[PATCH v2 0/3] Commonise address any check

[PATCH v2 0/3] Commonise address any check

[Sagi Grimberg]

Both iscsit and nvmet need to check the listening endpoint
portal is not INADDR_ANY (or IN6ADDR_ANY_INIT) when replying
to a discovery information query (iscsi sendtaragets or nvmet
discovery log pages). If indeed the nvmet port or iscsit portal
are listening on any address, they need to return the source
address of the socket that carried the query itslef.

iscsit already does this while nvmet did not.
instead of copying the inaddr_any check, place it in net-utils
and call it from both (patches 1+3).

In addition, fix the behavior in nvmet (patch 2).

Changes from v1:
- small alignment fix in inet_addr_is_any
- warn if got unexpected address family
- make in6_any static

Sagi Grimberg (3):
  net/utils: Introduce inet_addr_is_any
  nvmet: don't return "any" ip address in discovery log page
  iscsi-target: use common inet_addr_is_any

 drivers/nvme/target/discovery.c     | 30 +++++++++++++++++++++++++++---
 drivers/nvme/target/nvmet.h         |  2 ++
 drivers/nvme/target/rdma.c          | 18 ++++++++++++++++++
 drivers/target/iscsi/iscsi_target.c | 28 ++--------------------------
 include/linux/inet.h                |  1 +
 net/core/utils.c                    | 23 +++++++++++++++++++++++
 6 files changed, 73 insertions(+), 29 deletions(-)

-- 
2.14.1

[PATCH v2 0/3] Commonise address any check

[Sagi Grimberg]

Both iscsit and nvmet need to check the listening endpoint
portal is not INADDR_ANY (or IN6ADDR_ANY_INIT) when replying
to a discovery information query (iscsi sendtaragets or nvmet
discovery log pages). If indeed the nvmet port or iscsit portal
are listening on any address, they need to return the source
address of the socket that carried the query itslef.

iscsit already does this while nvmet did not.
instead of copying the inaddr_any check, place it in net-utils
and call it from both (patches 1+3).

In addition, fix the behavior in nvmet (patch 2).

Changes from v1:
- small alignment fix in inet_addr_is_any
- warn if got unexpected address family
- make in6_any static

Sagi Grimberg (3):
  net/utils: Introduce inet_addr_is_any
  nvmet: don't return "any" ip address in discovery log page
  iscsi-target: use common inet_addr_is_any

 drivers/nvme/target/discovery.c     | 30 +++++++++++++++++++++++++++---
 drivers/nvme/target/nvmet.h         |  2 ++
 drivers/nvme/target/rdma.c          | 18 ++++++++++++++++++
 drivers/target/iscsi/iscsi_target.c | 28 ++--------------------------
 include/linux/inet.h                |  1 +
 net/core/utils.c                    | 23 +++++++++++++++++++++++
 6 files changed, 73 insertions(+), 29 deletions(-)

-- 
2.14.1

[PATCH v2 1/3] net/utils: Introduce inet_addr_is_any

[Sagi Grimberg]

Can be useful to check INET_ANY address for both ipv4/ipv6 addresses.

Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
---
 include/linux/inet.h |  1 +
 net/core/utils.c     | 23 +++++++++++++++++++++++
 2 files changed, 24 insertions(+)

diff --git a/include/linux/inet.h b/include/linux/inet.h
index 636ebe87e6f8..97defc1139e9 100644
--- a/include/linux/inet.h
+++ b/include/linux/inet.h
@@ -59,5 +59,6 @@ extern int in6_pton(const char *src, int srclen, u8 *dst, int delim, const char
 
 extern int inet_pton_with_scope(struct net *net, unsigned short af,
 		const char *src, const char *port, struct sockaddr_storage *addr);
+extern bool inet_addr_is_any(struct sockaddr *addr);
 
 #endif	/* _LINUX_INET_H */
diff --git a/net/core/utils.c b/net/core/utils.c
index 93066bd0305a..6e331a7904e7 100644
--- a/net/core/utils.c
+++ b/net/core/utils.c
@@ -403,6 +403,29 @@ int inet_pton_with_scope(struct net *net, __kernel_sa_family_t af,
 }
 EXPORT_SYMBOL(inet_pton_with_scope);
 
+bool inet_addr_is_any(struct sockaddr *addr)
+{
+	if (addr->sa_family = AF_INET6) {
+		struct sockaddr_in6 *in6 = (struct sockaddr_in6 *)addr;
+		static const struct sockaddr_in6 in6_any +			{ .sin6_addr = IN6ADDR_ANY_INIT };
+
+		if (!memcmp(in6->sin6_addr.s6_addr,
+			    in6_any.sin6_addr.s6_addr, 16))
+			return true;
+	} else if (addr->sa_family = AF_INET) {
+		struct sockaddr_in *in = (struct sockaddr_in *)addr;
+
+		if (in->sin_addr.s_addr = htonl(INADDR_ANY))
+			return true;
+	} else {
+		pr_warn("unexpected address family %u\n", addr->sa_family);
+	}
+
+	return false;
+}
+EXPORT_SYMBOL(inet_addr_is_any);
+
 void inet_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb,
 			      __be32 from, __be32 to, bool pseudohdr)
 {
-- 
2.14.1

[PATCH v2 1/3] net/utils: Introduce inet_addr_is_any

[Sagi Grimberg]

Can be useful to check INET_ANY address for both ipv4/ipv6 addresses.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
Cc: "David S. Miller" <davem at davemloft.net>
Cc: netdev at vger.kernel.org
---
 include/linux/inet.h |  1 +
 net/core/utils.c     | 23 +++++++++++++++++++++++
 2 files changed, 24 insertions(+)

diff --git a/include/linux/inet.h b/include/linux/inet.h
index 636ebe87e6f8..97defc1139e9 100644
--- a/include/linux/inet.h
+++ b/include/linux/inet.h
@@ -59,5 +59,6 @@ extern int in6_pton(const char *src, int srclen, u8 *dst, int delim, const char
 
 extern int inet_pton_with_scope(struct net *net, unsigned short af,
 		const char *src, const char *port, struct sockaddr_storage *addr);
+extern bool inet_addr_is_any(struct sockaddr *addr);
 
 #endif	/* _LINUX_INET_H */
diff --git a/net/core/utils.c b/net/core/utils.c
index 93066bd0305a..6e331a7904e7 100644
--- a/net/core/utils.c
+++ b/net/core/utils.c
@@ -403,6 +403,29 @@ int inet_pton_with_scope(struct net *net, __kernel_sa_family_t af,
 }
 EXPORT_SYMBOL(inet_pton_with_scope);
 
+bool inet_addr_is_any(struct sockaddr *addr)
+{
+	if (addr->sa_family == AF_INET6) {
+		struct sockaddr_in6 *in6 = (struct sockaddr_in6 *)addr;
+		static const struct sockaddr_in6 in6_any =
+			{ .sin6_addr = IN6ADDR_ANY_INIT };
+
+		if (!memcmp(in6->sin6_addr.s6_addr,
+			    in6_any.sin6_addr.s6_addr, 16))
+			return true;
+	} else if (addr->sa_family == AF_INET) {
+		struct sockaddr_in *in = (struct sockaddr_in *)addr;
+
+		if (in->sin_addr.s_addr == htonl(INADDR_ANY))
+			return true;
+	} else {
+		pr_warn("unexpected address family %u\n", addr->sa_family);
+	}
+
+	return false;
+}
+EXPORT_SYMBOL(inet_addr_is_any);
+
 void inet_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb,
 			      __be32 from, __be32 to, bool pseudohdr)
 {
-- 
2.14.1

[PATCH v2 2/3] nvmet: don't return "any" ip address in discovery log page

[Sagi Grimberg]

Its perfectly valid to assign a nvmet port to listen on "any"
IP address (traddr 0.0.0.0 for ipv4 address family) for IP based
transport ports. However, we must not return this address in
discovery log entries. Instead we need to return the address
where the request was accepted on (req->port address).

Since this is nvme transport specific, introduce an optional
.disc_traddr interface that is designed to check that a
port in question is bound to "any" IP address and if so, set
the traddr from the port where the request came from.

Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
---
 drivers/nvme/target/discovery.c | 30 +++++++++++++++++++++++++++---
 drivers/nvme/target/nvmet.h     |  2 ++
 drivers/nvme/target/rdma.c      | 18 ++++++++++++++++++
 3 files changed, 47 insertions(+), 3 deletions(-)

diff --git a/drivers/nvme/target/discovery.c b/drivers/nvme/target/discovery.c
index 8f3b57b4c97b..a72425d8bce0 100644
--- a/drivers/nvme/target/discovery.c
+++ b/drivers/nvme/target/discovery.c
@@ -43,7 +43,8 @@ void nvmet_referral_disable(struct nvmet_port *port)
 }
 
 static void nvmet_format_discovery_entry(struct nvmf_disc_rsp_page_hdr *hdr,
-		struct nvmet_port *port, char *subsys_nqn, u8 type, u32 numrec)
+		struct nvmet_port *port, char *subsys_nqn, char *traddr,
+		u8 type, u32 numrec)
 {
 	struct nvmf_disc_rsp_page_entry *e = &hdr->entries[numrec];
 
@@ -56,11 +57,30 @@ static void nvmet_format_discovery_entry(struct nvmf_disc_rsp_page_hdr *hdr,
 	e->asqsz = cpu_to_le16(NVME_AQ_DEPTH);
 	e->subtype = type;
 	memcpy(e->trsvcid, port->disc_addr.trsvcid, NVMF_TRSVCID_SIZE);
-	memcpy(e->traddr, port->disc_addr.traddr, NVMF_TRADDR_SIZE);
+	memcpy(e->traddr, traddr, NVMF_TRADDR_SIZE);
 	memcpy(e->tsas.common, port->disc_addr.tsas.common, NVMF_TSAS_SIZE);
 	memcpy(e->subnqn, subsys_nqn, NVMF_NQN_SIZE);
 }
 
+/*
+ * nvmet_set_disc_traddr - set a correct discovery log entry traddr
+ *
+ * IP based transports (e.g RDMA) can listen on "any" ipv4/ipv6 addresses
+ * (INADDR_ANY or IN6ADDR_ANY_INIT). The discovery log page traddr reply
+ * must not contain that "any" IP address. If the transport implements
+ * .disc_traddr, use it. this callback will set the discovery traddr
+ * from the req->port address in case the port in question listens
+ * "any" IP address.
+ */
+static void nvmet_set_disc_traddr(struct nvmet_req *req, struct nvmet_port *port,
+		char *traddr)
+{
+	if (req->ops->disc_traddr)
+		req->ops->disc_traddr(req, port, traddr);
+	else
+		memcpy(traddr, port->disc_addr.traddr, NVMF_TRADDR_SIZE);
+}
+
 static void nvmet_execute_get_disc_log_page(struct nvmet_req *req)
 {
 	const int entry_size = sizeof(struct nvmf_disc_rsp_page_entry);
@@ -90,8 +110,11 @@ static void nvmet_execute_get_disc_log_page(struct nvmet_req *req)
 		if (!nvmet_host_allowed(req, p->subsys, ctrl->hostnqn))
 			continue;
 		if (residual_len >= entry_size) {
+			char traddr[NVMF_TRADDR_SIZE];
+
+			nvmet_set_disc_traddr(req, req->port, traddr);
 			nvmet_format_discovery_entry(hdr, req->port,
-					p->subsys->subsysnqn,
+					p->subsys->subsysnqn, traddr,
 					NVME_NQN_NVME, numrec);
 			residual_len -= entry_size;
 		}
@@ -102,6 +125,7 @@ static void nvmet_execute_get_disc_log_page(struct nvmet_req *req)
 		if (residual_len >= entry_size) {
 			nvmet_format_discovery_entry(hdr, r,
 					NVME_DISC_SUBSYS_NAME,
+					r->disc_addr.traddr,
 					NVME_NQN_DISC, numrec);
 			residual_len -= entry_size;
 		}
diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h
index 178fd9a1d128..a6e191ff34ba 100644
--- a/drivers/nvme/target/nvmet.h
+++ b/drivers/nvme/target/nvmet.h
@@ -215,6 +215,8 @@ struct nvmet_fabrics_ops {
 	void (*remove_port)(struct nvmet_port *port);
 	void (*delete_ctrl)(struct nvmet_ctrl *ctrl);
 	int (*install_queue)(struct nvmet_sq *nvme_sq);
+	void (*disc_traddr)(struct nvmet_req *req,
+			struct nvmet_port *port, char *traddr);
 };
 
 #define NVMET_MAX_INLINE_BIOVEC	8
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
index 978e169c11bf..d7831372e1f9 100644
--- a/drivers/nvme/target/rdma.c
+++ b/drivers/nvme/target/rdma.c
@@ -1445,6 +1445,23 @@ static void nvmet_rdma_remove_port(struct nvmet_port *port)
 		rdma_destroy_id(cm_id);
 }
 
+static void nvmet_rdma_disc_port_addr(struct nvmet_req *req,
+		struct nvmet_port *port, char *traddr)
+{
+	struct rdma_cm_id *cm_id = port->priv;
+
+	if (inet_addr_is_any((struct sockaddr *)&cm_id->route.addr.src_addr)) {
+		struct nvmet_rdma_rsp *rsp +			container_of(req, struct nvmet_rdma_rsp, req);
+		struct rdma_cm_id *req_cm_id = rsp->queue->cm_id;
+		struct sockaddr *addr = (void *)&req_cm_id->route.addr.src_addr;
+
+		sprintf(traddr, "%pISc", addr);
+	} else {
+		memcpy(traddr, port->disc_addr.traddr, NVMF_TRADDR_SIZE);
+	}
+}
+
 static struct nvmet_fabrics_ops nvmet_rdma_ops = {
 	.owner			= THIS_MODULE,
 	.type			= NVMF_TRTYPE_RDMA,
@@ -1455,6 +1472,7 @@ static struct nvmet_fabrics_ops nvmet_rdma_ops = {
 	.remove_port		= nvmet_rdma_remove_port,
 	.queue_response		= nvmet_rdma_queue_response,
 	.delete_ctrl		= nvmet_rdma_delete_ctrl,
+	.disc_traddr		= nvmet_rdma_disc_port_addr,
 };
 
 static void nvmet_rdma_remove_one(struct ib_device *ib_device, void *client_data)
-- 
2.14.1

[PATCH v2 2/3] nvmet: don't return "any" ip address in discovery log page

[Sagi Grimberg]

Its perfectly valid to assign a nvmet port to listen on "any"
IP address (traddr 0.0.0.0 for ipv4 address family) for IP based
transport ports. However, we must not return this address in
discovery log entries. Instead we need to return the address
where the request was accepted on (req->port address).

Since this is nvme transport specific, introduce an optional
.disc_traddr interface that is designed to check that a
port in question is bound to "any" IP address and if so, set
the traddr from the port where the request came from.

Reviewed-by: Johannes Thumshirn <jthumshirn at suse.de>
Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/target/discovery.c | 30 +++++++++++++++++++++++++++---
 drivers/nvme/target/nvmet.h     |  2 ++
 drivers/nvme/target/rdma.c      | 18 ++++++++++++++++++
 3 files changed, 47 insertions(+), 3 deletions(-)

diff --git a/drivers/nvme/target/discovery.c b/drivers/nvme/target/discovery.c
index 8f3b57b4c97b..a72425d8bce0 100644
--- a/drivers/nvme/target/discovery.c
+++ b/drivers/nvme/target/discovery.c
@@ -43,7 +43,8 @@ void nvmet_referral_disable(struct nvmet_port *port)
 }
 
 static void nvmet_format_discovery_entry(struct nvmf_disc_rsp_page_hdr *hdr,
-		struct nvmet_port *port, char *subsys_nqn, u8 type, u32 numrec)
+		struct nvmet_port *port, char *subsys_nqn, char *traddr,
+		u8 type, u32 numrec)
 {
 	struct nvmf_disc_rsp_page_entry *e = &hdr->entries[numrec];
 
@@ -56,11 +57,30 @@ static void nvmet_format_discovery_entry(struct nvmf_disc_rsp_page_hdr *hdr,
 	e->asqsz = cpu_to_le16(NVME_AQ_DEPTH);
 	e->subtype = type;
 	memcpy(e->trsvcid, port->disc_addr.trsvcid, NVMF_TRSVCID_SIZE);
-	memcpy(e->traddr, port->disc_addr.traddr, NVMF_TRADDR_SIZE);
+	memcpy(e->traddr, traddr, NVMF_TRADDR_SIZE);
 	memcpy(e->tsas.common, port->disc_addr.tsas.common, NVMF_TSAS_SIZE);
 	memcpy(e->subnqn, subsys_nqn, NVMF_NQN_SIZE);
 }
 
+/*
+ * nvmet_set_disc_traddr - set a correct discovery log entry traddr
+ *
+ * IP based transports (e.g RDMA) can listen on "any" ipv4/ipv6 addresses
+ * (INADDR_ANY or IN6ADDR_ANY_INIT). The discovery log page traddr reply
+ * must not contain that "any" IP address. If the transport implements
+ * .disc_traddr, use it. this callback will set the discovery traddr
+ * from the req->port address in case the port in question listens
+ * "any" IP address.
+ */
+static void nvmet_set_disc_traddr(struct nvmet_req *req, struct nvmet_port *port,
+		char *traddr)
+{
+	if (req->ops->disc_traddr)
+		req->ops->disc_traddr(req, port, traddr);
+	else
+		memcpy(traddr, port->disc_addr.traddr, NVMF_TRADDR_SIZE);
+}
+
 static void nvmet_execute_get_disc_log_page(struct nvmet_req *req)
 {
 	const int entry_size = sizeof(struct nvmf_disc_rsp_page_entry);
@@ -90,8 +110,11 @@ static void nvmet_execute_get_disc_log_page(struct nvmet_req *req)
 		if (!nvmet_host_allowed(req, p->subsys, ctrl->hostnqn))
 			continue;
 		if (residual_len >= entry_size) {
+			char traddr[NVMF_TRADDR_SIZE];
+
+			nvmet_set_disc_traddr(req, req->port, traddr);
 			nvmet_format_discovery_entry(hdr, req->port,
-					p->subsys->subsysnqn,
+					p->subsys->subsysnqn, traddr,
 					NVME_NQN_NVME, numrec);
 			residual_len -= entry_size;
 		}
@@ -102,6 +125,7 @@ static void nvmet_execute_get_disc_log_page(struct nvmet_req *req)
 		if (residual_len >= entry_size) {
 			nvmet_format_discovery_entry(hdr, r,
 					NVME_DISC_SUBSYS_NAME,
+					r->disc_addr.traddr,
 					NVME_NQN_DISC, numrec);
 			residual_len -= entry_size;
 		}
diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h
index 178fd9a1d128..a6e191ff34ba 100644
--- a/drivers/nvme/target/nvmet.h
+++ b/drivers/nvme/target/nvmet.h
@@ -215,6 +215,8 @@ struct nvmet_fabrics_ops {
 	void (*remove_port)(struct nvmet_port *port);
 	void (*delete_ctrl)(struct nvmet_ctrl *ctrl);
 	int (*install_queue)(struct nvmet_sq *nvme_sq);
+	void (*disc_traddr)(struct nvmet_req *req,
+			struct nvmet_port *port, char *traddr);
 };
 
 #define NVMET_MAX_INLINE_BIOVEC	8
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
index 978e169c11bf..d7831372e1f9 100644
--- a/drivers/nvme/target/rdma.c
+++ b/drivers/nvme/target/rdma.c
@@ -1445,6 +1445,23 @@ static void nvmet_rdma_remove_port(struct nvmet_port *port)
 		rdma_destroy_id(cm_id);
 }
 
+static void nvmet_rdma_disc_port_addr(struct nvmet_req *req,
+		struct nvmet_port *port, char *traddr)
+{
+	struct rdma_cm_id *cm_id = port->priv;
+
+	if (inet_addr_is_any((struct sockaddr *)&cm_id->route.addr.src_addr)) {
+		struct nvmet_rdma_rsp *rsp =
+			container_of(req, struct nvmet_rdma_rsp, req);
+		struct rdma_cm_id *req_cm_id = rsp->queue->cm_id;
+		struct sockaddr *addr = (void *)&req_cm_id->route.addr.src_addr;
+
+		sprintf(traddr, "%pISc", addr);
+	} else {
+		memcpy(traddr, port->disc_addr.traddr, NVMF_TRADDR_SIZE);
+	}
+}
+
 static struct nvmet_fabrics_ops nvmet_rdma_ops = {
 	.owner			= THIS_MODULE,
 	.type			= NVMF_TRTYPE_RDMA,
@@ -1455,6 +1472,7 @@ static struct nvmet_fabrics_ops nvmet_rdma_ops = {
 	.remove_port		= nvmet_rdma_remove_port,
 	.queue_response		= nvmet_rdma_queue_response,
 	.delete_ctrl		= nvmet_rdma_delete_ctrl,
+	.disc_traddr		= nvmet_rdma_disc_port_addr,
 };
 
 static void nvmet_rdma_remove_one(struct ib_device *ib_device, void *client_data)
-- 
2.14.1

[PATCH v2 3/3] iscsi-target: use common inet_addr_is_any

[Sagi Grimberg]

Instead of open-coding it.

Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
Cc: target-devel@vger.kernel.org
---
 drivers/target/iscsi/iscsi_target.c | 28 ++--------------------------
 1 file changed, 2 insertions(+), 26 deletions(-)

diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index 9eb10d34682c..8e223799347a 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -26,6 +26,7 @@
 #include <linux/delay.h>
 #include <linux/sched/signal.h>
 #include <asm/unaligned.h>
+#include <linux/inet.h>
 #include <net/ipv6.h>
 #include <scsi/scsi_proto.h>
 #include <scsi/iscsi_proto.h>
@@ -3291,30 +3292,6 @@ iscsit_send_task_mgt_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn)
 	return conn->conn_transport->iscsit_xmit_pdu(conn, cmd, NULL, NULL, 0);
 }
 
-static bool iscsit_check_inaddr_any(struct iscsi_np *np)
-{
-	bool ret = false;
-
-	if (np->np_sockaddr.ss_family = AF_INET6) {
-		const struct sockaddr_in6 sin6 = {
-			.sin6_addr = IN6ADDR_ANY_INIT };
-		struct sockaddr_in6 *sock_in6 -			 (struct sockaddr_in6 *)&np->np_sockaddr;
-
-		if (!memcmp(sock_in6->sin6_addr.s6_addr,
-				sin6.sin6_addr.s6_addr, 16))
-			ret = true;
-	} else {
-		struct sockaddr_in * sock_in -			(struct sockaddr_in *)&np->np_sockaddr;
-
-		if (sock_in->sin_addr.s_addr = htonl(INADDR_ANY))
-			ret = true;
-	}
-
-	return ret;
-}
-
 #define SENDTARGETS_BUF_LIMIT 32768U
 
 static int
@@ -3393,7 +3370,6 @@ iscsit_build_sendtargets_response(struct iscsi_cmd *cmd,
 			list_for_each_entry(tpg_np, &tpg->tpg_gnp_list,
 						tpg_np_list) {
 				struct iscsi_np *np = tpg_np->tpg_np;
-				bool inaddr_any = iscsit_check_inaddr_any(np);
 				struct sockaddr_storage *sockaddr;
 
 				if (np->np_network_transport != network_transport)
@@ -3422,7 +3398,7 @@ iscsit_build_sendtargets_response(struct iscsi_cmd *cmd,
 					}
 				}
 
-				if (inaddr_any)
+				if (inet_addr_is_any((struct sockaddr *)&np->np_sockaddr))
 					sockaddr = &conn->local_sockaddr;
 				else
 					sockaddr = &np->np_sockaddr;
-- 
2.14.1

[PATCH v2 3/3] iscsi-target: use common inet_addr_is_any

[Sagi Grimberg]

Instead of open-coding it.

Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
Cc: "Nicholas A. Bellinger" <nab at linux-iscsi.org>
Cc: target-devel at vger.kernel.org
---
 drivers/target/iscsi/iscsi_target.c | 28 ++--------------------------
 1 file changed, 2 insertions(+), 26 deletions(-)

diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index 9eb10d34682c..8e223799347a 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -26,6 +26,7 @@
 #include <linux/delay.h>
 #include <linux/sched/signal.h>
 #include <asm/unaligned.h>
+#include <linux/inet.h>
 #include <net/ipv6.h>
 #include <scsi/scsi_proto.h>
 #include <scsi/iscsi_proto.h>
@@ -3291,30 +3292,6 @@ iscsit_send_task_mgt_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn)
 	return conn->conn_transport->iscsit_xmit_pdu(conn, cmd, NULL, NULL, 0);
 }
 
-static bool iscsit_check_inaddr_any(struct iscsi_np *np)
-{
-	bool ret = false;
-
-	if (np->np_sockaddr.ss_family == AF_INET6) {
-		const struct sockaddr_in6 sin6 = {
-			.sin6_addr = IN6ADDR_ANY_INIT };
-		struct sockaddr_in6 *sock_in6 =
-			 (struct sockaddr_in6 *)&np->np_sockaddr;
-
-		if (!memcmp(sock_in6->sin6_addr.s6_addr,
-				sin6.sin6_addr.s6_addr, 16))
-			ret = true;
-	} else {
-		struct sockaddr_in * sock_in =
-			(struct sockaddr_in *)&np->np_sockaddr;
-
-		if (sock_in->sin_addr.s_addr == htonl(INADDR_ANY))
-			ret = true;
-	}
-
-	return ret;
-}
-
 #define SENDTARGETS_BUF_LIMIT 32768U
 
 static int
@@ -3393,7 +3370,6 @@ iscsit_build_sendtargets_response(struct iscsi_cmd *cmd,
 			list_for_each_entry(tpg_np, &tpg->tpg_gnp_list,
 						tpg_np_list) {
 				struct iscsi_np *np = tpg_np->tpg_np;
-				bool inaddr_any = iscsit_check_inaddr_any(np);
 				struct sockaddr_storage *sockaddr;
 
 				if (np->np_network_transport != network_transport)
@@ -3422,7 +3398,7 @@ iscsit_build_sendtargets_response(struct iscsi_cmd *cmd,
 					}
 				}
 
-				if (inaddr_any)
+				if (inet_addr_is_any((struct sockaddr *)&np->np_sockaddr))
 					sockaddr = &conn->local_sockaddr;
 				else
 					sockaddr = &np->np_sockaddr;
-- 
2.14.1

Re: [PATCH v2 1/3] net/utils: Introduce inet_addr_is_any

[Bart Van Assche]

T24gV2VkLCAyMDE4LTAyLTA3IGF0IDE1OjA1ICswMjAwLCBTYWdpIEdyaW1iZXJnIHdyb3RlOg0K
PiBDYW4gYmUgdXNlZnVsIHRvIGNoZWNrIElORVRfQU5ZIGFkZHJlc3MgZm9yIGJvdGggaXB2NC9p
cHY2IGFkZHJlc3Nlcy4NCg0KUmV2aWV3ZWQtYnk6IEJhcnQgVmFuIEFzc2NoZSA8YmFydC52YW5h
c3NjaGVAd2RjLmNvbT4NCg0KDQoNCg0K

[PATCH v2 1/3] net/utils: Introduce inet_addr_is_any

[Bart Van Assche]

On Wed, 2018-02-07@15:05 +0200, Sagi Grimberg wrote:
> Can be useful to check INET_ANY address for both ipv4/ipv6 addresses.

Reviewed-by: Bart Van Assche <bart.vanassche at wdc.com>

Re: [PATCH v2 3/3] iscsi-target: use common inet_addr_is_any

[Bart Van Assche]

On 02/07/18 05:05, Sagi Grimberg wrote:
> Instead of open-coding it.

Reviewed-by: Bart Van Assche <bart.vanassche@wdc.com>

[PATCH v2 3/3] iscsi-target: use common inet_addr_is_any

[Bart Van Assche]

On 02/07/18 05:05, Sagi Grimberg wrote:
> Instead of open-coding it.

Reviewed-by: Bart Van Assche <bart.vanassche at wdc.com>

Re: [PATCH v2 3/3] iscsi-target: use common inet_addr_is_any

[Johannes Thumshirn]

Looks good,
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn@suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

[PATCH v2 3/3] iscsi-target: use common inet_addr_is_any

[Johannes Thumshirn]

Looks good,
Reviewed-by: Johannes Thumshirn <jthumshirn at suse.de>
-- 
Johannes Thumshirn                                          Storage
jthumshirn at suse.de                                +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 N?rnberg
GF: Felix Imend?rffer, Jane Smithard, Graham Norton
HRB 21284 (AG N?rnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850