Merge branch 'for-3.8' of git://linux-nfs.org/~bfields/linux

Pull nfsd update from Bruce Fields:
 "Included this time:

   - more nfsd containerization work from Stanislav Kinsbursky: we're
     not quite there yet, but should be by 3.9.

   - NFSv4.1 progress: implementation of basic backchannel security
     negotiation and the mandatory BACKCHANNEL_CTL operation.  See

       http://wiki.linux-nfs.org/wiki/index.php/Server_4.0_and_4.1_issues

     for remaining TODO's

   - Fixes for some bugs that could be triggered by unusual compounds.
     Our xdr code wasn't designed with v4 compounds in mind, and it
     shows.  A more thorough rewrite is still a todo.

   - If you've ever seen "RPC: multiple fragments per record not
     supported" logged while using some sort of odd userland NFS client,
     that should now be fixed.

   - Further work from Jeff Layton on our mechanism for storing
     information about NFSv4 clients across reboots.

   - Further work from Bryan Schumaker on his fault-injection mechanism
     (which allows us to discard selective NFSv4 state, to excercise
     rarely-taken recovery code paths in the client.)

   - The usual mix of miscellaneous bugs and cleanup.

  Thanks to everyone who tested or contributed this cycle."

* 'for-3.8' of git://linux-nfs.org/~bfields/linux: (111 commits)
  nfsd4: don't leave freed stateid hashed
  nfsd4: free_stateid can use the current stateid
  nfsd4: cleanup: replace rq_resused count by rq_next_page pointer
  nfsd: warn on odd reply state in nfsd_vfs_read
  nfsd4: fix oops on unusual readlike compound
  nfsd4: disable zero-copy on non-final read ops
  svcrpc: fix some printks
  NFSD: Correct the size calculation in fault_inject_write
  NFSD: Pass correct buffer size to rpc_ntop
  nfsd: pass proper net to nfsd_destroy() from NFSd kthreads
  nfsd: simplify service shutdown
  nfsd: replace boolean nfsd_up flag by users counter
  nfsd: simplify NFSv4 state init and shutdown
  nfsd: introduce helpers for generic resources init and shutdown
  nfsd: make NFSd service structure allocated per net
  nfsd: make NFSd service boot time per-net
  nfsd: per-net NFSd up flag introduced
  nfsd: move per-net startup code to separated function
  nfsd: pass net to __write_ports() and down
  nfsd: pass net to nfsd_set_nrthreads()
  ...

5 files changed, 61 insertions, 60 deletions

diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index 411f332de0b3..795a0f4e920b 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -23,7 +23,6 @@
 #include <linux/errno.h>
 #include <linux/mutex.h>
 #include <linux/slab.h>
-#include <linux/nsproxy.h>
 #include <net/ipv6.h>
 
 #include <linux/sunrpc/clnt.h>
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index dfa4ba69ff45..dbf12ac5ecb7 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -20,7 +20,6 @@
 #include <linux/module.h>
 #include <linux/kthread.h>
 #include <linux/slab.h>
-#include <linux/nsproxy.h>
 
 #include <linux/sunrpc/types.h>
 #include <linux/sunrpc/xdr.h>
@@ -1041,7 +1040,7 @@ static void svc_unregister(const struct svc_serv *serv, struct net *net)
 }
 
 /*
- * Printk the given error with the address of the client that caused it.
+ * dprintk the given error with the address of the client that caused it.
  */
 static __printf(2, 3)
 void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
@@ -1055,8 +1054,7 @@ void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
 	vaf.fmt = fmt;
 	vaf.va = &args;
 
-	net_warn_ratelimited("svc: %s: %pV",
-			     svc_print_addr(rqstp, buf, sizeof(buf)), &vaf);
+	dprintk("svc: %s: %pV", svc_print_addr(rqstp, buf, sizeof(buf)), &vaf);
 
 	va_end(args);
 }
@@ -1305,7 +1303,7 @@ svc_process(struct svc_rqst *rqstp)
 	 * Setup response xdr_buf.
 	 * Initially it has just one page
 	 */
-	rqstp->rq_resused = 1;
+	rqstp->rq_next_page = &rqstp->rq_respages[1];
 	resv->iov_base = page_address(rqstp->rq_respages[0]);
 	resv->iov_len = 0;
 	rqstp->rq_res.pages = rqstp->rq_respages + 1;
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index cc3020d16789..0a148c9d2a5c 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -605,6 +605,7 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		rqstp->rq_respages = rqstp->rq_pages + 1 +
 			DIV_ROUND_UP(rqstp->rq_arg.page_len, PAGE_SIZE);
 	}
+	rqstp->rq_next_page = rqstp->rq_respages+1;
 
 	if (serv->sv_stats)
 		serv->sv_stats->netudpcnt++;
@@ -878,9 +879,9 @@ static unsigned int svc_tcp_restore_pages(struct svc_sock *svsk, struct svc_rqst
 {
 	unsigned int i, len, npages;
 
-	if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+	if (svsk->sk_datalen == 0)
 		return 0;
-	len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+	len = svsk->sk_datalen;
 	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
 	for (i = 0; i < npages; i++) {
 		if (rqstp->rq_pages[i] != NULL)
@@ -897,9 +898,9 @@ static void svc_tcp_save_pages(struct svc_sock *svsk, struct svc_rqst *rqstp)
 {
 	unsigned int i, len, npages;
 
-	if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+	if (svsk->sk_datalen == 0)
 		return;
-	len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+	len = svsk->sk_datalen;
 	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
 	for (i = 0; i < npages; i++) {
 		svsk->sk_pages[i] = rqstp->rq_pages[i];
@@ -911,9 +912,9 @@ static void svc_tcp_clear_pages(struct svc_sock *svsk)
 {
 	unsigned int i, len, npages;
 
-	if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+	if (svsk->sk_datalen == 0)
 		goto out;
-	len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+	len = svsk->sk_datalen;
 	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
 	for (i = 0; i < npages; i++) {
 		BUG_ON(svsk->sk_pages[i] == NULL);
@@ -922,13 +923,12 @@ static void svc_tcp_clear_pages(struct svc_sock *svsk)
 	}
 out:
 	svsk->sk_tcplen = 0;
+	svsk->sk_datalen = 0;
 }
 
 /*
- * Receive data.
+ * Receive fragment record header.
  * If we haven't gotten the record length yet, get the next four bytes.
- * Otherwise try to gobble up as much as possible up to the complete
- * record length.
  */
 static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
 {
@@ -954,32 +954,16 @@ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
 			return -EAGAIN;
 		}
 
-		svsk->sk_reclen = ntohl(svsk->sk_reclen);
-		if (!(svsk->sk_reclen & RPC_LAST_STREAM_FRAGMENT)) {
-			/* FIXME: technically, a record can be fragmented,
-			 *  and non-terminal fragments will not have the top
-			 *  bit set in the fragment length header.
-			 *  But apparently no known nfs clients send fragmented
-			 *  records. */
-			net_notice_ratelimited("RPC: multiple fragments per record not supported\n");
-			goto err_delete;
-		}
-
-		svsk->sk_reclen &= RPC_FRAGMENT_SIZE_MASK;
-		dprintk("svc: TCP record, %d bytes\n", svsk->sk_reclen);
-		if (svsk->sk_reclen > serv->sv_max_mesg) {
-			net_notice_ratelimited("RPC: fragment too large: 0x%08lx\n",
-					       (unsigned long)svsk->sk_reclen);
+		dprintk("svc: TCP record, %d bytes\n", svc_sock_reclen(svsk));
+		if (svc_sock_reclen(svsk) + svsk->sk_datalen >
+							serv->sv_max_mesg) {
+			net_notice_ratelimited("RPC: fragment too large: %d\n",
+					svc_sock_reclen(svsk));
 			goto err_delete;
 		}
 	}
 
-	if (svsk->sk_reclen < 8)
-		goto err_delete; /* client is nuts. */
-
-	len = svsk->sk_reclen;
-
-	return len;
+	return svc_sock_reclen(svsk);
 error:
 	dprintk("RPC: TCP recv_record got %d\n", len);
 	return len;
@@ -1023,7 +1007,7 @@ static int receive_cb_reply(struct svc_sock *svsk, struct svc_rqst *rqstp)
 	if (dst->iov_len < src->iov_len)
 		return -EAGAIN; /* whatever; just giving up. */
 	memcpy(dst->iov_base, src->iov_base, src->iov_len);
-	xprt_complete_rqst(req->rq_task, svsk->sk_reclen);
+	xprt_complete_rqst(req->rq_task, rqstp->rq_arg.len);
 	rqstp->rq_arg.len = 0;
 	return 0;
 }
@@ -1042,6 +1026,17 @@ static int copy_pages_to_kvecs(struct kvec *vec, struct page **pages, int len)
 	return i;
 }
 
+static void svc_tcp_fragment_received(struct svc_sock *svsk)
+{
+	/* If we have more data, signal svc_xprt_enqueue() to try again */
+	if (svc_recv_available(svsk) > sizeof(rpc_fraghdr))
+		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
+	dprintk("svc: TCP %s record (%d bytes)\n",
+		svc_sock_final_rec(svsk) ? "final" : "nonfinal",
+		svc_sock_reclen(svsk));
+	svsk->sk_tcplen = 0;
+	svsk->sk_reclen = 0;
+}
 
 /*
  * Receive data from a TCP socket.
@@ -1068,29 +1063,39 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 		goto error;
 
 	base = svc_tcp_restore_pages(svsk, rqstp);
-	want = svsk->sk_reclen - base;
+	want = svc_sock_reclen(svsk) - (svsk->sk_tcplen - sizeof(rpc_fraghdr));
 
 	vec = rqstp->rq_vec;
 
 	pnum = copy_pages_to_kvecs(&vec[0], &rqstp->rq_pages[0],
-						svsk->sk_reclen);
+						svsk->sk_datalen + want);
 
 	rqstp->rq_respages = &rqstp->rq_pages[pnum];
+	rqstp->rq_next_page = rqstp->rq_respages + 1;
 
 	/* Now receive data */
 	len = svc_partial_recvfrom(rqstp, vec, pnum, want, base);
-	if (len >= 0)
+	if (len >= 0) {
 		svsk->sk_tcplen += len;
-	if (len != want) {
+		svsk->sk_datalen += len;
+	}
+	if (len != want || !svc_sock_final_rec(svsk)) {
 		svc_tcp_save_pages(svsk, rqstp);
 		if (len < 0 && len != -EAGAIN)
-			goto err_other;
-		dprintk("svc: incomplete TCP record (%d of %d)\n",
-			svsk->sk_tcplen, svsk->sk_reclen);
+			goto err_delete;
+		if (len == want)
+			svc_tcp_fragment_received(svsk);
+		else
+			dprintk("svc: incomplete TCP record (%d of %d)\n",
+				(int)(svsk->sk_tcplen - sizeof(rpc_fraghdr)),
+				svc_sock_reclen(svsk));
 		goto err_noclose;
 	}
 
-	rqstp->rq_arg.len = svsk->sk_reclen;
+	if (svc_sock_reclen(svsk) < 8)
+		goto err_delete; /* client is nuts. */
+
+	rqstp->rq_arg.len = svsk->sk_datalen;
 	rqstp->rq_arg.page_base = 0;
 	if (rqstp->rq_arg.len <= rqstp->rq_arg.head[0].iov_len) {
 		rqstp->rq_arg.head[0].iov_len = rqstp->rq_arg.len;
@@ -1107,11 +1112,8 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 		len = receive_cb_reply(svsk, rqstp);
 
 	/* Reset TCP read info */
-	svsk->sk_reclen = 0;
-	svsk->sk_tcplen = 0;
-	/* If we have more data, signal svc_xprt_enqueue() to try again */
-	if (svc_recv_available(svsk) > sizeof(rpc_fraghdr))
-		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
+	svsk->sk_datalen = 0;
+	svc_tcp_fragment_received(svsk);
 
 	if (len < 0)
 		goto error;
@@ -1120,15 +1122,14 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 	if (serv->sv_stats)
 		serv->sv_stats->nettcpcnt++;
 
-	dprintk("svc: TCP complete record (%d bytes)\n", rqstp->rq_arg.len);
 	return rqstp->rq_arg.len;
 
 error:
 	if (len != -EAGAIN)
-		goto err_other;
+		goto err_delete;
 	dprintk("RPC: TCP recvfrom got EAGAIN\n");
 	return 0;
-err_other:
+err_delete:
 	printk(KERN_NOTICE "%s: recvfrom returned errno %d\n",
 	       svsk->sk_xprt.xpt_server->sv_name, -len);
 	set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
@@ -1305,6 +1306,7 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
 
 		svsk->sk_reclen = 0;
 		svsk->sk_tcplen = 0;
+		svsk->sk_datalen = 0;
 		memset(&svsk->sk_pages[0], 0, sizeof(svsk->sk_pages));
 
 		tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF;
diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index 41cb63b623df..0ce75524ed21 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -521,11 +521,11 @@ next_sge:
 		rqstp->rq_pages[ch_no] = NULL;
 
 	/*
-	 * Detach res pages. svc_release must see a resused count of
-	 * zero or it will attempt to put them.
+	 * Detach res pages. If svc_release sees any it will attempt to
+	 * put them.
 	 */
-	while (rqstp->rq_resused)
-		rqstp->rq_respages[--rqstp->rq_resused] = NULL;
+	while (rqstp->rq_next_page != rqstp->rq_respages)
+		*(--rqstp->rq_next_page) = NULL;
 
 	return err;
 }
@@ -550,7 +550,7 @@ static int rdma_read_complete(struct svc_rqst *rqstp,
 
 	/* rq_respages starts after the last arg page */
 	rqstp->rq_respages = &rqstp->rq_arg.pages[page_no];
-	rqstp->rq_resused = 0;
+	rqstp->rq_next_page = &rqstp->rq_arg.pages[page_no];
 
 	/* Rebuild rq_arg head and tail. */
 	rqstp->rq_arg.head[0] = head->arg.head[0];
diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
index 42eb7ba0b903..c1d124dc772b 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
@@ -548,6 +548,7 @@ static int send_reply(struct svcxprt_rdma *rdma,
 	int sge_no;
 	int sge_bytes;
 	int page_no;
+	int pages;
 	int ret;
 
 	/* Post a recv buffer to handle another request. */
@@ -611,7 +612,8 @@ static int send_reply(struct svcxprt_rdma *rdma,
 	 * respages array. They are our pages until the I/O
 	 * completes.
 	 */
-	for (page_no = 0; page_no < rqstp->rq_resused; page_no++) {
+	pages = rqstp->rq_next_page - rqstp->rq_respages;
+	for (page_no = 0; page_no < pages; page_no++) {
 		ctxt->pages[page_no+1] = rqstp->rq_respages[page_no];
 		ctxt->count++;
 		rqstp->rq_respages[page_no] = NULL;