From f0eede10fd401a1c5193e020cd372d7b7014d9f3 Mon Sep 17 00:00:00 2001
From: Nicholas Mc Guire <hofrat@osadl.org>
Date: Tue, 3 Mar 2015 04:24:25 -0500
Subject: SUNRPC: use jiffies_to_msecs for converting jiffies

Use jiffies_to_msecs for converting jiffies as it handles all of the corner
cases reliably and also helps readability.

Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/sched.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'net')

diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index b91fd9c597b4..c234e7ff2bf0 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -90,7 +90,7 @@ __rpc_add_timer(struct rpc_wait_queue *queue, struct rpc_task *task)
 		return;
 
 	dprintk("RPC: %5u setting alarm for %lu ms\n",
-			task->tk_pid, task->tk_timeout * 1000 / HZ);
+			task->tk_pid, jiffies_to_msecs(task->tk_timeout));
 
 	task->u.tk_wait.expires = jiffies + task->tk_timeout;
 	if (list_empty(&queue->timer_list.list) || time_before(task->u.tk_wait.expires, queue->timer_list.expires))
-- 
cgit v1.2.3


From 55cc1d7806cd4827d38a18ddf3bcf04706758218 Mon Sep 17 00:00:00 2001
From: Nicholas Mc Guire <hofrat@osadl.org>
Date: Fri, 13 Mar 2015 04:25:34 -0400
Subject: SUNRPC: fix build-warning due to format missmatch

fix build-warning introduced by commit: f0eede10fd4 ("SUNRPC: use
jiffies_to_msecs for converting jiffies") which did not fixup
the format properly (my bad).

Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/sched.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index c234e7ff2bf0..337ca851a350 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -89,8 +89,8 @@ __rpc_add_timer(struct rpc_wait_queue *queue, struct rpc_task *task)
 	if (!task->tk_timeout)
 		return;
 
-	dprintk("RPC: %5u setting alarm for %lu ms\n",
-			task->tk_pid, jiffies_to_msecs(task->tk_timeout));
+	dprintk("RPC: %5u setting alarm for %u ms\n",
+		task->tk_pid, jiffies_to_msecs(task->tk_timeout));
 
 	task->u.tk_wait.expires = jiffies + task->tk_timeout;
 	if (list_empty(&queue->timer_list.list) || time_before(task->u.tk_wait.expires, queue->timer_list.expires))
-- 
cgit v1.2.3


From 9bfc52c1091c871cbc58390874b5c4ebe266bee0 Mon Sep 17 00:00:00 2001
From: Fabian Frederick <fabf@skynet.be>
Date: Thu, 23 Oct 2014 18:19:35 +0200
Subject: 9p: remove unused variable in p9_fd_create()

p is initialized but unused.

Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
---
 net/9p/trans_fd.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'net')

diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c
index 80d08f6664cb..c73b894846c1 100644
--- a/net/9p/trans_fd.c
+++ b/net/9p/trans_fd.c
@@ -1013,7 +1013,6 @@ p9_fd_create(struct p9_client *client, const char *addr, char *args)
 {
 	int err;
 	struct p9_fd_opts opts;
-	struct p9_trans_fd *p;
 
 	parse_opts(args, &opts);
 
@@ -1026,7 +1025,6 @@ p9_fd_create(struct p9_client *client, const char *addr, char *args)
 	if (err < 0)
 		return err;
 
-	p = (struct p9_trans_fd *) client->trans;
 	p9_conn_create(client);
 
 	return 0;
-- 
cgit v1.2.3


From 6250a8badb311953a49bedb16ed17eb59d21c03a Mon Sep 17 00:00:00 2001
From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Date: Tue, 30 Dec 2014 02:48:09 +0200
Subject: 9p: use unsigned integers for nwqid/count

As specification says, all integers in messages are unsigned. Let's fix
behaviour of p9pdu_vreadf()/p9pdu_vwritef() accordingly.

Fix for p9pdu_vreadf() is critical. If server replies with Rwalk, where
nwqid > SHRT_MAX, the value will be interpreted as negative. kmalloc, in
its order, will cast the value to (very big) size_t.

It should never happen in normal situation: we never submit Twalk with
nwname > 16, but malicious or broken server can still produce
problematic Rwalk.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
---
 net/9p/protocol.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'net')

diff --git a/net/9p/protocol.c b/net/9p/protocol.c
index ab9127ec5b7a..305e4789f2cc 100644
--- a/net/9p/protocol.c
+++ b/net/9p/protocol.c
@@ -273,7 +273,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
 			}
 			break;
 		case 'R':{
-				int16_t *nwqid = va_arg(ap, int16_t *);
+				uint16_t *nwqid = va_arg(ap, uint16_t *);
 				struct p9_qid **wqids =
 				    va_arg(ap, struct p9_qid **);
 
@@ -448,7 +448,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
 			}
 			break;
 		case 'U':{
-				int32_t count = va_arg(ap, int32_t);
+				uint32_t count = va_arg(ap, uint32_t);
 				const char __user *udata =
 						va_arg(ap, const void __user *);
 				errcode = p9pdu_writef(pdu, proto_version, "d",
@@ -479,7 +479,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
 			}
 			break;
 		case 'R':{
-				int16_t nwqid = va_arg(ap, int);
+				uint16_t nwqid = va_arg(ap, int);
 				struct p9_qid *wqids =
 				    va_arg(ap, struct p9_qid *);
 
-- 
cgit v1.2.3


From 179a5bc4b8cbe68ca675057b960dd805867e41c4 Mon Sep 17 00:00:00 2001
From: Andrey Ryabinin <a.ryabinin@samsung.com>
Date: Tue, 27 Jan 2015 16:00:19 +0300
Subject: net/9p: use memcpy() instead of snprintf() in p9_mount_tag_show()

p9_mount_tag_show() uses '%s' format string to print
non-NULL terminated chan->tag string. This leads
to out of bounds memory read, because format '%s'
implies that string is NULL-terminated.

The length of string is know here, so its simpler and safer
to use memcpy instead of snprintf().

Signed-off-by: Andrey Ryabinin <a.ryabinin@samsung.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
---
 net/9p/trans_virtio.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'net')

diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c
index 36a1a739ad68..486df019f875 100644
--- a/net/9p/trans_virtio.c
+++ b/net/9p/trans_virtio.c
@@ -504,7 +504,10 @@ static ssize_t p9_mount_tag_show(struct device *dev,
 	vdev = dev_to_virtio(dev);
 	chan = vdev->priv;
 
-	return snprintf(buf, chan->tag_len + 1, "%s", chan->tag);
+	memcpy(buf, chan->tag, chan->tag_len);
+	buf[chan->tag_len] = 0;
+
+	return chan->tag_len + 1;
 }
 
 static DEVICE_ATTR(mount_tag, 0444, p9_mount_tag_show, NULL);
-- 
cgit v1.2.3


From b99baa43e533eb62a947e623d0ef844cfbf28d8e Mon Sep 17 00:00:00 2001
From: Dominique Martinet <dominique.martinet@cea.fr>
Date: Fri, 9 Jan 2015 13:05:56 +0100
Subject: net/9p: Initialize opts->privport as it should be.

We're currently using an uninitialized value if option privport is not set,
thus (almost) always using a privileged port.

Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
---
 net/9p/trans_fd.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'net')

diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c
index c73b894846c1..154479d2756d 100644
--- a/net/9p/trans_fd.c
+++ b/net/9p/trans_fd.c
@@ -734,6 +734,7 @@ static int parse_opts(char *params, struct p9_fd_opts *opts)
 	opts->port = P9_PORT;
 	opts->rfd = ~0;
 	opts->wfd = ~0;
+	opts->privport = 0;
 
 	if (!params)
 		return 0;
-- 
cgit v1.2.3


From f569d3ef8254d4b3b8daa4f131f9397d48bf296c Mon Sep 17 00:00:00 2001
From: Dominique Martinet <dominique.martinet@cea.fr>
Date: Fri, 9 Jan 2015 13:07:00 +0100
Subject: net/9p: add a privport option for RDMA transport.

RDMA can use the same kind of weak security as TCP by checking the
client can bind to a privileged port, which is better than nothing
if TAUTH isn't implemented.

Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
---
 net/9p/trans_rdma.c | 52 ++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 44 insertions(+), 8 deletions(-)

(limited to 'net')

diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c
index 14ad43b5cf89..3533d2a53ab6 100644
--- a/net/9p/trans_rdma.c
+++ b/net/9p/trans_rdma.c
@@ -139,6 +139,7 @@ struct p9_rdma_opts {
 	int sq_depth;
 	int rq_depth;
 	long timeout;
+	int privport;
 };
 
 /*
@@ -146,7 +147,10 @@ struct p9_rdma_opts {
  */
 enum {
 	/* Options that take integer arguments */
-	Opt_port, Opt_rq_depth, Opt_sq_depth, Opt_timeout, Opt_err,
+	Opt_port, Opt_rq_depth, Opt_sq_depth, Opt_timeout,
+	/* Options that take no argument */
+	Opt_privport,
+	Opt_err,
 };
 
 static match_table_t tokens = {
@@ -154,6 +158,7 @@ static match_table_t tokens = {
 	{Opt_sq_depth, "sq=%u"},
 	{Opt_rq_depth, "rq=%u"},
 	{Opt_timeout, "timeout=%u"},
+	{Opt_privport, "privport"},
 	{Opt_err, NULL},
 };
 
@@ -175,6 +180,7 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts)
 	opts->sq_depth = P9_RDMA_SQ_DEPTH;
 	opts->rq_depth = P9_RDMA_RQ_DEPTH;
 	opts->timeout = P9_RDMA_TIMEOUT;
+	opts->privport = 0;
 
 	if (!params)
 		return 0;
@@ -193,13 +199,13 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts)
 		if (!*p)
 			continue;
 		token = match_token(p, tokens, args);
-		if (token == Opt_err)
-			continue;
-		r = match_int(&args[0], &option);
-		if (r < 0) {
-			p9_debug(P9_DEBUG_ERROR,
-				 "integer field, but no integer?\n");
-			continue;
+		if ((token != Opt_err) && (token != Opt_privport)) {
+			r = match_int(&args[0], &option);
+			if (r < 0) {
+				p9_debug(P9_DEBUG_ERROR,
+					 "integer field, but no integer?\n");
+				continue;
+			}
 		}
 		switch (token) {
 		case Opt_port:
@@ -214,6 +220,9 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts)
 		case Opt_timeout:
 			opts->timeout = option;
 			break;
+		case Opt_privport:
+			opts->privport = 1;
+			break;
 		default:
 			continue;
 		}
@@ -607,6 +616,23 @@ static int rdma_cancelled(struct p9_client *client, struct p9_req_t *req)
 	return 0;
 }
 
+static int p9_rdma_bind_privport(struct p9_trans_rdma *rdma)
+{
+	struct sockaddr_in cl = {
+		.sin_family = AF_INET,
+		.sin_addr.s_addr = htonl(INADDR_ANY),
+	};
+	int port, err = -EINVAL;
+
+	for (port = P9_DEF_MAX_RESVPORT; port >= P9_DEF_MIN_RESVPORT; port--) {
+		cl.sin_port = htons((ushort)port);
+		err = rdma_bind_addr(rdma->cm_id, (struct sockaddr *)&cl);
+		if (err != -EADDRINUSE)
+			break;
+	}
+	return err;
+}
+
 /**
  * trans_create_rdma - Transport method for creating atransport instance
  * @client: client instance
@@ -642,6 +668,16 @@ rdma_create_trans(struct p9_client *client, const char *addr, char *args)
 	/* Associate the client with the transport */
 	client->trans = rdma;
 
+	/* Bind to a privileged port if we need to */
+	if (opts.privport) {
+		err = p9_rdma_bind_privport(rdma);
+		if (err < 0) {
+			pr_err("%s (%d): problem binding to privport: %d\n",
+			       __func__, task_pid_nr(current), -err);
+			goto error;
+		}
+	}
+
 	/* Resolve the server's address */
 	rdma->addr.sin_family = AF_INET;
 	rdma->addr.sin_addr.s_addr = in_aton(addr);
-- 
cgit v1.2.3


From 0695314ef0920b745423510f1e96bf60415a404a Mon Sep 17 00:00:00 2001
From: Trond Myklebust <trond.myklebust@primarydata.com>
Date: Mon, 23 Mar 2015 16:10:00 -0400
Subject: SUNRPC: Fix a regression when reconnecting

If the task needs to give up the socket lock in order to allow a
reconnect to occur, then it must also clear the 'rq_bytes_sent' field
so that when it retransmits, it knows to start from the beginning.

Fixes: 718ba5b87343 ("SUNRPC: Add helpers to prevent socket create from racing")
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 net/sunrpc/xprt.c | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index e3015aede0d9..ed4aa786c240 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -326,6 +326,15 @@ out_unlock:
 	xprt_clear_locked(xprt);
 }
 
+static void xprt_task_clear_bytes_sent(struct rpc_task *task)
+{
+	if (task != NULL) {
+		struct rpc_rqst *req = task->tk_rqstp;
+		if (req != NULL)
+			req->rq_bytes_sent = 0;
+	}
+}
+
 /**
  * xprt_release_xprt - allow other requests to use a transport
  * @xprt: transport with other tasks potentially waiting
@@ -336,11 +345,7 @@ out_unlock:
 void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
 {
 	if (xprt->snd_task == task) {
-		if (task != NULL) {
-			struct rpc_rqst *req = task->tk_rqstp;
-			if (req != NULL)
-				req->rq_bytes_sent = 0;
-		}
+		xprt_task_clear_bytes_sent(task);
 		xprt_clear_locked(xprt);
 		__xprt_lock_write_next(xprt);
 	}
@@ -358,11 +363,7 @@ EXPORT_SYMBOL_GPL(xprt_release_xprt);
 void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
 {
 	if (xprt->snd_task == task) {
-		if (task != NULL) {
-			struct rpc_rqst *req = task->tk_rqstp;
-			if (req != NULL)
-				req->rq_bytes_sent = 0;
-		}
+		xprt_task_clear_bytes_sent(task);
 		xprt_clear_locked(xprt);
 		__xprt_lock_write_next_cong(xprt);
 	}
@@ -700,6 +701,7 @@ bool xprt_lock_connect(struct rpc_xprt *xprt,
 		goto out;
 	if (xprt->snd_task != task)
 		goto out;
+	xprt_task_clear_bytes_sent(task);
 	xprt->snd_task = cookie;
 	ret = true;
 out:
-- 
cgit v1.2.3


From 0dd39cae26f3990789b4a558f9abafe59adc6fc1 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:33:43 -0400
Subject: xprtrdma: Display IPv6 addresses and port numbers correctly

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/transport.c | 47 +++++++++++++++++++++++++++++++++--------
 net/sunrpc/xprtrdma/verbs.c     | 21 ++++++++----------
 2 files changed, 47 insertions(+), 21 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 2e192baa59f3..9be7f97205ba 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -156,13 +156,48 @@ static struct ctl_table sunrpc_table[] = {
 
 static struct rpc_xprt_ops xprt_rdma_procs;	/* forward reference */
 
+static void
+xprt_rdma_format_addresses4(struct rpc_xprt *xprt, struct sockaddr *sap)
+{
+	struct sockaddr_in *sin = (struct sockaddr_in *)sap;
+	char buf[20];
+
+	snprintf(buf, sizeof(buf), "%08x", ntohl(sin->sin_addr.s_addr));
+	xprt->address_strings[RPC_DISPLAY_HEX_ADDR] = kstrdup(buf, GFP_KERNEL);
+
+	xprt->address_strings[RPC_DISPLAY_NETID] = RPCBIND_NETID_RDMA;
+}
+
+static void
+xprt_rdma_format_addresses6(struct rpc_xprt *xprt, struct sockaddr *sap)
+{
+	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
+	char buf[40];
+
+	snprintf(buf, sizeof(buf), "%pi6", &sin6->sin6_addr);
+	xprt->address_strings[RPC_DISPLAY_HEX_ADDR] = kstrdup(buf, GFP_KERNEL);
+
+	xprt->address_strings[RPC_DISPLAY_NETID] = RPCBIND_NETID_RDMA6;
+}
+
 static void
 xprt_rdma_format_addresses(struct rpc_xprt *xprt)
 {
 	struct sockaddr *sap = (struct sockaddr *)
 					&rpcx_to_rdmad(xprt).addr;
-	struct sockaddr_in *sin = (struct sockaddr_in *)sap;
-	char buf[64];
+	char buf[128];
+
+	switch (sap->sa_family) {
+	case AF_INET:
+		xprt_rdma_format_addresses4(xprt, sap);
+		break;
+	case AF_INET6:
+		xprt_rdma_format_addresses6(xprt, sap);
+		break;
+	default:
+		pr_err("rpcrdma: Unrecognized address family\n");
+		return;
+	}
 
 	(void)rpc_ntop(sap, buf, sizeof(buf));
 	xprt->address_strings[RPC_DISPLAY_ADDR] = kstrdup(buf, GFP_KERNEL);
@@ -170,16 +205,10 @@ xprt_rdma_format_addresses(struct rpc_xprt *xprt)
 	snprintf(buf, sizeof(buf), "%u", rpc_get_port(sap));
 	xprt->address_strings[RPC_DISPLAY_PORT] = kstrdup(buf, GFP_KERNEL);
 
-	xprt->address_strings[RPC_DISPLAY_PROTO] = "rdma";
-
-	snprintf(buf, sizeof(buf), "%08x", ntohl(sin->sin_addr.s_addr));
-	xprt->address_strings[RPC_DISPLAY_HEX_ADDR] = kstrdup(buf, GFP_KERNEL);
-
 	snprintf(buf, sizeof(buf), "%4hx", rpc_get_port(sap));
 	xprt->address_strings[RPC_DISPLAY_HEX_PORT] = kstrdup(buf, GFP_KERNEL);
 
-	/* netid */
-	xprt->address_strings[RPC_DISPLAY_NETID] = "rdma";
+	xprt->address_strings[RPC_DISPLAY_PROTO] = "rdma";
 }
 
 static void
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 124676c13780..1aa55b74b1b7 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -50,6 +50,7 @@
 #include <linux/interrupt.h>
 #include <linux/slab.h>
 #include <linux/prefetch.h>
+#include <linux/sunrpc/addr.h>
 #include <asm/bitops.h>
 
 #include "xprt_rdma.h"
@@ -424,7 +425,7 @@ rpcrdma_conn_upcall(struct rdma_cm_id *id, struct rdma_cm_event *event)
 	struct rpcrdma_ia *ia = &xprt->rx_ia;
 	struct rpcrdma_ep *ep = &xprt->rx_ep;
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
-	struct sockaddr_in *addr = (struct sockaddr_in *) &ep->rep_remote_addr;
+	struct sockaddr *sap = (struct sockaddr *)&ep->rep_remote_addr;
 #endif
 	struct ib_qp_attr *attr = &ia->ri_qp_attr;
 	struct ib_qp_init_attr *iattr = &ia->ri_qp_init_attr;
@@ -480,9 +481,8 @@ connected:
 		wake_up_all(&ep->rep_connect_wait);
 		/*FALLTHROUGH*/
 	default:
-		dprintk("RPC:       %s: %pI4:%u (ep 0x%p): %s\n",
-			__func__, &addr->sin_addr.s_addr,
-			ntohs(addr->sin_port), ep,
+		dprintk("RPC:       %s: %pIS:%u (ep 0x%p): %s\n",
+			__func__, sap, rpc_get_port(sap), ep,
 			CONNECTION_MSG(event->event));
 		break;
 	}
@@ -491,19 +491,16 @@ connected:
 	if (connstate == 1) {
 		int ird = attr->max_dest_rd_atomic;
 		int tird = ep->rep_remote_cma.responder_resources;
-		printk(KERN_INFO "rpcrdma: connection to %pI4:%u "
-			"on %s, memreg %d slots %d ird %d%s\n",
-			&addr->sin_addr.s_addr,
-			ntohs(addr->sin_port),
+
+		pr_info("rpcrdma: connection to %pIS:%u on %s, memreg %d slots %d ird %d%s\n",
+			sap, rpc_get_port(sap),
 			ia->ri_id->device->name,
 			ia->ri_memreg_strategy,
 			xprt->rx_buf.rb_max_requests,
 			ird, ird < 4 && ird < tird / 2 ? " (low!)" : "");
 	} else if (connstate < 0) {
-		printk(KERN_INFO "rpcrdma: connection to %pI4:%u closed (%d)\n",
-			&addr->sin_addr.s_addr,
-			ntohs(addr->sin_port),
-			connstate);
+		pr_info("rpcrdma: connection to %pIS:%u closed (%d)\n",
+			sap, rpc_get_port(sap), connstate);
 	}
 #endif
 
-- 
cgit v1.2.3


From e23779451ee05e824fedbb68bd17fc5c77e40166 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:33:53 -0400
Subject: xprtrdma: Perform a full marshal on retransmit

Commit 6ab59945f292 ("xprtrdma: Update rkeys after transport
reconnect" added logic in the ->send_request path to update the
chunk list when an RPC/RDMA request is retransmitted.

Note that rpc_xdr_encode() resets and re-encodes the entire RPC
send buffer for each retransmit of an RPC. The RPC send buffer
is not preserved from the previous transmission of an RPC.

Revert 6ab59945f292, and instead, just force each request to be
fully marshaled every time through ->send_request. This should
preserve the fix from 6ab59945f292, while also performing pullup
during retransmits.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/rpc_rdma.c  | 71 +++++++++++++++++++----------------------
 net/sunrpc/xprtrdma/transport.c |  5 +--
 net/sunrpc/xprtrdma/xprt_rdma.h | 10 ------
 3 files changed, 34 insertions(+), 52 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c
index 91ffde82fa0c..41456d9e5a7d 100644
--- a/net/sunrpc/xprtrdma/rpc_rdma.c
+++ b/net/sunrpc/xprtrdma/rpc_rdma.c
@@ -53,6 +53,14 @@
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
+enum rpcrdma_chunktype {
+	rpcrdma_noch = 0,
+	rpcrdma_readch,
+	rpcrdma_areadch,
+	rpcrdma_writech,
+	rpcrdma_replych
+};
+
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 static const char transfertypes[][12] = {
 	"pure inline",	/* no chunks */
@@ -283,28 +291,6 @@ out:
 	return n;
 }
 
-/*
- * Marshal chunks. This routine returns the header length
- * consumed by marshaling.
- *
- * Returns positive RPC/RDMA header size, or negative errno.
- */
-
-ssize_t
-rpcrdma_marshal_chunks(struct rpc_rqst *rqst, ssize_t result)
-{
-	struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
-	struct rpcrdma_msg *headerp = rdmab_to_msg(req->rl_rdmabuf);
-
-	if (req->rl_rtype != rpcrdma_noch)
-		result = rpcrdma_create_chunks(rqst, &rqst->rq_snd_buf,
-					       headerp, req->rl_rtype);
-	else if (req->rl_wtype != rpcrdma_noch)
-		result = rpcrdma_create_chunks(rqst, &rqst->rq_rcv_buf,
-					       headerp, req->rl_wtype);
-	return result;
-}
-
 /*
  * Copy write data inline.
  * This function is used for "small" requests. Data which is passed
@@ -397,6 +383,7 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 	char *base;
 	size_t rpclen, padlen;
 	ssize_t hdrlen;
+	enum rpcrdma_chunktype rtype, wtype;
 	struct rpcrdma_msg *headerp;
 
 	/*
@@ -433,13 +420,13 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 	 * into pages; otherwise use reply chunks.
 	 */
 	if (rqst->rq_rcv_buf.buflen <= RPCRDMA_INLINE_READ_THRESHOLD(rqst))
-		req->rl_wtype = rpcrdma_noch;
+		wtype = rpcrdma_noch;
 	else if (rqst->rq_rcv_buf.page_len == 0)
-		req->rl_wtype = rpcrdma_replych;
+		wtype = rpcrdma_replych;
 	else if (rqst->rq_rcv_buf.flags & XDRBUF_READ)
-		req->rl_wtype = rpcrdma_writech;
+		wtype = rpcrdma_writech;
 	else
-		req->rl_wtype = rpcrdma_replych;
+		wtype = rpcrdma_replych;
 
 	/*
 	 * Chunks needed for arguments?
@@ -456,16 +443,16 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 	 * TBD check NFSv4 setacl
 	 */
 	if (rqst->rq_snd_buf.len <= RPCRDMA_INLINE_WRITE_THRESHOLD(rqst))
-		req->rl_rtype = rpcrdma_noch;
+		rtype = rpcrdma_noch;
 	else if (rqst->rq_snd_buf.page_len == 0)
-		req->rl_rtype = rpcrdma_areadch;
+		rtype = rpcrdma_areadch;
 	else
-		req->rl_rtype = rpcrdma_readch;
+		rtype = rpcrdma_readch;
 
 	/* The following simplification is not true forever */
-	if (req->rl_rtype != rpcrdma_noch && req->rl_wtype == rpcrdma_replych)
-		req->rl_wtype = rpcrdma_noch;
-	if (req->rl_rtype != rpcrdma_noch && req->rl_wtype != rpcrdma_noch) {
+	if (rtype != rpcrdma_noch && wtype == rpcrdma_replych)
+		wtype = rpcrdma_noch;
+	if (rtype != rpcrdma_noch && wtype != rpcrdma_noch) {
 		dprintk("RPC:       %s: cannot marshal multiple chunk lists\n",
 			__func__);
 		return -EIO;
@@ -479,7 +466,7 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 	 * When padding is in use and applies to the transfer, insert
 	 * it and change the message type.
 	 */
-	if (req->rl_rtype == rpcrdma_noch) {
+	if (rtype == rpcrdma_noch) {
 
 		padlen = rpcrdma_inline_pullup(rqst,
 						RPCRDMA_INLINE_PAD_VALUE(rqst));
@@ -494,7 +481,7 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 			headerp->rm_body.rm_padded.rm_pempty[1] = xdr_zero;
 			headerp->rm_body.rm_padded.rm_pempty[2] = xdr_zero;
 			hdrlen += 2 * sizeof(u32); /* extra words in padhdr */
-			if (req->rl_wtype != rpcrdma_noch) {
+			if (wtype != rpcrdma_noch) {
 				dprintk("RPC:       %s: invalid chunk list\n",
 					__func__);
 				return -EIO;
@@ -515,18 +502,26 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst)
 			 * on receive. Therefore, we request a reply chunk
 			 * for non-writes wherever feasible and efficient.
 			 */
-			if (req->rl_wtype == rpcrdma_noch)
-				req->rl_wtype = rpcrdma_replych;
+			if (wtype == rpcrdma_noch)
+				wtype = rpcrdma_replych;
 		}
 	}
 
-	hdrlen = rpcrdma_marshal_chunks(rqst, hdrlen);
+	if (rtype != rpcrdma_noch) {
+		hdrlen = rpcrdma_create_chunks(rqst, &rqst->rq_snd_buf,
+					       headerp, rtype);
+		wtype = rtype;	/* simplify dprintk */
+
+	} else if (wtype != rpcrdma_noch) {
+		hdrlen = rpcrdma_create_chunks(rqst, &rqst->rq_rcv_buf,
+					       headerp, wtype);
+	}
 	if (hdrlen < 0)
 		return hdrlen;
 
 	dprintk("RPC:       %s: %s: hdrlen %zd rpclen %zd padlen %zd"
 		" headerp 0x%p base 0x%p lkey 0x%x\n",
-		__func__, transfertypes[req->rl_wtype], hdrlen, rpclen, padlen,
+		__func__, transfertypes[wtype], hdrlen, rpclen, padlen,
 		headerp, base, rdmab_lkey(req->rl_rdmabuf));
 
 	/*
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 9be7f97205ba..97f656292feb 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -608,10 +608,7 @@ xprt_rdma_send_request(struct rpc_task *task)
 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 	int rc = 0;
 
-	if (req->rl_niovs == 0)
-		rc = rpcrdma_marshal_req(rqst);
-	else if (r_xprt->rx_ia.ri_memreg_strategy != RPCRDMA_ALLPHYSICAL)
-		rc = rpcrdma_marshal_chunks(rqst, 0);
+	rc = rpcrdma_marshal_req(rqst);
 	if (rc < 0)
 		goto failed_marshal;
 
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 0a16fb6f0885..c8afd83e8b75 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -143,14 +143,6 @@ rdmab_to_msg(struct rpcrdma_regbuf *rb)
 	return (struct rpcrdma_msg *)rb->rg_base;
 }
 
-enum rpcrdma_chunktype {
-	rpcrdma_noch = 0,
-	rpcrdma_readch,
-	rpcrdma_areadch,
-	rpcrdma_writech,
-	rpcrdma_replych
-};
-
 /*
  * struct rpcrdma_rep -- this structure encapsulates state required to recv
  * and complete a reply, asychronously. It needs several pieces of
@@ -258,7 +250,6 @@ struct rpcrdma_req {
 	unsigned int	rl_niovs;	/* 0, 2 or 4 */
 	unsigned int	rl_nchunks;	/* non-zero if chunks */
 	unsigned int	rl_connect_cookie;	/* retry detection */
-	enum rpcrdma_chunktype	rl_rtype, rl_wtype;
 	struct rpcrdma_buffer *rl_buffer; /* home base for this structure */
 	struct rpcrdma_rep	*rl_reply;/* holder for reply buffer */
 	struct ib_sge	rl_send_iov[4];	/* for active requests */
@@ -418,7 +409,6 @@ void rpcrdma_reply_handler(struct rpcrdma_rep *);
 /*
  * RPC/RDMA protocol calls - xprtrdma/rpc_rdma.c
  */
-ssize_t rpcrdma_marshal_chunks(struct rpc_rqst *, ssize_t);
 int rpcrdma_marshal_req(struct rpc_rqst *);
 size_t rpcrdma_max_payload(struct rpcrdma_xprt *);
 
-- 
cgit v1.2.3


From 805272406a980cab0e11742e5423ba97b6f38836 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:34:02 -0400
Subject: xprtrdma: Byte-align FRWR registration

The RPC/RDMA transport's FRWR registration logic registers whole
pages. This means areas in the first and last pages that are not
involved in the RDMA I/O are needlessly exposed to the server.

Buffered I/O is typically page-aligned, so not a problem there. But
for direct I/O, which can be byte-aligned, and for reply chunks,
which are nearly always smaller than a page, the transport could
expose memory outside the I/O buffer.

FRWR allows byte-aligned memory registration, so let's use it as
it was intended.

Reported-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/verbs.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 1aa55b74b1b7..60f3317c90ee 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1924,23 +1924,19 @@ rpcrdma_register_frmr_external(struct rpcrdma_mr_seg *seg,
 		    offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len))
 			break;
 	}
-	dprintk("RPC:       %s: Using frmr %p to map %d segments\n",
-		__func__, mw, i);
+	dprintk("RPC:       %s: Using frmr %p to map %d segments (%d bytes)\n",
+		__func__, mw, i, len);
 
 	frmr->fr_state = FRMR_IS_VALID;
 
 	memset(&fastreg_wr, 0, sizeof(fastreg_wr));
 	fastreg_wr.wr_id = (unsigned long)(void *)mw;
 	fastreg_wr.opcode = IB_WR_FAST_REG_MR;
-	fastreg_wr.wr.fast_reg.iova_start = seg1->mr_dma;
+	fastreg_wr.wr.fast_reg.iova_start = seg1->mr_dma + pageoff;
 	fastreg_wr.wr.fast_reg.page_list = frmr->fr_pgl;
 	fastreg_wr.wr.fast_reg.page_list_len = page_no;
 	fastreg_wr.wr.fast_reg.page_shift = PAGE_SHIFT;
-	fastreg_wr.wr.fast_reg.length = page_no << PAGE_SHIFT;
-	if (fastreg_wr.wr.fast_reg.length < len) {
-		rc = -EIO;
-		goto out_err;
-	}
+	fastreg_wr.wr.fast_reg.length = len;
 
 	/* Bump the key */
 	key = (u8)(mr->rkey & 0x000000FF);
-- 
cgit v1.2.3


From 41f97028969e4c88efa5fcf58bc6125210413a6d Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:34:12 -0400
Subject: xprtrdma: Prevent infinite loop in rpcrdma_ep_create()

If a provider advertizes a zero max_fast_reg_page_list_len, FRWR
depth detection loops forever. Instead of just failing the mount,
try other memory registration modes.

Fixes: 0fc6c4e7bb28 ("xprtrdma: mind the device's max fast . . .")
Reported-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/verbs.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 60f3317c90ee..99752b5b7354 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -618,9 +618,10 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg)
 
 	if (memreg == RPCRDMA_FRMR) {
 		/* Requires both frmr reg and local dma lkey */
-		if ((devattr->device_cap_flags &
+		if (((devattr->device_cap_flags &
 		     (IB_DEVICE_MEM_MGT_EXTENSIONS|IB_DEVICE_LOCAL_DMA_LKEY)) !=
-		    (IB_DEVICE_MEM_MGT_EXTENSIONS|IB_DEVICE_LOCAL_DMA_LKEY)) {
+		    (IB_DEVICE_MEM_MGT_EXTENSIONS|IB_DEVICE_LOCAL_DMA_LKEY)) ||
+		      (devattr->max_fast_reg_page_list_len == 0)) {
 			dprintk("RPC:       %s: FRMR registration "
 				"not supported by HCA\n", __func__);
 			memreg = RPCRDMA_MTHCAFMR;
-- 
cgit v1.2.3


From a0ce85f595c22d28bf03c3fae8545b3077b7be1b Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:34:21 -0400
Subject: xprtrdma: Add vector of ops for each memory registration strategy

Instead of employing switch() statements, let's use the typical
Linux kernel idiom for handling behavioral variation: virtual
functions.

Start by defining a vector of operations for each supported memory
registration mode, and by adding a source file for each mode.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/Makefile       |  3 ++-
 net/sunrpc/xprtrdma/fmr_ops.c      | 22 ++++++++++++++++++++++
 net/sunrpc/xprtrdma/frwr_ops.c     | 22 ++++++++++++++++++++++
 net/sunrpc/xprtrdma/physical_ops.c | 24 ++++++++++++++++++++++++
 net/sunrpc/xprtrdma/verbs.c        | 11 +++++++----
 net/sunrpc/xprtrdma/xprt_rdma.h    | 12 ++++++++++++
 6 files changed, 89 insertions(+), 5 deletions(-)
 create mode 100644 net/sunrpc/xprtrdma/fmr_ops.c
 create mode 100644 net/sunrpc/xprtrdma/frwr_ops.c
 create mode 100644 net/sunrpc/xprtrdma/physical_ops.c

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/Makefile b/net/sunrpc/xprtrdma/Makefile
index da5136fd5694..579f72bbcf4b 100644
--- a/net/sunrpc/xprtrdma/Makefile
+++ b/net/sunrpc/xprtrdma/Makefile
@@ -1,6 +1,7 @@
 obj-$(CONFIG_SUNRPC_XPRT_RDMA_CLIENT) += xprtrdma.o
 
-xprtrdma-y := transport.o rpc_rdma.o verbs.o
+xprtrdma-y := transport.o rpc_rdma.o verbs.o \
+	fmr_ops.o frwr_ops.o physical_ops.o
 
 obj-$(CONFIG_SUNRPC_XPRT_RDMA_SERVER) += svcrdma.o
 
diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
new file mode 100644
index 000000000000..ffb7d9358480
--- /dev/null
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -0,0 +1,22 @@
+/*
+ * Copyright (c) 2015 Oracle.  All rights reserved.
+ * Copyright (c) 2003-2007 Network Appliance, Inc. All rights reserved.
+ */
+
+/* Lightweight memory registration using Fast Memory Regions (FMR).
+ * Referred to sometimes as MTHCAFMR mode.
+ *
+ * FMR uses synchronous memory registration and deregistration.
+ * FMR registration is known to be fast, but FMR deregistration
+ * can take tens of usecs to complete.
+ */
+
+#include "xprt_rdma.h"
+
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
+# define RPCDBG_FACILITY	RPCDBG_TRANS
+#endif
+
+const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
+	.ro_displayname			= "fmr",
+};
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
new file mode 100644
index 000000000000..79173f98e09a
--- /dev/null
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -0,0 +1,22 @@
+/*
+ * Copyright (c) 2015 Oracle.  All rights reserved.
+ * Copyright (c) 2003-2007 Network Appliance, Inc. All rights reserved.
+ */
+
+/* Lightweight memory registration using Fast Registration Work
+ * Requests (FRWR). Also referred to sometimes as FRMR mode.
+ *
+ * FRWR features ordered asynchronous registration and deregistration
+ * of arbitrarily sized memory regions. This is the fastest and safest
+ * but most complex memory registration mode.
+ */
+
+#include "xprt_rdma.h"
+
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
+# define RPCDBG_FACILITY	RPCDBG_TRANS
+#endif
+
+const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
+	.ro_displayname			= "frwr",
+};
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
new file mode 100644
index 000000000000..b0922accabcf
--- /dev/null
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -0,0 +1,24 @@
+/*
+ * Copyright (c) 2015 Oracle.  All rights reserved.
+ * Copyright (c) 2003-2007 Network Appliance, Inc. All rights reserved.
+ */
+
+/* No-op chunk preparation. All client memory is pre-registered.
+ * Sometimes referred to as ALLPHYSICAL mode.
+ *
+ * Physical registration is simple because all client memory is
+ * pre-registered and never deregistered. This mode is good for
+ * adapter bring up, but is considered not safe: the server is
+ * trusted not to abuse its access to client memory not involved
+ * in RDMA I/O.
+ */
+
+#include "xprt_rdma.h"
+
+#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
+# define RPCDBG_FACILITY	RPCDBG_TRANS
+#endif
+
+const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
+	.ro_displayname			= "physical",
+};
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 99752b5b7354..c3319e12551c 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -492,10 +492,10 @@ connected:
 		int ird = attr->max_dest_rd_atomic;
 		int tird = ep->rep_remote_cma.responder_resources;
 
-		pr_info("rpcrdma: connection to %pIS:%u on %s, memreg %d slots %d ird %d%s\n",
+		pr_info("rpcrdma: connection to %pIS:%u on %s, memreg '%s', %d credits, %d responders%s\n",
 			sap, rpc_get_port(sap),
 			ia->ri_id->device->name,
-			ia->ri_memreg_strategy,
+			ia->ri_ops->ro_displayname,
 			xprt->rx_buf.rb_max_requests,
 			ird, ird < 4 && ird < tird / 2 ? " (low!)" : "");
 	} else if (connstate < 0) {
@@ -650,13 +650,16 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg)
 	 */
 	switch (memreg) {
 	case RPCRDMA_FRMR:
+		ia->ri_ops = &rpcrdma_frwr_memreg_ops;
 		break;
 	case RPCRDMA_ALLPHYSICAL:
+		ia->ri_ops = &rpcrdma_physical_memreg_ops;
 		mem_priv = IB_ACCESS_LOCAL_WRITE |
 				IB_ACCESS_REMOTE_WRITE |
 				IB_ACCESS_REMOTE_READ;
 		goto register_setup;
 	case RPCRDMA_MTHCAFMR:
+		ia->ri_ops = &rpcrdma_fmr_memreg_ops;
 		if (ia->ri_have_dma_lkey)
 			break;
 		mem_priv = IB_ACCESS_LOCAL_WRITE;
@@ -676,8 +679,8 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg)
 		rc = -ENOMEM;
 		goto out3;
 	}
-	dprintk("RPC:       %s: memory registration strategy is %d\n",
-		__func__, memreg);
+	dprintk("RPC:       %s: memory registration strategy is '%s'\n",
+		__func__, ia->ri_ops->ro_displayname);
 
 	/* Else will do memory reg/dereg for each chunk */
 	ia->ri_memreg_strategy = memreg;
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index c8afd83e8b75..ef3cf4aeecd6 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -60,6 +60,7 @@
  * Interface Adapter -- one per transport instance
  */
 struct rpcrdma_ia {
+	const struct rpcrdma_memreg_ops	*ri_ops;
 	rwlock_t		ri_qplock;
 	struct rdma_cm_id 	*ri_id;
 	struct ib_pd		*ri_pd;
@@ -330,6 +331,17 @@ struct rpcrdma_stats {
 	unsigned long		bad_reply_count;
 };
 
+/*
+ * Per-registration mode operations
+ */
+struct rpcrdma_memreg_ops {
+	const char	*ro_displayname;
+};
+
+extern const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops;
+extern const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops;
+extern const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops;
+
 /*
  * RPCRDMA transport -- encapsulates the structures above for
  * integration with RPC.
-- 
cgit v1.2.3


From 1c9351ee0e346ec1b3c700a4bc8f881923e1808e Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:34:30 -0400
Subject: xprtrdma: Add a "max_payload" op for each memreg mode

The max_payload computation is generalized to ensure that the
payload maximum is the lesser of RPC_MAX_DATA_SEGS and the number of
data segments that can be transmitted in an inline buffer.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      | 13 ++++++++++
 net/sunrpc/xprtrdma/frwr_ops.c     | 13 ++++++++++
 net/sunrpc/xprtrdma/physical_ops.c | 10 ++++++++
 net/sunrpc/xprtrdma/transport.c    |  5 +++-
 net/sunrpc/xprtrdma/verbs.c        | 49 ++++++++++++--------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    |  5 +++-
 6 files changed, 59 insertions(+), 36 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index ffb7d9358480..eec266055b28 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -17,6 +17,19 @@
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
+/* Maximum scatter/gather per FMR */
+#define RPCRDMA_MAX_FMR_SGES	(64)
+
+/* FMR mode conveys up to 64 pages of payload per chunk segment.
+ */
+static size_t
+fmr_op_maxpages(struct rpcrdma_xprt *r_xprt)
+{
+	return min_t(unsigned int, RPCRDMA_MAX_DATA_SEGS,
+		     rpcrdma_max_segments(r_xprt) * RPCRDMA_MAX_FMR_SGES);
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
+	.ro_maxpages			= fmr_op_maxpages,
 	.ro_displayname			= "fmr",
 };
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 79173f98e09a..73a5ac898efc 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -17,6 +17,19 @@
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
+/* FRWR mode conveys a list of pages per chunk segment. The
+ * maximum length of that list is the FRWR page list depth.
+ */
+static size_t
+frwr_op_maxpages(struct rpcrdma_xprt *r_xprt)
+{
+	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+
+	return min_t(unsigned int, RPCRDMA_MAX_DATA_SEGS,
+		     rpcrdma_max_segments(r_xprt) * ia->ri_max_frmr_depth);
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
+	.ro_maxpages			= frwr_op_maxpages,
 	.ro_displayname			= "frwr",
 };
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index b0922accabcf..28ade1943a57 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -19,6 +19,16 @@
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
+/* PHYSICAL memory registration conveys one page per chunk segment.
+ */
+static size_t
+physical_op_maxpages(struct rpcrdma_xprt *r_xprt)
+{
+	return min_t(unsigned int, RPCRDMA_MAX_DATA_SEGS,
+		     rpcrdma_max_segments(r_xprt));
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
+	.ro_maxpages			= physical_op_maxpages,
 	.ro_displayname			= "physical",
 };
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 97f656292feb..da71a24641e3 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -406,7 +406,10 @@ xprt_setup_rdma(struct xprt_create *args)
 			  xprt_rdma_connect_worker);
 
 	xprt_rdma_format_addresses(xprt);
-	xprt->max_payload = rpcrdma_max_payload(new_xprt);
+	xprt->max_payload = new_xprt->rx_ia.ri_ops->ro_maxpages(new_xprt);
+	if (xprt->max_payload == 0)
+		goto out4;
+	xprt->max_payload <<= PAGE_SHIFT;
 	dprintk("RPC:       %s: transport data payload maximum: %zu bytes\n",
 		__func__, xprt->max_payload);
 
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index c3319e12551c..da55cda30568 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -2212,43 +2212,24 @@ rpcrdma_ep_post_recv(struct rpcrdma_ia *ia,
 	return rc;
 }
 
-/* Physical mapping means one Read/Write list entry per-page.
- * All list entries must fit within an inline buffer
- *
- * NB: The server must return a Write list for NFS READ,
- *     which has the same constraint. Factor in the inline
- *     rsize as well.
+/* How many chunk list items fit within our inline buffers?
  */
-static size_t
-rpcrdma_physical_max_payload(struct rpcrdma_xprt *r_xprt)
+unsigned int
+rpcrdma_max_segments(struct rpcrdma_xprt *r_xprt)
 {
 	struct rpcrdma_create_data_internal *cdata = &r_xprt->rx_data;
-	unsigned int inline_size, pages;
-
-	inline_size = min_t(unsigned int,
-			    cdata->inline_wsize, cdata->inline_rsize);
-	inline_size -= RPCRDMA_HDRLEN_MIN;
-	pages = inline_size / sizeof(struct rpcrdma_segment);
-	return pages << PAGE_SHIFT;
-}
+	int bytes, segments;
 
-static size_t
-rpcrdma_mr_max_payload(struct rpcrdma_xprt *r_xprt)
-{
-	return RPCRDMA_MAX_DATA_SEGS << PAGE_SHIFT;
-}
-
-size_t
-rpcrdma_max_payload(struct rpcrdma_xprt *r_xprt)
-{
-	size_t result;
-
-	switch (r_xprt->rx_ia.ri_memreg_strategy) {
-	case RPCRDMA_ALLPHYSICAL:
-		result = rpcrdma_physical_max_payload(r_xprt);
-		break;
-	default:
-		result = rpcrdma_mr_max_payload(r_xprt);
+	bytes = min_t(unsigned int, cdata->inline_wsize, cdata->inline_rsize);
+	bytes -= RPCRDMA_HDRLEN_MIN;
+	if (bytes < sizeof(struct rpcrdma_segment) * 2) {
+		pr_warn("RPC:       %s: inline threshold too small\n",
+			__func__);
+		return 0;
 	}
-	return result;
+
+	segments = 1 << (fls(bytes / sizeof(struct rpcrdma_segment)) - 1);
+	dprintk("RPC:       %s: max chunk list size = %d segments\n",
+		__func__, segments);
+	return segments;
 }
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index ef3cf4aeecd6..59e627e96a0b 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -334,7 +334,9 @@ struct rpcrdma_stats {
 /*
  * Per-registration mode operations
  */
+struct rpcrdma_xprt;
 struct rpcrdma_memreg_ops {
+	size_t		(*ro_maxpages)(struct rpcrdma_xprt *);
 	const char	*ro_displayname;
 };
 
@@ -411,6 +413,8 @@ struct rpcrdma_regbuf *rpcrdma_alloc_regbuf(struct rpcrdma_ia *,
 void rpcrdma_free_regbuf(struct rpcrdma_ia *,
 			 struct rpcrdma_regbuf *);
 
+unsigned int rpcrdma_max_segments(struct rpcrdma_xprt *);
+
 /*
  * RPC/RDMA connection management calls - xprtrdma/rpc_rdma.c
  */
@@ -422,7 +426,6 @@ void rpcrdma_reply_handler(struct rpcrdma_rep *);
  * RPC/RDMA protocol calls - xprtrdma/rpc_rdma.c
  */
 int rpcrdma_marshal_req(struct rpc_rqst *);
-size_t rpcrdma_max_payload(struct rpcrdma_xprt *);
 
 /* Temporary NFS request map cache. Created in svc_rdma.c  */
 extern struct kmem_cache *svc_rdma_map_cachep;
-- 
cgit v1.2.3


From 9c1b4d775f2d7dd5bb806e3de2f3e1244a7cbd16 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:34:39 -0400
Subject: xprtrdma: Add a "register_external" op for each memreg mode

There is very little common processing among the different external
memory registration functions. Have rpcrdma_create_chunks() call
the registration method directly. This removes a stack frame and a
switch statement from the external registration path.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      |  51 +++++++++++
 net/sunrpc/xprtrdma/frwr_ops.c     |  82 ++++++++++++++++++
 net/sunrpc/xprtrdma/physical_ops.c |  17 ++++
 net/sunrpc/xprtrdma/rpc_rdma.c     |   5 +-
 net/sunrpc/xprtrdma/verbs.c        | 168 +------------------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    |   6 +-
 6 files changed, 160 insertions(+), 169 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index eec266055b28..45fb646a6d36 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -29,7 +29,58 @@ fmr_op_maxpages(struct rpcrdma_xprt *r_xprt)
 		     rpcrdma_max_segments(r_xprt) * RPCRDMA_MAX_FMR_SGES);
 }
 
+/* Use the ib_map_phys_fmr() verb to register a memory region
+ * for remote access via RDMA READ or RDMA WRITE.
+ */
+static int
+fmr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
+	   int nsegs, bool writing)
+{
+	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+	struct rpcrdma_mr_seg *seg1 = seg;
+	struct rpcrdma_mw *mw = seg1->rl_mw;
+	u64 physaddrs[RPCRDMA_MAX_DATA_SEGS];
+	int len, pageoff, i, rc;
+
+	pageoff = offset_in_page(seg1->mr_offset);
+	seg1->mr_offset -= pageoff;	/* start of page */
+	seg1->mr_len += pageoff;
+	len = -pageoff;
+	if (nsegs > RPCRDMA_MAX_FMR_SGES)
+		nsegs = RPCRDMA_MAX_FMR_SGES;
+	for (i = 0; i < nsegs;) {
+		rpcrdma_map_one(ia, seg, writing);
+		physaddrs[i] = seg->mr_dma;
+		len += seg->mr_len;
+		++seg;
+		++i;
+		/* Check for holes */
+		if ((i < nsegs && offset_in_page(seg->mr_offset)) ||
+		    offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len))
+			break;
+	}
+
+	rc = ib_map_phys_fmr(mw->r.fmr, physaddrs, i, seg1->mr_dma);
+	if (rc)
+		goto out_maperr;
+
+	seg1->mr_rkey = mw->r.fmr->rkey;
+	seg1->mr_base = seg1->mr_dma + pageoff;
+	seg1->mr_nsegs = i;
+	seg1->mr_len = len;
+	return i;
+
+out_maperr:
+	dprintk("RPC:       %s: ib_map_phys_fmr %u@0x%llx+%i (%d) status %i\n",
+		__func__, len, (unsigned long long)seg1->mr_dma,
+		pageoff, i, rc);
+	while (i--)
+		rpcrdma_unmap_one(ia, --seg);
+	return rc;
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
+	.ro_map				= fmr_op_map,
 	.ro_maxpages			= fmr_op_maxpages,
 	.ro_displayname			= "fmr",
 };
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 73a5ac898efc..23e4d99a2097 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -29,7 +29,89 @@ frwr_op_maxpages(struct rpcrdma_xprt *r_xprt)
 		     rpcrdma_max_segments(r_xprt) * ia->ri_max_frmr_depth);
 }
 
+/* Post a FAST_REG Work Request to register a memory region
+ * for remote access via RDMA READ or RDMA WRITE.
+ */
+static int
+frwr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
+	    int nsegs, bool writing)
+{
+	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+	struct rpcrdma_mr_seg *seg1 = seg;
+	struct rpcrdma_mw *mw = seg1->rl_mw;
+	struct rpcrdma_frmr *frmr = &mw->r.frmr;
+	struct ib_mr *mr = frmr->fr_mr;
+	struct ib_send_wr fastreg_wr, *bad_wr;
+	u8 key;
+	int len, pageoff;
+	int i, rc;
+	int seg_len;
+	u64 pa;
+	int page_no;
+
+	pageoff = offset_in_page(seg1->mr_offset);
+	seg1->mr_offset -= pageoff;	/* start of page */
+	seg1->mr_len += pageoff;
+	len = -pageoff;
+	if (nsegs > ia->ri_max_frmr_depth)
+		nsegs = ia->ri_max_frmr_depth;
+	for (page_no = i = 0; i < nsegs;) {
+		rpcrdma_map_one(ia, seg, writing);
+		pa = seg->mr_dma;
+		for (seg_len = seg->mr_len; seg_len > 0; seg_len -= PAGE_SIZE) {
+			frmr->fr_pgl->page_list[page_no++] = pa;
+			pa += PAGE_SIZE;
+		}
+		len += seg->mr_len;
+		++seg;
+		++i;
+		/* Check for holes */
+		if ((i < nsegs && offset_in_page(seg->mr_offset)) ||
+		    offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len))
+			break;
+	}
+	dprintk("RPC:       %s: Using frmr %p to map %d segments (%d bytes)\n",
+		__func__, mw, i, len);
+
+	frmr->fr_state = FRMR_IS_VALID;
+
+	memset(&fastreg_wr, 0, sizeof(fastreg_wr));
+	fastreg_wr.wr_id = (unsigned long)(void *)mw;
+	fastreg_wr.opcode = IB_WR_FAST_REG_MR;
+	fastreg_wr.wr.fast_reg.iova_start = seg1->mr_dma + pageoff;
+	fastreg_wr.wr.fast_reg.page_list = frmr->fr_pgl;
+	fastreg_wr.wr.fast_reg.page_shift = PAGE_SHIFT;
+	fastreg_wr.wr.fast_reg.page_list_len = page_no;
+	fastreg_wr.wr.fast_reg.length = len;
+	fastreg_wr.wr.fast_reg.access_flags = writing ?
+				IB_ACCESS_REMOTE_WRITE | IB_ACCESS_LOCAL_WRITE :
+				IB_ACCESS_REMOTE_READ;
+	key = (u8)(mr->rkey & 0x000000FF);
+	ib_update_fast_reg_key(mr, ++key);
+	fastreg_wr.wr.fast_reg.rkey = mr->rkey;
+
+	DECR_CQCOUNT(&r_xprt->rx_ep);
+	rc = ib_post_send(ia->ri_id->qp, &fastreg_wr, &bad_wr);
+	if (rc)
+		goto out_senderr;
+
+	seg1->mr_rkey = mr->rkey;
+	seg1->mr_base = seg1->mr_dma + pageoff;
+	seg1->mr_nsegs = i;
+	seg1->mr_len = len;
+	return i;
+
+out_senderr:
+	dprintk("RPC:       %s: ib_post_send status %i\n", __func__, rc);
+	ib_update_fast_reg_key(mr, --key);
+	frmr->fr_state = FRMR_IS_INVALID;
+	while (i--)
+		rpcrdma_unmap_one(ia, --seg);
+	return rc;
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
+	.ro_map				= frwr_op_map,
 	.ro_maxpages			= frwr_op_maxpages,
 	.ro_displayname			= "frwr",
 };
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index 28ade1943a57..5a284ee0e058 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -28,7 +28,24 @@ physical_op_maxpages(struct rpcrdma_xprt *r_xprt)
 		     rpcrdma_max_segments(r_xprt));
 }
 
+/* The client's physical memory is already exposed for
+ * remote access via RDMA READ or RDMA WRITE.
+ */
+static int
+physical_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
+		int nsegs, bool writing)
+{
+	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+
+	rpcrdma_map_one(ia, seg, writing);
+	seg->mr_rkey = ia->ri_bind_mem->rkey;
+	seg->mr_base = seg->mr_dma;
+	seg->mr_nsegs = 1;
+	return 1;
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
+	.ro_map				= physical_op_map,
 	.ro_maxpages			= physical_op_maxpages,
 	.ro_displayname			= "physical",
 };
diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c
index 41456d9e5a7d..6ab1d03d7f3e 100644
--- a/net/sunrpc/xprtrdma/rpc_rdma.c
+++ b/net/sunrpc/xprtrdma/rpc_rdma.c
@@ -187,6 +187,7 @@ rpcrdma_create_chunks(struct rpc_rqst *rqst, struct xdr_buf *target,
 	struct rpcrdma_write_array *warray = NULL;
 	struct rpcrdma_write_chunk *cur_wchunk = NULL;
 	__be32 *iptr = headerp->rm_body.rm_chunks;
+	int (*map)(struct rpcrdma_xprt *, struct rpcrdma_mr_seg *, int, bool);
 
 	if (type == rpcrdma_readch || type == rpcrdma_areadch) {
 		/* a read chunk - server will RDMA Read our memory */
@@ -209,9 +210,9 @@ rpcrdma_create_chunks(struct rpc_rqst *rqst, struct xdr_buf *target,
 	if (nsegs < 0)
 		return nsegs;
 
+	map = r_xprt->rx_ia.ri_ops->ro_map;
 	do {
-		n = rpcrdma_register_external(seg, nsegs,
-						cur_wchunk != NULL, r_xprt);
+		n = map(r_xprt, seg, nsegs, cur_wchunk != NULL);
 		if (n <= 0)
 			goto out;
 		if (cur_rchunk) {	/* read */
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index da55cda30568..4318c04e095c 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1858,8 +1858,8 @@ rpcrdma_free_regbuf(struct rpcrdma_ia *ia, struct rpcrdma_regbuf *rb)
  * Wrappers for chunk registration, shared by read/write chunk code.
  */
 
-static void
-rpcrdma_map_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg, int writing)
+void
+rpcrdma_map_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg, bool writing)
 {
 	seg->mr_dir = writing ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
 	seg->mr_dmalen = seg->mr_len;
@@ -1879,7 +1879,7 @@ rpcrdma_map_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg, int writing)
 	}
 }
 
-static void
+void
 rpcrdma_unmap_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg)
 {
 	if (seg->mr_page)
@@ -1890,89 +1890,6 @@ rpcrdma_unmap_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg)
 				seg->mr_dma, seg->mr_dmalen, seg->mr_dir);
 }
 
-static int
-rpcrdma_register_frmr_external(struct rpcrdma_mr_seg *seg,
-			int *nsegs, int writing, struct rpcrdma_ia *ia,
-			struct rpcrdma_xprt *r_xprt)
-{
-	struct rpcrdma_mr_seg *seg1 = seg;
-	struct rpcrdma_mw *mw = seg1->rl_mw;
-	struct rpcrdma_frmr *frmr = &mw->r.frmr;
-	struct ib_mr *mr = frmr->fr_mr;
-	struct ib_send_wr fastreg_wr, *bad_wr;
-	u8 key;
-	int len, pageoff;
-	int i, rc;
-	int seg_len;
-	u64 pa;
-	int page_no;
-
-	pageoff = offset_in_page(seg1->mr_offset);
-	seg1->mr_offset -= pageoff;	/* start of page */
-	seg1->mr_len += pageoff;
-	len = -pageoff;
-	if (*nsegs > ia->ri_max_frmr_depth)
-		*nsegs = ia->ri_max_frmr_depth;
-	for (page_no = i = 0; i < *nsegs;) {
-		rpcrdma_map_one(ia, seg, writing);
-		pa = seg->mr_dma;
-		for (seg_len = seg->mr_len; seg_len > 0; seg_len -= PAGE_SIZE) {
-			frmr->fr_pgl->page_list[page_no++] = pa;
-			pa += PAGE_SIZE;
-		}
-		len += seg->mr_len;
-		++seg;
-		++i;
-		/* Check for holes */
-		if ((i < *nsegs && offset_in_page(seg->mr_offset)) ||
-		    offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len))
-			break;
-	}
-	dprintk("RPC:       %s: Using frmr %p to map %d segments (%d bytes)\n",
-		__func__, mw, i, len);
-
-	frmr->fr_state = FRMR_IS_VALID;
-
-	memset(&fastreg_wr, 0, sizeof(fastreg_wr));
-	fastreg_wr.wr_id = (unsigned long)(void *)mw;
-	fastreg_wr.opcode = IB_WR_FAST_REG_MR;
-	fastreg_wr.wr.fast_reg.iova_start = seg1->mr_dma + pageoff;
-	fastreg_wr.wr.fast_reg.page_list = frmr->fr_pgl;
-	fastreg_wr.wr.fast_reg.page_list_len = page_no;
-	fastreg_wr.wr.fast_reg.page_shift = PAGE_SHIFT;
-	fastreg_wr.wr.fast_reg.length = len;
-
-	/* Bump the key */
-	key = (u8)(mr->rkey & 0x000000FF);
-	ib_update_fast_reg_key(mr, ++key);
-
-	fastreg_wr.wr.fast_reg.access_flags = (writing ?
-				IB_ACCESS_REMOTE_WRITE | IB_ACCESS_LOCAL_WRITE :
-				IB_ACCESS_REMOTE_READ);
-	fastreg_wr.wr.fast_reg.rkey = mr->rkey;
-	DECR_CQCOUNT(&r_xprt->rx_ep);
-
-	rc = ib_post_send(ia->ri_id->qp, &fastreg_wr, &bad_wr);
-	if (rc) {
-		dprintk("RPC:       %s: failed ib_post_send for register,"
-			" status %i\n", __func__, rc);
-		ib_update_fast_reg_key(mr, --key);
-		goto out_err;
-	} else {
-		seg1->mr_rkey = mr->rkey;
-		seg1->mr_base = seg1->mr_dma + pageoff;
-		seg1->mr_nsegs = i;
-		seg1->mr_len = len;
-	}
-	*nsegs = i;
-	return 0;
-out_err:
-	frmr->fr_state = FRMR_IS_INVALID;
-	while (i--)
-		rpcrdma_unmap_one(ia, --seg);
-	return rc;
-}
-
 static int
 rpcrdma_deregister_frmr_external(struct rpcrdma_mr_seg *seg,
 			struct rpcrdma_ia *ia, struct rpcrdma_xprt *r_xprt)
@@ -2003,49 +1920,6 @@ rpcrdma_deregister_frmr_external(struct rpcrdma_mr_seg *seg,
 	return rc;
 }
 
-static int
-rpcrdma_register_fmr_external(struct rpcrdma_mr_seg *seg,
-			int *nsegs, int writing, struct rpcrdma_ia *ia)
-{
-	struct rpcrdma_mr_seg *seg1 = seg;
-	u64 physaddrs[RPCRDMA_MAX_DATA_SEGS];
-	int len, pageoff, i, rc;
-
-	pageoff = offset_in_page(seg1->mr_offset);
-	seg1->mr_offset -= pageoff;	/* start of page */
-	seg1->mr_len += pageoff;
-	len = -pageoff;
-	if (*nsegs > RPCRDMA_MAX_DATA_SEGS)
-		*nsegs = RPCRDMA_MAX_DATA_SEGS;
-	for (i = 0; i < *nsegs;) {
-		rpcrdma_map_one(ia, seg, writing);
-		physaddrs[i] = seg->mr_dma;
-		len += seg->mr_len;
-		++seg;
-		++i;
-		/* Check for holes */
-		if ((i < *nsegs && offset_in_page(seg->mr_offset)) ||
-		    offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len))
-			break;
-	}
-	rc = ib_map_phys_fmr(seg1->rl_mw->r.fmr, physaddrs, i, seg1->mr_dma);
-	if (rc) {
-		dprintk("RPC:       %s: failed ib_map_phys_fmr "
-			"%u@0x%llx+%i (%d)... status %i\n", __func__,
-			len, (unsigned long long)seg1->mr_dma,
-			pageoff, i, rc);
-		while (i--)
-			rpcrdma_unmap_one(ia, --seg);
-	} else {
-		seg1->mr_rkey = seg1->rl_mw->r.fmr->rkey;
-		seg1->mr_base = seg1->mr_dma + pageoff;
-		seg1->mr_nsegs = i;
-		seg1->mr_len = len;
-	}
-	*nsegs = i;
-	return rc;
-}
-
 static int
 rpcrdma_deregister_fmr_external(struct rpcrdma_mr_seg *seg,
 			struct rpcrdma_ia *ia)
@@ -2066,42 +1940,6 @@ rpcrdma_deregister_fmr_external(struct rpcrdma_mr_seg *seg,
 	return rc;
 }
 
-int
-rpcrdma_register_external(struct rpcrdma_mr_seg *seg,
-			int nsegs, int writing, struct rpcrdma_xprt *r_xprt)
-{
-	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
-	int rc = 0;
-
-	switch (ia->ri_memreg_strategy) {
-
-	case RPCRDMA_ALLPHYSICAL:
-		rpcrdma_map_one(ia, seg, writing);
-		seg->mr_rkey = ia->ri_bind_mem->rkey;
-		seg->mr_base = seg->mr_dma;
-		seg->mr_nsegs = 1;
-		nsegs = 1;
-		break;
-
-	/* Registration using frmr registration */
-	case RPCRDMA_FRMR:
-		rc = rpcrdma_register_frmr_external(seg, &nsegs, writing, ia, r_xprt);
-		break;
-
-	/* Registration using fmr memory registration */
-	case RPCRDMA_MTHCAFMR:
-		rc = rpcrdma_register_fmr_external(seg, &nsegs, writing, ia);
-		break;
-
-	default:
-		return -EIO;
-	}
-	if (rc)
-		return rc;
-
-	return nsegs;
-}
-
 int
 rpcrdma_deregister_external(struct rpcrdma_mr_seg *seg,
 		struct rpcrdma_xprt *r_xprt)
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 59e627e96a0b..7bf077bf751d 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -336,6 +336,8 @@ struct rpcrdma_stats {
  */
 struct rpcrdma_xprt;
 struct rpcrdma_memreg_ops {
+	int		(*ro_map)(struct rpcrdma_xprt *,
+				  struct rpcrdma_mr_seg *, int, bool);
 	size_t		(*ro_maxpages)(struct rpcrdma_xprt *);
 	const char	*ro_displayname;
 };
@@ -403,8 +405,6 @@ void rpcrdma_buffer_put(struct rpcrdma_req *);
 void rpcrdma_recv_buffer_get(struct rpcrdma_req *);
 void rpcrdma_recv_buffer_put(struct rpcrdma_rep *);
 
-int rpcrdma_register_external(struct rpcrdma_mr_seg *,
-				int, int, struct rpcrdma_xprt *);
 int rpcrdma_deregister_external(struct rpcrdma_mr_seg *,
 				struct rpcrdma_xprt *);
 
@@ -414,6 +414,8 @@ void rpcrdma_free_regbuf(struct rpcrdma_ia *,
 			 struct rpcrdma_regbuf *);
 
 unsigned int rpcrdma_max_segments(struct rpcrdma_xprt *);
+void rpcrdma_map_one(struct rpcrdma_ia *, struct rpcrdma_mr_seg *, bool);
+void rpcrdma_unmap_one(struct rpcrdma_ia *, struct rpcrdma_mr_seg *);
 
 /*
  * RPC/RDMA connection management calls - xprtrdma/rpc_rdma.c
-- 
cgit v1.2.3


From 6814baead86b5d44096ddfbb6f944163578e68c3 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:34:48 -0400
Subject: xprtrdma: Add a "deregister_external" op for each memreg mode

There is very little common processing among the different external
memory deregistration functions.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      | 27 +++++++++++++
 net/sunrpc/xprtrdma/frwr_ops.c     | 36 +++++++++++++++++
 net/sunrpc/xprtrdma/physical_ops.c | 10 +++++
 net/sunrpc/xprtrdma/rpc_rdma.c     | 11 +++---
 net/sunrpc/xprtrdma/transport.c    |  4 +-
 net/sunrpc/xprtrdma/verbs.c        | 81 +-------------------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    |  5 +--
 7 files changed, 84 insertions(+), 90 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index 45fb646a6d36..888aa107cf0b 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -79,8 +79,35 @@ out_maperr:
 	return rc;
 }
 
+/* Use the ib_unmap_fmr() verb to prevent further remote
+ * access via RDMA READ or RDMA WRITE.
+ */
+static int
+fmr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
+{
+	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+	struct rpcrdma_mr_seg *seg1 = seg;
+	int rc, nsegs = seg->mr_nsegs;
+	LIST_HEAD(l);
+
+	list_add(&seg1->rl_mw->r.fmr->list, &l);
+	rc = ib_unmap_fmr(&l);
+	read_lock(&ia->ri_qplock);
+	while (seg1->mr_nsegs--)
+		rpcrdma_unmap_one(ia, seg++);
+	read_unlock(&ia->ri_qplock);
+	if (rc)
+		goto out_err;
+	return nsegs;
+
+out_err:
+	dprintk("RPC:       %s: ib_unmap_fmr status %i\n", __func__, rc);
+	return nsegs;
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
 	.ro_map				= fmr_op_map,
+	.ro_unmap			= fmr_op_unmap,
 	.ro_maxpages			= fmr_op_maxpages,
 	.ro_displayname			= "fmr",
 };
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 23e4d99a2097..35b725bf0afb 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -110,8 +110,44 @@ out_senderr:
 	return rc;
 }
 
+/* Post a LOCAL_INV Work Request to prevent further remote access
+ * via RDMA READ or RDMA WRITE.
+ */
+static int
+frwr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
+{
+	struct rpcrdma_mr_seg *seg1 = seg;
+	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+	struct ib_send_wr invalidate_wr, *bad_wr;
+	int rc, nsegs = seg->mr_nsegs;
+
+	seg1->rl_mw->r.frmr.fr_state = FRMR_IS_INVALID;
+
+	memset(&invalidate_wr, 0, sizeof(invalidate_wr));
+	invalidate_wr.wr_id = (unsigned long)(void *)seg1->rl_mw;
+	invalidate_wr.opcode = IB_WR_LOCAL_INV;
+	invalidate_wr.ex.invalidate_rkey = seg1->rl_mw->r.frmr.fr_mr->rkey;
+	DECR_CQCOUNT(&r_xprt->rx_ep);
+
+	read_lock(&ia->ri_qplock);
+	while (seg1->mr_nsegs--)
+		rpcrdma_unmap_one(ia, seg++);
+	rc = ib_post_send(ia->ri_id->qp, &invalidate_wr, &bad_wr);
+	read_unlock(&ia->ri_qplock);
+	if (rc)
+		goto out_err;
+	return nsegs;
+
+out_err:
+	/* Force rpcrdma_buffer_get() to retry */
+	seg1->rl_mw->r.frmr.fr_state = FRMR_IS_STALE;
+	dprintk("RPC:       %s: ib_post_send status %i\n", __func__, rc);
+	return nsegs;
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
 	.ro_map				= frwr_op_map,
+	.ro_unmap			= frwr_op_unmap,
 	.ro_maxpages			= frwr_op_maxpages,
 	.ro_displayname			= "frwr",
 };
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index 5a284ee0e058..5b5a63adfdf3 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -44,8 +44,18 @@ physical_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 	return 1;
 }
 
+/* Unmap a memory region, but leave it registered.
+ */
+static int
+physical_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
+{
+	rpcrdma_unmap_one(&r_xprt->rx_ia, seg);
+	return 1;
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
 	.ro_map				= physical_op_map,
+	.ro_unmap			= physical_op_unmap,
 	.ro_maxpages			= physical_op_maxpages,
 	.ro_displayname			= "physical",
 };
diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c
index 6ab1d03d7f3e..2c53ea9e1b83 100644
--- a/net/sunrpc/xprtrdma/rpc_rdma.c
+++ b/net/sunrpc/xprtrdma/rpc_rdma.c
@@ -284,11 +284,12 @@ rpcrdma_create_chunks(struct rpc_rqst *rqst, struct xdr_buf *target,
 	return (unsigned char *)iptr - (unsigned char *)headerp;
 
 out:
-	if (r_xprt->rx_ia.ri_memreg_strategy != RPCRDMA_FRMR) {
-		for (pos = 0; nchunks--;)
-			pos += rpcrdma_deregister_external(
-					&req->rl_segments[pos], r_xprt);
-	}
+	if (r_xprt->rx_ia.ri_memreg_strategy == RPCRDMA_FRMR)
+		return n;
+
+	for (pos = 0; nchunks--;)
+		pos += r_xprt->rx_ia.ri_ops->ro_unmap(r_xprt,
+						      &req->rl_segments[pos]);
 	return n;
 }
 
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index da71a24641e3..54f23b1be986 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -584,8 +584,8 @@ xprt_rdma_free(void *buffer)
 
 	for (i = 0; req->rl_nchunks;) {
 		--req->rl_nchunks;
-		i += rpcrdma_deregister_external(
-			&req->rl_segments[i], r_xprt);
+		i += r_xprt->rx_ia.ri_ops->ro_unmap(r_xprt,
+						    &req->rl_segments[i]);
 	}
 
 	rpcrdma_buffer_put(req);
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 4318c04e095c..b167c99fbfb6 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1510,7 +1510,7 @@ rpcrdma_buffer_put_sendbuf(struct rpcrdma_req *req, struct rpcrdma_buffer *buf)
 	}
 }
 
-/* rpcrdma_unmap_one() was already done by rpcrdma_deregister_frmr_external().
+/* rpcrdma_unmap_one() was already done during deregistration.
  * Redo only the ib_post_send().
  */
 static void
@@ -1890,85 +1890,6 @@ rpcrdma_unmap_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg)
 				seg->mr_dma, seg->mr_dmalen, seg->mr_dir);
 }
 
-static int
-rpcrdma_deregister_frmr_external(struct rpcrdma_mr_seg *seg,
-			struct rpcrdma_ia *ia, struct rpcrdma_xprt *r_xprt)
-{
-	struct rpcrdma_mr_seg *seg1 = seg;
-	struct ib_send_wr invalidate_wr, *bad_wr;
-	int rc;
-
-	seg1->rl_mw->r.frmr.fr_state = FRMR_IS_INVALID;
-
-	memset(&invalidate_wr, 0, sizeof invalidate_wr);
-	invalidate_wr.wr_id = (unsigned long)(void *)seg1->rl_mw;
-	invalidate_wr.opcode = IB_WR_LOCAL_INV;
-	invalidate_wr.ex.invalidate_rkey = seg1->rl_mw->r.frmr.fr_mr->rkey;
-	DECR_CQCOUNT(&r_xprt->rx_ep);
-
-	read_lock(&ia->ri_qplock);
-	while (seg1->mr_nsegs--)
-		rpcrdma_unmap_one(ia, seg++);
-	rc = ib_post_send(ia->ri_id->qp, &invalidate_wr, &bad_wr);
-	read_unlock(&ia->ri_qplock);
-	if (rc) {
-		/* Force rpcrdma_buffer_get() to retry */
-		seg1->rl_mw->r.frmr.fr_state = FRMR_IS_STALE;
-		dprintk("RPC:       %s: failed ib_post_send for invalidate,"
-			" status %i\n", __func__, rc);
-	}
-	return rc;
-}
-
-static int
-rpcrdma_deregister_fmr_external(struct rpcrdma_mr_seg *seg,
-			struct rpcrdma_ia *ia)
-{
-	struct rpcrdma_mr_seg *seg1 = seg;
-	LIST_HEAD(l);
-	int rc;
-
-	list_add(&seg1->rl_mw->r.fmr->list, &l);
-	rc = ib_unmap_fmr(&l);
-	read_lock(&ia->ri_qplock);
-	while (seg1->mr_nsegs--)
-		rpcrdma_unmap_one(ia, seg++);
-	read_unlock(&ia->ri_qplock);
-	if (rc)
-		dprintk("RPC:       %s: failed ib_unmap_fmr,"
-			" status %i\n", __func__, rc);
-	return rc;
-}
-
-int
-rpcrdma_deregister_external(struct rpcrdma_mr_seg *seg,
-		struct rpcrdma_xprt *r_xprt)
-{
-	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
-	int nsegs = seg->mr_nsegs, rc;
-
-	switch (ia->ri_memreg_strategy) {
-
-	case RPCRDMA_ALLPHYSICAL:
-		read_lock(&ia->ri_qplock);
-		rpcrdma_unmap_one(ia, seg);
-		read_unlock(&ia->ri_qplock);
-		break;
-
-	case RPCRDMA_FRMR:
-		rc = rpcrdma_deregister_frmr_external(seg, ia, r_xprt);
-		break;
-
-	case RPCRDMA_MTHCAFMR:
-		rc = rpcrdma_deregister_fmr_external(seg, ia);
-		break;
-
-	default:
-		break;
-	}
-	return nsegs;
-}
-
 /*
  * Prepost any receive buffer, then post send.
  *
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 7bf077bf751d..9a727f960df9 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -338,6 +338,8 @@ struct rpcrdma_xprt;
 struct rpcrdma_memreg_ops {
 	int		(*ro_map)(struct rpcrdma_xprt *,
 				  struct rpcrdma_mr_seg *, int, bool);
+	int		(*ro_unmap)(struct rpcrdma_xprt *,
+				    struct rpcrdma_mr_seg *);
 	size_t		(*ro_maxpages)(struct rpcrdma_xprt *);
 	const char	*ro_displayname;
 };
@@ -405,9 +407,6 @@ void rpcrdma_buffer_put(struct rpcrdma_req *);
 void rpcrdma_recv_buffer_get(struct rpcrdma_req *);
 void rpcrdma_recv_buffer_put(struct rpcrdma_rep *);
 
-int rpcrdma_deregister_external(struct rpcrdma_mr_seg *,
-				struct rpcrdma_xprt *);
-
 struct rpcrdma_regbuf *rpcrdma_alloc_regbuf(struct rpcrdma_ia *,
 					    size_t, gfp_t);
 void rpcrdma_free_regbuf(struct rpcrdma_ia *,
-- 
cgit v1.2.3


From 91e70e70e47b3355bb0a8b3b196c93897dcdb440 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:34:58 -0400
Subject: xprtrdma: Add "init MRs" memreg op

This method is used when setting up a new transport instance to
create a pool of Memory Region objects that will be used to register
memory during operation.

Memory Regions are not needed for "physical" registration, since
->prepare and ->release are no-ops for that mode.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      |  42 +++++++++++++++
 net/sunrpc/xprtrdma/frwr_ops.c     |  66 +++++++++++++++++++++++
 net/sunrpc/xprtrdma/physical_ops.c |   7 +++
 net/sunrpc/xprtrdma/verbs.c        | 104 ++-----------------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    |   1 +
 5 files changed, 119 insertions(+), 101 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index 888aa107cf0b..825ce9641350 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -29,6 +29,47 @@ fmr_op_maxpages(struct rpcrdma_xprt *r_xprt)
 		     rpcrdma_max_segments(r_xprt) * RPCRDMA_MAX_FMR_SGES);
 }
 
+static int
+fmr_op_init(struct rpcrdma_xprt *r_xprt)
+{
+	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
+	int mr_access_flags = IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_READ;
+	struct ib_fmr_attr fmr_attr = {
+		.max_pages	= RPCRDMA_MAX_FMR_SGES,
+		.max_maps	= 1,
+		.page_shift	= PAGE_SHIFT
+	};
+	struct ib_pd *pd = r_xprt->rx_ia.ri_pd;
+	struct rpcrdma_mw *r;
+	int i, rc;
+
+	INIT_LIST_HEAD(&buf->rb_mws);
+	INIT_LIST_HEAD(&buf->rb_all);
+
+	i = (buf->rb_max_requests + 1) * RPCRDMA_MAX_SEGS;
+	dprintk("RPC:       %s: initalizing %d FMRs\n", __func__, i);
+
+	while (i--) {
+		r = kzalloc(sizeof(*r), GFP_KERNEL);
+		if (!r)
+			return -ENOMEM;
+
+		r->r.fmr = ib_alloc_fmr(pd, mr_access_flags, &fmr_attr);
+		if (IS_ERR(r->r.fmr))
+			goto out_fmr_err;
+
+		list_add(&r->mw_list, &buf->rb_mws);
+		list_add(&r->mw_all, &buf->rb_all);
+	}
+	return 0;
+
+out_fmr_err:
+	rc = PTR_ERR(r->r.fmr);
+	dprintk("RPC:       %s: ib_alloc_fmr status %i\n", __func__, rc);
+	kfree(r);
+	return rc;
+}
+
 /* Use the ib_map_phys_fmr() verb to register a memory region
  * for remote access via RDMA READ or RDMA WRITE.
  */
@@ -109,5 +150,6 @@ const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
 	.ro_map				= fmr_op_map,
 	.ro_unmap			= fmr_op_unmap,
 	.ro_maxpages			= fmr_op_maxpages,
+	.ro_init			= fmr_op_init,
 	.ro_displayname			= "fmr",
 };
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 35b725bf0afb..9168c15faafa 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -17,6 +17,35 @@
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
+static int
+__frwr_init(struct rpcrdma_mw *r, struct ib_pd *pd, struct ib_device *device,
+	    unsigned int depth)
+{
+	struct rpcrdma_frmr *f = &r->r.frmr;
+	int rc;
+
+	f->fr_mr = ib_alloc_fast_reg_mr(pd, depth);
+	if (IS_ERR(f->fr_mr))
+		goto out_mr_err;
+	f->fr_pgl = ib_alloc_fast_reg_page_list(device, depth);
+	if (IS_ERR(f->fr_pgl))
+		goto out_list_err;
+	return 0;
+
+out_mr_err:
+	rc = PTR_ERR(f->fr_mr);
+	dprintk("RPC:       %s: ib_alloc_fast_reg_mr status %i\n",
+		__func__, rc);
+	return rc;
+
+out_list_err:
+	rc = PTR_ERR(f->fr_pgl);
+	dprintk("RPC:       %s: ib_alloc_fast_reg_page_list status %i\n",
+		__func__, rc);
+	ib_dereg_mr(f->fr_mr);
+	return rc;
+}
+
 /* FRWR mode conveys a list of pages per chunk segment. The
  * maximum length of that list is the FRWR page list depth.
  */
@@ -29,6 +58,42 @@ frwr_op_maxpages(struct rpcrdma_xprt *r_xprt)
 		     rpcrdma_max_segments(r_xprt) * ia->ri_max_frmr_depth);
 }
 
+static int
+frwr_op_init(struct rpcrdma_xprt *r_xprt)
+{
+	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
+	struct ib_device *device = r_xprt->rx_ia.ri_id->device;
+	unsigned int depth = r_xprt->rx_ia.ri_max_frmr_depth;
+	struct ib_pd *pd = r_xprt->rx_ia.ri_pd;
+	int i;
+
+	INIT_LIST_HEAD(&buf->rb_mws);
+	INIT_LIST_HEAD(&buf->rb_all);
+
+	i = (buf->rb_max_requests + 1) * RPCRDMA_MAX_SEGS;
+	dprintk("RPC:       %s: initalizing %d FRMRs\n", __func__, i);
+
+	while (i--) {
+		struct rpcrdma_mw *r;
+		int rc;
+
+		r = kzalloc(sizeof(*r), GFP_KERNEL);
+		if (!r)
+			return -ENOMEM;
+
+		rc = __frwr_init(r, pd, device, depth);
+		if (rc) {
+			kfree(r);
+			return rc;
+		}
+
+		list_add(&r->mw_list, &buf->rb_mws);
+		list_add(&r->mw_all, &buf->rb_all);
+	}
+
+	return 0;
+}
+
 /* Post a FAST_REG Work Request to register a memory region
  * for remote access via RDMA READ or RDMA WRITE.
  */
@@ -149,5 +214,6 @@ const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
 	.ro_map				= frwr_op_map,
 	.ro_unmap			= frwr_op_unmap,
 	.ro_maxpages			= frwr_op_maxpages,
+	.ro_init			= frwr_op_init,
 	.ro_displayname			= "frwr",
 };
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index 5b5a63adfdf3..c37205190a2f 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -28,6 +28,12 @@ physical_op_maxpages(struct rpcrdma_xprt *r_xprt)
 		     rpcrdma_max_segments(r_xprt));
 }
 
+static int
+physical_op_init(struct rpcrdma_xprt *r_xprt)
+{
+	return 0;
+}
+
 /* The client's physical memory is already exposed for
  * remote access via RDMA READ or RDMA WRITE.
  */
@@ -57,5 +63,6 @@ const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
 	.ro_map				= physical_op_map,
 	.ro_unmap			= physical_op_unmap,
 	.ro_maxpages			= physical_op_maxpages,
+	.ro_init			= physical_op_init,
 	.ro_displayname			= "physical",
 };
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index b167c99fbfb6..e89a57d4e4f2 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1124,91 +1124,6 @@ out:
 	return ERR_PTR(rc);
 }
 
-static int
-rpcrdma_init_fmrs(struct rpcrdma_ia *ia, struct rpcrdma_buffer *buf)
-{
-	int mr_access_flags = IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_READ;
-	struct ib_fmr_attr fmr_attr = {
-		.max_pages	= RPCRDMA_MAX_DATA_SEGS,
-		.max_maps	= 1,
-		.page_shift	= PAGE_SHIFT
-	};
-	struct rpcrdma_mw *r;
-	int i, rc;
-
-	i = (buf->rb_max_requests + 1) * RPCRDMA_MAX_SEGS;
-	dprintk("RPC:       %s: initalizing %d FMRs\n", __func__, i);
-
-	while (i--) {
-		r = kzalloc(sizeof(*r), GFP_KERNEL);
-		if (r == NULL)
-			return -ENOMEM;
-
-		r->r.fmr = ib_alloc_fmr(ia->ri_pd, mr_access_flags, &fmr_attr);
-		if (IS_ERR(r->r.fmr)) {
-			rc = PTR_ERR(r->r.fmr);
-			dprintk("RPC:       %s: ib_alloc_fmr failed %i\n",
-				__func__, rc);
-			goto out_free;
-		}
-
-		list_add(&r->mw_list, &buf->rb_mws);
-		list_add(&r->mw_all, &buf->rb_all);
-	}
-	return 0;
-
-out_free:
-	kfree(r);
-	return rc;
-}
-
-static int
-rpcrdma_init_frmrs(struct rpcrdma_ia *ia, struct rpcrdma_buffer *buf)
-{
-	struct rpcrdma_frmr *f;
-	struct rpcrdma_mw *r;
-	int i, rc;
-
-	i = (buf->rb_max_requests + 1) * RPCRDMA_MAX_SEGS;
-	dprintk("RPC:       %s: initalizing %d FRMRs\n", __func__, i);
-
-	while (i--) {
-		r = kzalloc(sizeof(*r), GFP_KERNEL);
-		if (r == NULL)
-			return -ENOMEM;
-		f = &r->r.frmr;
-
-		f->fr_mr = ib_alloc_fast_reg_mr(ia->ri_pd,
-						ia->ri_max_frmr_depth);
-		if (IS_ERR(f->fr_mr)) {
-			rc = PTR_ERR(f->fr_mr);
-			dprintk("RPC:       %s: ib_alloc_fast_reg_mr "
-				"failed %i\n", __func__, rc);
-			goto out_free;
-		}
-
-		f->fr_pgl = ib_alloc_fast_reg_page_list(ia->ri_id->device,
-							ia->ri_max_frmr_depth);
-		if (IS_ERR(f->fr_pgl)) {
-			rc = PTR_ERR(f->fr_pgl);
-			dprintk("RPC:       %s: ib_alloc_fast_reg_page_list "
-				"failed %i\n", __func__, rc);
-
-			ib_dereg_mr(f->fr_mr);
-			goto out_free;
-		}
-
-		list_add(&r->mw_list, &buf->rb_mws);
-		list_add(&r->mw_all, &buf->rb_all);
-	}
-
-	return 0;
-
-out_free:
-	kfree(r);
-	return rc;
-}
-
 int
 rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt)
 {
@@ -1245,22 +1160,9 @@ rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt)
 	buf->rb_recv_bufs = (struct rpcrdma_rep **) p;
 	p = (char *) &buf->rb_recv_bufs[buf->rb_max_requests];
 
-	INIT_LIST_HEAD(&buf->rb_mws);
-	INIT_LIST_HEAD(&buf->rb_all);
-	switch (ia->ri_memreg_strategy) {
-	case RPCRDMA_FRMR:
-		rc = rpcrdma_init_frmrs(ia, buf);
-		if (rc)
-			goto out;
-		break;
-	case RPCRDMA_MTHCAFMR:
-		rc = rpcrdma_init_fmrs(ia, buf);
-		if (rc)
-			goto out;
-		break;
-	default:
-		break;
-	}
+	rc = ia->ri_ops->ro_init(r_xprt);
+	if (rc)
+		goto out;
 
 	for (i = 0; i < buf->rb_max_requests; i++) {
 		struct rpcrdma_req *req;
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 9a727f960df9..90b60feab45a 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -341,6 +341,7 @@ struct rpcrdma_memreg_ops {
 	int		(*ro_unmap)(struct rpcrdma_xprt *,
 				    struct rpcrdma_mr_seg *);
 	size_t		(*ro_maxpages)(struct rpcrdma_xprt *);
+	int		(*ro_init)(struct rpcrdma_xprt *);
 	const char	*ro_displayname;
 };
 
-- 
cgit v1.2.3


From 31a701a94751509bb72e13d851f18ddcf22ff722 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:35:07 -0400
Subject: xprtrdma: Add "reset MRs" memreg op

This method is invoked when a transport instance is about to be
reconnected. Each Memory Region object is reset to its initial
state.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      |  23 +++++++++
 net/sunrpc/xprtrdma/frwr_ops.c     |  51 ++++++++++++++++++
 net/sunrpc/xprtrdma/physical_ops.c |   6 +++
 net/sunrpc/xprtrdma/verbs.c        | 103 +------------------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    |   1 +
 5 files changed, 83 insertions(+), 101 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index 825ce9641350..93261b05891e 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -146,10 +146,33 @@ out_err:
 	return nsegs;
 }
 
+/* After a disconnect, unmap all FMRs.
+ *
+ * This is invoked only in the transport connect worker in order
+ * to serialize with rpcrdma_register_fmr_external().
+ */
+static void
+fmr_op_reset(struct rpcrdma_xprt *r_xprt)
+{
+	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
+	struct rpcrdma_mw *r;
+	LIST_HEAD(list);
+	int rc;
+
+	list_for_each_entry(r, &buf->rb_all, mw_all)
+		list_add(&r->r.fmr->list, &list);
+
+	rc = ib_unmap_fmr(&list);
+	if (rc)
+		dprintk("RPC:       %s: ib_unmap_fmr failed %i\n",
+			__func__, rc);
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
 	.ro_map				= fmr_op_map,
 	.ro_unmap			= fmr_op_unmap,
 	.ro_maxpages			= fmr_op_maxpages,
 	.ro_init			= fmr_op_init,
+	.ro_reset			= fmr_op_reset,
 	.ro_displayname			= "fmr",
 };
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 9168c15faafa..c2bb29d4df3c 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -46,6 +46,18 @@ out_list_err:
 	return rc;
 }
 
+static void
+__frwr_release(struct rpcrdma_mw *r)
+{
+	int rc;
+
+	rc = ib_dereg_mr(r->r.frmr.fr_mr);
+	if (rc)
+		dprintk("RPC:       %s: ib_dereg_mr status %i\n",
+			__func__, rc);
+	ib_free_fast_reg_page_list(r->r.frmr.fr_pgl);
+}
+
 /* FRWR mode conveys a list of pages per chunk segment. The
  * maximum length of that list is the FRWR page list depth.
  */
@@ -210,10 +222,49 @@ out_err:
 	return nsegs;
 }
 
+/* After a disconnect, a flushed FAST_REG_MR can leave an FRMR in
+ * an unusable state. Find FRMRs in this state and dereg / reg
+ * each.  FRMRs that are VALID and attached to an rpcrdma_req are
+ * also torn down.
+ *
+ * This gives all in-use FRMRs a fresh rkey and leaves them INVALID.
+ *
+ * This is invoked only in the transport connect worker in order
+ * to serialize with rpcrdma_register_frmr_external().
+ */
+static void
+frwr_op_reset(struct rpcrdma_xprt *r_xprt)
+{
+	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
+	struct ib_device *device = r_xprt->rx_ia.ri_id->device;
+	unsigned int depth = r_xprt->rx_ia.ri_max_frmr_depth;
+	struct ib_pd *pd = r_xprt->rx_ia.ri_pd;
+	struct rpcrdma_mw *r;
+	int rc;
+
+	list_for_each_entry(r, &buf->rb_all, mw_all) {
+		if (r->r.frmr.fr_state == FRMR_IS_INVALID)
+			continue;
+
+		__frwr_release(r);
+		rc = __frwr_init(r, pd, device, depth);
+		if (rc) {
+			dprintk("RPC:       %s: mw %p left %s\n",
+				__func__, r,
+				(r->r.frmr.fr_state == FRMR_IS_STALE ?
+					"stale" : "valid"));
+			continue;
+		}
+
+		r->r.frmr.fr_state = FRMR_IS_INVALID;
+	}
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
 	.ro_map				= frwr_op_map,
 	.ro_unmap			= frwr_op_unmap,
 	.ro_maxpages			= frwr_op_maxpages,
 	.ro_init			= frwr_op_init,
+	.ro_reset			= frwr_op_reset,
 	.ro_displayname			= "frwr",
 };
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index c37205190a2f..e0607136655b 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -59,10 +59,16 @@ physical_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
 	return 1;
 }
 
+static void
+physical_op_reset(struct rpcrdma_xprt *r_xprt)
+{
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
 	.ro_map				= physical_op_map,
 	.ro_unmap			= physical_op_unmap,
 	.ro_maxpages			= physical_op_maxpages,
 	.ro_init			= physical_op_init,
+	.ro_reset			= physical_op_reset,
 	.ro_displayname			= "physical",
 };
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index e89a57d4e4f2..1b2c1f4ec4c3 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -63,9 +63,6 @@
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
-static void rpcrdma_reset_frmrs(struct rpcrdma_ia *);
-static void rpcrdma_reset_fmrs(struct rpcrdma_ia *);
-
 /*
  * internal functions
  */
@@ -945,21 +942,9 @@ retry:
 		rpcrdma_ep_disconnect(ep, ia);
 		rpcrdma_flush_cqs(ep);
 
-		switch (ia->ri_memreg_strategy) {
-		case RPCRDMA_FRMR:
-			rpcrdma_reset_frmrs(ia);
-			break;
-		case RPCRDMA_MTHCAFMR:
-			rpcrdma_reset_fmrs(ia);
-			break;
-		case RPCRDMA_ALLPHYSICAL:
-			break;
-		default:
-			rc = -EIO;
-			goto out;
-		}
-
 		xprt = container_of(ia, struct rpcrdma_xprt, rx_ia);
+		ia->ri_ops->ro_reset(xprt);
+
 		id = rpcrdma_create_id(xprt, ia,
 				(struct sockaddr *)&xprt->rx_data.addr);
 		if (IS_ERR(id)) {
@@ -1289,90 +1274,6 @@ rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf)
 	kfree(buf->rb_pool);
 }
 
-/* After a disconnect, unmap all FMRs.
- *
- * This is invoked only in the transport connect worker in order
- * to serialize with rpcrdma_register_fmr_external().
- */
-static void
-rpcrdma_reset_fmrs(struct rpcrdma_ia *ia)
-{
-	struct rpcrdma_xprt *r_xprt =
-				container_of(ia, struct rpcrdma_xprt, rx_ia);
-	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
-	struct list_head *pos;
-	struct rpcrdma_mw *r;
-	LIST_HEAD(l);
-	int rc;
-
-	list_for_each(pos, &buf->rb_all) {
-		r = list_entry(pos, struct rpcrdma_mw, mw_all);
-
-		INIT_LIST_HEAD(&l);
-		list_add(&r->r.fmr->list, &l);
-		rc = ib_unmap_fmr(&l);
-		if (rc)
-			dprintk("RPC:       %s: ib_unmap_fmr failed %i\n",
-				__func__, rc);
-	}
-}
-
-/* After a disconnect, a flushed FAST_REG_MR can leave an FRMR in
- * an unusable state. Find FRMRs in this state and dereg / reg
- * each.  FRMRs that are VALID and attached to an rpcrdma_req are
- * also torn down.
- *
- * This gives all in-use FRMRs a fresh rkey and leaves them INVALID.
- *
- * This is invoked only in the transport connect worker in order
- * to serialize with rpcrdma_register_frmr_external().
- */
-static void
-rpcrdma_reset_frmrs(struct rpcrdma_ia *ia)
-{
-	struct rpcrdma_xprt *r_xprt =
-				container_of(ia, struct rpcrdma_xprt, rx_ia);
-	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
-	struct list_head *pos;
-	struct rpcrdma_mw *r;
-	int rc;
-
-	list_for_each(pos, &buf->rb_all) {
-		r = list_entry(pos, struct rpcrdma_mw, mw_all);
-
-		if (r->r.frmr.fr_state == FRMR_IS_INVALID)
-			continue;
-
-		rc = ib_dereg_mr(r->r.frmr.fr_mr);
-		if (rc)
-			dprintk("RPC:       %s: ib_dereg_mr failed %i\n",
-				__func__, rc);
-		ib_free_fast_reg_page_list(r->r.frmr.fr_pgl);
-
-		r->r.frmr.fr_mr = ib_alloc_fast_reg_mr(ia->ri_pd,
-					ia->ri_max_frmr_depth);
-		if (IS_ERR(r->r.frmr.fr_mr)) {
-			rc = PTR_ERR(r->r.frmr.fr_mr);
-			dprintk("RPC:       %s: ib_alloc_fast_reg_mr"
-				" failed %i\n", __func__, rc);
-			continue;
-		}
-		r->r.frmr.fr_pgl = ib_alloc_fast_reg_page_list(
-					ia->ri_id->device,
-					ia->ri_max_frmr_depth);
-		if (IS_ERR(r->r.frmr.fr_pgl)) {
-			rc = PTR_ERR(r->r.frmr.fr_pgl);
-			dprintk("RPC:       %s: "
-				"ib_alloc_fast_reg_page_list "
-				"failed %i\n", __func__, rc);
-
-			ib_dereg_mr(r->r.frmr.fr_mr);
-			continue;
-		}
-		r->r.frmr.fr_state = FRMR_IS_INVALID;
-	}
-}
-
 /* "*mw" can be NULL when rpcrdma_buffer_get_mrs() fails, leaving
  * some req segments uninitialized.
  */
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 90b60feab45a..06802394cf89 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -342,6 +342,7 @@ struct rpcrdma_memreg_ops {
 				    struct rpcrdma_mr_seg *);
 	size_t		(*ro_maxpages)(struct rpcrdma_xprt *);
 	int		(*ro_init)(struct rpcrdma_xprt *);
+	void		(*ro_reset)(struct rpcrdma_xprt *);
 	const char	*ro_displayname;
 };
 
-- 
cgit v1.2.3


From 4561f347d49c645fd81d1f47b0fb460e8a6e4587 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:35:17 -0400
Subject: xprtrdma: Add "destroy MRs" memreg op

Memory Region objects associated with a transport instance are
destroyed before the instance is shutdown and destroyed.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      | 18 +++++++++++++
 net/sunrpc/xprtrdma/frwr_ops.c     | 14 ++++++++++
 net/sunrpc/xprtrdma/physical_ops.c |  6 +++++
 net/sunrpc/xprtrdma/verbs.c        | 52 +-------------------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    |  1 +
 5 files changed, 40 insertions(+), 51 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index 93261b05891e..e9ca5944ac1e 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -168,11 +168,29 @@ fmr_op_reset(struct rpcrdma_xprt *r_xprt)
 			__func__, rc);
 }
 
+static void
+fmr_op_destroy(struct rpcrdma_buffer *buf)
+{
+	struct rpcrdma_mw *r;
+	int rc;
+
+	while (!list_empty(&buf->rb_all)) {
+		r = list_entry(buf->rb_all.next, struct rpcrdma_mw, mw_all);
+		list_del(&r->mw_all);
+		rc = ib_dealloc_fmr(r->r.fmr);
+		if (rc)
+			dprintk("RPC:       %s: ib_dealloc_fmr failed %i\n",
+				__func__, rc);
+		kfree(r);
+	}
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
 	.ro_map				= fmr_op_map,
 	.ro_unmap			= fmr_op_unmap,
 	.ro_maxpages			= fmr_op_maxpages,
 	.ro_init			= fmr_op_init,
 	.ro_reset			= fmr_op_reset,
+	.ro_destroy			= fmr_op_destroy,
 	.ro_displayname			= "fmr",
 };
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index c2bb29d4df3c..121e400d0565 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -260,11 +260,25 @@ frwr_op_reset(struct rpcrdma_xprt *r_xprt)
 	}
 }
 
+static void
+frwr_op_destroy(struct rpcrdma_buffer *buf)
+{
+	struct rpcrdma_mw *r;
+
+	while (!list_empty(&buf->rb_all)) {
+		r = list_entry(buf->rb_all.next, struct rpcrdma_mw, mw_all);
+		list_del(&r->mw_all);
+		__frwr_release(r);
+		kfree(r);
+	}
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
 	.ro_map				= frwr_op_map,
 	.ro_unmap			= frwr_op_unmap,
 	.ro_maxpages			= frwr_op_maxpages,
 	.ro_init			= frwr_op_init,
 	.ro_reset			= frwr_op_reset,
+	.ro_destroy			= frwr_op_destroy,
 	.ro_displayname			= "frwr",
 };
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index e0607136655b..eb39011e3129 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -64,11 +64,17 @@ physical_op_reset(struct rpcrdma_xprt *r_xprt)
 {
 }
 
+static void
+physical_op_destroy(struct rpcrdma_buffer *buf)
+{
+}
+
 const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
 	.ro_map				= physical_op_map,
 	.ro_unmap			= physical_op_unmap,
 	.ro_maxpages			= physical_op_maxpages,
 	.ro_init			= physical_op_init,
 	.ro_reset			= physical_op_reset,
+	.ro_destroy			= physical_op_destroy,
 	.ro_displayname			= "physical",
 };
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 1b2c1f4ec4c3..a7fb31441069 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1199,47 +1199,6 @@ rpcrdma_destroy_req(struct rpcrdma_ia *ia, struct rpcrdma_req *req)
 	kfree(req);
 }
 
-static void
-rpcrdma_destroy_fmrs(struct rpcrdma_buffer *buf)
-{
-	struct rpcrdma_mw *r;
-	int rc;
-
-	while (!list_empty(&buf->rb_all)) {
-		r = list_entry(buf->rb_all.next, struct rpcrdma_mw, mw_all);
-		list_del(&r->mw_all);
-		list_del(&r->mw_list);
-
-		rc = ib_dealloc_fmr(r->r.fmr);
-		if (rc)
-			dprintk("RPC:       %s: ib_dealloc_fmr failed %i\n",
-				__func__, rc);
-
-		kfree(r);
-	}
-}
-
-static void
-rpcrdma_destroy_frmrs(struct rpcrdma_buffer *buf)
-{
-	struct rpcrdma_mw *r;
-	int rc;
-
-	while (!list_empty(&buf->rb_all)) {
-		r = list_entry(buf->rb_all.next, struct rpcrdma_mw, mw_all);
-		list_del(&r->mw_all);
-		list_del(&r->mw_list);
-
-		rc = ib_dereg_mr(r->r.frmr.fr_mr);
-		if (rc)
-			dprintk("RPC:       %s: ib_dereg_mr failed %i\n",
-				__func__, rc);
-		ib_free_fast_reg_page_list(r->r.frmr.fr_pgl);
-
-		kfree(r);
-	}
-}
-
 void
 rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf)
 {
@@ -1260,16 +1219,7 @@ rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf)
 			rpcrdma_destroy_req(ia, buf->rb_send_bufs[i]);
 	}
 
-	switch (ia->ri_memreg_strategy) {
-	case RPCRDMA_FRMR:
-		rpcrdma_destroy_frmrs(buf);
-		break;
-	case RPCRDMA_MTHCAFMR:
-		rpcrdma_destroy_fmrs(buf);
-		break;
-	default:
-		break;
-	}
+	ia->ri_ops->ro_destroy(buf);
 
 	kfree(buf->rb_pool);
 }
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 06802394cf89..b95e223d3d69 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -343,6 +343,7 @@ struct rpcrdma_memreg_ops {
 	size_t		(*ro_maxpages)(struct rpcrdma_xprt *);
 	int		(*ro_init)(struct rpcrdma_xprt *);
 	void		(*ro_reset)(struct rpcrdma_xprt *);
+	void		(*ro_destroy)(struct rpcrdma_buffer *);
 	const char	*ro_displayname;
 };
 
-- 
cgit v1.2.3


From 3968cb58501bf526eed1441f4ef237028aa9cd2d Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:35:26 -0400
Subject: xprtrdma: Add "open" memreg op

The open op determines the size of various transport data structures
based on device capabilities and memory registration mode.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      |  8 +++++++
 net/sunrpc/xprtrdma/frwr_ops.c     | 48 +++++++++++++++++++++++++++++++++++++
 net/sunrpc/xprtrdma/physical_ops.c |  8 +++++++
 net/sunrpc/xprtrdma/verbs.c        | 49 +++-----------------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    |  3 +++
 5 files changed, 70 insertions(+), 46 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index e9ca5944ac1e..e8a9837f8d63 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -20,6 +20,13 @@
 /* Maximum scatter/gather per FMR */
 #define RPCRDMA_MAX_FMR_SGES	(64)
 
+static int
+fmr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep,
+	    struct rpcrdma_create_data_internal *cdata)
+{
+	return 0;
+}
+
 /* FMR mode conveys up to 64 pages of payload per chunk segment.
  */
 static size_t
@@ -188,6 +195,7 @@ fmr_op_destroy(struct rpcrdma_buffer *buf)
 const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
 	.ro_map				= fmr_op_map,
 	.ro_unmap			= fmr_op_unmap,
+	.ro_open			= fmr_op_open,
 	.ro_maxpages			= fmr_op_maxpages,
 	.ro_init			= fmr_op_init,
 	.ro_reset			= fmr_op_reset,
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 121e400d0565..e17d54d473a7 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -58,6 +58,53 @@ __frwr_release(struct rpcrdma_mw *r)
 	ib_free_fast_reg_page_list(r->r.frmr.fr_pgl);
 }
 
+static int
+frwr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep,
+	     struct rpcrdma_create_data_internal *cdata)
+{
+	struct ib_device_attr *devattr = &ia->ri_devattr;
+	int depth, delta;
+
+	ia->ri_max_frmr_depth =
+			min_t(unsigned int, RPCRDMA_MAX_DATA_SEGS,
+			      devattr->max_fast_reg_page_list_len);
+	dprintk("RPC:       %s: device's max FR page list len = %u\n",
+		__func__, ia->ri_max_frmr_depth);
+
+	/* Add room for frmr register and invalidate WRs.
+	 * 1. FRMR reg WR for head
+	 * 2. FRMR invalidate WR for head
+	 * 3. N FRMR reg WRs for pagelist
+	 * 4. N FRMR invalidate WRs for pagelist
+	 * 5. FRMR reg WR for tail
+	 * 6. FRMR invalidate WR for tail
+	 * 7. The RDMA_SEND WR
+	 */
+	depth = 7;
+
+	/* Calculate N if the device max FRMR depth is smaller than
+	 * RPCRDMA_MAX_DATA_SEGS.
+	 */
+	if (ia->ri_max_frmr_depth < RPCRDMA_MAX_DATA_SEGS) {
+		delta = RPCRDMA_MAX_DATA_SEGS - ia->ri_max_frmr_depth;
+		do {
+			depth += 2; /* FRMR reg + invalidate */
+			delta -= ia->ri_max_frmr_depth;
+		} while (delta > 0);
+	}
+
+	ep->rep_attr.cap.max_send_wr *= depth;
+	if (ep->rep_attr.cap.max_send_wr > devattr->max_qp_wr) {
+		cdata->max_requests = devattr->max_qp_wr / depth;
+		if (!cdata->max_requests)
+			return -EINVAL;
+		ep->rep_attr.cap.max_send_wr = cdata->max_requests *
+					       depth;
+	}
+
+	return 0;
+}
+
 /* FRWR mode conveys a list of pages per chunk segment. The
  * maximum length of that list is the FRWR page list depth.
  */
@@ -276,6 +323,7 @@ frwr_op_destroy(struct rpcrdma_buffer *buf)
 const struct rpcrdma_memreg_ops rpcrdma_frwr_memreg_ops = {
 	.ro_map				= frwr_op_map,
 	.ro_unmap			= frwr_op_unmap,
+	.ro_open			= frwr_op_open,
 	.ro_maxpages			= frwr_op_maxpages,
 	.ro_init			= frwr_op_init,
 	.ro_reset			= frwr_op_reset,
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index eb39011e3129..0ba130bed1fc 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -19,6 +19,13 @@
 # define RPCDBG_FACILITY	RPCDBG_TRANS
 #endif
 
+static int
+physical_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep,
+		 struct rpcrdma_create_data_internal *cdata)
+{
+	return 0;
+}
+
 /* PHYSICAL memory registration conveys one page per chunk segment.
  */
 static size_t
@@ -72,6 +79,7 @@ physical_op_destroy(struct rpcrdma_buffer *buf)
 const struct rpcrdma_memreg_ops rpcrdma_physical_memreg_ops = {
 	.ro_map				= physical_op_map,
 	.ro_unmap			= physical_op_unmap,
+	.ro_open			= physical_op_open,
 	.ro_maxpages			= physical_op_maxpages,
 	.ro_init			= physical_op_init,
 	.ro_reset			= physical_op_reset,
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index a7fb31441069..b697b3ed6273 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -622,11 +622,6 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg)
 			dprintk("RPC:       %s: FRMR registration "
 				"not supported by HCA\n", __func__);
 			memreg = RPCRDMA_MTHCAFMR;
-		} else {
-			/* Mind the ia limit on FRMR page list depth */
-			ia->ri_max_frmr_depth = min_t(unsigned int,
-				RPCRDMA_MAX_DATA_SEGS,
-				devattr->max_fast_reg_page_list_len);
 		}
 	}
 	if (memreg == RPCRDMA_MTHCAFMR) {
@@ -741,49 +736,11 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia,
 
 	ep->rep_attr.event_handler = rpcrdma_qp_async_error_upcall;
 	ep->rep_attr.qp_context = ep;
-	/* send_cq and recv_cq initialized below */
 	ep->rep_attr.srq = NULL;
 	ep->rep_attr.cap.max_send_wr = cdata->max_requests;
-	switch (ia->ri_memreg_strategy) {
-	case RPCRDMA_FRMR: {
-		int depth = 7;
-
-		/* Add room for frmr register and invalidate WRs.
-		 * 1. FRMR reg WR for head
-		 * 2. FRMR invalidate WR for head
-		 * 3. N FRMR reg WRs for pagelist
-		 * 4. N FRMR invalidate WRs for pagelist
-		 * 5. FRMR reg WR for tail
-		 * 6. FRMR invalidate WR for tail
-		 * 7. The RDMA_SEND WR
-		 */
-
-		/* Calculate N if the device max FRMR depth is smaller than
-		 * RPCRDMA_MAX_DATA_SEGS.
-		 */
-		if (ia->ri_max_frmr_depth < RPCRDMA_MAX_DATA_SEGS) {
-			int delta = RPCRDMA_MAX_DATA_SEGS -
-				    ia->ri_max_frmr_depth;
-
-			do {
-				depth += 2; /* FRMR reg + invalidate */
-				delta -= ia->ri_max_frmr_depth;
-			} while (delta > 0);
-
-		}
-		ep->rep_attr.cap.max_send_wr *= depth;
-		if (ep->rep_attr.cap.max_send_wr > devattr->max_qp_wr) {
-			cdata->max_requests = devattr->max_qp_wr / depth;
-			if (!cdata->max_requests)
-				return -EINVAL;
-			ep->rep_attr.cap.max_send_wr = cdata->max_requests *
-						       depth;
-		}
-		break;
-	}
-	default:
-		break;
-	}
+	rc = ia->ri_ops->ro_open(ia, ep, cdata);
+	if (rc)
+		return rc;
 	ep->rep_attr.cap.max_recv_wr = cdata->max_requests;
 	ep->rep_attr.cap.max_send_sge = (cdata->padding ? 4 : 2);
 	ep->rep_attr.cap.max_recv_sge = 1;
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index b95e223d3d69..9036fb4174d5 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -340,6 +340,9 @@ struct rpcrdma_memreg_ops {
 				  struct rpcrdma_mr_seg *, int, bool);
 	int		(*ro_unmap)(struct rpcrdma_xprt *,
 				    struct rpcrdma_mr_seg *);
+	int		(*ro_open)(struct rpcrdma_ia *,
+				   struct rpcrdma_ep *,
+				   struct rpcrdma_create_data_internal *);
 	size_t		(*ro_maxpages)(struct rpcrdma_xprt *);
 	int		(*ro_init)(struct rpcrdma_xprt *);
 	void		(*ro_reset)(struct rpcrdma_xprt *);
-- 
cgit v1.2.3


From e46ac34c3c34e408435656a5fed605c4c787d081 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:35:35 -0400
Subject: xprtrdma: Handle non-SEND completions via a callout

Allow each memory registration mode to plug in a callout that handles
the completion of a memory registration operation.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/frwr_ops.c  | 17 +++++++++++++++++
 net/sunrpc/xprtrdma/verbs.c     | 16 ++++++----------
 net/sunrpc/xprtrdma/xprt_rdma.h |  5 +++++
 3 files changed, 28 insertions(+), 10 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index e17d54d473a7..ea59c1b435ff 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -117,6 +117,22 @@ frwr_op_maxpages(struct rpcrdma_xprt *r_xprt)
 		     rpcrdma_max_segments(r_xprt) * ia->ri_max_frmr_depth);
 }
 
+/* If FAST_REG or LOCAL_INV failed, indicate the frmr needs to be reset. */
+static void
+frwr_sendcompletion(struct ib_wc *wc)
+{
+	struct rpcrdma_mw *r;
+
+	if (likely(wc->status == IB_WC_SUCCESS))
+		return;
+
+	/* WARNING: Only wr_id and status are reliable at this point */
+	r = (struct rpcrdma_mw *)(unsigned long)wc->wr_id;
+	dprintk("RPC:       %s: frmr %p (stale), status %d\n",
+		__func__, r, wc->status);
+	r->r.frmr.fr_state = FRMR_IS_STALE;
+}
+
 static int
 frwr_op_init(struct rpcrdma_xprt *r_xprt)
 {
@@ -148,6 +164,7 @@ frwr_op_init(struct rpcrdma_xprt *r_xprt)
 
 		list_add(&r->mw_list, &buf->rb_mws);
 		list_add(&r->mw_all, &buf->rb_all);
+		r->mw_sendcompletion = frwr_sendcompletion;
 	}
 
 	return 0;
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index b697b3ed6273..cac06f290a26 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -186,7 +186,7 @@ static const char * const wc_status[] = {
 	"remote access error",
 	"remote operation error",
 	"transport retry counter exceeded",
-	"RNR retrycounter exceeded",
+	"RNR retry counter exceeded",
 	"local RDD violation error",
 	"remove invalid RD request",
 	"operation aborted",
@@ -204,21 +204,17 @@ static const char * const wc_status[] = {
 static void
 rpcrdma_sendcq_process_wc(struct ib_wc *wc)
 {
-	if (likely(wc->status == IB_WC_SUCCESS))
-		return;
-
 	/* WARNING: Only wr_id and status are reliable at this point */
-	if (wc->wr_id == 0ULL) {
-		if (wc->status != IB_WC_WR_FLUSH_ERR)
+	if (wc->wr_id == RPCRDMA_IGNORE_COMPLETION) {
+		if (wc->status != IB_WC_SUCCESS &&
+		    wc->status != IB_WC_WR_FLUSH_ERR)
 			pr_err("RPC:       %s: SEND: %s\n",
 			       __func__, COMPLETION_MSG(wc->status));
 	} else {
 		struct rpcrdma_mw *r;
 
 		r = (struct rpcrdma_mw *)(unsigned long)wc->wr_id;
-		r->r.frmr.fr_state = FRMR_IS_STALE;
-		pr_err("RPC:       %s: frmr %p (stale): %s\n",
-		       __func__, r, COMPLETION_MSG(wc->status));
+		r->mw_sendcompletion(wc);
 	}
 }
 
@@ -1622,7 +1618,7 @@ rpcrdma_ep_post(struct rpcrdma_ia *ia,
 	}
 
 	send_wr.next = NULL;
-	send_wr.wr_id = 0ULL;	/* no send cookie */
+	send_wr.wr_id = RPCRDMA_IGNORE_COMPLETION;
 	send_wr.sg_list = req->rl_send_iov;
 	send_wr.num_sge = req->rl_niovs;
 	send_wr.opcode = IB_WR_SEND;
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 9036fb4174d5..54bcbe47075d 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -106,6 +106,10 @@ struct rpcrdma_ep {
 #define INIT_CQCOUNT(ep) atomic_set(&(ep)->rep_cqcount, (ep)->rep_cqinit)
 #define DECR_CQCOUNT(ep) atomic_sub_return(1, &(ep)->rep_cqcount)
 
+/* Force completion handler to ignore the signal
+ */
+#define RPCRDMA_IGNORE_COMPLETION	(0ULL)
+
 /* Registered buffer -- registered kmalloc'd memory for RDMA SEND/RECV
  *
  * The below structure appears at the front of a large region of kmalloc'd
@@ -206,6 +210,7 @@ struct rpcrdma_mw {
 		struct ib_fmr		*fmr;
 		struct rpcrdma_frmr	frmr;
 	} r;
+	void			(*mw_sendcompletion)(struct ib_wc *);
 	struct list_head	mw_list;
 	struct list_head	mw_all;
 };
-- 
cgit v1.2.3


From d654788e98f74f2df8dfc6079fa314938f739486 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Mar 2015 14:35:44 -0400
Subject: xprtrdma: Make rpcrdma_{un}map_one() into inline functions

These functions are called in a loop for each page transferred via
RDMA READ or WRITE. Extract loop invariants and inline them to
reduce CPU overhead.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Devesh Sharma <Devesh.Sharma@Emulex.Com>
Tested-by: Meghana Cheripady <Meghana.Cheripady@Emulex.Com>
Tested-by: Veeresh U. Kokatnur <veereshuk@chelsio.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 net/sunrpc/xprtrdma/fmr_ops.c      | 10 ++++++---
 net/sunrpc/xprtrdma/frwr_ops.c     | 10 ++++++---
 net/sunrpc/xprtrdma/physical_ops.c | 10 +++++++--
 net/sunrpc/xprtrdma/verbs.c        | 44 +++++++------------------------------
 net/sunrpc/xprtrdma/xprt_rdma.h    | 45 ++++++++++++++++++++++++++++++++++++--
 5 files changed, 73 insertions(+), 46 deletions(-)

(limited to 'net')

diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c
index e8a9837f8d63..a91ba2c8ef1e 100644
--- a/net/sunrpc/xprtrdma/fmr_ops.c
+++ b/net/sunrpc/xprtrdma/fmr_ops.c
@@ -85,6 +85,8 @@ fmr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 	   int nsegs, bool writing)
 {
 	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+	struct ib_device *device = ia->ri_id->device;
+	enum dma_data_direction direction = rpcrdma_data_dir(writing);
 	struct rpcrdma_mr_seg *seg1 = seg;
 	struct rpcrdma_mw *mw = seg1->rl_mw;
 	u64 physaddrs[RPCRDMA_MAX_DATA_SEGS];
@@ -97,7 +99,7 @@ fmr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 	if (nsegs > RPCRDMA_MAX_FMR_SGES)
 		nsegs = RPCRDMA_MAX_FMR_SGES;
 	for (i = 0; i < nsegs;) {
-		rpcrdma_map_one(ia, seg, writing);
+		rpcrdma_map_one(device, seg, direction);
 		physaddrs[i] = seg->mr_dma;
 		len += seg->mr_len;
 		++seg;
@@ -123,7 +125,7 @@ out_maperr:
 		__func__, len, (unsigned long long)seg1->mr_dma,
 		pageoff, i, rc);
 	while (i--)
-		rpcrdma_unmap_one(ia, --seg);
+		rpcrdma_unmap_one(device, --seg);
 	return rc;
 }
 
@@ -135,14 +137,16 @@ fmr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
 {
 	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
 	struct rpcrdma_mr_seg *seg1 = seg;
+	struct ib_device *device;
 	int rc, nsegs = seg->mr_nsegs;
 	LIST_HEAD(l);
 
 	list_add(&seg1->rl_mw->r.fmr->list, &l);
 	rc = ib_unmap_fmr(&l);
 	read_lock(&ia->ri_qplock);
+	device = ia->ri_id->device;
 	while (seg1->mr_nsegs--)
-		rpcrdma_unmap_one(ia, seg++);
+		rpcrdma_unmap_one(device, seg++);
 	read_unlock(&ia->ri_qplock);
 	if (rc)
 		goto out_err;
diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index ea59c1b435ff..0a7b9df70133 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -178,6 +178,8 @@ frwr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 	    int nsegs, bool writing)
 {
 	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+	struct ib_device *device = ia->ri_id->device;
+	enum dma_data_direction direction = rpcrdma_data_dir(writing);
 	struct rpcrdma_mr_seg *seg1 = seg;
 	struct rpcrdma_mw *mw = seg1->rl_mw;
 	struct rpcrdma_frmr *frmr = &mw->r.frmr;
@@ -197,7 +199,7 @@ frwr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 	if (nsegs > ia->ri_max_frmr_depth)
 		nsegs = ia->ri_max_frmr_depth;
 	for (page_no = i = 0; i < nsegs;) {
-		rpcrdma_map_one(ia, seg, writing);
+		rpcrdma_map_one(device, seg, direction);
 		pa = seg->mr_dma;
 		for (seg_len = seg->mr_len; seg_len > 0; seg_len -= PAGE_SIZE) {
 			frmr->fr_pgl->page_list[page_no++] = pa;
@@ -247,7 +249,7 @@ out_senderr:
 	ib_update_fast_reg_key(mr, --key);
 	frmr->fr_state = FRMR_IS_INVALID;
 	while (i--)
-		rpcrdma_unmap_one(ia, --seg);
+		rpcrdma_unmap_one(device, --seg);
 	return rc;
 }
 
@@ -261,6 +263,7 @@ frwr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
 	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
 	struct ib_send_wr invalidate_wr, *bad_wr;
 	int rc, nsegs = seg->mr_nsegs;
+	struct ib_device *device;
 
 	seg1->rl_mw->r.frmr.fr_state = FRMR_IS_INVALID;
 
@@ -271,8 +274,9 @@ frwr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
 	DECR_CQCOUNT(&r_xprt->rx_ep);
 
 	read_lock(&ia->ri_qplock);
+	device = ia->ri_id->device;
 	while (seg1->mr_nsegs--)
-		rpcrdma_unmap_one(ia, seg++);
+		rpcrdma_unmap_one(device, seg++);
 	rc = ib_post_send(ia->ri_id->qp, &invalidate_wr, &bad_wr);
 	read_unlock(&ia->ri_qplock);
 	if (rc)
diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c
index 0ba130bed1fc..ba518af16787 100644
--- a/net/sunrpc/xprtrdma/physical_ops.c
+++ b/net/sunrpc/xprtrdma/physical_ops.c
@@ -50,7 +50,8 @@ physical_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 {
 	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
 
-	rpcrdma_map_one(ia, seg, writing);
+	rpcrdma_map_one(ia->ri_id->device, seg,
+			rpcrdma_data_dir(writing));
 	seg->mr_rkey = ia->ri_bind_mem->rkey;
 	seg->mr_base = seg->mr_dma;
 	seg->mr_nsegs = 1;
@@ -62,7 +63,12 @@ physical_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 static int
 physical_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
 {
-	rpcrdma_unmap_one(&r_xprt->rx_ia, seg);
+	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
+
+	read_lock(&ia->ri_qplock);
+	rpcrdma_unmap_one(ia->ri_id->device, seg);
+	read_unlock(&ia->ri_qplock);
+
 	return 1;
 }
 
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index cac06f290a26..4870d272e006 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -1436,6 +1436,14 @@ rpcrdma_recv_buffer_put(struct rpcrdma_rep *rep)
  * Wrappers for internal-use kmalloc memory registration, used by buffer code.
  */
 
+void
+rpcrdma_mapping_error(struct rpcrdma_mr_seg *seg)
+{
+	dprintk("RPC:       map_one: offset %p iova %llx len %zu\n",
+		seg->mr_offset,
+		(unsigned long long)seg->mr_dma, seg->mr_dmalen);
+}
+
 static int
 rpcrdma_register_internal(struct rpcrdma_ia *ia, void *va, int len,
 				struct ib_mr **mrp, struct ib_sge *iov)
@@ -1560,42 +1568,6 @@ rpcrdma_free_regbuf(struct rpcrdma_ia *ia, struct rpcrdma_regbuf *rb)
 	}
 }
 
-/*
- * Wrappers for chunk registration, shared by read/write chunk code.
- */
-
-void
-rpcrdma_map_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg, bool writing)
-{
-	seg->mr_dir = writing ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
-	seg->mr_dmalen = seg->mr_len;
-	if (seg->mr_page)
-		seg->mr_dma = ib_dma_map_page(ia->ri_id->device,
-				seg->mr_page, offset_in_page(seg->mr_offset),
-				seg->mr_dmalen, seg->mr_dir);
-	else
-		seg->mr_dma = ib_dma_map_single(ia->ri_id->device,
-				seg->mr_offset,
-				seg->mr_dmalen, seg->mr_dir);
-	if (ib_dma_mapping_error(ia->ri_id->device, seg->mr_dma)) {
-		dprintk("RPC:       %s: mr_dma %llx mr_offset %p mr_dma_len %zu\n",
-			__func__,
-			(unsigned long long)seg->mr_dma,
-			seg->mr_offset, seg->mr_dmalen);
-	}
-}
-
-void
-rpcrdma_unmap_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg)
-{
-	if (seg->mr_page)
-		ib_dma_unmap_page(ia->ri_id->device,
-				seg->mr_dma, seg->mr_dmalen, seg->mr_dir);
-	else
-		ib_dma_unmap_single(ia->ri_id->device,
-				seg->mr_dma, seg->mr_dmalen, seg->mr_dir);
-}
-
 /*
  * Prepost any receive buffer, then post send.
  *
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index 54bcbe47075d..78e0b8beaa36 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -424,8 +424,49 @@ void rpcrdma_free_regbuf(struct rpcrdma_ia *,
 			 struct rpcrdma_regbuf *);
 
 unsigned int rpcrdma_max_segments(struct rpcrdma_xprt *);
-void rpcrdma_map_one(struct rpcrdma_ia *, struct rpcrdma_mr_seg *, bool);
-void rpcrdma_unmap_one(struct rpcrdma_ia *, struct rpcrdma_mr_seg *);
+
+/*
+ * Wrappers for chunk registration, shared by read/write chunk code.
+ */
+
+void rpcrdma_mapping_error(struct rpcrdma_mr_seg *);
+
+static inline enum dma_data_direction
+rpcrdma_data_dir(bool writing)
+{
+	return writing ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
+}
+
+static inline void
+rpcrdma_map_one(struct ib_device *device, struct rpcrdma_mr_seg *seg,
+		enum dma_data_direction direction)
+{
+	seg->mr_dir = direction;
+	seg->mr_dmalen = seg->mr_len;
+
+	if (seg->mr_page)
+		seg->mr_dma = ib_dma_map_page(device,
+				seg->mr_page, offset_in_page(seg->mr_offset),
+				seg->mr_dmalen, seg->mr_dir);
+	else
+		seg->mr_dma = ib_dma_map_single(device,
+				seg->mr_offset,
+				seg->mr_dmalen, seg->mr_dir);
+
+	if (ib_dma_mapping_error(device, seg->mr_dma))
+		rpcrdma_mapping_error(seg);
+}
+
+static inline void
+rpcrdma_unmap_one(struct ib_device *device, struct rpcrdma_mr_seg *seg)
+{
+	if (seg->mr_page)
+		ib_dma_unmap_page(device,
+				  seg->mr_dma, seg->mr_dmalen, seg->mr_dir);
+	else
+		ib_dma_unmap_single(device,
+				    seg->mr_dma, seg->mr_dmalen, seg->mr_dir);
+}
 
 /*
  * RPC/RDMA connection management calls - xprtrdma/rpc_rdma.c
-- 
cgit v1.2.3


From ee8ac4d61c2cf43bdd427e70db97ac330e61570d Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 6 Mar 2015 14:05:26 +0000
Subject: VFS: AF_UNIX sockets should call mknod on the top layer only

AF_UNIX sockets should call mknod on the top layer only and should not attempt
to modify the lower layer in a layered filesystem such as overlayfs.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 net/unix/af_unix.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'net')

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 433f287ee548..8b8f5e6d20bc 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -839,7 +839,7 @@ static int unix_mknod(const char *sun_path, umode_t mode, struct path *res)
 	 */
 	err = security_path_mknod(&path, dentry, mode, 0);
 	if (!err) {
-		err = vfs_mknod(path.dentry->d_inode, dentry, mode, 0);
+		err = vfs_mknod(d_inode(path.dentry), dentry, mode, 0);
 		if (!err) {
 			res->mnt = mntget(path.mnt);
 			res->dentry = dget(dentry);
-- 
cgit v1.2.3


From a25b376bded1ba7fd1d455e140d723b7de2e343c Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 17 Mar 2015 22:26:21 +0000
Subject: VFS: net/unix: d_backing_inode() annotations

places where we are dealing with S_ISSOCK file creation/lookups.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 net/unix/af_unix.c | 6 +++---
 net/unix/diag.c    | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'net')

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 8b8f5e6d20bc..5266ea7b922b 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -305,7 +305,7 @@ static struct sock *unix_find_socket_byinode(struct inode *i)
 		    &unix_socket_table[i->i_ino & (UNIX_HASH_SIZE - 1)]) {
 		struct dentry *dentry = unix_sk(s)->path.dentry;
 
-		if (dentry && dentry->d_inode == i) {
+		if (dentry && d_backing_inode(dentry) == i) {
 			sock_hold(s);
 			goto found;
 		}
@@ -778,7 +778,7 @@ static struct sock *unix_find_other(struct net *net,
 		err = kern_path(sunname->sun_path, LOOKUP_FOLLOW, &path);
 		if (err)
 			goto fail;
-		inode = path.dentry->d_inode;
+		inode = d_backing_inode(path.dentry);
 		err = inode_permission(inode, MAY_WRITE);
 		if (err)
 			goto put_fail;
@@ -905,7 +905,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 			goto out_up;
 		}
 		addr->hash = UNIX_HASH_SIZE;
-		hash = path.dentry->d_inode->i_ino & (UNIX_HASH_SIZE-1);
+		hash = d_backing_inode(path.dentry)->i_ino & (UNIX_HASH_SIZE-1);
 		spin_lock(&unix_table_lock);
 		u->path = path;
 		list = &unix_socket_table[hash];
diff --git a/net/unix/diag.c b/net/unix/diag.c
index ef542fbca9fe..c512f64d5287 100644
--- a/net/unix/diag.c
+++ b/net/unix/diag.c
@@ -25,7 +25,7 @@ static int sk_diag_dump_vfs(struct sock *sk, struct sk_buff *nlskb)
 
 	if (dentry) {
 		struct unix_diag_vfs uv = {
-			.udiag_vfs_ino = dentry->d_inode->i_ino,
+			.udiag_vfs_ino = d_backing_inode(dentry)->i_ino,
 			.udiag_vfs_dev = dentry->d_sb->s_dev,
 		};
 
-- 
cgit v1.2.3


From c5ef60352893b139147b7c033354e8e028e7f52a Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 17 Mar 2015 22:26:16 +0000
Subject: VFS: net/: d_inode() annotations

socket inodes and sunrpc filesystems - inodes owned by that code

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 net/socket.c          |  6 +++---
 net/sunrpc/rpc_pipe.c | 32 ++++++++++++++++----------------
 2 files changed, 19 insertions(+), 19 deletions(-)

(limited to 'net')

diff --git a/net/socket.c b/net/socket.c
index 3e33959f3ce5..884e32997698 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -312,7 +312,7 @@ static const struct super_operations sockfs_ops = {
 static char *sockfs_dname(struct dentry *dentry, char *buffer, int buflen)
 {
 	return dynamic_dname(dentry, buffer, buflen, "socket:[%lu]",
-				dentry->d_inode->i_ino);
+				d_inode(dentry)->i_ino);
 }
 
 static const struct dentry_operations sockfs_dentry_operations = {
@@ -375,7 +375,7 @@ struct file *sock_alloc_file(struct socket *sock, int flags, const char *dname)
 		  &socket_file_ops);
 	if (unlikely(IS_ERR(file))) {
 		/* drop dentry, keep inode */
-		ihold(path.dentry->d_inode);
+		ihold(d_inode(path.dentry));
 		path_put(&path);
 		return file;
 	}
@@ -497,7 +497,7 @@ static ssize_t sockfs_listxattr(struct dentry *dentry, char *buffer,
 	ssize_t len;
 	ssize_t used = 0;
 
-	len = security_inode_listsecurity(dentry->d_inode, buffer, size);
+	len = security_inode_listsecurity(d_inode(dentry), buffer, size);
 	if (len < 0)
 		return len;
 	used += len;
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index 2d12b76b5a64..d81186d34558 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -94,7 +94,7 @@ rpc_timeout_upcall_queue(struct work_struct *work)
 	}
 	dentry = dget(pipe->dentry);
 	spin_unlock(&pipe->lock);
-	rpc_purge_list(dentry ? &RPC_I(dentry->d_inode)->waitq : NULL,
+	rpc_purge_list(dentry ? &RPC_I(d_inode(dentry))->waitq : NULL,
 			&free_list, destroy_msg, -ETIMEDOUT);
 	dput(dentry);
 }
@@ -152,7 +152,7 @@ rpc_queue_upcall(struct rpc_pipe *pipe, struct rpc_pipe_msg *msg)
 	dentry = dget(pipe->dentry);
 	spin_unlock(&pipe->lock);
 	if (dentry) {
-		wake_up(&RPC_I(dentry->d_inode)->waitq);
+		wake_up(&RPC_I(d_inode(dentry))->waitq);
 		dput(dentry);
 	}
 	return res;
@@ -591,7 +591,7 @@ static int __rpc_mkpipe_dentry(struct inode *dir, struct dentry *dentry,
 	err = __rpc_create_common(dir, dentry, S_IFIFO | mode, i_fop, private);
 	if (err)
 		return err;
-	rpci = RPC_I(dentry->d_inode);
+	rpci = RPC_I(d_inode(dentry));
 	rpci->private = private;
 	rpci->pipe = pipe;
 	fsnotify_create(dir, dentry);
@@ -616,7 +616,7 @@ int rpc_rmdir(struct dentry *dentry)
 	int error;
 
 	parent = dget_parent(dentry);
-	dir = parent->d_inode;
+	dir = d_inode(parent);
 	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
 	error = __rpc_rmdir(dir, dentry);
 	mutex_unlock(&dir->i_mutex);
@@ -638,7 +638,7 @@ static int __rpc_unlink(struct inode *dir, struct dentry *dentry)
 
 static int __rpc_rmpipe(struct inode *dir, struct dentry *dentry)
 {
-	struct inode *inode = dentry->d_inode;
+	struct inode *inode = d_inode(dentry);
 
 	rpc_close_pipes(inode);
 	return __rpc_unlink(dir, dentry);
@@ -654,7 +654,7 @@ static struct dentry *__rpc_lookup_create_exclusive(struct dentry *parent,
 		if (!dentry)
 			return ERR_PTR(-ENOMEM);
 	}
-	if (dentry->d_inode == NULL)
+	if (d_really_is_negative(dentry))
 		return dentry;
 	dput(dentry);
 	return ERR_PTR(-EEXIST);
@@ -667,7 +667,7 @@ static void __rpc_depopulate(struct dentry *parent,
 			     const struct rpc_filelist *files,
 			     int start, int eof)
 {
-	struct inode *dir = parent->d_inode;
+	struct inode *dir = d_inode(parent);
 	struct dentry *dentry;
 	struct qstr name;
 	int i;
@@ -679,9 +679,9 @@ static void __rpc_depopulate(struct dentry *parent,
 
 		if (dentry == NULL)
 			continue;
-		if (dentry->d_inode == NULL)
+		if (d_really_is_negative(dentry))
 			goto next;
-		switch (dentry->d_inode->i_mode & S_IFMT) {
+		switch (d_inode(dentry)->i_mode & S_IFMT) {
 			default:
 				BUG();
 			case S_IFREG:
@@ -699,7 +699,7 @@ static void rpc_depopulate(struct dentry *parent,
 			   const struct rpc_filelist *files,
 			   int start, int eof)
 {
-	struct inode *dir = parent->d_inode;
+	struct inode *dir = d_inode(parent);
 
 	mutex_lock_nested(&dir->i_mutex, I_MUTEX_CHILD);
 	__rpc_depopulate(parent, files, start, eof);
@@ -711,7 +711,7 @@ static int rpc_populate(struct dentry *parent,
 			int start, int eof,
 			void *private)
 {
-	struct inode *dir = parent->d_inode;
+	struct inode *dir = d_inode(parent);
 	struct dentry *dentry;
 	int i, err;
 
@@ -754,7 +754,7 @@ static struct dentry *rpc_mkdir_populate(struct dentry *parent,
 		int (*populate)(struct dentry *, void *), void *args_populate)
 {
 	struct dentry *dentry;
-	struct inode *dir = parent->d_inode;
+	struct inode *dir = d_inode(parent);
 	int error;
 
 	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
@@ -787,7 +787,7 @@ static int rpc_rmdir_depopulate(struct dentry *dentry,
 	int error;
 
 	parent = dget_parent(dentry);
-	dir = parent->d_inode;
+	dir = d_inode(parent);
 	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
 	if (depopulate != NULL)
 		depopulate(dentry);
@@ -819,7 +819,7 @@ struct dentry *rpc_mkpipe_dentry(struct dentry *parent, const char *name,
 				 void *private, struct rpc_pipe *pipe)
 {
 	struct dentry *dentry;
-	struct inode *dir = parent->d_inode;
+	struct inode *dir = d_inode(parent);
 	umode_t umode = S_IFIFO | S_IRUSR | S_IWUSR;
 	int err;
 
@@ -864,7 +864,7 @@ rpc_unlink(struct dentry *dentry)
 	int error = 0;
 
 	parent = dget_parent(dentry);
-	dir = parent->d_inode;
+	dir = d_inode(parent);
 	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
 	error = __rpc_rmpipe(dir, dentry);
 	mutex_unlock(&dir->i_mutex);
@@ -1375,7 +1375,7 @@ rpc_gssd_dummy_depopulate(struct dentry *pipe_dentry)
 	struct dentry *clnt_dir = pipe_dentry->d_parent;
 	struct dentry *gssd_dir = clnt_dir->d_parent;
 
-	__rpc_rmpipe(clnt_dir->d_inode, pipe_dentry);
+	__rpc_rmpipe(d_inode(clnt_dir), pipe_dentry);
 	__rpc_depopulate(clnt_dir, gssd_dummy_info_file, 0, 1);
 	__rpc_depopulate(gssd_dir, gssd_dummy_clnt_dir, 0, 1);
 	dput(pipe_dentry);
-- 
cgit v1.2.3


From 540207ae69777b85d167df28f469e77f0fcbb8f9 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Wed, 15 Apr 2015 11:48:49 -0700
Subject: fou: avoid missing unlock in failure path

Fixes: 7a6c8c34e5b7 ("fou: implement FOU_CMD_GET")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fou.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'net')

diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c
index af150b43b214..34968cd5c146 100644
--- a/net/ipv4/fou.c
+++ b/net/ipv4/fou.c
@@ -711,11 +711,10 @@ static int fou_nl_dump(struct sk_buff *skb, struct netlink_callback *cb)
 				    cb->nlh->nlmsg_seq, NLM_F_MULTI,
 				    skb, FOU_CMD_GET);
 		if (ret)
-			goto done;
+			break;
 	}
 	mutex_unlock(&fn->fou_lock);
 
-done:
 	cb->args[0] = idx;
 	return skb->len;
 }
-- 
cgit v1.2.3


From 5a950ad58d412d76d33f4f4399d69308d511c1a4 Mon Sep 17 00:00:00 2001
From: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Date: Thu, 16 Apr 2015 21:17:35 +0800
Subject: netns: remove duplicated include from net_namespace.c

Remove duplicated include.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/net_namespace.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'net')

diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index a3abb719221f..78fc04ad36fc 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -16,7 +16,6 @@
 #include <linux/export.h>
 #include <linux/user_namespace.h>
 #include <linux/net_namespace.h>
-#include <linux/rtnetlink.h>
 #include <net/sock.h>
 #include <net/netlink.h>
 #include <net/net_namespace.h>
-- 
cgit v1.2.3


From a166151cbe33b53221c24259e4a7201064b3ba79 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@plumgrid.com>
Date: Wed, 15 Apr 2015 12:55:45 -0700
Subject: bpf: fix bpf helpers to use skb->mac_header relative offsets

For the short-term solution, lets fix bpf helper functions to use
skb->mac_header relative offsets instead of skb->data in order to
get the same eBPF programs with cls_bpf and act_bpf work on ingress
and egress qdisc path. We need to ensure that mac_header is set
before calling into programs. This is effectively the first option
from below referenced discussion.

More long term solution for LD_ABS|LD_IND instructions will be more
intrusive but also more beneficial than this, and implemented later
as it's too risky at this point in time.

I.e., we plan to look into the option of moving skb_pull() out of
eth_type_trans() and into netif_receive_skb() as has been suggested
as second option. Meanwhile, this solution ensures ingress can be
used with eBPF, too, and that we won't run into ABI troubles later.
For dealing with negative offsets inside eBPF helper functions,
we've implemented bpf_skb_clone_unwritable() to test for unwriteable
headers.

Reference: http://thread.gmane.org/gmane.linux.network/359129/focus=359694
Fixes: 608cd71a9c7c ("tc: bpf: generalize pedit action")
Fixes: 91bc4822c3d6 ("tc: bpf: add checksum helpers")
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/bpf.h    |  2 +-
 include/uapi/linux/filter.h |  7 +++++--
 net/core/filter.c           | 41 ++++++++++++++++++++++++++++++++---------
 net/sched/act_bpf.c         |  3 +++
 net/sched/cls_bpf.c         |  3 +++
 samples/bpf/tcbpf1_kern.c   | 16 ++++++----------
 6 files changed, 50 insertions(+), 22 deletions(-)

(limited to 'net')

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 5c1cee11f777..a9ebdf5701e8 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -177,7 +177,7 @@ enum bpf_func_id {
 	/**
 	 * skb_store_bytes(skb, offset, from, len, flags) - store bytes into packet
 	 * @skb: pointer to skb
-	 * @offset: offset within packet from skb->data
+	 * @offset: offset within packet from skb->mac_header
 	 * @from: pointer where to copy bytes from
 	 * @len: number of bytes to store into packet
 	 * @flags: bit 0 - if true, recompute skb->csum
diff --git a/include/uapi/linux/filter.h b/include/uapi/linux/filter.h
index 34c7936ca114..c97340e43dd6 100644
--- a/include/uapi/linux/filter.h
+++ b/include/uapi/linux/filter.h
@@ -79,8 +79,11 @@ struct sock_fprog {	/* Required for SO_ATTACH_FILTER. */
 #define SKF_AD_RANDOM	56
 #define SKF_AD_VLAN_TPID	60
 #define SKF_AD_MAX	64
-#define SKF_NET_OFF   (-0x100000)
-#define SKF_LL_OFF    (-0x200000)
 
+#define SKF_NET_OFF	(-0x100000)
+#define SKF_LL_OFF	(-0x200000)
+
+#define BPF_NET_OFF	SKF_NET_OFF
+#define BPF_LL_OFF	SKF_LL_OFF
 
 #endif /* _UAPI__LINUX_FILTER_H__ */
diff --git a/net/core/filter.c b/net/core/filter.c
index b669e75d2b36..bf831a85c315 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1175,12 +1175,27 @@ int sk_attach_bpf(u32 ufd, struct sock *sk)
 	return 0;
 }
 
+/**
+ *	bpf_skb_clone_not_writable - is the header of a clone not writable
+ *	@skb: buffer to check
+ *	@len: length up to which to write, can be negative
+ *
+ *	Returns true if modifying the header part of the cloned buffer
+ *	does require the data to be copied. I.e. this version works with
+ *	negative lengths needed for eBPF case!
+ */
+static bool bpf_skb_clone_unwritable(const struct sk_buff *skb, int len)
+{
+	return skb_header_cloned(skb) ||
+	       (int) skb_headroom(skb) + len > skb->hdr_len;
+}
+
 #define BPF_RECOMPUTE_CSUM(flags)	((flags) & 1)
 
 static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 {
 	struct sk_buff *skb = (struct sk_buff *) (long) r1;
-	unsigned int offset = (unsigned int) r2;
+	int offset = (int) r2;
 	void *from = (void *) (long) r3;
 	unsigned int len = (unsigned int) r4;
 	char buf[16];
@@ -1194,10 +1209,12 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 	 *
 	 * so check for invalid 'offset' and too large 'len'
 	 */
-	if (unlikely(offset > 0xffff || len > sizeof(buf)))
+	if (unlikely((u32) offset > 0xffff || len > sizeof(buf)))
 		return -EFAULT;
 
-	if (skb_cloned(skb) && !skb_clone_writable(skb, offset + len))
+	offset -= skb->data - skb_mac_header(skb);
+	if (unlikely(skb_cloned(skb) &&
+		     bpf_skb_clone_unwritable(skb, offset + len)))
 		return -EFAULT;
 
 	ptr = skb_header_pointer(skb, offset, len, buf);
@@ -1232,15 +1249,18 @@ const struct bpf_func_proto bpf_skb_store_bytes_proto = {
 #define BPF_HEADER_FIELD_SIZE(flags)	((flags) & 0x0f)
 #define BPF_IS_PSEUDO_HEADER(flags)	((flags) & 0x10)
 
-static u64 bpf_l3_csum_replace(u64 r1, u64 offset, u64 from, u64 to, u64 flags)
+static u64 bpf_l3_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 {
 	struct sk_buff *skb = (struct sk_buff *) (long) r1;
+	int offset = (int) r2;
 	__sum16 sum, *ptr;
 
-	if (unlikely(offset > 0xffff))
+	if (unlikely((u32) offset > 0xffff))
 		return -EFAULT;
 
-	if (skb_cloned(skb) && !skb_clone_writable(skb, offset + sizeof(sum)))
+	offset -= skb->data - skb_mac_header(skb);
+	if (unlikely(skb_cloned(skb) &&
+		     bpf_skb_clone_unwritable(skb, offset + sizeof(sum))))
 		return -EFAULT;
 
 	ptr = skb_header_pointer(skb, offset, sizeof(sum), &sum);
@@ -1276,16 +1296,19 @@ const struct bpf_func_proto bpf_l3_csum_replace_proto = {
 	.arg5_type	= ARG_ANYTHING,
 };
 
-static u64 bpf_l4_csum_replace(u64 r1, u64 offset, u64 from, u64 to, u64 flags)
+static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 {
 	struct sk_buff *skb = (struct sk_buff *) (long) r1;
 	u32 is_pseudo = BPF_IS_PSEUDO_HEADER(flags);
+	int offset = (int) r2;
 	__sum16 sum, *ptr;
 
-	if (unlikely(offset > 0xffff))
+	if (unlikely((u32) offset > 0xffff))
 		return -EFAULT;
 
-	if (skb_cloned(skb) && !skb_clone_writable(skb, offset + sizeof(sum)))
+	offset -= skb->data - skb_mac_header(skb);
+	if (unlikely(skb_cloned(skb) &&
+		     bpf_skb_clone_unwritable(skb, offset + sizeof(sum))))
 		return -EFAULT;
 
 	ptr = skb_header_pointer(skb, offset, sizeof(sum), &sum);
diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c
index 4d2cede17468..dc6a2d324bd8 100644
--- a/net/sched/act_bpf.c
+++ b/net/sched/act_bpf.c
@@ -38,6 +38,9 @@ static int tcf_bpf(struct sk_buff *skb, const struct tc_action *act,
 	struct tcf_bpf *prog = act->priv;
 	int action, filter_res;
 
+	if (unlikely(!skb_mac_header_was_set(skb)))
+		return TC_ACT_UNSPEC;
+
 	spin_lock(&prog->tcf_lock);
 
 	prog->tcf_tm.lastuse = jiffies;
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index 5c4171c5d2bd..91bd9c19471d 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -66,6 +66,9 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 	struct cls_bpf_prog *prog;
 	int ret = -1;
 
+	if (unlikely(!skb_mac_header_was_set(skb)))
+		return -1;
+
 	/* Needed here for accessing maps. */
 	rcu_read_lock();
 	list_for_each_entry_rcu(prog, &head->plist, link) {
diff --git a/samples/bpf/tcbpf1_kern.c b/samples/bpf/tcbpf1_kern.c
index 7cf3f42a6e39..7c27710f8296 100644
--- a/samples/bpf/tcbpf1_kern.c
+++ b/samples/bpf/tcbpf1_kern.c
@@ -4,6 +4,8 @@
 #include <uapi/linux/ip.h>
 #include <uapi/linux/in.h>
 #include <uapi/linux/tcp.h>
+#include <uapi/linux/filter.h>
+
 #include "bpf_helpers.h"
 
 /* compiler workaround */
@@ -14,18 +16,12 @@ static inline void set_dst_mac(struct __sk_buff *skb, char *mac)
 	bpf_skb_store_bytes(skb, 0, mac, ETH_ALEN, 1);
 }
 
-/* use 1 below for ingress qdisc and 0 for egress */
-#if 0
-#undef ETH_HLEN
-#define ETH_HLEN 0
-#endif
-
 #define IP_CSUM_OFF (ETH_HLEN + offsetof(struct iphdr, check))
 #define TOS_OFF (ETH_HLEN + offsetof(struct iphdr, tos))
 
 static inline void set_ip_tos(struct __sk_buff *skb, __u8 new_tos)
 {
-	__u8 old_tos = load_byte(skb, TOS_OFF);
+	__u8 old_tos = load_byte(skb, BPF_LL_OFF + TOS_OFF);
 
 	bpf_l3_csum_replace(skb, IP_CSUM_OFF, htons(old_tos), htons(new_tos), 2);
 	bpf_skb_store_bytes(skb, TOS_OFF, &new_tos, sizeof(new_tos), 0);
@@ -38,7 +34,7 @@ static inline void set_ip_tos(struct __sk_buff *skb, __u8 new_tos)
 
 static inline void set_tcp_ip_src(struct __sk_buff *skb, __u32 new_ip)
 {
-	__u32 old_ip = _htonl(load_word(skb, IP_SRC_OFF));
+	__u32 old_ip = _htonl(load_word(skb, BPF_LL_OFF + IP_SRC_OFF));
 
 	bpf_l4_csum_replace(skb, TCP_CSUM_OFF, old_ip, new_ip, IS_PSEUDO | sizeof(new_ip));
 	bpf_l3_csum_replace(skb, IP_CSUM_OFF, old_ip, new_ip, sizeof(new_ip));
@@ -48,7 +44,7 @@ static inline void set_tcp_ip_src(struct __sk_buff *skb, __u32 new_ip)
 #define TCP_DPORT_OFF (ETH_HLEN + sizeof(struct iphdr) + offsetof(struct tcphdr, dest))
 static inline void set_tcp_dest_port(struct __sk_buff *skb, __u16 new_port)
 {
-	__u16 old_port = htons(load_half(skb, TCP_DPORT_OFF));
+	__u16 old_port = htons(load_half(skb, BPF_LL_OFF + TCP_DPORT_OFF));
 
 	bpf_l4_csum_replace(skb, TCP_CSUM_OFF, old_port, new_port, sizeof(new_port));
 	bpf_skb_store_bytes(skb, TCP_DPORT_OFF, &new_port, sizeof(new_port), 0);
@@ -57,7 +53,7 @@ static inline void set_tcp_dest_port(struct __sk_buff *skb, __u16 new_port)
 SEC("classifier")
 int bpf_prog1(struct __sk_buff *skb)
 {
-	__u8 proto = load_byte(skb, ETH_HLEN + offsetof(struct iphdr, protocol));
+	__u8 proto = load_byte(skb, BPF_LL_OFF + ETH_HLEN + offsetof(struct iphdr, protocol));
 	long *value;
 
 	if (proto == IPPROTO_TCP) {
-- 
cgit v1.2.3


From 4c0ee414e877b899f7fc80aafb98d9425c02797f Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 16 Apr 2015 16:12:53 +0800
Subject: Revert "net: Reset secmark when scrubbing packet"

This patch reverts commit b8fb4e0648a2ab3734140342002f68fb0c7d1602
because the secmark must be preserved even when a packet crosses
namespace boundaries.  The reason is that security labels apply to
the system as a whole and is not per-namespace.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/skbuff.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'net')

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 3b6e5830256e..f9800f4059b4 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4133,7 +4133,6 @@ void skb_scrub_packet(struct sk_buff *skb, bool xnet)
 	skb_dst_drop(skb);
 	skb->mark = 0;
 	skb_sender_cpu_clear(skb);
-	skb_init_secmark(skb);
 	secpath_reset(skb);
 	nf_reset(skb);
 	nf_reset_trace(skb);
-- 
cgit v1.2.3


From 213dd74aee765d4e5f3f4b9607fef0cf97faa2af Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 16 Apr 2015 09:03:27 +0800
Subject: skbuff: Do not scrub skb mark within the same name space
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On Wed, Apr 15, 2015 at 05:41:26PM +0200, Nicolas Dichtel wrote:
> Le 15/04/2015 15:57, Herbert Xu a écrit :
> >On Wed, Apr 15, 2015 at 06:22:29PM +0800, Herbert Xu wrote:
> [snip]
> >Subject: skbuff: Do not scrub skb mark within the same name space
> >
> >The commit ea23192e8e577dfc51e0f4fc5ca113af334edff9 ("tunnels:
> Maybe add a Fixes tag?
> Fixes: ea23192e8e57 ("tunnels: harmonize cleanup done on skb on rx path")
>
> >harmonize cleanup done on skb on rx path") broke anyone trying to
> >use netfilter marking across IPv4 tunnels.  While most of the
> >fields that are cleared by skb_scrub_packet don't matter, the
> >netfilter mark must be preserved.
> >
> >This patch rearranges skb_scurb_packet to preserve the mark field.
> nit: s/scurb/scrub
>
> Else it's fine for me.

Sure.

PS I used the wrong email for James the first time around.  So
let me repeat the question here.  Should secmark be preserved
or cleared across tunnels within the same name space? In fact,
do our security models even support name spaces?

---8<---
The commit ea23192e8e577dfc51e0f4fc5ca113af334edff9 ("tunnels:
harmonize cleanup done on skb on rx path") broke anyone trying to
use netfilter marking across IPv4 tunnels.  While most of the
fields that are cleared by skb_scrub_packet don't matter, the
netfilter mark must be preserved.

This patch rearranges skb_scrub_packet to preserve the mark field.

Fixes: ea23192e8e57 ("tunnels: harmonize cleanup done on skb on rx path")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/skbuff.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

(limited to 'net')

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index f9800f4059b4..d1967dab9cc6 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4124,18 +4124,21 @@ EXPORT_SYMBOL(skb_try_coalesce);
  */
 void skb_scrub_packet(struct sk_buff *skb, bool xnet)
 {
-	if (xnet)
-		skb_orphan(skb);
 	skb->tstamp.tv64 = 0;
 	skb->pkt_type = PACKET_HOST;
 	skb->skb_iif = 0;
 	skb->ignore_df = 0;
 	skb_dst_drop(skb);
-	skb->mark = 0;
 	skb_sender_cpu_clear(skb);
 	secpath_reset(skb);
 	nf_reset(skb);
 	nf_reset_trace(skb);
+
+	if (!xnet)
+		return;
+
+	skb_orphan(skb);
+	skb->mark = 0;
 }
 EXPORT_SYMBOL_GPL(skb_scrub_packet);
 
-- 
cgit v1.2.3


From fad9dfefea6405039491e7e4fc21fb6e59e7d26c Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Apr 2015 16:12:28 -0700
Subject: tcp: tcp_get_info() should fetch socket fields once

tcp_get_info() can be called without holding socket lock,
so any socket fields can change under us.

Use READ_ONCE() to fetch sk_pacing_rate and sk_max_pacing_rate

Fixes: 977cb0ecf82e ("tcp: add pacing_rate information into tcp_info")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

(limited to 'net')

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 18e3a12eb1b2..59c8a027721b 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2595,6 +2595,7 @@ void tcp_get_info(const struct sock *sk, struct tcp_info *info)
 	const struct tcp_sock *tp = tcp_sk(sk);
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	u32 now = tcp_time_stamp;
+	u32 rate;
 
 	memset(info, 0, sizeof(*info));
 
@@ -2655,10 +2656,11 @@ void tcp_get_info(const struct sock *sk, struct tcp_info *info)
 
 	info->tcpi_total_retrans = tp->total_retrans;
 
-	info->tcpi_pacing_rate = sk->sk_pacing_rate != ~0U ?
-					sk->sk_pacing_rate : ~0ULL;
-	info->tcpi_max_pacing_rate = sk->sk_max_pacing_rate != ~0U ?
-					sk->sk_max_pacing_rate : ~0ULL;
+	rate = READ_ONCE(sk->sk_pacing_rate);
+	info->tcpi_pacing_rate = rate != ~0U ? rate : ~0ULL;
+
+	rate = READ_ONCE(sk->sk_max_pacing_rate);
+	info->tcpi_max_pacing_rate = rate != ~0U ? rate : ~0ULL;
 }
 EXPORT_SYMBOL_GPL(tcp_get_info);
 
-- 
cgit v1.2.3


From 521f1cf1dbb9d5ad858dca5dc75d1b45f64b6589 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 16 Apr 2015 18:10:35 -0700
Subject: inet_diag: fix access to tcp cc information

Two different problems are fixed here :

1) inet_sk_diag_fill() might be called without socket lock held.
   icsk->icsk_ca_ops can change under us and module be unloaded.
   -> Access to freed memory.
   Fix this using rcu_read_lock() to prevent module unload.

2) Some TCP Congestion Control modules provide information
   but again this is not safe against icsk->icsk_ca_ops
   change and nla_put() errors were ignored. Some sockets
   could not get the additional info if skb was almost full.

Fix this by returning a status from get_info() handlers and
using rcu protection as well.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h       |  2 +-
 net/ipv4/inet_diag.c    | 28 ++++++++++++++++++++++------
 net/ipv4/tcp_dctcp.c    |  5 +++--
 net/ipv4/tcp_illinois.c |  6 +++---
 net/ipv4/tcp_vegas.c    |  5 +++--
 net/ipv4/tcp_vegas.h    |  2 +-
 net/ipv4/tcp_westwood.c |  6 +++---
 7 files changed, 36 insertions(+), 18 deletions(-)

(limited to 'net')

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 9598871485ce..051dc5c2802d 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -829,7 +829,7 @@ struct tcp_congestion_ops {
 	/* hook for packet ack accounting (optional) */
 	void (*pkts_acked)(struct sock *sk, u32 num_acked, s32 rtt_us);
 	/* get info for inet_diag (optional) */
-	void (*get_info)(struct sock *sk, u32 ext, struct sk_buff *skb);
+	int (*get_info)(struct sock *sk, u32 ext, struct sk_buff *skb);
 
 	char 		name[TCP_CA_NAME_MAX];
 	struct module 	*owner;
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 70e8b3c308ec..bb77ebdae3b3 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -111,6 +111,7 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 		      const struct nlmsghdr *unlh)
 {
 	const struct inet_sock *inet = inet_sk(sk);
+	const struct tcp_congestion_ops *ca_ops;
 	const struct inet_diag_handler *handler;
 	int ext = req->idiag_ext;
 	struct inet_diag_msg *r;
@@ -208,16 +209,31 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 		info = nla_data(attr);
 	}
 
-	if ((ext & (1 << (INET_DIAG_CONG - 1))) && icsk->icsk_ca_ops)
-		if (nla_put_string(skb, INET_DIAG_CONG,
-				   icsk->icsk_ca_ops->name) < 0)
+	if (ext & (1 << (INET_DIAG_CONG - 1))) {
+		int err = 0;
+
+		rcu_read_lock();
+		ca_ops = READ_ONCE(icsk->icsk_ca_ops);
+		if (ca_ops)
+			err = nla_put_string(skb, INET_DIAG_CONG, ca_ops->name);
+		rcu_read_unlock();
+		if (err < 0)
 			goto errout;
+	}
 
 	handler->idiag_get_info(sk, r, info);
 
-	if (sk->sk_state < TCP_TIME_WAIT &&
-	    icsk->icsk_ca_ops && icsk->icsk_ca_ops->get_info)
-		icsk->icsk_ca_ops->get_info(sk, ext, skb);
+	if (sk->sk_state < TCP_TIME_WAIT) {
+		int err = 0;
+
+		rcu_read_lock();
+		ca_ops = READ_ONCE(icsk->icsk_ca_ops);
+		if (ca_ops && ca_ops->get_info)
+			err = ca_ops->get_info(sk, ext, skb);
+		rcu_read_unlock();
+		if (err < 0)
+			goto errout;
+	}
 
 out:
 	nlmsg_end(skb, nlh);
diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c
index b504371af742..4376016f7fa5 100644
--- a/net/ipv4/tcp_dctcp.c
+++ b/net/ipv4/tcp_dctcp.c
@@ -277,7 +277,7 @@ static void dctcp_cwnd_event(struct sock *sk, enum tcp_ca_event ev)
 	}
 }
 
-static void dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
+static int dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
 {
 	const struct dctcp *ca = inet_csk_ca(sk);
 
@@ -297,8 +297,9 @@ static void dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
 			info.dctcp_ab_tot = ca->acked_bytes_total;
 		}
 
-		nla_put(skb, INET_DIAG_DCTCPINFO, sizeof(info), &info);
+		return nla_put(skb, INET_DIAG_DCTCPINFO, sizeof(info), &info);
 	}
+	return 0;
 }
 
 static struct tcp_congestion_ops dctcp __read_mostly = {
diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c
index 1d5a30a90adf..67476f085e48 100644
--- a/net/ipv4/tcp_illinois.c
+++ b/net/ipv4/tcp_illinois.c
@@ -300,8 +300,7 @@ static u32 tcp_illinois_ssthresh(struct sock *sk)
 }
 
 /* Extract info for Tcp socket info provided via netlink. */
-static void tcp_illinois_info(struct sock *sk, u32 ext,
-			      struct sk_buff *skb)
+static int tcp_illinois_info(struct sock *sk, u32 ext, struct sk_buff *skb)
 {
 	const struct illinois *ca = inet_csk_ca(sk);
 
@@ -318,8 +317,9 @@ static void tcp_illinois_info(struct sock *sk, u32 ext,
 			do_div(t, info.tcpv_rttcnt);
 			info.tcpv_rtt = t;
 		}
-		nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
+		return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
 	}
+	return 0;
 }
 
 static struct tcp_congestion_ops tcp_illinois __read_mostly = {
diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c
index a6afde666ab1..c71a1b8f7bde 100644
--- a/net/ipv4/tcp_vegas.c
+++ b/net/ipv4/tcp_vegas.c
@@ -286,7 +286,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked)
 }
 
 /* Extract info for Tcp socket info provided via netlink. */
-void tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
+int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
 {
 	const struct vegas *ca = inet_csk_ca(sk);
 	if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) {
@@ -297,8 +297,9 @@ void tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb)
 			.tcpv_minrtt = ca->minRTT,
 		};
 
-		nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
+		return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
 	}
+	return 0;
 }
 EXPORT_SYMBOL_GPL(tcp_vegas_get_info);
 
diff --git a/net/ipv4/tcp_vegas.h b/net/ipv4/tcp_vegas.h
index 0531b99d8637..e8a6b33cc61d 100644
--- a/net/ipv4/tcp_vegas.h
+++ b/net/ipv4/tcp_vegas.h
@@ -19,6 +19,6 @@ void tcp_vegas_init(struct sock *sk);
 void tcp_vegas_state(struct sock *sk, u8 ca_state);
 void tcp_vegas_pkts_acked(struct sock *sk, u32 cnt, s32 rtt_us);
 void tcp_vegas_cwnd_event(struct sock *sk, enum tcp_ca_event event);
-void tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb);
+int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb);
 
 #endif	/* __TCP_VEGAS_H */
diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c
index bb63fba47d47..b3c57cceb990 100644
--- a/net/ipv4/tcp_westwood.c
+++ b/net/ipv4/tcp_westwood.c
@@ -256,8 +256,7 @@ static void tcp_westwood_event(struct sock *sk, enum tcp_ca_event event)
 }
 
 /* Extract info for Tcp socket info provided via netlink. */
-static void tcp_westwood_info(struct sock *sk, u32 ext,
-			      struct sk_buff *skb)
+static int tcp_westwood_info(struct sock *sk, u32 ext, struct sk_buff *skb)
 {
 	const struct westwood *ca = inet_csk_ca(sk);
 
@@ -268,8 +267,9 @@ static void tcp_westwood_info(struct sock *sk, u32 ext,
 			.tcpv_minrtt = jiffies_to_usecs(ca->rtt_min),
 		};
 
-		nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
+		return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info);
 	}
+	return 0;
 }
 
 static struct tcp_congestion_ops tcp_westwood __read_mostly = {
-- 
cgit v1.2.3


From f40ae91307c275fc8b17420fa74145e9937c3c0b Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 17 Apr 2015 13:32:09 +0800
Subject: act_mirred: Fix bogus header when redirecting from VLAN

When you redirect a VLAN device to any device, you end up with
crap in af_packet on the xmit path because hard_header_len is
not equal to skb->mac_len.  So the redirected packet contains
four extra bytes at the start which then gets interpreted as
part of the MAC address.

This patch fixes this by only pushing skb->mac_len.  We also
need to fix ifb because it tries to undo the pushing done by
act_mirred.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c      | 2 +-
 net/sched/act_mirred.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'net')

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index 34f846b4bd05..94570aace241 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -105,7 +105,7 @@ static void ri_tasklet(unsigned long dev)
 		if (from & AT_EGRESS) {
 			dev_queue_xmit(skb);
 		} else if (from & AT_INGRESS) {
-			skb_pull(skb, skb->dev->hard_header_len);
+			skb_pull(skb, skb->mac_len);
 			netif_receive_skb(skb);
 		} else
 			BUG();
diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index 5953517ec059..3f63ceac8e01 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -157,7 +157,7 @@ static int tcf_mirred(struct sk_buff *skb, const struct tc_action *a,
 
 	if (!(at & AT_EGRESS)) {
 		if (m->tcfm_ok_push)
-			skb_push(skb2, skb2->dev->hard_header_len);
+			skb_push(skb2, skb->mac_len);
 	}
 
 	/* mirror is always swallowed */
-- 
cgit v1.2.3


From 8b86a61da37cbbcf4bd6e87fda494a59b1cf16c4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 17 Apr 2015 15:45:04 +0200
Subject: net: remove unused 'dev' argument from netif_needs_gso()

In commit 04ffcb255f22 ("net: Add ndo_gso_check") Tom originally
added the 'dev' argument to be able to call ndo_gso_check().

Then later, when generalizing this in commit 5f35227ea34b
("net: Generalize ndo_gso_check to ndo_features_check")
Jesse removed the call to ndo_gso_check() in netif_needs_gso()
by calling the new ndo_features_check() in a different place.
This made the 'dev' argument unused.

Remove the unused argument and go back to the code as before.

Cc: Tom Herbert <therbert@google.com>
Cc: Jesse Gross <jesse@nicira.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macvtap.c      | 2 +-
 drivers/net/xen-netfront.c | 2 +-
 include/linux/netdevice.h  | 2 +-
 net/core/dev.c             | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'net')

diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 8362aef0c15e..58c6ba5746d5 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -313,7 +313,7 @@ static rx_handler_result_t macvtap_handle_frame(struct sk_buff **pskb)
 	 */
 	if (q->flags & IFF_VNET_HDR)
 		features |= vlan->tap_features;
-	if (netif_needs_gso(dev, skb, features)) {
+	if (netif_needs_gso(skb, features)) {
 		struct sk_buff *segs = __skb_gso_segment(skb, features, false);
 
 		if (IS_ERR(segs))
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index 720aaf6313d2..8dcf31063e9a 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -560,7 +560,7 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (unlikely(!netif_carrier_ok(dev) ||
 		     (slots > 1 && !xennet_can_sg(dev)) ||
-		     netif_needs_gso(dev, skb, netif_skb_features(skb)))) {
+		     netif_needs_gso(skb, netif_skb_features(skb)))) {
 		spin_unlock_irqrestore(&queue->tx_lock, flags);
 		goto drop;
 	}
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b5679aed660b..bcbde799ec69 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3713,7 +3713,7 @@ static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
 	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
 }
 
-static inline bool netif_needs_gso(struct net_device *dev, struct sk_buff *skb,
+static inline bool netif_needs_gso(struct sk_buff *skb,
 				   netdev_features_t features)
 {
 	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
diff --git a/net/core/dev.c b/net/core/dev.c
index af4a1b0adc10..1796cef55ab5 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2713,7 +2713,7 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device
 	if (unlikely(!skb))
 		goto out_null;
 
-	if (netif_needs_gso(dev, skb, features)) {
+	if (netif_needs_gso(skb, features)) {
 		struct sk_buff *segs;
 
 		segs = skb_gso_segment(skb, features);
-- 
cgit v1.2.3


From e3122b7fae7b4e3d1d49fa84f6515bcbe6cbc6fc Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 17 Apr 2015 15:12:25 -0400
Subject: net: dsa: use DEVICE_ATTR_RW to declare temp1_max

Since commit da4759c (sysfs: Use only return value from is_visible for
the file mode), it is possible to reduce the permissions of a file.

So declare temp1_max with the DEVICE_ATTR_RW macro and remove the write
permission in dsa_hwmon_attrs_visible if set_temp_limit isn't provided.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/dsa.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'net')

diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 5eaadabe23a1..079a224471e7 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -124,7 +124,7 @@ static ssize_t temp1_max_store(struct device *dev,
 
 	return count;
 }
-static DEVICE_ATTR(temp1_max, S_IRUGO, temp1_max_show, temp1_max_store);
+static DEVICE_ATTR_RW(temp1_max);
 
 static ssize_t temp1_max_alarm_show(struct device *dev,
 				    struct device_attribute *attr, char *buf)
@@ -159,8 +159,8 @@ static umode_t dsa_hwmon_attrs_visible(struct kobject *kobj,
 	if (index == 1) {
 		if (!drv->get_temp_limit)
 			mode = 0;
-		else if (drv->set_temp_limit)
-			mode |= S_IWUSR;
+		else if (!drv->set_temp_limit)
+			mode &= ~S_IWUSR;
 	} else if (index == 2 && !drv->get_temp_alarm) {
 		mode = 0;
 	}
-- 
cgit v1.2.3


From 1f5014d6a77513fa7cefe30eb7791d5856c04384 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Fri, 17 Apr 2015 22:34:18 +0200
Subject: Bluetooth: hidp: Fix regression with older userspace and flags
 validation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

While it is not used by newer userspace anymore, the older userspace was
utilizing HIDP_VIRTUAL_CABLE_UNPLUG and HIDP_BOOT_PROTOCOL_MODE flags
when adding a new HIDP connection.

The flags validation is important, but we can not break older userspace
and with that allow providing these flags even if newer userspace does
not use them anymore.

Reported-and-tested-by: Jörg Otte <jrg.otte@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 net/bluetooth/hidp/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'net')

diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c
index a05b9dbf14c9..9070dfd6b4ad 100644
--- a/net/bluetooth/hidp/core.c
+++ b/net/bluetooth/hidp/core.c
@@ -1313,7 +1313,8 @@ int hidp_connection_add(struct hidp_connadd_req *req,
 			struct socket *ctrl_sock,
 			struct socket *intr_sock)
 {
-	u32 valid_flags = 0;
+	u32 valid_flags = BIT(HIDP_VIRTUAL_CABLE_UNPLUG) |
+			  BIT(HIDP_BOOT_PROTOCOL_MODE);
 	struct hidp_session *session;
 	struct l2cap_conn *conn;
 	struct l2cap_chan *chan;
-- 
cgit v1.2.3


From 67c64eb742a49d3d3f5dcef75d0c32a3394e5519 Mon Sep 17 00:00:00 2001
From: Ilya Dryomov <idryomov@gmail.com>
Date: Mon, 23 Mar 2015 14:52:40 +0300
Subject: libceph: don't overwrite specific con error msgs

- specific con->error_msg messages (e.g. "protocol version mismatch")
  end up getting overwritten by a catch-all "socket error on read
  / write", introduced in commit 3a140a0d5c4b ("libceph: report socket
  read/write error message")
- "bad message sequence # for incoming message" loses to "bad crc" due
  to the fact that -EBADMSG is used for both

Fix it, and tidy up con->error_msg assignments and pr_errs while at it.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
---
 net/ceph/messenger.c | 25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)

(limited to 'net')

diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c
index a9f4ae45b7fb..967080a9f043 100644
--- a/net/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -505,8 +505,6 @@ static int ceph_tcp_connect(struct ceph_connection *con)
 		pr_err("connect %s error %d\n",
 		       ceph_pr_addr(&con->peer_addr.in_addr), ret);
 		sock_release(sock);
-		con->error_msg = "connect error";
-
 		return ret;
 	}
 
@@ -2145,12 +2143,10 @@ static int process_connect(struct ceph_connection *con)
 		 * to WAIT.  This shouldn't happen if we are the
 		 * client.
 		 */
-		pr_err("process_connect got WAIT as client\n");
 		con->error_msg = "protocol error, got WAIT as client";
 		return -1;
 
 	default:
-		pr_err("connect protocol error, will retry\n");
 		con->error_msg = "protocol error, garbage tag during connect";
 		return -1;
 	}
@@ -2282,8 +2278,7 @@ static int read_partial_message(struct ceph_connection *con)
 
 	crc = crc32c(0, &con->in_hdr, offsetof(struct ceph_msg_header, crc));
 	if (cpu_to_le32(crc) != con->in_hdr.crc) {
-		pr_err("read_partial_message bad hdr "
-		       " crc %u != expected %u\n",
+		pr_err("read_partial_message bad hdr crc %u != expected %u\n",
 		       crc, con->in_hdr.crc);
 		return -EBADMSG;
 	}
@@ -2313,7 +2308,7 @@ static int read_partial_message(struct ceph_connection *con)
 		pr_err("read_partial_message bad seq %lld expected %lld\n",
 		       seq, con->in_seq + 1);
 		con->error_msg = "bad message sequence # for incoming message";
-		return -EBADMSG;
+		return -EBADE;
 	}
 
 	/* allocate message? */
@@ -2660,6 +2655,8 @@ more:
 			switch (ret) {
 			case -EBADMSG:
 				con->error_msg = "bad crc";
+				/* fall through */
+			case -EBADE:
 				ret = -EIO;
 				break;
 			case -EIO:
@@ -2838,7 +2835,8 @@ static void con_work(struct work_struct *work)
 		if (ret < 0) {
 			if (ret == -EAGAIN)
 				continue;
-			con->error_msg = "socket error on read";
+			if (!con->error_msg)
+				con->error_msg = "socket error on read";
 			fault = true;
 			break;
 		}
@@ -2847,7 +2845,8 @@ static void con_work(struct work_struct *work)
 		if (ret < 0) {
 			if (ret == -EAGAIN)
 				continue;
-			con->error_msg = "socket error on write";
+			if (!con->error_msg)
+				con->error_msg = "socket error on write";
 			fault = true;
 		}
 
@@ -2869,11 +2868,13 @@ static void con_work(struct work_struct *work)
  */
 static void con_fault(struct ceph_connection *con)
 {
-	pr_warn("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
-		ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg);
 	dout("fault %p state %lu to peer %s\n",
 	     con, con->state, ceph_pr_addr(&con->peer_addr.in_addr));
 
+	pr_warn("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
+		ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg);
+	con->error_msg = NULL;
+
 	WARN_ON(con->state != CON_STATE_CONNECTING &&
 	       con->state != CON_STATE_NEGOTIATING &&
 	       con->state != CON_STATE_OPEN);
@@ -3295,8 +3296,8 @@ static int ceph_con_in_msg_alloc(struct ceph_connection *con, int *skip)
 		 */
 		if (*skip)
 			return 0;
-		con->error_msg = "error allocating memory for incoming message";
 
+		con->error_msg = "error allocating memory for incoming message";
 		return -ENOMEM;
 	}
 	memcpy(&con->in_msg->hdr, &con->in_hdr, sizeof(con->in_hdr));
-- 
cgit v1.2.3


From ff40f9ae95917b72b6acb6057471c99054b6ee24 Mon Sep 17 00:00:00 2001
From: Ilya Dryomov <idryomov@gmail.com>
Date: Wed, 25 Mar 2015 21:02:16 +0300
Subject: libceph, ceph: split ceph_show_options()

Split ceph_show_options() into two pieces and move the piece
responsible for printing client (libceph) options into net/ceph.  This
way people adding a libceph option wouldn't have to remember to update
code in fs/ceph.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
---
 fs/ceph/super.c              | 40 +++++++++++++++-------------------------
 include/linux/ceph/libceph.h |  1 +
 net/ceph/ceph_common.c       | 37 +++++++++++++++++++++++++++++++++++++
 3 files changed, 53 insertions(+), 25 deletions(-)

(limited to 'net')

diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 9f035ccb6191..34a779edd421 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -411,31 +411,20 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root)
 {
 	struct ceph_fs_client *fsc = ceph_sb_to_client(root->d_sb);
 	struct ceph_mount_options *fsopt = fsc->mount_options;
-	struct ceph_options *opt = fsc->client->options;
-
-	if (opt->flags & CEPH_OPT_FSID)
-		seq_printf(m, ",fsid=%pU", &opt->fsid);
-	if (opt->flags & CEPH_OPT_NOSHARE)
-		seq_puts(m, ",noshare");
-	if (opt->flags & CEPH_OPT_NOCRC)
-		seq_puts(m, ",nocrc");
-	if (opt->flags & CEPH_OPT_NOMSGAUTH)
-		seq_puts(m, ",nocephx_require_signatures");
-	if ((opt->flags & CEPH_OPT_TCP_NODELAY) == 0)
-		seq_puts(m, ",notcp_nodelay");
-
-	if (opt->name)
-		seq_printf(m, ",name=%s", opt->name);
-	if (opt->key)
-		seq_puts(m, ",secret=<hidden>");
-
-	if (opt->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
-		seq_printf(m, ",mount_timeout=%d", opt->mount_timeout);
-	if (opt->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT)
-		seq_printf(m, ",osd_idle_ttl=%d", opt->osd_idle_ttl);
-	if (opt->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT)
-		seq_printf(m, ",osdkeepalivetimeout=%d",
-			   opt->osd_keepalive_timeout);
+	size_t pos;
+	int ret;
+
+	/* a comma between MNT/MS and client options */
+	seq_putc(m, ',');
+	pos = m->count;
+
+	ret = ceph_print_client_options(m, fsc->client);
+	if (ret)
+		return ret;
+
+	/* retract our comma if no client options */
+	if (m->count == pos)
+		m->count--;
 
 	if (fsopt->flags & CEPH_MOUNT_OPT_DIRSTAT)
 		seq_puts(m, ",dirstat");
@@ -482,6 +471,7 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root)
 		seq_printf(m, ",readdir_max_bytes=%d", fsopt->max_readdir_bytes);
 	if (strcmp(fsopt->snapdir_name, CEPH_SNAPDIRNAME_DEFAULT))
 		seq_printf(m, ",snapdirname=%s", fsopt->snapdir_name);
+
 	return 0;
 }
 
diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h
index 16fff9608848..303be6ef7f94 100644
--- a/include/linux/ceph/libceph.h
+++ b/include/linux/ceph/libceph.h
@@ -191,6 +191,7 @@ extern struct ceph_options *ceph_parse_options(char *options,
 			      const char *dev_name, const char *dev_name_end,
 			      int (*parse_extra_token)(char *c, void *private),
 			      void *private);
+int ceph_print_client_options(struct seq_file *m, struct ceph_client *client);
 extern void ceph_destroy_options(struct ceph_options *opt);
 extern int ceph_compare_options(struct ceph_options *new_opt,
 				struct ceph_client *client);
diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c
index ec565508e904..79e8f71aef5b 100644
--- a/net/ceph/ceph_common.c
+++ b/net/ceph/ceph_common.c
@@ -490,6 +490,43 @@ out:
 }
 EXPORT_SYMBOL(ceph_parse_options);
 
+int ceph_print_client_options(struct seq_file *m, struct ceph_client *client)
+{
+	struct ceph_options *opt = client->options;
+	size_t pos = m->count;
+
+	if (opt->name)
+		seq_printf(m, "name=%s,", opt->name);
+	if (opt->key)
+		seq_puts(m, "secret=<hidden>,");
+
+	if (opt->flags & CEPH_OPT_FSID)
+		seq_printf(m, "fsid=%pU,", &opt->fsid);
+	if (opt->flags & CEPH_OPT_NOSHARE)
+		seq_puts(m, "noshare,");
+	if (opt->flags & CEPH_OPT_NOCRC)
+		seq_puts(m, "nocrc,");
+	if (opt->flags & CEPH_OPT_NOMSGAUTH)
+		seq_puts(m, "nocephx_require_signatures,");
+	if ((opt->flags & CEPH_OPT_TCP_NODELAY) == 0)
+		seq_puts(m, "notcp_nodelay,");
+
+	if (opt->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
+		seq_printf(m, "mount_timeout=%d,", opt->mount_timeout);
+	if (opt->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT)
+		seq_printf(m, "osd_idle_ttl=%d,", opt->osd_idle_ttl);
+	if (opt->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT)
+		seq_printf(m, "osdkeepalivetimeout=%d,",
+			   opt->osd_keepalive_timeout);
+
+	/* drop redundant comma */
+	if (m->count != pos)
+		m->count--;
+
+	return 0;
+}
+EXPORT_SYMBOL(ceph_print_client_options);
+
 u64 ceph_client_id(struct ceph_client *client)
 {
 	return client->monc.auth->global_id;
-- 
cgit v1.2.3


From 5cf7bd30120ead3db43ef9074be38018d9acf22f Mon Sep 17 00:00:00 2001
From: Ilya Dryomov <idryomov@gmail.com>
Date: Wed, 25 Mar 2015 21:07:41 +0300
Subject: libceph: expose client options through debugfs

Add a client_options attribute for showing libceph options.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
---
 include/linux/ceph/libceph.h |  1 +
 net/ceph/debugfs.c           | 24 ++++++++++++++++++++++++
 2 files changed, 25 insertions(+)

(limited to 'net')

diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h
index 303be6ef7f94..30f92cefaa72 100644
--- a/include/linux/ceph/libceph.h
+++ b/include/linux/ceph/libceph.h
@@ -135,6 +135,7 @@ struct ceph_client {
 	struct dentry *debugfs_dir;
 	struct dentry *debugfs_monmap;
 	struct dentry *debugfs_osdmap;
+	struct dentry *debugfs_options;
 #endif
 };
 
diff --git a/net/ceph/debugfs.c b/net/ceph/debugfs.c
index 14d9995097cc..593dc2eabcc8 100644
--- a/net/ceph/debugfs.c
+++ b/net/ceph/debugfs.c
@@ -22,6 +22,7 @@
  *      .../monmap      - current monmap
  *      .../osdc        - active osd requests
  *      .../monc        - mon client state
+ *      .../client_options - libceph-only (i.e. not rbd or cephfs) options
  *      .../dentry_lru  - dump contents of dentry lru
  *      .../caps        - expose cap (reservation) stats
  *      .../bdi         - symlink to ../../bdi/something
@@ -177,10 +178,24 @@ static int osdc_show(struct seq_file *s, void *pp)
 	return 0;
 }
 
+static int client_options_show(struct seq_file *s, void *p)
+{
+	struct ceph_client *client = s->private;
+	int ret;
+
+	ret = ceph_print_client_options(s, client);
+	if (ret)
+		return ret;
+
+	seq_putc(s, '\n');
+	return 0;
+}
+
 CEPH_DEFINE_SHOW_FUNC(monmap_show)
 CEPH_DEFINE_SHOW_FUNC(osdmap_show)
 CEPH_DEFINE_SHOW_FUNC(monc_show)
 CEPH_DEFINE_SHOW_FUNC(osdc_show)
+CEPH_DEFINE_SHOW_FUNC(client_options_show)
 
 int ceph_debugfs_init(void)
 {
@@ -242,6 +257,14 @@ int ceph_debugfs_client_init(struct ceph_client *client)
 	if (!client->debugfs_osdmap)
 		goto out;
 
+	client->debugfs_options = debugfs_create_file("client_options",
+					0600,
+					client->debugfs_dir,
+					client,
+					&client_options_show_fops);
+	if (!client->debugfs_options)
+		goto out;
+
 	return 0;
 
 out:
@@ -252,6 +275,7 @@ out:
 void ceph_debugfs_client_cleanup(struct ceph_client *client)
 {
 	dout("ceph_debugfs_client_cleanup %p\n", client);
+	debugfs_remove(client->debugfs_options);
 	debugfs_remove(client->debugfs_osdmap);
 	debugfs_remove(client->debugfs_monmap);
 	debugfs_remove(client->osdc.debugfs_file);
-- 
cgit v1.2.3


From 2ab957492d13bb819400ac29ae55911d50a82a13 Mon Sep 17 00:00:00 2001
From: Sebastian Pöhn <sebastian.poehn@gmail.com>
Date: Mon, 20 Apr 2015 09:19:20 +0200
Subject: ip_forward: Drop frames with attached skb->sk

Initial discussion was:
[FYI] xfrm: Don't lookup sk_policy for timewait sockets

Forwarded frames should not have a socket attached. Especially
tw sockets will lead to panics later-on in the stack.

This was observed with TPROXY assigning a tw socket and broken
policy routing (misconfigured). As a result frame enters
forwarding path instead of input. We cannot solve this in
TPROXY as it cannot know that policy routing is broken.

v2:
Remove useless comment

Signed-off-by: Sebastian Poehn <sebastian.poehn@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/ip_forward.c | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'net')

diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 939992c456f3..3674484946a5 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -82,6 +82,9 @@ int ip_forward(struct sk_buff *skb)
 	if (skb->pkt_type != PACKET_HOST)
 		goto drop;
 
+	if (unlikely(skb->sk))
+		goto drop;
+
 	if (skb_warn_if_lro(skb))
 		goto drop;
 
-- 
cgit v1.2.3


From 3c7151275c0c9a80c3375f9874b1c7129a105eea Mon Sep 17 00:00:00 2001
From: "jbaron@akamai.com" <jbaron@akamai.com>
Date: Mon, 20 Apr 2015 20:05:07 +0000
Subject: tcp: add memory barriers to write space paths

Ensure that we either see that the buffer has write space
in tcp_poll() or that we perform a wakeup from the input
side. Did not run into any actual problem here, but thought
that we should make things explicit.

Signed-off-by: Jason Baron <jbaron@akamai.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp.c       | 4 +++-
 net/ipv4/tcp_input.c | 2 ++
 2 files changed, 5 insertions(+), 1 deletion(-)

(limited to 'net')

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 59c8a027721b..8c5cd9efebbc 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -520,8 +520,10 @@ unsigned int tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 
 				/* Race breaker. If space is freed after
 				 * wspace test but before the flags are set,
-				 * IO signal will be lost.
+				 * IO signal will be lost. Memory barrier
+				 * pairs with the input side.
 				 */
+				smp_mb__after_atomic();
 				if (sk_stream_is_writeable(sk))
 					mask |= POLLOUT | POLLWRNORM;
 			}
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a7ef679dd3ea..3a4d9b34bed4 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4845,6 +4845,8 @@ static void tcp_check_space(struct sock *sk)
 {
 	if (sock_flag(sk, SOCK_QUEUE_SHRUNK)) {
 		sock_reset_flag(sk, SOCK_QUEUE_SHRUNK);
+		/* pairs with tcp_poll() */
+		smp_mb__after_atomic();
 		if (sk->sk_socket &&
 		    test_bit(SOCK_NOSPACE, &sk->sk_socket->flags))
 			tcp_new_space(sk);
-- 
cgit v1.2.3


From 9be6df215a1baa57ccad0c388daeb2ad8f8d99ee Mon Sep 17 00:00:00 2001
From: Ilya Dryomov <idryomov@gmail.com>
Date: Tue, 14 Apr 2015 20:55:13 +0300
Subject: crush: drop unnecessary include from mapper.c

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
---
 net/ceph/crush/mapper.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'net')

diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c
index a1ef53c04415..5549fb609358 100644
--- a/net/ceph/crush/mapper.c
+++ b/net/ceph/crush/mapper.c
@@ -20,7 +20,6 @@
 
 #include <linux/crush/crush.h>
 #include <linux/crush/hash.h>
-#include <linux/crush/mapper.h>
 
 /*
  * Implement the core CRUSH mapping algorithm.
-- 
cgit v1.2.3


From 45002267e8d2699bf9b022315bee3dd13b044843 Mon Sep 17 00:00:00 2001
From: Ilya Dryomov <idryomov@gmail.com>
Date: Tue, 14 Apr 2015 16:04:23 +0300
Subject: crush: ensuring at most num-rep osds are selected

Crush temporary buffers are allocated as per replica size configured
by the user.  When there are more final osds (to be selected as per
rule) than the replicas, buffer overlaps and it causes crash.  Now, it
ensures that at most num-rep osds are selected even if more number of
osds are allowed by the rule.

Reflects ceph.git commits 6b4d1aa99718e3b367496326c1e64551330fabc0,
                          234b066ba04976783d15ff2abc3e81b6cc06fb10.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
---
 net/ceph/crush/mapper.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

(limited to 'net')

diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c
index 5549fb609358..91c41fe83113 100644
--- a/net/ceph/crush/mapper.c
+++ b/net/ceph/crush/mapper.c
@@ -289,6 +289,7 @@ static int is_out(const struct crush_map *map,
  * @type: the type of item to choose
  * @out: pointer to output vector
  * @outpos: our position in that vector
+ * @out_size: size of the out vector
  * @tries: number of attempts to make
  * @recurse_tries: number of attempts to have recursive chooseleaf make
  * @local_retries: localized retries
@@ -303,6 +304,7 @@ static int crush_choose_firstn(const struct crush_map *map,
 			       const __u32 *weight, int weight_max,
 			       int x, int numrep, int type,
 			       int *out, int outpos,
+			       int out_size,
 			       unsigned int tries,
 			       unsigned int recurse_tries,
 			       unsigned int local_retries,
@@ -321,6 +323,7 @@ static int crush_choose_firstn(const struct crush_map *map,
 	int item = 0;
 	int itemtype;
 	int collide, reject;
+	int count = out_size;
 
 	dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d tries %d recurse_tries %d local_retries %d local_fallback_retries %d parent_r %d\n",
 		recurse_to_leaf ? "_LEAF" : "",
@@ -328,7 +331,7 @@ static int crush_choose_firstn(const struct crush_map *map,
 		tries, recurse_tries, local_retries, local_fallback_retries,
 		parent_r);
 
-	for (rep = outpos; rep < numrep; rep++) {
+	for (rep = outpos; rep < numrep && count > 0 ; rep++) {
 		/* keep trying until we get a non-out, non-colliding item */
 		ftotal = 0;
 		skip_rep = 0;
@@ -402,7 +405,7 @@ static int crush_choose_firstn(const struct crush_map *map,
 							 map->buckets[-1-item],
 							 weight, weight_max,
 							 x, outpos+1, 0,
-							 out2, outpos,
+							 out2, outpos, count,
 							 recurse_tries, 0,
 							 local_retries,
 							 local_fallback_retries,
@@ -462,6 +465,7 @@ reject:
 		dprintk("CHOOSE got %d\n", item);
 		out[outpos] = item;
 		outpos++;
+		count--;
 	}
 
 	dprintk("CHOOSE returns %d\n", outpos);
@@ -653,6 +657,7 @@ int crush_do_rule(const struct crush_map *map,
 	__u32 step;
 	int i, j;
 	int numrep;
+	int out_size;
 	/*
 	 * the original choose_total_tries value was off by one (it
 	 * counted "retries" and not "tries").  add one.
@@ -760,6 +765,7 @@ int crush_do_rule(const struct crush_map *map,
 						x, numrep,
 						curstep->arg2,
 						o+osize, j,
+						result_max-osize,
 						choose_tries,
 						recurse_tries,
 						choose_local_retries,
@@ -769,11 +775,13 @@ int crush_do_rule(const struct crush_map *map,
 						c+osize,
 						0);
 				} else {
+					out_size = ((numrep < (result_max-osize)) ?
+                                                    numrep : (result_max-osize));
 					crush_choose_indep(
 						map,
 						map->buckets[-1-w[i]],
 						weight, weight_max,
-						x, numrep, numrep,
+						x, out_size, numrep,
 						curstep->arg2,
 						o+osize, j,
 						choose_tries,
@@ -782,7 +790,7 @@ int crush_do_rule(const struct crush_map *map,
 						recurse_to_leaf,
 						c+osize,
 						0);
-					osize += numrep;
+					osize += out_size;
 				}
 			}
 
-- 
cgit v1.2.3


From 958a27658d94cf212caeb0ffb04ee0b0bc89cc40 Mon Sep 17 00:00:00 2001
From: Ilya Dryomov <idryomov@gmail.com>
Date: Tue, 14 Apr 2015 16:54:52 +0300
Subject: crush: straw2 bucket type with an efficient 64-bit crush_ln()

This is an improved straw bucket that correctly avoids any data movement
between items A and B when neither A nor B's weights are changed.  Said
differently, if we adjust the weight of item C (including adding it anew
or removing it completely), we will only see inputs move to or from C,
never between other items in the bucket.

Notably, there is not intermediate scaling factor that needs to be
calculated.  The mapping function is a simple function of the item weights.

The below commits were squashed together into this one (mostly to avoid
adding and then yanking a ~6000 lines worth of crush_ln_table):

- crush: add a straw2 bucket type
- crush: add crush_ln to calculate nature log efficently
- crush: improve straw2 adjustment slightly
- crush: change crush_ln to provide 32 more digits
- crush: fix crush_get_bucket_item_weight and bucket destroy for straw2
- crush/mapper: fix divide-by-0 in straw2
  (with div64_s64() for draw = ln / w and INT64_MIN -> S64_MIN - need
   to create a proper compat.h in ceph.git)

Reflects ceph.git commits 242293c908e923d474910f2b8203fa3b41eb5a53,
                          32a1ead92efcd351822d22a5fc37d159c65c1338,
                          6289912418c4a3597a11778bcf29ed5415117ad9,
                          35fcb04e2945717cf5cfe150b9fa89cb3d2303a1,
                          6445d9ee7290938de1e4ee9563912a6ab6d8ee5f,
                          b5921d55d16796e12d66ad2c4add7305f9ce2353.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
---
 include/linux/crush/crush.h     |  12 ++-
 net/ceph/crush/crush.c          |  14 ++++
 net/ceph/crush/crush_ln_table.h | 166 ++++++++++++++++++++++++++++++++++++++++
 net/ceph/crush/mapper.c         | 101 ++++++++++++++++++++++++
 net/ceph/osdmap.c               |  25 ++++++
 5 files changed, 316 insertions(+), 2 deletions(-)
 create mode 100644 net/ceph/crush/crush_ln_table.h

(limited to 'net')

diff --git a/include/linux/crush/crush.h b/include/linux/crush/crush.h
index 4fad5f8ee01d..48a1a7d100f1 100644
--- a/include/linux/crush/crush.h
+++ b/include/linux/crush/crush.h
@@ -96,13 +96,15 @@ struct crush_rule {
  *  uniform         O(1)       poor         poor
  *  list            O(n)       optimal      poor
  *  tree            O(log n)   good         good
- *  straw           O(n)       optimal      optimal
+ *  straw           O(n)       better       better
+ *  straw2          O(n)       optimal      optimal
  */
 enum {
 	CRUSH_BUCKET_UNIFORM = 1,
 	CRUSH_BUCKET_LIST = 2,
 	CRUSH_BUCKET_TREE = 3,
-	CRUSH_BUCKET_STRAW = 4
+	CRUSH_BUCKET_STRAW = 4,
+	CRUSH_BUCKET_STRAW2 = 5,
 };
 extern const char *crush_bucket_alg_name(int alg);
 
@@ -149,6 +151,11 @@ struct crush_bucket_straw {
 	__u32 *straws;         /* 16-bit fixed point */
 };
 
+struct crush_bucket_straw2 {
+	struct crush_bucket h;
+	__u32 *item_weights;   /* 16-bit fixed point */
+};
+
 
 
 /*
@@ -189,6 +196,7 @@ extern void crush_destroy_bucket_uniform(struct crush_bucket_uniform *b);
 extern void crush_destroy_bucket_list(struct crush_bucket_list *b);
 extern void crush_destroy_bucket_tree(struct crush_bucket_tree *b);
 extern void crush_destroy_bucket_straw(struct crush_bucket_straw *b);
+extern void crush_destroy_bucket_straw2(struct crush_bucket_straw2 *b);
 extern void crush_destroy_bucket(struct crush_bucket *b);
 extern void crush_destroy_rule(struct crush_rule *r);
 extern void crush_destroy(struct crush_map *map);
diff --git a/net/ceph/crush/crush.c b/net/ceph/crush/crush.c
index 16bc199d9a62..9d84ce4ea0df 100644
--- a/net/ceph/crush/crush.c
+++ b/net/ceph/crush/crush.c
@@ -17,6 +17,7 @@ const char *crush_bucket_alg_name(int alg)
 	case CRUSH_BUCKET_LIST: return "list";
 	case CRUSH_BUCKET_TREE: return "tree";
 	case CRUSH_BUCKET_STRAW: return "straw";
+	case CRUSH_BUCKET_STRAW2: return "straw2";
 	default: return "unknown";
 	}
 }
@@ -40,6 +41,8 @@ int crush_get_bucket_item_weight(const struct crush_bucket *b, int p)
 		return ((struct crush_bucket_tree *)b)->node_weights[crush_calc_tree_node(p)];
 	case CRUSH_BUCKET_STRAW:
 		return ((struct crush_bucket_straw *)b)->item_weights[p];
+	case CRUSH_BUCKET_STRAW2:
+		return ((struct crush_bucket_straw2 *)b)->item_weights[p];
 	}
 	return 0;
 }
@@ -77,6 +80,14 @@ void crush_destroy_bucket_straw(struct crush_bucket_straw *b)
 	kfree(b);
 }
 
+void crush_destroy_bucket_straw2(struct crush_bucket_straw2 *b)
+{
+	kfree(b->item_weights);
+	kfree(b->h.perm);
+	kfree(b->h.items);
+	kfree(b);
+}
+
 void crush_destroy_bucket(struct crush_bucket *b)
 {
 	switch (b->alg) {
@@ -92,6 +103,9 @@ void crush_destroy_bucket(struct crush_bucket *b)
 	case CRUSH_BUCKET_STRAW:
 		crush_destroy_bucket_straw((struct crush_bucket_straw *)b);
 		break;
+	case CRUSH_BUCKET_STRAW2:
+		crush_destroy_bucket_straw2((struct crush_bucket_straw2 *)b);
+		break;
 	}
 }
 
diff --git a/net/ceph/crush/crush_ln_table.h b/net/ceph/crush/crush_ln_table.h
new file mode 100644
index 000000000000..6192c7fc958c
--- /dev/null
+++ b/net/ceph/crush/crush_ln_table.h
@@ -0,0 +1,166 @@
+/*
+ * Ceph - scalable distributed file system
+ *
+ * Copyright (C) 2015 Intel Corporation All Rights Reserved
+ *
+ * This is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License version 2.1, as published by the Free Software
+ * Foundation.  See file COPYING.
+ *
+ */
+
+#if defined(__linux__)
+#include <linux/types.h>
+#elif defined(__FreeBSD__)
+#include <sys/types.h>
+#endif
+
+#ifndef CEPH_CRUSH_LN_H
+#define CEPH_CRUSH_LN_H
+
+
+// RH_LH_tbl[2*k] = 2^48/(1.0+k/128.0)
+// RH_LH_tbl[2*k+1] = 2^48*log2(1.0+k/128.0)
+
+static int64_t __RH_LH_tbl[128*2+2] = {
+  0x0001000000000000ll, 0x0000000000000000ll, 0x0000fe03f80fe040ll, 0x000002dfca16dde1ll,
+  0x0000fc0fc0fc0fc1ll, 0x000005b9e5a170b4ll, 0x0000fa232cf25214ll, 0x0000088e68ea899all,
+  0x0000f83e0f83e0f9ll, 0x00000b5d69bac77ell, 0x0000f6603d980f67ll, 0x00000e26fd5c8555ll,
+  0x0000f4898d5f85bcll, 0x000010eb389fa29fll, 0x0000f2b9d6480f2cll, 0x000013aa2fdd27f1ll,
+  0x0000f0f0f0f0f0f1ll, 0x00001663f6fac913ll, 0x0000ef2eb71fc435ll, 0x00001918a16e4633ll,
+  0x0000ed7303b5cc0fll, 0x00001bc84240adabll, 0x0000ebbdb2a5c162ll, 0x00001e72ec117fa5ll,
+  0x0000ea0ea0ea0ea1ll, 0x00002118b119b4f3ll, 0x0000e865ac7b7604ll, 0x000023b9a32eaa56ll,
+  0x0000e6c2b4481cd9ll, 0x00002655d3c4f15cll, 0x0000e525982af70dll, 0x000028ed53f307eell,
+  0x0000e38e38e38e39ll, 0x00002b803473f7adll, 0x0000e1fc780e1fc8ll, 0x00002e0e85a9de04ll,
+  0x0000e070381c0e08ll, 0x0000309857a05e07ll, 0x0000dee95c4ca038ll, 0x0000331dba0efce1ll,
+  0x0000dd67c8a60dd7ll, 0x0000359ebc5b69d9ll, 0x0000dbeb61eed19dll, 0x0000381b6d9bb29bll,
+  0x0000da740da740dbll, 0x00003a93dc9864b2ll, 0x0000d901b2036407ll, 0x00003d0817ce9cd4ll,
+  0x0000d79435e50d7all, 0x00003f782d7204d0ll, 0x0000d62b80d62b81ll, 0x000041e42b6ec0c0ll,
+  0x0000d4c77b03531ell, 0x0000444c1f6b4c2dll, 0x0000d3680d3680d4ll, 0x000046b016ca47c1ll,
+  0x0000d20d20d20d21ll, 0x000049101eac381cll, 0x0000d0b69fcbd259ll, 0x00004b6c43f1366all,
+  0x0000cf6474a8819fll, 0x00004dc4933a9337ll, 0x0000ce168a772509ll, 0x0000501918ec6c11ll,
+  0x0000cccccccccccdll, 0x00005269e12f346ell, 0x0000cb8727c065c4ll, 0x000054b6f7f1325all,
+  0x0000ca4587e6b750ll, 0x0000570068e7ef5all, 0x0000c907da4e8712ll, 0x000059463f919deell,
+  0x0000c7ce0c7ce0c8ll, 0x00005b8887367433ll, 0x0000c6980c6980c7ll, 0x00005dc74ae9fbecll,
+  0x0000c565c87b5f9ell, 0x00006002958c5871ll, 0x0000c4372f855d83ll, 0x0000623a71cb82c8ll,
+  0x0000c30c30c30c31ll, 0x0000646eea247c5cll, 0x0000c1e4bbd595f7ll, 0x000066a008e4788cll,
+  0x0000c0c0c0c0c0c1ll, 0x000068cdd829fd81ll, 0x0000bfa02fe80bfbll, 0x00006af861e5fc7dll,
+  0x0000be82fa0be830ll, 0x00006d1fafdce20all, 0x0000bd6910470767ll, 0x00006f43cba79e40ll,
+  0x0000bc52640bc527ll, 0x00007164beb4a56dll, 0x0000bb3ee721a54ell, 0x000073829248e961ll,
+  0x0000ba2e8ba2e8bbll, 0x0000759d4f80cba8ll, 0x0000b92143fa36f6ll, 0x000077b4ff5108d9ll,
+  0x0000b81702e05c0cll, 0x000079c9aa879d53ll, 0x0000b70fbb5a19bfll, 0x00007bdb59cca388ll,
+  0x0000b60b60b60b61ll, 0x00007dea15a32c1bll, 0x0000b509e68a9b95ll, 0x00007ff5e66a0ffell,
+  0x0000b40b40b40b41ll, 0x000081fed45cbccbll, 0x0000b30f63528918ll, 0x00008404e793fb81ll,
+  0x0000b21642c8590cll, 0x000086082806b1d5ll, 0x0000b11fd3b80b12ll, 0x000088089d8a9e47ll,
+  0x0000b02c0b02c0b1ll, 0x00008a064fd50f2all, 0x0000af3addc680b0ll, 0x00008c01467b94bbll,
+  0x0000ae4c415c9883ll, 0x00008df988f4ae80ll, 0x0000ad602b580ad7ll, 0x00008fef1e987409ll,
+  0x0000ac7691840ac8ll, 0x000091e20ea1393ell, 0x0000ab8f69e2835all, 0x000093d2602c2e5fll,
+  0x0000aaaaaaaaaaabll, 0x000095c01a39fbd6ll, 0x0000a9c84a47a080ll, 0x000097ab43af59f9ll,
+  0x0000a8e83f5717c1ll, 0x00009993e355a4e5ll, 0x0000a80a80a80a81ll, 0x00009b79ffdb6c8bll,
+  0x0000a72f0539782all, 0x00009d5d9fd5010bll, 0x0000a655c4392d7cll, 0x00009f3ec9bcfb80ll,
+  0x0000a57eb50295fbll, 0x0000a11d83f4c355ll, 0x0000a4a9cf1d9684ll, 0x0000a2f9d4c51039ll,
+  0x0000a3d70a3d70a4ll, 0x0000a4d3c25e68dcll, 0x0000a3065e3fae7dll, 0x0000a6ab52d99e76ll,
+  0x0000a237c32b16d0ll, 0x0000a8808c384547ll, 0x0000a16b312ea8fdll, 0x0000aa5374652a1cll,
+  0x0000a0a0a0a0a0a1ll, 0x0000ac241134c4e9ll, 0x00009fd809fd80a0ll, 0x0000adf26865a8a1ll,
+  0x00009f1165e72549ll, 0x0000afbe7fa0f04dll, 0x00009e4cad23dd60ll, 0x0000b1885c7aa982ll,
+  0x00009d89d89d89d9ll, 0x0000b35004723c46ll, 0x00009cc8e160c3fcll, 0x0000b5157cf2d078ll,
+  0x00009c09c09c09c1ll, 0x0000b6d8cb53b0call, 0x00009b4c6f9ef03bll, 0x0000b899f4d8ab63ll,
+  0x00009a90e7d95bc7ll, 0x0000ba58feb2703all, 0x000099d722dabde6ll, 0x0000bc15edfeed32ll,
+  0x0000991f1a515886ll, 0x0000bdd0c7c9a817ll, 0x00009868c809868dll, 0x0000bf89910c1678ll,
+  0x000097b425ed097cll, 0x0000c1404eadf383ll, 0x000097012e025c05ll, 0x0000c2f5058593d9ll,
+  0x0000964fda6c0965ll, 0x0000c4a7ba58377cll, 0x000095a02568095bll, 0x0000c65871da59ddll,
+  0x000094f2094f2095ll, 0x0000c80730b00016ll, 0x0000944580944581ll, 0x0000c9b3fb6d0559ll,
+  0x0000939a85c4093all, 0x0000cb5ed69565afll, 0x000092f113840498ll, 0x0000cd07c69d8702ll,
+  0x0000924924924925ll, 0x0000ceaecfea8085ll, 0x000091a2b3c4d5e7ll, 0x0000d053f6d26089ll,
+  0x000090fdbc090fdcll, 0x0000d1f73f9c70c0ll, 0x0000905a38633e07ll, 0x0000d398ae817906ll,
+  0x00008fb823ee08fcll, 0x0000d53847ac00a6ll, 0x00008f1779d9fdc4ll, 0x0000d6d60f388e41ll,
+  0x00008e78356d1409ll, 0x0000d8720935e643ll, 0x00008dda5202376all, 0x0000da0c39a54804ll,
+  0x00008d3dcb08d3ddll, 0x0000dba4a47aa996ll, 0x00008ca29c046515ll, 0x0000dd3b4d9cf24bll,
+  0x00008c08c08c08c1ll, 0x0000ded038e633f3ll, 0x00008b70344a139cll, 0x0000e0636a23e2eell,
+  0x00008ad8f2fba939ll, 0x0000e1f4e5170d02ll, 0x00008a42f870566all, 0x0000e384ad748f0ell,
+  0x000089ae4089ae41ll, 0x0000e512c6e54998ll, 0x0000891ac73ae982ll, 0x0000e69f35065448ll,
+  0x0000888888888889ll, 0x0000e829fb693044ll, 0x000087f78087f781ll, 0x0000e9b31d93f98ell,
+  0x00008767ab5f34e5ll, 0x0000eb3a9f019750ll, 0x000086d905447a35ll, 0x0000ecc08321eb30ll,
+  0x0000864b8a7de6d2ll, 0x0000ee44cd59ffabll, 0x000085bf37612cefll, 0x0000efc781043579ll,
+  0x0000853408534086ll, 0x0000f148a170700all, 0x000084a9f9c8084bll, 0x0000f2c831e44116ll,
+  0x0000842108421085ll, 0x0000f446359b1353ll, 0x0000839930523fbfll, 0x0000f5c2afc65447ll,
+  0x000083126e978d50ll, 0x0000f73da38d9d4all, 0x0000828cbfbeb9a1ll, 0x0000f8b7140edbb1ll,
+  0x0000820820820821ll, 0x0000fa2f045e7832ll, 0x000081848da8faf1ll, 0x0000fba577877d7dll,
+  0x0000810204081021ll, 0x0000fd1a708bbe11ll, 0x0000808080808081ll, 0x0000fe8df263f957ll,
+  0x0000800000000000ll, 0x0000ffff00000000ll,
+  };
+
+
+    // LL_tbl[k] = 2^48*log2(1.0+k/2^15);
+static int64_t __LL_tbl[256] = {
+  0x0000000000000000ull, 0x00000002e2a60a00ull, 0x000000070cb64ec5ull, 0x00000009ef50ce67ull,
+  0x0000000cd1e588fdull, 0x0000000fb4747e9cull, 0x0000001296fdaf5eull, 0x0000001579811b58ull,
+  0x000000185bfec2a1ull, 0x0000001b3e76a552ull, 0x0000001e20e8c380ull, 0x0000002103551d43ull,
+  0x00000023e5bbb2b2ull, 0x00000026c81c83e4ull, 0x00000029aa7790f0ull, 0x0000002c8cccd9edull,
+  0x0000002f6f1c5ef2ull, 0x0000003251662017ull, 0x0000003533aa1d71ull, 0x0000003815e8571aull,
+  0x0000003af820cd26ull, 0x0000003dda537faeull, 0x00000040bc806ec8ull, 0x000000439ea79a8cull,
+  0x0000004680c90310ull, 0x0000004962e4a86cull, 0x0000004c44fa8ab6ull, 0x0000004f270aaa06ull,
+  0x0000005209150672ull, 0x00000054eb19a013ull, 0x00000057cd1876fdull, 0x0000005aaf118b4aull,
+  0x0000005d9104dd0full, 0x0000006072f26c64ull, 0x0000006354da3960ull, 0x0000006636bc441aull,
+  0x0000006918988ca8ull, 0x0000006bfa6f1322ull, 0x0000006edc3fd79full, 0x00000071be0ada35ull,
+  0x000000749fd01afdull, 0x00000077818f9a0cull, 0x0000007a6349577aull, 0x0000007d44fd535eull,
+  0x0000008026ab8dceull, 0x00000083085406e3ull, 0x00000085e9f6beb2ull, 0x00000088cb93b552ull,
+  0x0000008bad2aeadcull, 0x0000008e8ebc5f65ull, 0x0000009170481305ull, 0x0000009451ce05d3ull,
+  0x00000097334e37e5ull, 0x0000009a14c8a953ull, 0x0000009cf63d5a33ull, 0x0000009fd7ac4a9dull,
+  0x000000a2b07f3458ull, 0x000000a59a78ea6aull, 0x000000a87bd699fbull, 0x000000ab5d2e8970ull,
+  0x000000ae3e80b8e3ull, 0x000000b11fcd2869ull, 0x000000b40113d818ull, 0x000000b6e254c80aull,
+  0x000000b9c38ff853ull, 0x000000bca4c5690cull, 0x000000bf85f51a4aull, 0x000000c2671f0c26ull,
+  0x000000c548433eb6ull, 0x000000c82961b211ull, 0x000000cb0a7a664dull, 0x000000cdeb8d5b82ull,
+  0x000000d0cc9a91c8ull, 0x000000d3ada20933ull, 0x000000d68ea3c1ddull, 0x000000d96f9fbbdbull,
+  0x000000dc5095f744ull, 0x000000df31867430ull, 0x000000e2127132b5ull, 0x000000e4f35632eaull,
+  0x000000e7d43574e6ull, 0x000000eab50ef8c1ull, 0x000000ed95e2be90ull, 0x000000f076b0c66cull,
+  0x000000f35779106aull, 0x000000f6383b9ca2ull, 0x000000f918f86b2aull, 0x000000fbf9af7c1aull,
+  0x000000feda60cf88ull, 0x00000101bb0c658cull, 0x000001049bb23e3cull, 0x000001077c5259afull,
+  0x0000010a5cecb7fcull, 0x0000010d3d81593aull, 0x000001101e103d7full, 0x00000112fe9964e4ull,
+  0x00000115df1ccf7eull, 0x00000118bf9a7d64ull, 0x0000011ba0126eadull, 0x0000011e8084a371ull,
+  0x0000012160f11bc6ull, 0x000001244157d7c3ull, 0x0000012721b8d77full, 0x0000012a02141b10ull,
+  0x0000012ce269a28eull, 0x0000012fc2b96e0full, 0x00000132a3037daaull, 0x000001358347d177ull,
+  0x000001386386698cull, 0x0000013b43bf45ffull, 0x0000013e23f266e9ull, 0x00000141041fcc5eull,
+  0x00000143e4477678ull, 0x00000146c469654bull, 0x00000149a48598f0ull, 0x0000014c849c117cull,
+  0x0000014f64accf08ull, 0x0000015244b7d1a9ull, 0x0000015524bd1976ull, 0x0000015804bca687ull,
+  0x0000015ae4b678f2ull, 0x0000015dc4aa90ceull, 0x00000160a498ee31ull, 0x0000016384819134ull,
+  0x00000166646479ecull, 0x000001694441a870ull, 0x0000016c24191cd7ull, 0x0000016df6ca19bdull,
+  0x00000171e3b6d7aaull, 0x00000174c37d1e44ull, 0x00000177a33dab1cull, 0x0000017a82f87e49ull,
+  0x0000017d62ad97e2ull, 0x00000180425cf7feull, 0x00000182b07f3458ull, 0x0000018601aa8c19ull,
+  0x00000188e148c046ull, 0x0000018bc0e13b52ull, 0x0000018ea073fd52ull, 0x000001918001065dull,
+  0x000001945f88568bull, 0x000001973f09edf2ull, 0x0000019a1e85ccaaull, 0x0000019cfdfbf2c8ull,
+  0x0000019fdd6c6063ull, 0x000001a2bcd71593ull, 0x000001a59c3c126eull, 0x000001a87b9b570bull,
+  0x000001ab5af4e380ull, 0x000001ae3a48b7e5ull, 0x000001b11996d450ull, 0x000001b3f8df38d9ull,
+  0x000001b6d821e595ull, 0x000001b9b75eda9bull, 0x000001bc96961803ull, 0x000001bf75c79de3ull,
+  0x000001c254f36c51ull, 0x000001c534198365ull, 0x000001c81339e336ull, 0x000001caf2548bd9ull,
+  0x000001cdd1697d67ull, 0x000001d0b078b7f5ull, 0x000001d38f823b9aull, 0x000001d66e86086dull,
+  0x000001d94d841e86ull, 0x000001dc2c7c7df9ull, 0x000001df0b6f26dfull, 0x000001e1ea5c194eull,
+  0x000001e4c943555dull, 0x000001e7a824db23ull, 0x000001ea8700aab5ull, 0x000001ed65d6c42bull,
+  0x000001f044a7279dull, 0x000001f32371d51full, 0x000001f60236cccaull, 0x000001f8e0f60eb3ull,
+  0x000001fbbfaf9af3ull, 0x000001fe9e63719eull, 0x000002017d1192ccull, 0x000002045bb9fe94ull,
+  0x000002073a5cb50dull, 0x00000209c06e6212ull, 0x0000020cf791026aull, 0x0000020fd622997cull,
+  0x00000212b07f3458ull, 0x000002159334a8d8ull, 0x0000021871b52150ull, 0x0000021b502fe517ull,
+  0x0000021d6a73a78full, 0x000002210d144eeeull, 0x00000223eb7df52cull, 0x00000226c9e1e713ull,
+  0x00000229a84024bbull, 0x0000022c23679b4eull, 0x0000022f64eb83a8ull, 0x000002324338a51bull,
+  0x00000235218012a9ull, 0x00000237ffc1cc69ull, 0x0000023a2c3b0ea4ull, 0x0000023d13ee805bull,
+  0x0000024035e9221full, 0x00000243788faf25ull, 0x0000024656b4e735ull, 0x00000247ed646bfeull,
+  0x0000024c12ee3d98ull, 0x0000024ef1025c1aull, 0x00000251cf10c799ull, 0x0000025492644d65ull,
+  0x000002578b1c85eeull, 0x0000025a6919d8f0ull, 0x0000025d13ee805bull, 0x0000026025036716ull,
+  0x0000026296453882ull, 0x00000265e0d62b53ull, 0x00000268beb701f3ull, 0x0000026b9c92265eull,
+  0x0000026d32f798a9ull, 0x00000271583758ebull, 0x000002743601673bull, 0x0000027713c5c3b0ull,
+  0x00000279f1846e5full, 0x0000027ccf3d6761ull, 0x0000027e6580aecbull, 0x000002828a9e44b3ull,
+  0x0000028568462932ull, 0x00000287bdbf5255ull, 0x0000028b2384de4aull, 0x0000028d13ee805bull,
+  0x0000029035e9221full, 0x0000029296453882ull, 0x0000029699bdfb61ull, 0x0000029902a37aabull,
+  0x0000029c54b864c9ull, 0x0000029deabd1083ull, 0x000002a20f9c0bb5ull, 0x000002a4c7605d61ull,
+  0x000002a7bdbf5255ull, 0x000002a96056dafcull, 0x000002ac3daf14efull, 0x000002af1b019ecaull,
+  0x000002b296453882ull, 0x000002b5d022d80full, 0x000002b8fa471cb3ull, 0x000002ba9012e713ull,
+  0x000002bd6d4901ccull, 0x000002c04a796cf6ull, 0x000002c327a428a6ull, 0x000002c61a5e8f4cull,
+  0x000002c8e1e891f6ull, 0x000002cbbf023fc2ull, 0x000002ce9c163e6eull, 0x000002d179248e13ull,
+  0x000002d4562d2ec6ull, 0x000002d73330209dull, 0x000002da102d63b0ull, 0x000002dced24f814ull,
+};
+
+
+
+
+#endif
diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c
index 91c41fe83113..5b47736d27d9 100644
--- a/net/ceph/crush/mapper.c
+++ b/net/ceph/crush/mapper.c
@@ -20,6 +20,7 @@
 
 #include <linux/crush/crush.h>
 #include <linux/crush/hash.h>
+#include "crush_ln_table.h"
 
 /*
  * Implement the core CRUSH mapping algorithm.
@@ -237,6 +238,102 @@ static int bucket_straw_choose(struct crush_bucket_straw *bucket,
 	return bucket->h.items[high];
 }
 
+// compute 2^44*log2(input+1)
+uint64_t crush_ln(unsigned xin)
+{
+    unsigned x=xin, x1;
+    int iexpon, index1, index2;
+    uint64_t RH, LH, LL, xl64, result;
+
+    x++;
+
+    // normalize input
+    iexpon = 15;
+    while(!(x&0x18000)) { x<<=1; iexpon--; }
+
+    index1 = (x>>8)<<1;
+    // RH ~ 2^56/index1
+    RH = __RH_LH_tbl[index1 - 256];
+    // LH ~ 2^48 * log2(index1/256)
+    LH = __RH_LH_tbl[index1 + 1 - 256];
+
+    // RH*x ~ 2^48 * (2^15 + xf), xf<2^8
+    xl64 = (int64_t)x * RH;
+    xl64 >>= 48;
+    x1 = xl64;
+
+    result = iexpon;
+    result <<= (12 + 32);
+
+    index2 = x1 & 0xff;
+    // LL ~ 2^48*log2(1.0+index2/2^15)
+    LL = __LL_tbl[index2];
+
+    LH = LH + LL;
+
+    LH >>= (48-12 - 32);
+    result += LH;
+
+    return result;
+}
+
+
+/*
+ * straw2
+ *
+ * for reference, see:
+ *
+ * http://en.wikipedia.org/wiki/Exponential_distribution#Distribution_of_the_minimum_of_exponential_random_variables
+ *
+ */
+
+static int bucket_straw2_choose(struct crush_bucket_straw2 *bucket,
+				int x, int r)
+{
+	unsigned i, high = 0;
+	unsigned u;
+	unsigned w;
+	__s64 ln, draw, high_draw = 0;
+
+	for (i = 0; i < bucket->h.size; i++) {
+		w = bucket->item_weights[i];
+		if (w) {
+			u = crush_hash32_3(bucket->h.hash, x,
+					   bucket->h.items[i], r);
+			u &= 0xffff;
+
+			/*
+			 * for some reason slightly less than 0x10000 produces
+			 * a slightly more accurate distribution... probably a
+			 * rounding effect.
+			 *
+			 * the natural log lookup table maps [0,0xffff]
+			 * (corresponding to real numbers [1/0x10000, 1] to
+			 * [0, 0xffffffffffff] (corresponding to real numbers
+			 * [-11.090355,0]).
+			 */
+			ln = crush_ln(u) - 0x1000000000000ll;
+
+			/*
+			 * divide by 16.16 fixed-point weight.  note
+			 * that the ln value is negative, so a larger
+			 * weight means a larger (less negative) value
+			 * for draw.
+			 */
+			draw = div64_s64(ln, w);
+		} else {
+			draw = S64_MIN;
+		}
+
+		if (i == 0 || draw > high_draw) {
+			high = i;
+			high_draw = draw;
+		}
+	}
+	return bucket->h.items[high];
+}
+
+
 static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
 {
 	dprintk(" crush_bucket_choose %d x=%d r=%d\n", in->id, x, r);
@@ -254,12 +351,16 @@ static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
 	case CRUSH_BUCKET_STRAW:
 		return bucket_straw_choose((struct crush_bucket_straw *)in,
 					   x, r);
+	case CRUSH_BUCKET_STRAW2:
+		return bucket_straw2_choose((struct crush_bucket_straw2 *)in,
+					    x, r);
 	default:
 		dprintk("unknown bucket %d alg %d\n", in->id, in->alg);
 		return in->items[0];
 	}
 }
 
+
 /*
  * true if device is marked "out" (failed, fully offloaded)
  * of the cluster
diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c
index b8c3fde5b04f..15796696d64e 100644
--- a/net/ceph/osdmap.c
+++ b/net/ceph/osdmap.c
@@ -122,6 +122,22 @@ bad:
 	return -EINVAL;
 }
 
+static int crush_decode_straw2_bucket(void **p, void *end,
+				      struct crush_bucket_straw2 *b)
+{
+	int j;
+	dout("crush_decode_straw2_bucket %p to %p\n", *p, end);
+	b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
+	if (b->item_weights == NULL)
+		return -ENOMEM;
+	ceph_decode_need(p, end, b->h.size * sizeof(u32), bad);
+	for (j = 0; j < b->h.size; j++)
+		b->item_weights[j] = ceph_decode_32(p);
+	return 0;
+bad:
+	return -EINVAL;
+}
+
 static int skip_name_map(void **p, void *end)
 {
         int len;
@@ -204,6 +220,9 @@ static struct crush_map *crush_decode(void *pbyval, void *end)
 		case CRUSH_BUCKET_STRAW:
 			size = sizeof(struct crush_bucket_straw);
 			break;
+		case CRUSH_BUCKET_STRAW2:
+			size = sizeof(struct crush_bucket_straw2);
+			break;
 		default:
 			err = -EINVAL;
 			goto bad;
@@ -261,6 +280,12 @@ static struct crush_map *crush_decode(void *pbyval, void *end)
 			if (err < 0)
 				goto bad;
 			break;
+		case CRUSH_BUCKET_STRAW2:
+			err = crush_decode_straw2_bucket(p, end,
+				(struct crush_bucket_straw2 *)b);
+			if (err < 0)
+				goto bad;
+			break;
 		}
 	}
 
-- 
cgit v1.2.3


From 3f9400981691f6845e5c22b962500742b80a5484 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@poochiereds.net>
Date: Tue, 31 Mar 2015 12:03:28 -0400
Subject: sunrpc: make debugfs file creation failure non-fatal

v2: gracefully handle the case where some dentry pointers end up NULL
    and be more dilligent about zeroing out dentry pointers

We currently have a problem that SELinux policy is being enforced when
creating debugfs files. If a debugfs file is created as a side effect of
doing some syscall, then that creation can fail if the SELinux policy
for that process prevents it.

This seems wrong. We don't do that for files under /proc, for instance,
so Bruce has proposed a patch to fix that.

While discussing that patch however, Greg K.H. stated:

    "No kernel code should care / fail if a debugfs function fails, so
     please fix up the sunrpc code first."

This patch converts all of the sunrpc debugfs setup code to be void
return functins, and the callers to not look for errors from those
functions.

This should allow rpc_clnt and rpc_xprt creation to work, even if the
kernel fails to create debugfs files for some reason.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
 include/linux/sunrpc/debug.h | 18 +++++++--------
 net/sunrpc/clnt.c            |  4 +---
 net/sunrpc/debugfs.c         | 52 ++++++++++++++++++++++++--------------------
 net/sunrpc/sunrpc_syms.c     |  7 +-----
 net/sunrpc/xprt.c            |  7 +-----
 5 files changed, 41 insertions(+), 47 deletions(-)

(limited to 'net')

diff --git a/include/linux/sunrpc/debug.h b/include/linux/sunrpc/debug.h
index c57d8ea0716c..59a7889e15db 100644
--- a/include/linux/sunrpc/debug.h
+++ b/include/linux/sunrpc/debug.h
@@ -60,17 +60,17 @@ struct rpc_xprt;
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 void		rpc_register_sysctl(void);
 void		rpc_unregister_sysctl(void);
-int		sunrpc_debugfs_init(void);
+void		sunrpc_debugfs_init(void);
 void		sunrpc_debugfs_exit(void);
-int		rpc_clnt_debugfs_register(struct rpc_clnt *);
+void		rpc_clnt_debugfs_register(struct rpc_clnt *);
 void		rpc_clnt_debugfs_unregister(struct rpc_clnt *);
-int		rpc_xprt_debugfs_register(struct rpc_xprt *);
+void		rpc_xprt_debugfs_register(struct rpc_xprt *);
 void		rpc_xprt_debugfs_unregister(struct rpc_xprt *);
 #else
-static inline int
+static inline void
 sunrpc_debugfs_init(void)
 {
-	return 0;
+	return;
 }
 
 static inline void
@@ -79,10 +79,10 @@ sunrpc_debugfs_exit(void)
 	return;
 }
 
-static inline int
+static inline void
 rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 {
-	return 0;
+	return;
 }
 
 static inline void
@@ -91,10 +91,10 @@ rpc_clnt_debugfs_unregister(struct rpc_clnt *clnt)
 	return;
 }
 
-static inline int
+static inline void
 rpc_xprt_debugfs_register(struct rpc_xprt *xprt)
 {
-	return 0;
+	return;
 }
 
 static inline void
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 612aa73bbc60..e6ce1517367f 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -303,9 +303,7 @@ static int rpc_client_register(struct rpc_clnt *clnt,
 	struct super_block *pipefs_sb;
 	int err;
 
-	err = rpc_clnt_debugfs_register(clnt);
-	if (err)
-		return err;
+	rpc_clnt_debugfs_register(clnt);
 
 	pipefs_sb = rpc_get_sb_net(net);
 	if (pipefs_sb) {
diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
index e811f390f9f6..82962f7e6e88 100644
--- a/net/sunrpc/debugfs.c
+++ b/net/sunrpc/debugfs.c
@@ -129,48 +129,52 @@ static const struct file_operations tasks_fops = {
 	.release	= tasks_release,
 };
 
-int
+void
 rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 {
-	int len, err;
+	int len;
 	char name[24]; /* enough for "../../rpc_xprt/ + 8 hex digits + NULL */
+	struct rpc_xprt *xprt;
 
 	/* Already registered? */
-	if (clnt->cl_debugfs)
-		return 0;
+	if (clnt->cl_debugfs || !rpc_clnt_dir)
+		return;
 
 	len = snprintf(name, sizeof(name), "%x", clnt->cl_clid);
 	if (len >= sizeof(name))
-		return -EINVAL;
+		return;
 
 	/* make the per-client dir */
 	clnt->cl_debugfs = debugfs_create_dir(name, rpc_clnt_dir);
 	if (!clnt->cl_debugfs)
-		return -ENOMEM;
+		return;
 
 	/* make tasks file */
-	err = -ENOMEM;
 	if (!debugfs_create_file("tasks", S_IFREG | S_IRUSR, clnt->cl_debugfs,
 				 clnt, &tasks_fops))
 		goto out_err;
 
-	err = -EINVAL;
 	rcu_read_lock();
+	xprt = rcu_dereference(clnt->cl_xprt);
+	/* no "debugfs" dentry? Don't bother with the symlink. */
+	if (!xprt->debugfs) {
+		rcu_read_unlock();
+		return;
+	}
 	len = snprintf(name, sizeof(name), "../../rpc_xprt/%s",
-			rcu_dereference(clnt->cl_xprt)->debugfs->d_name.name);
+			xprt->debugfs->d_name.name);
 	rcu_read_unlock();
+
 	if (len >= sizeof(name))
 		goto out_err;
 
-	err = -ENOMEM;
 	if (!debugfs_create_symlink("xprt", clnt->cl_debugfs, name))
 		goto out_err;
 
-	return 0;
+	return;
 out_err:
 	debugfs_remove_recursive(clnt->cl_debugfs);
 	clnt->cl_debugfs = NULL;
-	return err;
 }
 
 void
@@ -226,33 +230,33 @@ static const struct file_operations xprt_info_fops = {
 	.release	= xprt_info_release,
 };
 
-int
+void
 rpc_xprt_debugfs_register(struct rpc_xprt *xprt)
 {
 	int len, id;
 	static atomic_t	cur_id;
 	char		name[9]; /* 8 hex digits + NULL term */
 
+	if (!rpc_xprt_dir)
+		return;
+
 	id = (unsigned int)atomic_inc_return(&cur_id);
 
 	len = snprintf(name, sizeof(name), "%x", id);
 	if (len >= sizeof(name))
-		return -EINVAL;
+		return;
 
 	/* make the per-client dir */
 	xprt->debugfs = debugfs_create_dir(name, rpc_xprt_dir);
 	if (!xprt->debugfs)
-		return -ENOMEM;
+		return;
 
 	/* make tasks file */
 	if (!debugfs_create_file("info", S_IFREG | S_IRUSR, xprt->debugfs,
 				 xprt, &xprt_info_fops)) {
 		debugfs_remove_recursive(xprt->debugfs);
 		xprt->debugfs = NULL;
-		return -ENOMEM;
 	}
-
-	return 0;
 }
 
 void
@@ -266,14 +270,17 @@ void __exit
 sunrpc_debugfs_exit(void)
 {
 	debugfs_remove_recursive(topdir);
+	topdir = NULL;
+	rpc_clnt_dir = NULL;
+	rpc_xprt_dir = NULL;
 }
 
-int __init
+void __init
 sunrpc_debugfs_init(void)
 {
 	topdir = debugfs_create_dir("sunrpc", NULL);
 	if (!topdir)
-		goto out;
+		return;
 
 	rpc_clnt_dir = debugfs_create_dir("rpc_clnt", topdir);
 	if (!rpc_clnt_dir)
@@ -283,10 +290,9 @@ sunrpc_debugfs_init(void)
 	if (!rpc_xprt_dir)
 		goto out_remove;
 
-	return 0;
+	return;
 out_remove:
 	debugfs_remove_recursive(topdir);
 	topdir = NULL;
-out:
-	return -ENOMEM;
+	rpc_clnt_dir = NULL;
 }
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index e37fbed87956..ee5d3d253102 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -98,10 +98,7 @@ init_sunrpc(void)
 	if (err)
 		goto out4;
 
-	err = sunrpc_debugfs_init();
-	if (err)
-		goto out5;
-
+	sunrpc_debugfs_init();
 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 	rpc_register_sysctl();
 #endif
@@ -109,8 +106,6 @@ init_sunrpc(void)
 	init_socket_xprt();	/* clnt sock transport */
 	return 0;
 
-out5:
-	unregister_rpc_pipefs();
 out4:
 	unregister_pernet_subsys(&sunrpc_net_ops);
 out3:
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index e3015aede0d9..9949722d99ce 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -1331,7 +1331,6 @@ static void xprt_init(struct rpc_xprt *xprt, struct net *net)
  */
 struct rpc_xprt *xprt_create_transport(struct xprt_create *args)
 {
-	int err;
 	struct rpc_xprt	*xprt;
 	struct xprt_class *t;
 
@@ -1372,11 +1371,7 @@ found:
 		return ERR_PTR(-ENOMEM);
 	}
 
-	err = rpc_xprt_debugfs_register(xprt);
-	if (err) {
-		xprt_destroy(xprt);
-		return ERR_PTR(err);
-	}
+	rpc_xprt_debugfs_register(xprt);
 
 	dprintk("RPC:       created transport %p with %u slots\n", xprt,
 			xprt->max_reqs);
-- 
cgit v1.2.3