[SCSI] SCSI and FC Transport: add netlink support for posting of transport events

This patch formally adds support for the posting of FC events via netlink.
It is a followup to the original RFC at:
  http://marc.theaimsgroup.com/?l=linux-scsi&m=114530667923464&w=2
and the initial posting at:
  http://marc.theaimsgroup.com/?l=linux-scsi&m=115507374832500&w=2

The patch has been updated to optimize the send path, per the discussions
in the initial posting.

Per discussions at the Storage Summit and at OLS, we are to use netlink for
async events from transports. Also per discussions, to avoid a netlink
protocol per transport, I've create a single NETLINK_SCSITRANSPORT protocol,
which can then be used by all transports.

This patch:
- Creates new files scsi_netlink.c and scsi_netlink.h, which contains the
  single and shared definitions for the SCSI Transport. It is tied into the
  base SCSI subsystem intialization.
  Contains a single interface routine, scsi_send_transport_event(), for a
  transport to send an event (via multicast to a protocol specific group).
- Creates a new scsi_netlink_fc.h file, which contains the FC netlink event
  messages
- Adds 3 new routines to the fc transport:
   fc_get_event_number() -  to get a FC event #
   fc_host_post_event()  -  to send a simple FC event (32 bits of data)
   fc_host_post_vendor_event() - to send a Vendor unique event, with
                                 arbitrary amounts of data.

   Note: the separation of event number allows for a LLD to send a standard
     event, followed by vendor-specific data for the event.

Note: This patch assumes 2 prior fc transport patches have been installed:
   http://marc.theaimsgroup.com/?l=linux-scsi&m=115555807316329&w=2
   http://marc.theaimsgroup.com/?l=linux-scsi&m=115581614930261&w=2

   Sorry - next time I'll do something like making these individual
   patches of the same posting when I know they'll be posted closely
   together.

Signed-off-by: James Smart <James.Smart@emulex.com>

Tidy up configuration not to make SCSI always select NET

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>

6 files changed, 419 insertions, 1 deletions

diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index c8c6065..4d1998d 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -27,6 +27,11 @@ config SCSI
 	  However, do not compile this as a module if your root file system
 	  (the one containing the directory /) is located on a SCSI device.
 
+config SCSI_NETLINK
+	tristate
+	default	n
+	select NET
+
 config SCSI_PROC_FS
 	bool "legacy /proc/scsi/ support"
 	depends on SCSI && PROC_FS
@@ -222,6 +227,7 @@ config SCSI_SPI_ATTRS
 config SCSI_FC_ATTRS
 	tristate "FiberChannel Transport Attributes"
 	depends on SCSI
+	select SCSI_NETLINK
 	help
 	  If you wish to export transport-specific information about
 	  each attached FiberChannel device to sysfs, say Y.
diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile
index fd9aeb1..8fc2c59 100644
--- a/drivers/scsi/Makefile
+++ b/drivers/scsi/Makefile
@@ -159,6 +159,7 @@ scsi_mod-y			+= scsi.o hosts.o scsi_ioctl.o constants.o \
 				   scsicam.o scsi_error.o scsi_lib.o \
 				   scsi_scan.o scsi_sysfs.o \
 				   scsi_devinfo.o
+scsi_mod-$(CONFIG_SCSI_NETLINK)	+= scsi_netlink.o
 scsi_mod-$(CONFIG_SYSCTL)	+= scsi_sysctl.o
 scsi_mod-$(CONFIG_SCSI_PROC_FS)	+= scsi_proc.o
 
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 3784392..eedfd05 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -1118,6 +1118,8 @@ static int __init init_scsi(void)
 	for_each_possible_cpu(i)
 		INIT_LIST_HEAD(&per_cpu(scsi_done_q, i));
 
+	scsi_netlink_init();
+
 	printk(KERN_NOTICE "SCSI subsystem initialized\n");
 	return 0;
 
@@ -1138,6 +1140,7 @@ cleanup_queue:
 
 static void __exit exit_scsi(void)
 {
+	scsi_netlink_exit();
 	scsi_sysfs_unregister();
 	scsi_exit_sysctl();
 	scsi_exit_hosts();
diff --git a/drivers/scsi/scsi_netlink.c b/drivers/scsi/scsi_netlink.c
new file mode 100644
index 0000000..1b59b27
--- /dev/null
+++ b/drivers/scsi/scsi_netlink.c
@@ -0,0 +1,199 @@
+/*
+ *  scsi_netlink.c  - SCSI Transport Netlink Interface
+ *
+ *  Copyright (C) 2006   James Smart, Emulex Corporation
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+#include <linux/time.h>
+#include <linux/jiffies.h>
+#include <linux/security.h>
+#include <net/sock.h>
+#include <net/netlink.h>
+
+#include <scsi/scsi_netlink.h>
+#include "scsi_priv.h"
+
+struct sock *scsi_nl_sock = NULL;
+EXPORT_SYMBOL_GPL(scsi_nl_sock);
+
+
+/**
+ * scsi_nl_rcv_msg -
+ *    Receive message handler. Extracts message from a receive buffer.
+ *    Validates message header and calls appropriate transport message handler
+ *
+ * @skb:		socket receive buffer
+ *
+ **/
+static void
+scsi_nl_rcv_msg(struct sk_buff *skb)
+{
+	struct nlmsghdr *nlh;
+	struct scsi_nl_hdr *hdr;
+	uint32_t rlen;
+	int err;
+
+	while (skb->len >= NLMSG_SPACE(0)) {
+		err = 0;
+
+		nlh = (struct nlmsghdr *) skb->data;
+		if ((nlh->nlmsg_len < (sizeof(*nlh) + sizeof(*hdr))) ||
+		    (skb->len < nlh->nlmsg_len)) {
+			printk(KERN_WARNING "%s: discarding partial skb\n",
+				 __FUNCTION__);
+			return;
+		}
+
+		rlen = NLMSG_ALIGN(nlh->nlmsg_len);
+		if (rlen > skb->len)
+			rlen = skb->len;
+
+		if (nlh->nlmsg_type != SCSI_TRANSPORT_MSG) {
+			err = -EBADMSG;
+			goto next_msg;
+		}
+
+		hdr = NLMSG_DATA(nlh);
+		if ((hdr->version != SCSI_NL_VERSION) ||
+		    (hdr->magic != SCSI_NL_MAGIC)) {
+			err = -EPROTOTYPE;
+			goto next_msg;
+		}
+
+		if (security_netlink_recv(skb, CAP_SYS_ADMIN)) {
+			err = -EPERM;
+			goto next_msg;
+		}
+
+		if (nlh->nlmsg_len < (sizeof(*nlh) + hdr->msglen)) {
+			printk(KERN_WARNING "%s: discarding partial message\n",
+				 __FUNCTION__);
+			return;
+		}
+
+		/*
+		 * We currently don't support anyone sending us a message
+		 */
+
+next_msg:
+		if ((err) || (nlh->nlmsg_flags & NLM_F_ACK))
+			netlink_ack(skb, nlh, err);
+
+		skb_pull(skb, rlen);
+	}
+}
+
+
+/**
+ * scsi_nl_rcv_msg -
+ *    Receive handler for a socket. Extracts a received message buffer from
+ *    the socket, and starts message processing.
+ *
+ * @sk:		socket
+ * @len:	unused
+ *
+ **/
+static void
+scsi_nl_rcv(struct sock *sk, int len)
+{
+	struct sk_buff *skb;
+
+	while ((skb = skb_dequeue(&sk->sk_receive_queue))) {
+		scsi_nl_rcv_msg(skb);
+		kfree_skb(skb);
+	}
+}
+
+
+/**
+ * scsi_nl_rcv_event -
+ *    Event handler for a netlink socket.
+ *
+ * @this:		event notifier block
+ * @event:		event type
+ * @ptr:		event payload
+ *
+ **/
+static int
+scsi_nl_rcv_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+	struct netlink_notify *n = ptr;
+
+	if (n->protocol != NETLINK_SCSITRANSPORT)
+		return NOTIFY_DONE;
+
+	/*
+	 * Currently, we are not tracking PID's, etc. There is nothing
+	 * to handle.
+	 */
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block scsi_netlink_notifier = {
+	.notifier_call  = scsi_nl_rcv_event,
+};
+
+
+/**
+ * scsi_netlink_init -
+ *    Called by SCSI subsystem to intialize the SCSI transport netlink
+ *    interface
+ *
+ **/
+void
+scsi_netlink_init(void)
+{
+	int error;
+
+	error = netlink_register_notifier(&scsi_netlink_notifier);
+	if (error) {
+		printk(KERN_ERR "%s: register of event handler failed - %d\n",
+				__FUNCTION__, error);
+		return;
+	}
+
+	scsi_nl_sock = netlink_kernel_create(NETLINK_SCSITRANSPORT,
+				SCSI_NL_GRP_CNT, scsi_nl_rcv, THIS_MODULE);
+	if (!scsi_nl_sock) {
+		printk(KERN_ERR "%s: register of recieve handler failed\n",
+				__FUNCTION__);
+		netlink_unregister_notifier(&scsi_netlink_notifier);
+	}
+
+	return;
+}
+
+
+/**
+ * scsi_netlink_exit -
+ *    Called by SCSI subsystem to disable the SCSI transport netlink
+ *    interface
+ *
+ **/
+void
+scsi_netlink_exit(void)
+{
+	if (scsi_nl_sock) {
+		sock_release(scsi_nl_sock->sk_socket);
+		netlink_unregister_notifier(&scsi_netlink_notifier);
+	}
+
+	return;
+}
+
+
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index ae24c85..5d023d4 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -8,6 +8,7 @@ struct scsi_cmnd;
 struct scsi_device;
 struct scsi_host_template;
 struct Scsi_Host;
+struct scsi_nl_hdr;
 
 
 /*
@@ -110,6 +111,16 @@ extern void __scsi_remove_device(struct scsi_device *);
 
 extern struct bus_type scsi_bus_type;
 
+/* scsi_netlink.c */
+#ifdef CONFIG_SCSI_NETLINK
+extern void scsi_netlink_init(void);
+extern void scsi_netlink_exit(void);
+extern struct sock *scsi_nl_sock;
+#else
+static inline void scsi_netlink_init(void) {}
+static inline void scsi_netlink_exit(void) {}
+#endif
+
 /* 
  * internal scsi timeout functions: for use by mid-layer and transport
  * classes.
diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 79d31ca..05989f1 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -32,6 +32,9 @@
 #include <scsi/scsi_transport.h>
 #include <scsi/scsi_transport_fc.h>
 #include <scsi/scsi_cmnd.h>
+#include <linux/netlink.h>
+#include <net/netlink.h>
+#include <scsi/scsi_netlink_fc.h>
 #include "scsi_priv.h"
 
 static int fc_queue_work(struct Scsi_Host *, struct work_struct *);
@@ -93,6 +96,29 @@ fc_enum_name_search(port_type, fc_port_type, fc_port_type_names)
 #define FC_PORTTYPE_MAX_NAMELEN		50
 
 
+/* Convert fc_host_event_code values to ascii string name */
+static const struct {
+	enum fc_host_event_code		value;
+	char				*name;
+} fc_host_event_code_names[] = {
+	{ FCH_EVT_LIP,			"lip" },
+	{ FCH_EVT_LINKUP,		"link_up" },
+	{ FCH_EVT_LINKDOWN,		"link_down" },
+	{ FCH_EVT_LIPRESET,		"lip_reset" },
+	{ FCH_EVT_RSCN,			"rscn" },
+	{ FCH_EVT_ADAPTER_CHANGE,	"adapter_chg" },
+	{ FCH_EVT_PORT_UNKNOWN,		"port_unknown" },
+	{ FCH_EVT_PORT_ONLINE,		"port_online" },
+	{ FCH_EVT_PORT_OFFLINE,		"port_offline" },
+	{ FCH_EVT_PORT_FABRIC,		"port_fabric" },
+	{ FCH_EVT_LINK_UNKNOWN,		"link_unknown" },
+	{ FCH_EVT_VENDOR_UNIQUE,	"vendor_unique" },
+};
+fc_enum_name_search(host_event_code, fc_host_event_code,
+		fc_host_event_code_names)
+#define FC_HOST_EVENT_CODE_MAX_NAMELEN	30
+
+
 /* Convert fc_port_state values to ascii string name */
 static struct {
 	enum fc_port_state	value;
@@ -377,10 +403,182 @@ MODULE_PARM_DESC(dev_loss_tmo,
 		 " exceeded, the scsi target is removed. Value should be"
 		 " between 1 and SCSI_DEVICE_BLOCK_MAX_TIMEOUT.");
 
+/**
+ * Netlink Infrastructure
+ **/
+
+static atomic_t fc_event_seq;
+
+/**
+ * fc_get_event_number - Obtain the next sequential FC event number
+ *
+ * Notes:
+ *   We could have inline'd this, but it would have required fc_event_seq to
+ *   be exposed. For now, live with the subroutine call.
+ *   Atomic used to avoid lock/unlock...
+ **/
+u32
+fc_get_event_number(void)
+{
+	return atomic_add_return(1, &fc_event_seq);
+}
+EXPORT_SYMBOL(fc_get_event_number);
+
+
+/**
+ * fc_host_post_event - called to post an even on an fc_host.
+ *
+ * @shost:		host the event occurred on
+ * @event_number:	fc event number obtained from get_fc_event_number()
+ * @event_code:		fc_host event being posted
+ * @event_data:		32bits of data for the event being posted
+ *
+ * Notes:
+ *	This routine assumes no locks are held on entry.
+ **/
+void
+fc_host_post_event(struct Scsi_Host *shost, u32 event_number,
+		enum fc_host_event_code event_code, u32 event_data)
+{
+	struct sk_buff *skb;
+	struct nlmsghdr	*nlh;
+	struct fc_nl_event *event;
+	const char *name;
+	u32 len, skblen;
+	int err;
+
+	if (!scsi_nl_sock) {
+		err = -ENOENT;
+		goto send_fail;
+	}
+
+	len = FC_NL_MSGALIGN(sizeof(*event));
+	skblen = NLMSG_SPACE(len);
+
+	skb = alloc_skb(skblen, GFP_KERNEL);
+	if (!skb) {
+		err = -ENOBUFS;
+		goto send_fail;
+	}
+
+	nlh = nlmsg_put(skb, 0, 0, SCSI_TRANSPORT_MSG,
+				skblen - sizeof(*nlh), 0);
+	if (!nlh) {
+		err = -ENOBUFS;
+		goto send_fail_skb;
+	}
+	event = NLMSG_DATA(nlh);
+
+	INIT_SCSI_NL_HDR(&event->snlh, SCSI_NL_TRANSPORT_FC,
+				FC_NL_ASYNC_EVENT, len);
+	event->seconds = get_seconds();
+	event->vendor_id = 0;
+	event->host_no = shost->host_no;
+	event->event_datalen = sizeof(u32);	/* bytes */
+	event->event_num = event_number;
+	event->event_code = event_code;
+	event->event_data = event_data;
+
+	err = nlmsg_multicast(scsi_nl_sock, skb, 0, SCSI_NL_GRP_FC_EVENTS);
+	if (err && (err != -ESRCH))	/* filter no recipient errors */
+		/* nlmsg_multicast already kfree_skb'd */
+		goto send_fail;
+
+	return;
+
+send_fail_skb:
+	kfree_skb(skb);
+send_fail:
+	name = get_fc_host_event_code_name(event_code);
+	printk(KERN_WARNING
+		"%s: Dropped Event : host %d %s data 0x%08x - err %d\n",
+		__FUNCTION__, shost->host_no,
+		(name) ? name : "<unknown>", event_data, err);
+	return;
+}
+EXPORT_SYMBOL(fc_host_post_event);
+
+
+/**
+ * fc_host_post_vendor_event - called to post a vendor unique event on
+ *                             a fc_host
+ *
+ * @shost:		host the event occurred on
+ * @event_number:	fc event number obtained from get_fc_event_number()
+ * @data_len:		amount, in bytes, of vendor unique data
+ * @data_buf:		pointer to vendor unique data
+ *
+ * Notes:
+ *	This routine assumes no locks are held on entry.
+ **/
+void
+fc_host_post_vendor_event(struct Scsi_Host *shost, u32 event_number,
+		u32 data_len, char * data_buf, u32 vendor_id)
+{
+	struct sk_buff *skb;
+	struct nlmsghdr	*nlh;
+	struct fc_nl_event *event;
+	u32 len, skblen;
+	int err;
+
+	if (!scsi_nl_sock) {
+		err = -ENOENT;
+		goto send_vendor_fail;
+	}
+
+	len = FC_NL_MSGALIGN(sizeof(*event) + data_len);
+	skblen = NLMSG_SPACE(len);
+
+	skb = alloc_skb(skblen, GFP_KERNEL);
+	if (!skb) {
+		err = -ENOBUFS;
+		goto send_vendor_fail;
+	}
+
+	nlh = nlmsg_put(skb, 0, 0, SCSI_TRANSPORT_MSG,
+				skblen - sizeof(*nlh), 0);
+	if (!nlh) {
+		err = -ENOBUFS;
+		goto send_vendor_fail_skb;
+	}
+	event = NLMSG_DATA(nlh);
+
+	INIT_SCSI_NL_HDR(&event->snlh, SCSI_NL_TRANSPORT_FC,
+				FC_NL_ASYNC_EVENT, len);
+	event->seconds = get_seconds();
+	event->vendor_id = vendor_id;
+	event->host_no = shost->host_no;
+	event->event_datalen = data_len;	/* bytes */
+	event->event_num = event_number;
+	event->event_code = FCH_EVT_VENDOR_UNIQUE;
+	memcpy(&event->event_data, data_buf, data_len);
+
+	err = nlmsg_multicast(scsi_nl_sock, skb, 0, SCSI_NL_GRP_FC_EVENTS);
+	if (err && (err != -ESRCH))	/* filter no recipient errors */
+		/* nlmsg_multicast already kfree_skb'd */
+		goto send_vendor_fail;
+
+	return;
+
+send_vendor_fail_skb:
+	kfree_skb(skb);
+send_vendor_fail:
+	printk(KERN_WARNING
+		"%s: Dropped Event : host %d vendor_unique - err %d\n",
+		__FUNCTION__, shost->host_no, err);
+	return;
+}
+EXPORT_SYMBOL(fc_host_post_vendor_event);
+
+
 
 static __init int fc_transport_init(void)
 {
-	int error = transport_class_register(&fc_host_class);
+	int error;
+
+	atomic_set(&fc_event_seq, 0);
+
+	error = transport_class_register(&fc_host_class);
 	if (error)
 		return error;
 	error = transport_class_register(&fc_rport_class);