x310
[   51.042739]  #1: 00000000c7d6c034 (&of->mutex){+.+.}, at: kernfs_fop_write+0x25b/0x500
[   51.043400]  #2: 00000000ac662005 (kn->count#43){++++}, at: kernfs_fop_write+0x286/0x500

Reported-by: kernel test robot <lkp@intel.com>
Fixes: 3ba75830ce17 "drc containerization"
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/cache.h    |  2 ++
 fs/nfsd/netns.h    |  1 -
 fs/nfsd/nfscache.c | 29 +++++++++++++++++------------
 fs/nfsd/nfsctl.c   |  6 ++++++
 4 files changed, 25 insertions(+), 13 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/cache.h b/fs/nfsd/cache.h
index 10ec5ecdf117..65c331f75e9c 100644
--- a/fs/nfsd/cache.h
+++ b/fs/nfsd/cache.h
@@ -78,6 +78,8 @@ enum {
 /* Checksum this amount of the request */
 #define RC_CSUMLEN		(256U)
 
+int	nfsd_drc_slab_create(void);
+void	nfsd_drc_slab_free(void);
 int	nfsd_reply_cache_init(struct nfsd_net *);
 void	nfsd_reply_cache_shutdown(struct nfsd_net *);
 int	nfsd_cache_lookup(struct svc_rqst *);
diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h
index 09aa545825bd..9217cb64bf0e 100644
--- a/fs/nfsd/netns.h
+++ b/fs/nfsd/netns.h
@@ -139,7 +139,6 @@ struct nfsd_net {
 	 * Duplicate reply cache
 	 */
 	struct nfsd_drc_bucket   *drc_hashtbl;
-	struct kmem_cache        *drc_slab;
 
 	/* max number of entries allowed in the cache */
 	unsigned int             max_drc_entries;
diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index 945d2f5e760e..f30a7def7899 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -35,6 +35,8 @@ struct nfsd_drc_bucket {
 	spinlock_t cache_lock;
 };
 
+static struct kmem_cache	*drc_slab;
+
 static int	nfsd_cache_append(struct svc_rqst *rqstp, struct kvec *vec);
 static unsigned long nfsd_reply_cache_count(struct shrinker *shrink,
 					    struct shrink_control *sc);
@@ -94,7 +96,7 @@ nfsd_reply_cache_alloc(struct svc_rqst *rqstp, __wsum csum,
 {
 	struct svc_cacherep	*rp;
 
-	rp = kmem_cache_alloc(nn->drc_slab, GFP_KERNEL);
+	rp = kmem_cache_alloc(drc_slab, GFP_KERNEL);
 	if (rp) {
 		rp->c_state = RC_UNUSED;
 		rp->c_type = RC_NOCACHE;
@@ -128,7 +130,7 @@ nfsd_reply_cache_free_locked(struct nfsd_drc_bucket *b, struct svc_cacherep *rp,
 		atomic_dec(&nn->num_drc_entries);
 		nn->drc_mem_usage -= sizeof(*rp);
 	}
-	kmem_cache_free(nn->drc_slab, rp);
+	kmem_cache_free(drc_slab, rp);
 }
 
 static void
@@ -140,6 +142,18 @@ nfsd_reply_cache_free(struct nfsd_drc_bucket *b, struct svc_cacherep *rp,
 	spin_unlock(&b->cache_lock);
 }
 
+int nfsd_drc_slab_create(void)
+{
+	drc_slab = kmem_cache_create("nfsd_drc",
+				sizeof(struct svc_cacherep), 0, 0, NULL);
+	return drc_slab ? 0: -ENOMEM;
+}
+
+void nfsd_drc_slab_free(void)
+{
+	kmem_cache_destroy(drc_slab);
+}
+
 int nfsd_reply_cache_init(struct nfsd_net *nn)
 {
 	unsigned int hashsize;
@@ -158,18 +172,13 @@ int nfsd_reply_cache_init(struct nfsd_net *nn)
 	if (status)
 		goto out_nomem;
 
-	nn->drc_slab = kmem_cache_create("nfsd_drc",
-				sizeof(struct svc_cacherep), 0, 0, NULL);
-	if (!nn->drc_slab)
-		goto out_shrinker;
-
 	nn->drc_hashtbl = kcalloc(hashsize,
 				sizeof(*nn->drc_hashtbl), GFP_KERNEL);
 	if (!nn->drc_hashtbl) {
 		nn->drc_hashtbl = vzalloc(array_size(hashsize,
 						 sizeof(*nn->drc_hashtbl)));
 		if (!nn->drc_hashtbl)
-			goto out_slab;
+			goto out_shrinker;
 	}
 
 	for (i = 0; i < hashsize; i++) {
@@ -179,8 +188,6 @@ int nfsd_reply_cache_init(struct nfsd_net *nn)
 	nn->drc_hashsize = hashsize;
 
 	return 0;
-out_slab:
-	kmem_cache_destroy(nn->drc_slab);
 out_shrinker:
 	unregister_shrinker(&nn->nfsd_reply_cache_shrinker);
 out_nomem:
@@ -208,8 +215,6 @@ void nfsd_reply_cache_shutdown(struct nfsd_net *nn)
 	nn->drc_hashtbl = NULL;
 	nn->drc_hashsize = 0;
 
-	kmem_cache_destroy(nn->drc_slab);
-	nn->drc_slab = NULL;
 }
 
 /*
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index b48eac3bb72b..b68e96681522 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1533,6 +1533,9 @@ static int __init init_nfsd(void)
 		goto out_free_slabs;
 	nfsd_fault_inject_init(); /* nfsd fault injection controls */
 	nfsd_stat_init();	/* Statistics */
+	retval = nfsd_drc_slab_create();
+	if (retval)
+		goto out_free_stat;
 	nfsd_lockd_init();	/* lockd->nfsd callbacks */
 	retval = create_proc_exports_entry();
 	if (retval)
@@ -1546,6 +1549,8 @@ out_free_all:
 	remove_proc_entry("fs/nfs", NULL);
 out_free_lockd:
 	nfsd_lockd_shutdown();
+	nfsd_drc_slab_free();
+out_free_stat:
 	nfsd_stat_shutdown();
 	nfsd_fault_inject_cleanup();
 	nfsd4_exit_pnfs();
@@ -1560,6 +1565,7 @@ out_unregister_pernet:
 
 static void __exit exit_nfsd(void)
 {
+	nfsd_drc_slab_free();
 	remove_proc_entry("fs/nfs/exports", NULL);
 	remove_proc_entry("fs/nfs", NULL);
 	nfsd_stat_shutdown();
-- 
cgit v1.2.3


From c25bf185e57213b54ea0d632ac04907310993433 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@redhat.com>
Date: Wed, 3 Jun 2020 11:12:32 -0400
Subject: nfsd: safer handling of corrupted c_type

This can only happen if there's a bug somewhere, so let's make it a WARN
not a printk.  Also, I think it's safest to ignore the corruption rather
than trying to fix it by removing a cache entry.

Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfsd/nfscache.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'fs/nfsd')

diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index f30a7def7899..0a0cf1fd77d3 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -477,8 +477,7 @@ found_entry:
 		rtn = RC_REPLY;
 		break;
 	default:
-		printk(KERN_WARNING "nfsd: bad repcache type %d\n", rp->c_type);
-		nfsd_reply_cache_free_locked(b, rp, nn);
+		WARN_ONCE(1, "nfsd: bad repcache type %d\n", rp->c_type);
 	}
 
 out_trace:
-- 
cgit v1.2.3