Merge tag 'nfsd-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux

Pull nfsd updates from Chuck Lever:
 "We introduce 'courteous server' in this release. Previously NFSD would
  purge open and lock state for an unresponsive client after one lease
  period (typically 90 seconds). Now, after one lease period, another
  client can open and lock those files and the unresponsive client's
  lease is purged; otherwise if the unresponsive client's open and lock
  state is uncontended, the server retains that open and lock state for
  up to 24 hours, allowing the client's workload to resume after a
  lengthy network partition.

  A longstanding issue with NFSv4 file creation is also addressed.
  Previously a file creation can fail internally, returning an error to
  the client, but leave the newly created file in place as an artifact.
  The file creation code path has been reorganized so that internal
  failures and race conditions are less likely to result in an unwanted
  file creation.

  A fault injector has been added to help exercise paths that are run
  during kernel metadata cache invalidation. These caches contain
  information maintained by user space about exported filesystems. Many
  of our test workloads do not trigger cache invalidation.

  There is one patch that is needed to support PREEMPT_RT and a fix for
  an ancient 'sleep while spin-locked' splat that seems to have become
  easier to hit since v5.18-rc3"

* tag 'nfsd-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux: (36 commits)
  NFSD: nfsd_file_put() can sleep
  NFSD: Add documenting comment for nfsd4_release_lockowner()
  NFSD: Modernize nfsd4_release_lockowner()
  NFSD: Fix possible sleep during nfsd4_release_lockowner()
  nfsd: destroy percpu stats counters after reply cache shutdown
  nfsd: Fix null-ptr-deref in nfsd_fill_super()
  nfsd: Unregister the cld notifier when laundry_wq create failed
  SUNRPC: Use RMW bitops in single-threaded hot paths
  NFSD: Clean up the show_nf_flags() macro
  NFSD: Trace filecache opens
  NFSD: Move documenting comment for nfsd4_process_open2()
  NFSD: Fix whitespace
  NFSD: Remove dprintk call sites from tail of nfsd4_open()
  NFSD: Instantiate a struct file when creating a regular NFSv4 file
  NFSD: Clean up nfsd_open_verified()
  NFSD: Remove do_nfsd_create()
  NFSD: Refactor NFSv4 OPEN(CREATE)
  NFSD: Refactor NFSv3 CREATE
  NFSD: Refactor nfsd_create_setattr()
  NFSD: Avoid calling fh_drop_write() twice in do_nfsd_create()
  ...

9 files changed, 49 insertions, 41 deletions

diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c
index c2ba9d4cd2c7..bcd74dddbe2d 100644
--- a/net/sunrpc/auth_gss/svcauth_gss.c
+++ b/net/sunrpc/auth_gss/svcauth_gss.c
@@ -900,7 +900,7 @@ unwrap_integ_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct g
 	 * rejecting the server-computed MIC in this somewhat rare case,
 	 * do not use splice with the GSS integrity service.
 	 */
-	clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
+	__clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
 
 	/* Did we already verify the signature on the original pass through? */
 	if (rqstp->rq_deferred)
@@ -972,7 +972,7 @@ unwrap_priv_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct gs
 	int pad, remaining_len, offset;
 	u32 rseqno;
 
-	clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
+	__clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
 
 	priv_len = svc_getnl(&buf->head[0]);
 	if (rqstp->rq_deferred) {
diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index bb1177395b99..c3c693b51c94 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -33,7 +33,9 @@
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/rpc_pipe_fs.h>
 #include <trace/events/sunrpc.h>
+
 #include "netns.h"
+#include "fail.h"
 
 #define	 RPCDBG_FACILITY RPCDBG_CACHE
 
@@ -688,16 +690,30 @@ static void cache_limit_defers(void)
 		discard->revisit(discard, 1);
 }
 
+#if IS_ENABLED(CONFIG_FAIL_SUNRPC)
+static inline bool cache_defer_immediately(void)
+{
+	return !fail_sunrpc.ignore_cache_wait &&
+		should_fail(&fail_sunrpc.attr, 1);
+}
+#else
+static inline bool cache_defer_immediately(void)
+{
+	return false;
+}
+#endif
+
 /* Return true if and only if a deferred request is queued. */
 static bool cache_defer_req(struct cache_req *req, struct cache_head *item)
 {
 	struct cache_deferred_req *dreq;
 
-	if (req->thread_wait) {
+	if (!cache_defer_immediately()) {
 		cache_wait_req(req, item);
 		if (!test_bit(CACHE_PENDING, &item->flags))
 			return false;
 	}
+
 	dreq = req->defer(req);
 	if (dreq == NULL)
 		return false;
diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
index 7dc9cc929bfd..a176d5a0b0ee 100644
--- a/net/sunrpc/debugfs.c
+++ b/net/sunrpc/debugfs.c
@@ -262,6 +262,9 @@ static void fail_sunrpc_init(void)
 
 	debugfs_create_bool("ignore-server-disconnect", S_IFREG | 0600, dir,
 			    &fail_sunrpc.ignore_server_disconnect);
+
+	debugfs_create_bool("ignore-cache-wait", S_IFREG | 0600, dir,
+			    &fail_sunrpc.ignore_cache_wait);
 }
 #else
 static void fail_sunrpc_init(void)
diff --git a/net/sunrpc/fail.h b/net/sunrpc/fail.h
index 69dc30cc44b8..4b4b500df428 100644
--- a/net/sunrpc/fail.h
+++ b/net/sunrpc/fail.h
@@ -14,8 +14,8 @@ struct fail_sunrpc_attr {
 	struct fault_attr	attr;
 
 	bool			ignore_client_disconnect;
-
 	bool			ignore_server_disconnect;
+	bool			ignore_cache_wait;
 };
 
 extern struct fail_sunrpc_attr fail_sunrpc;
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 557004017548..7c9a0d0b1230 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -356,15 +356,21 @@ svc_pool_map_set_cpumask(struct task_struct *task, unsigned int pidx)
 	}
 }
 
-/*
- * Use the mapping mode to choose a pool for a given CPU.
- * Used when enqueueing an incoming RPC.  Always returns
- * a non-NULL pool pointer.
+/**
+ * svc_pool_for_cpu - Select pool to run a thread on this cpu
+ * @serv: An RPC service
+ *
+ * Use the active CPU and the svc_pool_map's mode setting to
+ * select the svc thread pool to use. Once initialized, the
+ * svc_pool_map does not change.
+ *
+ * Return value:
+ *   A pointer to an svc_pool
  */
-struct svc_pool *
-svc_pool_for_cpu(struct svc_serv *serv, int cpu)
+struct svc_pool *svc_pool_for_cpu(struct svc_serv *serv)
 {
 	struct svc_pool_map *m = &svc_pool_map;
+	int cpu = raw_smp_processor_id();
 	unsigned int pidx = 0;
 
 	if (serv->sv_nrpools <= 1)
@@ -1238,10 +1244,10 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
 		goto err_short_len;
 
 	/* Will be turned off by GSS integrity and privacy services */
-	set_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
+	__set_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
 	/* Will be turned off only when NFSv4 Sessions are used */
-	set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
-	clear_bit(RQ_DROPME, &rqstp->rq_flags);
+	__set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
+	__clear_bit(RQ_DROPME, &rqstp->rq_flags);
 
 	svc_putu32(resv, rqstp->rq_xid);
 
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index 5b59e2103526..2c4dd7ca95b0 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -448,7 +448,6 @@ void svc_xprt_enqueue(struct svc_xprt *xprt)
 {
 	struct svc_pool *pool;
 	struct svc_rqst	*rqstp = NULL;
-	int cpu;
 
 	if (!svc_xprt_ready(xprt))
 		return;
@@ -461,8 +460,7 @@ void svc_xprt_enqueue(struct svc_xprt *xprt)
 	if (test_and_set_bit(XPT_BUSY, &xprt->xpt_flags))
 		return;
 
-	cpu = get_cpu();
-	pool = svc_pool_for_cpu(xprt->xpt_server, cpu);
+	pool = svc_pool_for_cpu(xprt->xpt_server);
 
 	atomic_long_inc(&pool->sp_stats.packets);
 
@@ -485,7 +483,6 @@ void svc_xprt_enqueue(struct svc_xprt *xprt)
 	rqstp = NULL;
 out_unlock:
 	rcu_read_unlock();
-	put_cpu();
 	trace_svc_xprt_enqueue(xprt, rqstp);
 }
 EXPORT_SYMBOL_GPL(svc_xprt_enqueue);
@@ -1230,7 +1227,6 @@ static struct cache_deferred_req *svc_defer(struct cache_req *req)
 		dr->addrlen = rqstp->rq_addrlen;
 		dr->daddr = rqstp->rq_daddr;
 		dr->argslen = rqstp->rq_arg.len >> 2;
-		dr->xprt_hlen = rqstp->rq_xprt_hlen;
 		dr->xprt_ctxt = rqstp->rq_xprt_ctxt;
 		rqstp->rq_xprt_ctxt = NULL;
 
@@ -1242,7 +1238,7 @@ static struct cache_deferred_req *svc_defer(struct cache_req *req)
 	trace_svc_defer(rqstp);
 	svc_xprt_get(rqstp->rq_xprt);
 	dr->xprt = rqstp->rq_xprt;
-	set_bit(RQ_DROPME, &rqstp->rq_flags);
+	__set_bit(RQ_DROPME, &rqstp->rq_flags);
 
 	dr->handle.revisit = svc_revisit;
 	return &dr->handle;
@@ -1258,22 +1254,21 @@ static noinline int svc_deferred_recv(struct svc_rqst *rqstp)
 	trace_svc_defer_recv(dr);
 
 	/* setup iov_base past transport header */
-	rqstp->rq_arg.head[0].iov_base = dr->args + (dr->xprt_hlen>>2);
+	rqstp->rq_arg.head[0].iov_base = dr->args;
 	/* The iov_len does not include the transport header bytes */
-	rqstp->rq_arg.head[0].iov_len = (dr->argslen<<2) - dr->xprt_hlen;
+	rqstp->rq_arg.head[0].iov_len = dr->argslen << 2;
 	rqstp->rq_arg.page_len = 0;
 	/* The rq_arg.len includes the transport header bytes */
-	rqstp->rq_arg.len     = dr->argslen<<2;
+	rqstp->rq_arg.len     = dr->argslen << 2;
 	rqstp->rq_prot        = dr->prot;
 	memcpy(&rqstp->rq_addr, &dr->addr, dr->addrlen);
 	rqstp->rq_addrlen     = dr->addrlen;
 	/* Save off transport header len in case we get deferred again */
-	rqstp->rq_xprt_hlen   = dr->xprt_hlen;
 	rqstp->rq_daddr       = dr->daddr;
 	rqstp->rq_respages    = rqstp->rq_pages;
 	rqstp->rq_xprt_ctxt   = dr->xprt_ctxt;
 	svc_xprt_received(rqstp->rq_xprt);
-	return (dr->argslen<<2) - dr->xprt_hlen;
+	return dr->argslen << 2;
 }
 
 
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 45336e68bf79..2fc98fea59b4 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -117,15 +117,6 @@ static void svc_reclassify_socket(struct socket *sock)
  */
 static void svc_tcp_release_rqst(struct svc_rqst *rqstp)
 {
-	struct sk_buff *skb = rqstp->rq_xprt_ctxt;
-
-	if (skb) {
-		struct svc_sock *svsk =
-			container_of(rqstp->rq_xprt, struct svc_sock, sk_xprt);
-
-		rqstp->rq_xprt_ctxt = NULL;
-		skb_free_datagram_locked(svsk->sk_sk, skb);
-	}
 }
 
 /**
@@ -259,8 +250,6 @@ static ssize_t svc_tcp_read_msg(struct svc_rqst *rqstp, size_t buflen,
 	ssize_t len;
 	size_t t;
 
-	rqstp->rq_xprt_hlen = 0;
-
 	clear_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 
 	for (i = 0, t = 0; t < buflen; i++, t += PAGE_SIZE) {
@@ -309,9 +298,9 @@ static void svc_sock_setbufsize(struct svc_sock *svsk, unsigned int nreqs)
 static void svc_sock_secure_port(struct svc_rqst *rqstp)
 {
 	if (svc_port_is_privileged(svc_addr(rqstp)))
-		set_bit(RQ_SECURE, &rqstp->rq_flags);
+		__set_bit(RQ_SECURE, &rqstp->rq_flags);
 	else
-		clear_bit(RQ_SECURE, &rqstp->rq_flags);
+		__clear_bit(RQ_SECURE, &rqstp->rq_flags);
 }
 
 /*
@@ -1019,9 +1008,9 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 	rqstp->rq_xprt_ctxt   = NULL;
 	rqstp->rq_prot	      = IPPROTO_TCP;
 	if (test_bit(XPT_LOCAL, &svsk->sk_xprt.xpt_flags))
-		set_bit(RQ_LOCAL, &rqstp->rq_flags);
+		__set_bit(RQ_LOCAL, &rqstp->rq_flags);
 	else
-		clear_bit(RQ_LOCAL, &rqstp->rq_flags);
+		__clear_bit(RQ_LOCAL, &rqstp->rq_flags);
 
 	p = (__be32 *)rqstp->rq_arg.head[0].iov_base;
 	calldir = p[1];
diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index 864131a9fc6e..5242ad121450 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -831,7 +831,6 @@ int svc_rdma_recvfrom(struct svc_rqst *rqstp)
 		goto out_err;
 	if (ret == 0)
 		goto out_drop;
-	rqstp->rq_xprt_hlen = 0;
 
 	if (svc_rdma_is_reverse_direction_reply(xprt, ctxt))
 		goto out_backchannel;
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index 94b20fb47135..199fa012f18a 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -602,7 +602,7 @@ static int svc_rdma_has_wspace(struct svc_xprt *xprt)
 
 static void svc_rdma_secure_port(struct svc_rqst *rqstp)
 {
-	set_bit(RQ_SECURE, &rqstp->rq_flags);
+	__set_bit(RQ_SECURE, &rqstp->rq_flags);
 }
 
 static void svc_rdma_kill_temp_xprt(struct svc_xprt *xprt)