summaryrefslogtreecommitdiff
path: root/net
diff options
context:
space:
mode:
Diffstat (limited to 'net')
-rw-r--r--net/core/skbuff.c19
-rw-r--r--net/mctp/Kconfig12
-rw-r--r--net/mctp/Makefile3
-rw-r--r--net/mctp/af_mctp.c315
-rw-r--r--net/mctp/device.c174
-rw-r--r--net/mctp/neigh.c15
-rw-r--r--net/mctp/route.c459
-rw-r--r--net/mctp/test/route-test.c684
-rw-r--r--net/mctp/test/utils.c66
-rw-r--r--net/mctp/test/utils.h20
10 files changed, 1627 insertions, 140 deletions
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index e4badc189e37..0db93122adeb 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -70,6 +70,7 @@
#include <net/xfrm.h>
#include <net/mpls.h>
#include <net/mptcp.h>
+#include <net/mctp.h>
#include <net/page_pool.h>
#include <linux/uaccess.h>
@@ -4460,6 +4461,9 @@ static const u8 skb_ext_type_len[] = {
#if IS_ENABLED(CONFIG_MPTCP)
[SKB_EXT_MPTCP] = SKB_EXT_CHUNKSIZEOF(struct mptcp_ext),
#endif
+#if IS_ENABLED(CONFIG_MCTP_FLOWS)
+ [SKB_EXT_MCTP] = SKB_EXT_CHUNKSIZEOF(struct mctp_flow),
+#endif
};
static __always_inline unsigned int skb_ext_total_length(void)
@@ -4477,6 +4481,9 @@ static __always_inline unsigned int skb_ext_total_length(void)
#if IS_ENABLED(CONFIG_MPTCP)
skb_ext_type_len[SKB_EXT_MPTCP] +
#endif
+#if IS_ENABLED(CONFIG_MCTP_FLOWS)
+ skb_ext_type_len[SKB_EXT_MCTP] +
+#endif
0;
}
@@ -6548,6 +6555,14 @@ static void skb_ext_put_sp(struct sec_path *sp)
}
#endif
+#ifdef CONFIG_MCTP_FLOWS
+static void skb_ext_put_mctp(struct mctp_flow *flow)
+{
+ if (flow->key)
+ mctp_key_unref(flow->key);
+}
+#endif
+
void __skb_ext_del(struct sk_buff *skb, enum skb_ext_id id)
{
struct skb_ext *ext = skb->extensions;
@@ -6583,6 +6598,10 @@ free_now:
if (__skb_ext_exist(ext, SKB_EXT_SEC_PATH))
skb_ext_put_sp(skb_ext_get_ptr(ext, SKB_EXT_SEC_PATH));
#endif
+#ifdef CONFIG_MCTP_FLOWS
+ if (__skb_ext_exist(ext, SKB_EXT_MCTP))
+ skb_ext_put_mctp(skb_ext_get_ptr(ext, SKB_EXT_MCTP));
+#endif
kmem_cache_free(skbuff_ext_cache, ext);
}
diff --git a/net/mctp/Kconfig b/net/mctp/Kconfig
index 2cdf3d0a28c9..3a5c0e70da77 100644
--- a/net/mctp/Kconfig
+++ b/net/mctp/Kconfig
@@ -1,7 +1,7 @@
menuconfig MCTP
depends on NET
- tristate "MCTP core protocol support"
+ bool "MCTP core protocol support"
help
Management Component Transport Protocol (MCTP) is an in-system
protocol for communicating between management controllers and
@@ -11,3 +11,13 @@ menuconfig MCTP
This option enables core MCTP support. For communicating with other
devices, you'll want to enable a driver for a specific hardware
channel.
+
+config MCTP_TEST
+ bool "MCTP core tests" if !KUNIT_ALL_TESTS
+ depends on MCTP=y && KUNIT=y
+ default KUNIT_ALL_TESTS
+
+config MCTP_FLOWS
+ bool
+ depends on MCTP
+ select SKB_EXTENSIONS
diff --git a/net/mctp/Makefile b/net/mctp/Makefile
index 0171333384d7..6cd55233e685 100644
--- a/net/mctp/Makefile
+++ b/net/mctp/Makefile
@@ -1,3 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
obj-$(CONFIG_MCTP) += mctp.o
mctp-objs := af_mctp.o device.o route.o neigh.o
+
+# tests
+obj-$(CONFIG_MCTP_TEST) += test/utils.o
diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c
index 85cc1a28cbe9..f0702d920d8d 100644
--- a/net/mctp/af_mctp.c
+++ b/net/mctp/af_mctp.c
@@ -6,6 +6,7 @@
* Copyright (c) 2021 Google
*/
+#include <linux/compat.h>
#include <linux/if_arp.h>
#include <linux/net.h>
#include <linux/mctp.h>
@@ -16,8 +17,13 @@
#include <net/mctpdevice.h>
#include <net/sock.h>
+#define CREATE_TRACE_POINTS
+#include <trace/events/mctp.h>
+
/* socket implementation */
+static void mctp_sk_expire_keys(struct timer_list *timer);
+
static int mctp_release(struct socket *sock)
{
struct sock *sk = sock->sk;
@@ -36,6 +42,13 @@ static bool mctp_sockaddr_is_ok(const struct sockaddr_mctp *addr)
return !addr->__smctp_pad0 && !addr->__smctp_pad1;
}
+static bool mctp_sockaddr_ext_is_ok(const struct sockaddr_mctp_ext *addr)
+{
+ return !addr->__smctp_pad0[0] &&
+ !addr->__smctp_pad0[1] &&
+ !addr->__smctp_pad0[2];
+}
+
static int mctp_bind(struct socket *sock, struct sockaddr *addr, int addrlen)
{
struct sock *sk = sock->sk;
@@ -83,18 +96,26 @@ static int mctp_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
const int hlen = MCTP_HEADER_MAXLEN + sizeof(struct mctp_hdr);
int rc, addrlen = msg->msg_namelen;
struct sock *sk = sock->sk;
+ struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
struct mctp_skb_cb *cb;
struct mctp_route *rt;
struct sk_buff *skb;
if (addr) {
+ const u8 tagbits = MCTP_TAG_MASK | MCTP_TAG_OWNER |
+ MCTP_TAG_PREALLOC;
+
if (addrlen < sizeof(struct sockaddr_mctp))
return -EINVAL;
if (addr->smctp_family != AF_MCTP)
return -EINVAL;
if (!mctp_sockaddr_is_ok(addr))
return -EINVAL;
- if (addr->smctp_tag & ~(MCTP_TAG_MASK | MCTP_TAG_OWNER))
+ if (addr->smctp_tag & ~tagbits)
+ return -EINVAL;
+ /* can't preallocate a non-owned tag */
+ if (addr->smctp_tag & MCTP_TAG_PREALLOC &&
+ !(addr->smctp_tag & MCTP_TAG_OWNER))
return -EINVAL;
} else {
@@ -108,11 +129,6 @@ static int mctp_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
if (addr->smctp_network == MCTP_NET_ANY)
addr->smctp_network = mctp_default_net(sock_net(sk));
- rt = mctp_route_lookup(sock_net(sk), addr->smctp_network,
- addr->smctp_addr.s_addr);
- if (!rt)
- return -EHOSTUNREACH;
-
skb = sock_alloc_send_skb(sk, hlen + 1 + len,
msg->msg_flags & MSG_DONTWAIT, &rc);
if (!skb)
@@ -124,19 +140,46 @@ static int mctp_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
*(u8 *)skb_put(skb, 1) = addr->smctp_type;
rc = memcpy_from_msg((void *)skb_put(skb, len), msg, len);
- if (rc < 0) {
- kfree_skb(skb);
- return rc;
- }
+ if (rc < 0)
+ goto err_free;
/* set up cb */
cb = __mctp_cb(skb);
cb->net = addr->smctp_network;
+ /* direct addressing */
+ if (msk->addr_ext && addrlen >= sizeof(struct sockaddr_mctp_ext)) {
+ DECLARE_SOCKADDR(struct sockaddr_mctp_ext *,
+ extaddr, msg->msg_name);
+
+ if (!mctp_sockaddr_ext_is_ok(extaddr) ||
+ extaddr->smctp_halen > sizeof(cb->haddr)) {
+ rc = -EINVAL;
+ goto err_free;
+ }
+
+ cb->ifindex = extaddr->smctp_ifindex;
+ cb->halen = extaddr->smctp_halen;
+ memcpy(cb->haddr, extaddr->smctp_haddr, cb->halen);
+
+ rt = NULL;
+ } else {
+ rt = mctp_route_lookup(sock_net(sk), addr->smctp_network,
+ addr->smctp_addr.s_addr);
+ if (!rt) {
+ rc = -EHOSTUNREACH;
+ goto err_free;
+ }
+ }
+
rc = mctp_local_output(sk, rt, skb, addr->smctp_addr.s_addr,
addr->smctp_tag);
return rc ? : len;
+
+err_free:
+ kfree_skb(skb);
+ return rc;
}
static int mctp_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
@@ -144,6 +187,7 @@ static int mctp_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
{
DECLARE_SOCKADDR(struct sockaddr_mctp *, addr, msg->msg_name);
struct sock *sk = sock->sk;
+ struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
struct sk_buff *skb;
size_t msglen;
u8 type;
@@ -191,6 +235,17 @@ static int mctp_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
(MCTP_HDR_TAG_MASK | MCTP_HDR_FLAG_TO);
addr->__smctp_pad1 = 0;
msg->msg_namelen = sizeof(*addr);
+
+ if (msk->addr_ext) {
+ DECLARE_SOCKADDR(struct sockaddr_mctp_ext *, ae,
+ msg->msg_name);
+ msg->msg_namelen = sizeof(*ae);
+ ae->smctp_ifindex = cb->ifindex;
+ ae->smctp_halen = cb->halen;
+ memset(ae->__smctp_pad0, 0x0, sizeof(ae->__smctp_pad0));
+ memset(ae->smctp_haddr, 0x0, sizeof(ae->smctp_haddr));
+ memcpy(ae->smctp_haddr, cb->haddr, cb->halen);
+ }
}
rc = len;
@@ -203,18 +258,186 @@ out_free:
return rc;
}
+/* We're done with the key; invalidate, stop reassembly, and remove from lists.
+ */
+static void __mctp_key_remove(struct mctp_sk_key *key, struct net *net,
+ unsigned long flags, unsigned long reason)
+__releases(&key->lock)
+__must_hold(&net->mctp.keys_lock)
+{
+ struct sk_buff *skb;
+
+ trace_mctp_key_release(key, reason);
+ skb = key->reasm_head;
+ key->reasm_head = NULL;
+ key->reasm_dead = true;
+ key->valid = false;
+ mctp_dev_release_key(key->dev, key);
+ spin_unlock_irqrestore(&key->lock, flags);
+
+ hlist_del(&key->hlist);
+ hlist_del(&key->sklist);
+
+ /* unref for the lists */
+ mctp_key_unref(key);
+
+ kfree_skb(skb);
+}
+
static int mctp_setsockopt(struct socket *sock, int level, int optname,
sockptr_t optval, unsigned int optlen)
{
- return -EINVAL;
+ struct mctp_sock *msk = container_of(sock->sk, struct mctp_sock, sk);
+ int val;
+
+ if (level != SOL_MCTP)
+ return -EINVAL;
+
+ if (optname == MCTP_OPT_ADDR_EXT) {
+ if (optlen != sizeof(int))
+ return -EINVAL;
+ if (copy_from_sockptr(&val, optval, sizeof(int)))
+ return -EFAULT;
+ msk->addr_ext = val;
+ return 0;
+ }
+
+ return -ENOPROTOOPT;
}
static int mctp_getsockopt(struct socket *sock, int level, int optname,
char __user *optval, int __user *optlen)
{
+ struct mctp_sock *msk = container_of(sock->sk, struct mctp_sock, sk);
+ int len, val;
+
+ if (level != SOL_MCTP)
+ return -EINVAL;
+
+ if (get_user(len, optlen))
+ return -EFAULT;
+
+ if (optname == MCTP_OPT_ADDR_EXT) {
+ if (len != sizeof(int))
+ return -EINVAL;
+ val = !!msk->addr_ext;
+ if (copy_to_user(optval, &val, len))
+ return -EFAULT;
+ return 0;
+ }
+
return -EINVAL;
}
+static int mctp_ioctl_alloctag(struct mctp_sock *msk, unsigned long arg)
+{
+ struct net *net = sock_net(&msk->sk);
+ struct mctp_sk_key *key = NULL;
+ struct mctp_ioc_tag_ctl ctl;
+ unsigned long flags;
+ u8 tag;
+
+ if (copy_from_user(&ctl, (void __user *)arg, sizeof(ctl)))
+ return -EFAULT;
+
+ if (ctl.tag)
+ return -EINVAL;
+
+ if (ctl.flags)
+ return -EINVAL;
+
+ key = mctp_alloc_local_tag(msk, ctl.peer_addr, MCTP_ADDR_ANY,
+ true, &tag);
+ if (IS_ERR(key))
+ return PTR_ERR(key);
+
+ ctl.tag = tag | MCTP_TAG_OWNER | MCTP_TAG_PREALLOC;
+ if (copy_to_user((void __user *)arg, &ctl, sizeof(ctl))) {
+ spin_lock_irqsave(&key->lock, flags);
+ __mctp_key_remove(key, net, flags, MCTP_TRACE_KEY_DROPPED);
+ mctp_key_unref(key);
+ return -EFAULT;
+ }
+
+ mctp_key_unref(key);
+ return 0;
+}
+
+static int mctp_ioctl_droptag(struct mctp_sock *msk, unsigned long arg)
+{
+ struct net *net = sock_net(&msk->sk);
+ struct mctp_ioc_tag_ctl ctl;
+ unsigned long flags, fl2;
+ struct mctp_sk_key *key;
+ struct hlist_node *tmp;
+ int rc;
+ u8 tag;
+
+ if (copy_from_user(&ctl, (void __user *)arg, sizeof(ctl)))
+ return -EFAULT;
+
+ if (ctl.flags)
+ return -EINVAL;
+
+ /* Must be a local tag, TO set, preallocated */
+ if ((ctl.tag & ~MCTP_TAG_MASK) != (MCTP_TAG_OWNER | MCTP_TAG_PREALLOC))
+ return -EINVAL;
+
+ tag = ctl.tag & MCTP_TAG_MASK;
+ rc = -EINVAL;
+
+ spin_lock_irqsave(&net->mctp.keys_lock, flags);
+ hlist_for_each_entry_safe(key, tmp, &msk->keys, sklist) {
+ /* we do an irqsave here, even though we know the irq state,
+ * so we have the flags to pass to __mctp_key_remove
+ */
+ spin_lock_irqsave(&key->lock, fl2);
+ if (key->manual_alloc &&
+ ctl.peer_addr == key->peer_addr &&
+ tag == key->tag) {
+ __mctp_key_remove(key, net, fl2,
+ MCTP_TRACE_KEY_DROPPED);
+ rc = 0;
+ } else {
+ spin_unlock_irqrestore(&key->lock, fl2);
+ }
+ }
+ spin_unlock_irqrestore(&net->mctp.keys_lock, flags);
+
+ return rc;
+}
+
+static int mctp_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+ struct mctp_sock *msk = container_of(sock->sk, struct mctp_sock, sk);
+
+ switch (cmd) {
+ case SIOCMCTPALLOCTAG:
+ return mctp_ioctl_alloctag(msk, arg);
+ case SIOCMCTPDROPTAG:
+ return mctp_ioctl_droptag(msk, arg);
+ }
+
+ return -EINVAL;
+}
+
+#ifdef CONFIG_COMPAT
+static int mctp_compat_ioctl(struct socket *sock, unsigned int cmd,
+ unsigned long arg)
+{
+ void __user *argp = compat_ptr(arg);
+
+ switch (cmd) {
+ /* These have compatible ptr layouts */
+ case SIOCMCTPALLOCTAG:
+ case SIOCMCTPDROPTAG:
+ return mctp_ioctl(sock, cmd, (unsigned long)argp);
+ }
+
+ return -ENOIOCTLCMD;
+}
+#endif
+
static const struct proto_ops mctp_dgram_ops = {
.family = PF_MCTP,
.release = mctp_release,
@@ -224,7 +447,7 @@ static const struct proto_ops mctp_dgram_ops = {
.accept = sock_no_accept,
.getname = sock_no_getname,
.poll = datagram_poll,
- .ioctl = sock_no_ioctl,
+ .ioctl = mctp_ioctl,
.gettstamp = sock_gettstamp,
.listen = sock_no_listen,
.shutdown = sock_no_shutdown,
@@ -234,18 +457,67 @@ static const struct proto_ops mctp_dgram_ops = {
.recvmsg = mctp_recvmsg,
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage,
+#ifdef CONFIG_COMPAT
+ .compat_ioctl = mctp_compat_ioctl,
+#endif
};
+static void mctp_sk_expire_keys(struct timer_list *timer)
+{
+ struct mctp_sock *msk = container_of(timer, struct mctp_sock,
+ key_expiry);
+ struct net *net = sock_net(&msk->sk);
+ unsigned long next_expiry, flags, fl2;
+ struct mctp_sk_key *key;
+ struct hlist_node *tmp;
+ bool next_expiry_valid = false;
+
+ spin_lock_irqsave(&net->mctp.keys_lock, flags);
+
+ hlist_for_each_entry_safe(key, tmp, &msk->keys, sklist) {
+ /* don't expire. manual_alloc is immutable, no locking
+ * required.
+ */
+ if (key->manual_alloc)
+ continue;
+
+ spin_lock_irqsave(&key->lock, fl2);
+ if (!time_after_eq(key->expiry, jiffies)) {
+ __mctp_key_remove(key, net, fl2,
+ MCTP_TRACE_KEY_TIMEOUT);
+ continue;
+ }
+
+ if (next_expiry_valid) {
+ if (time_before(key->expiry, next_expiry))
+ next_expiry = key->expiry;
+ } else {
+ next_expiry = key->expiry;
+ next_expiry_valid = true;
+ }
+ spin_unlock_irqrestore(&key->lock, fl2);
+ }
+
+ spin_unlock_irqrestore(&net->mctp.keys_lock, flags);
+
+ if (next_expiry_valid)
+ mod_timer(timer, next_expiry);
+}
+
static int mctp_sk_init(struct sock *sk)
{
struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
INIT_HLIST_HEAD(&msk->keys);
+ timer_setup(&msk->key_expiry, mctp_sk_expire_keys, 0);
return 0;
}
static void mctp_sk_close(struct sock *sk, long timeout)
{
+ struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
+
+ del_timer_sync(&msk->key_expiry);
sk_common_release(sk);
}
@@ -264,9 +536,9 @@ static void mctp_sk_unhash(struct sock *sk)
{
struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
struct net *net = sock_net(sk);
+ unsigned long flags, fl2;
struct mctp_sk_key *key;
struct hlist_node *tmp;
- unsigned long flags;
/* remove from any type-based binds */
mutex_lock(&net->mctp.bind_lock);
@@ -276,21 +548,10 @@ static void mctp_sk_unhash(struct sock *sk)
/* remove tag allocations */
spin_lock_irqsave(&net->mctp.keys_lock, flags);
hlist_for_each_entry_safe(key, tmp, &msk->keys, sklist) {
- hlist_del_rcu(&key->sklist);
- hlist_del_rcu(&key->hlist);
-
- spin_lock(&key->reasm_lock);
- if (key->reasm_head)
- kfree_skb(key->reasm_head);
- key->reasm_head = NULL;
- key->reasm_dead = true;
- spin_unlock(&key->reasm_lock);
-
- kfree_rcu(key, rcu);
+ spin_lock_irqsave(&key->lock, fl2);
+ __mctp_key_remove(key, net, fl2, MCTP_TRACE_KEY_CLOSED);
}
spin_unlock_irqrestore(&net->mctp.keys_lock, flags);
-
- synchronize_rcu();
}
static struct proto mctp_proto = {
@@ -398,7 +659,7 @@ static __exit void mctp_exit(void)
sock_unregister(PF_MCTP);
}
-module_init(mctp_init);
+subsys_initcall(mctp_init);
module_exit(mctp_exit);
MODULE_DESCRIPTION("MCTP core");
diff --git a/net/mctp/device.c b/net/mctp/device.c
index b9f38e765f61..ffcd7d8d2fb6 100644
--- a/net/mctp/device.c
+++ b/net/mctp/device.c
@@ -24,33 +24,48 @@ struct mctp_dump_cb {
size_t a_idx;
};
-/* unlocked: caller must hold rcu_read_lock */
+/* unlocked: caller must hold rcu_read_lock.
+ * Returned mctp_dev has its refcount incremented, or NULL if unset.
+ */
struct mctp_dev *__mctp_dev_get(const struct net_device *dev)
{
- return rcu_dereference(dev->mctp_ptr);
+ struct mctp_dev *mdev = rcu_dereference(dev->mctp_ptr);
+
+ /* RCU guarantees that any mdev is still live.
+ * Zero refcount implies a pending free, return NULL.
+ */
+ if (mdev)
+ if (!refcount_inc_not_zero(&mdev->refs))
+ return NULL;
+ return mdev;
}
+/* Returned mctp_dev does not have refcount incremented. The returned pointer
+ * remains live while rtnl_lock is held, as that prevents mctp_unregister()
+ */
struct mctp_dev *mctp_dev_get_rtnl(const struct net_device *dev)
{
return rtnl_dereference(dev->mctp_ptr);
}
-static void mctp_dev_destroy(struct mctp_dev *mdev)
+static int mctp_addrinfo_size(void)
{
- struct net_device *dev = mdev->dev;
-
- dev_put(dev);
- kfree_rcu(mdev, rcu);
+ return NLMSG_ALIGN(sizeof(struct ifaddrmsg))
+ + nla_total_size(1) // IFA_LOCAL
+ + nla_total_size(1) // IFA_ADDRESS
+ ;
}
-static int mctp_fill_addrinfo(struct sk_buff *skb, struct netlink_callback *cb,
- struct mctp_dev *mdev, mctp_eid_t eid)
+/* flag should be NLM_F_MULTI for dump calls */
+static int mctp_fill_addrinfo(struct sk_buff *skb,
+ struct mctp_dev *mdev, mctp_eid_t eid,
+ int msg_type, u32 portid, u32 seq, int flag)
{
struct ifaddrmsg *hdr;
struct nlmsghdr *nlh;
- nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
- RTM_NEWADDR, sizeof(*hdr), NLM_F_MULTI);
+ nlh = nlmsg_put(skb, portid, seq,
+ msg_type, sizeof(*hdr), flag);
if (!nlh)
return -EMSGSIZE;
@@ -80,10 +95,14 @@ static int mctp_dump_dev_addrinfo(struct mctp_dev *mdev, struct sk_buff *skb,
struct netlink_callback *cb)
{
struct mctp_dump_cb *mcb = (void *)cb->ctx;
+ u32 portid, seq;
int rc = 0;
+ portid = NETLINK_CB(cb->skb).portid;
+ seq = cb->nlh->nlmsg_seq;
for (; mcb->a_idx < mdev->num_addrs; mcb->a_idx++) {
- rc = mctp_fill_addrinfo(skb, cb, mdev, mdev->addrs[mcb->a_idx]);
+ rc = mctp_fill_addrinfo(skb, mdev, mdev->addrs[mcb->a_idx],
+ RTM_NEWADDR, portid, seq, NLM_F_MULTI);
if (rc < 0)
break;
}
@@ -100,7 +119,7 @@ static int mctp_dump_addrinfo(struct sk_buff *skb, struct netlink_callback *cb)
struct ifaddrmsg *hdr;
struct mctp_dev *mdev;
int ifindex;
- int idx, rc;
+ int idx = 0, rc;
hdr = nlmsg_data(cb->nlh);
// filter by ifindex if requested
@@ -117,6 +136,7 @@ static int mctp_dump_addrinfo(struct sk_buff *skb, struct netlink_callback *cb)
if (mdev) {
rc = mctp_dump_dev_addrinfo(mdev,
skb, cb);
+ mctp_dev_put(mdev);
// Error indicates full buffer, this
// callback will get retried.
if (rc < 0)
@@ -135,6 +155,32 @@ out:
return skb->len;
}
+static void mctp_addr_notify(struct mctp_dev *mdev, mctp_eid_t eid, int msg_type,
+ struct sk_buff *req_skb, struct nlmsghdr *req_nlh)
+{
+ u32 portid = NETLINK_CB(req_skb).portid;
+ struct net *net = dev_net(mdev->dev);
+ struct sk_buff *skb;
+ int rc = -ENOBUFS;
+
+ skb = nlmsg_new(mctp_addrinfo_size(), GFP_KERNEL);
+ if (!skb)
+ goto out;
+
+ rc = mctp_fill_addrinfo(skb, mdev, eid, msg_type,
+ portid, req_nlh->nlmsg_seq, 0);
+ if (rc < 0) {
+ WARN_ON_ONCE(rc == -EMSGSIZE);
+ goto out;
+ }
+
+ rtnl_notify(skb, net, portid, RTNLGRP_MCTP_IFADDR, req_nlh, GFP_KERNEL);
+ return;
+out:
+ kfree_skb(skb);
+ rtnl_set_sk_err(net, RTNLGRP_MCTP_IFADDR, rc);
+}
+
static const struct nla_policy ifa_mctp_policy[IFA_MAX + 1] = {
[IFA_ADDRESS] = { .type = NLA_U8 },
[IFA_LOCAL] = { .type = NLA_U8 },
@@ -176,7 +222,7 @@ static int mctp_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh,
if (!mdev)
return -ENODEV;
- if (!mctp_address_ok(addr->s_addr))
+ if (!mctp_address_unicast(addr->s_addr))
return -EINVAL;
/* Prevent duplicates. Under RTNL so don't need to lock for reading */
@@ -197,6 +243,7 @@ static int mctp_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh,
kfree(tmp_addrs);
+ mctp_addr_notify(mdev, addr->s_addr, RTM_NEWADDR, skb, nlh);
mctp_route_add_local(mdev, addr->s_addr);
return 0;
@@ -252,9 +299,42 @@ static int mctp_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh,
mdev->num_addrs--;
spin_unlock_irqrestore(&mdev->addrs_lock, flags);
+ mctp_addr_notify(mdev, addr->s_addr, RTM_DELADDR, skb, nlh);
+
return 0;
}
+void mctp_dev_hold(struct mctp_dev *mdev)
+{
+ refcount_inc(&mdev->refs);
+}
+
+void mctp_dev_put(struct mctp_dev *mdev)
+{
+ if (mdev && refcount_dec_and_test(&mdev->refs)) {
+ dev_put(mdev->dev);
+ kfree_rcu(mdev, rcu);
+ }
+}
+
+void mctp_dev_release_key(struct mctp_dev *dev, struct mctp_sk_key *key)
+ __must_hold(&key->lock)
+{
+ if (!dev)
+ return;
+ if (dev->ops && dev->ops->release_flow)
+ dev->ops->release_flow(dev, key);
+ key->dev = NULL;
+ mctp_dev_put(dev);
+}
+
+void mctp_dev_set_key(struct mctp_dev *dev, struct mctp_sk_key *key)
+ __must_hold(&key->lock)
+{
+ mctp_dev_hold(dev);
+ key->dev = dev;
+}
+
static struct mctp_dev *mctp_add_dev(struct net_device *dev)
{
struct mctp_dev *mdev;
@@ -270,7 +350,9 @@ static struct mctp_dev *mctp_add_dev(struct net_device *dev)
mdev->net = mctp_default_net(dev_net(dev));
/* associate to net_device */
+ refcount_set(&mdev->refs, 1);
rcu_assign_pointer(dev->mctp_ptr, mdev);
+
dev_hold(dev);
mdev->dev = dev;
@@ -301,6 +383,7 @@ static size_t mctp_get_link_af_size(const struct net_device *dev,
if (!mdev)
return 0;
ret = nla_total_size(4); /* IFLA_MCTP_NET */
+ mctp_dev_put(mdev);
return ret;
}
@@ -330,12 +413,26 @@ static int mctp_set_link_af(struct net_device *dev, const struct nlattr *attr,
return 0;
}
+/* Matches netdev types that should have MCTP handling */
+static bool mctp_known(struct net_device *dev)
+{
+ /* only register specific types (inc. NONE for TUN devices) */
+ return dev->type == ARPHRD_MCTP ||
+ dev->type == ARPHRD_LOOPBACK ||
+ dev->type == ARPHRD_NONE;
+}
+
static void mctp_unregister(struct net_device *dev)
{
struct mctp_dev *mdev;
mdev = mctp_dev_get_rtnl(dev);
-
+ if (mdev && !mctp_known(dev)) {
+ // Sanity check, should match what was set in mctp_register
+ netdev_warn(dev, "%s: BUG mctp_ptr set for unknown type %d",
+ __func__, dev->type);
+ return;
+ }
if (!mdev)
return;
@@ -345,7 +442,7 @@ static void mctp_unregister(struct net_device *dev)
mctp_neigh_remove_dev(mdev);
kfree(mdev->addrs);
- mctp_dev_destroy(mdev);
+ mctp_dev_put(mdev);
}
static int mctp_register(struct net_device *dev)
@@ -353,11 +450,17 @@ static int mctp_register(struct net_device *dev)
struct mctp_dev *mdev;
/* Already registered? */
- if (rtnl_dereference(dev->mctp_ptr))
+ mdev = rtnl_dereference(dev->mctp_ptr);
+
+ if (mdev) {
+ if (!mctp_known(dev))
+ netdev_warn(dev, "%s: BUG mctp_ptr set for unknown type %d",
+ __func__, dev->type);
return 0;
+ }
- /* only register specific types; MCTP-specific and loopback for now */
- if (dev->type != ARPHRD_MCTP && dev->type != ARPHRD_LOOPBACK)
+ /* only register specific types */
+ if (!mctp_known(dev))
return 0;
mdev = mctp_add_dev(dev);
@@ -387,6 +490,39 @@ static int mctp_dev_notify(struct notifier_block *this, unsigned long event,
return NOTIFY_OK;
}
+static int mctp_register_netdevice(struct net_device *dev,
+ const struct mctp_netdev_ops *ops)
+{
+ struct mctp_dev *mdev;
+
+ mdev = mctp_add_dev(dev);
+ if (IS_ERR(mdev))
+ return PTR_ERR(mdev);
+
+ mdev->ops = ops;
+
+ return register_netdevice(dev);
+}
+
+int mctp_register_netdev(struct net_device *dev,
+ const struct mctp_netdev_ops *ops)
+{
+ int rc;
+
+ rtnl_lock();
+ rc = mctp_register_netdevice(dev, ops);
+ rtnl_unlock();
+
+ return rc;
+}
+EXPORT_SYMBOL_GPL(mctp_register_netdev);
+
+void mctp_unregister_netdev(struct net_device *dev)
+{
+ unregister_netdev(dev);
+}
+EXPORT_SYMBOL_GPL(mctp_unregister_netdev);
+
static struct rtnl_af_ops mctp_af_ops = {
.family = AF_MCTP,
.fill_link_af = mctp_fill_link_af,
diff --git a/net/mctp/neigh.c b/net/mctp/neigh.c
index 90ed2f02d1fb..ffa0f9e0983f 100644
--- a/net/mctp/neigh.c
+++ b/net/mctp/neigh.c
@@ -47,7 +47,7 @@ static int mctp_neigh_add(struct mctp_dev *mdev, mctp_eid_t eid,
}
INIT_LIST_HEAD(&neigh->list);
neigh->dev = mdev;
- dev_hold(neigh->dev->dev);
+ mctp_dev_hold(neigh->dev);
neigh->eid = eid;
neigh->source = source;
memcpy(neigh->ha, lladdr, lladdr_len);
@@ -63,7 +63,7 @@ static void __mctp_neigh_free(struct rcu_head *rcu)
{
struct mctp_neigh *neigh = container_of(rcu, struct mctp_neigh, rcu);
- dev_put(neigh->dev->dev);
+ mctp_dev_put(neigh->dev);
kfree(neigh);
}
@@ -85,8 +85,8 @@ void mctp_neigh_remove_dev(struct mctp_dev *mdev)
mutex_unlock(&net->mctp.neigh_lock);
}
-// TODO: add a "source" flag so netlink can only delete static neighbours?
-static int mctp_neigh_remove(struct mctp_dev *mdev, mctp_eid_t eid)
+static int mctp_neigh_remove(struct mctp_dev *mdev, mctp_eid_t eid,
+ enum mctp_neigh_source source)
{
struct net *net = dev_net(mdev->dev);
struct mctp_neigh *neigh, *tmp;
@@ -94,7 +94,8 @@ static int mctp_neigh_remove(struct mctp_dev *mdev, mctp_eid_t eid)
mutex_lock(&net->mctp.neigh_lock);
list_for_each_entry_safe(neigh, tmp, &net->mctp.neighbours, list) {
- if (neigh->dev == mdev && neigh->eid == eid) {
+ if (neigh->dev == mdev && neigh->eid == eid &&
+ neigh->source == source) {
list_del_rcu(&neigh->list);
/* TODO: immediate RTM_DELNEIGH */
call_rcu(&neigh->rcu, __mctp_neigh_free);
@@ -142,7 +143,7 @@ static int mctp_rtm_newneigh(struct sk_buff *skb, struct nlmsghdr *nlh,
}
eid = nla_get_u8(tb[NDA_DST]);
- if (!mctp_address_ok(eid)) {
+ if (!mctp_address_unicast(eid)) {
NL_SET_ERR_MSG(extack, "Invalid neighbour EID");
return -EINVAL;
}
@@ -202,7 +203,7 @@ static int mctp_rtm_delneigh(struct sk_buff *skb, struct nlmsghdr *nlh,
if (!mdev)
return -ENODEV;
- return mctp_neigh_remove(mdev, eid);
+ return mctp_neigh_remove(mdev, eid, MCTP_NEIGH_STATIC);
}
static int mctp_fill_neigh(struct sk_buff *skb, u32 portid, u32 seq, int event,
diff --git a/net/mctp/route.c b/net/mctp/route.c
index bbb13dbc9227..ee548c46c78f 100644
--- a/net/mctp/route.c
+++ b/net/mctp/route.c
@@ -11,6 +11,7 @@
*/
#include <linux/idr.h>
+#include <linux/kconfig.h>
#include <linux/mctp.h>
#include <linux/netdevice.h>
#include <linux/rtnetlink.h>
@@ -23,7 +24,12 @@
#include <net/netlink.h>
#include <net/sock.h>
+#include <trace/events/mctp.h>
+
static const unsigned int mctp_message_maxlen = 64 * 1024;
+static const unsigned long mctp_key_lifetime = 6 * CONFIG_HZ;
+
+static void mctp_flow_prepare_output(struct sk_buff *skb, struct mctp_dev *dev);
/* route output callbacks */
static int mctp_route_discard(struct mctp_route *route, struct sk_buff *skb)
@@ -58,8 +64,7 @@ static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
if (msk->bind_type != type)
continue;
- if (msk->bind_addr != MCTP_ADDR_ANY &&
- msk->bind_addr != mh->dest)
+ if (!mctp_address_matches(msk->bind_addr, mh->dest))
continue;
return msk;
@@ -71,7 +76,7 @@ static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
static bool mctp_key_match(struct mctp_sk_key *key, mctp_eid_t local,
mctp_eid_t peer, u8 tag)
{
- if (key->local_addr != local)
+ if (!mctp_address_matches(key->local_addr, local))
return false;
if (key->peer_addr != peer)
@@ -83,25 +88,43 @@ static bool mctp_key_match(struct mctp_sk_key *key, mctp_eid_t local,
return true;
}
+/* returns a key (with key->lock held, and refcounted), or NULL if no such
+ * key exists.
+ */
static struct mctp_sk_key *mctp_lookup_key(struct net *net, struct sk_buff *skb,
- mctp_eid_t peer)
+ mctp_eid_t peer,
+ unsigned long *irqflags)
+ __acquires(&key->lock)
{
struct mctp_sk_key *key, *ret;
+ unsigned long flags;
struct mctp_hdr *mh;
u8 tag;
- WARN_ON(!rcu_read_lock_held());
-
mh = mctp_hdr(skb);
tag = mh->flags_seq_tag & (MCTP_HDR_TAG_MASK | MCTP_HDR_FLAG_TO);
ret = NULL;
+ spin_lock_irqsave(&net->mctp.keys_lock, flags);
- hlist_for_each_entry_rcu(key, &net->mctp.keys, hlist) {
- if (mctp_key_match(key, mh->dest, peer, tag)) {
+ hlist_for_each_entry(key, &net->mctp.keys, hlist) {
+ if (!mctp_key_match(key, mh->dest, peer, tag))
+ continue;
+
+ spin_lock(&key->lock);
+ if (key->valid) {
+ refcount_inc(&key->refs);
ret = key;
break;
}
+ spin_unlock(&key->lock);
+ }
+
+ if (ret) {
+ spin_unlock(&net->mctp.keys_lock);
+ *irqflags = flags;
+ } else {
+ spin_unlock_irqrestore(&net->mctp.keys_lock, flags);
}
return ret;
@@ -121,11 +144,30 @@ static struct mctp_sk_key *mctp_key_alloc(struct mctp_sock *msk,
key->local_addr = local;
key->tag = tag;
key->sk = &msk->sk;
- spin_lock_init(&key->reasm_lock);
+ key->valid = true;
+ spin_lock_init(&key->lock);
+ refcount_set(&key->refs, 1);
return key;
}
+void mctp_key_unref(struct mctp_sk_key *key)
+{
+ unsigned long flags;
+
+ if (!refcount_dec_and_test(&key->refs))
+ return;
+
+ /* even though no refs exist here, the lock allows us to stay
+ * consistent with the locking requirement of mctp_dev_release_key
+ */
+ spin_lock_irqsave(&key->lock, flags);
+ mctp_dev_release_key(key->dev, key);
+ spin_unlock_irqrestore(&key->lock, flags);
+
+ kfree(key);
+}
+
static int mctp_key_add(struct mctp_sk_key *key, struct mctp_sock *msk)
{
struct net *net = sock_net(&msk->sk);
@@ -138,12 +180,20 @@ static int mctp_key_add(struct mctp_sk_key *key, struct mctp_sock *msk)
hlist_for_each_entry(tmp, &net->mctp.keys, hlist) {
if (mctp_key_match(tmp, key->local_addr, key->peer_addr,
key->tag)) {
- rc = -EEXIST;
- break;
+ spin_lock(&tmp->lock);
+ if (tmp->valid)
+ rc = -EEXIST;
+ spin_unlock(&tmp->lock);
+ if (rc)
+ break;
}
}
if (!rc) {
+ refcount_inc(&key->refs);
+ key->expiry = jiffies + mctp_key_lifetime;
+ timer_reduce(&msk->key_expiry, key->expiry);
+
hlist_add_head(&key->hlist, &net->mctp.keys);
hlist_add_head(&key->sklist, &msk->keys);
}
@@ -153,30 +203,79 @@ static int mctp_key_add(struct mctp_sk_key *key, struct mctp_sock *msk)
return rc;
}
-/* Must be called with key->reasm_lock, which it will release. Will schedule
- * the key for an RCU free.
+/* Helper for mctp_route_input().
+ * We're done with the key; unlock and unref the key.
+ * For the usual case of automatic expiry we remove the key from lists.
+ * In the case that manual allocation is set on a key we release the lock
+ * and local ref, reset reassembly, but don't remove from lists.
*/
-static void __mctp_key_unlock_drop(struct mctp_sk_key *key, struct net *net,
- unsigned long flags)
- __releases(&key->reasm_lock)
+static void __mctp_key_done_in(struct mctp_sk_key *key, struct net *net,
+ unsigned long flags, unsigned long reason)
+__releases(&key->lock)
{
struct sk_buff *skb;
+ trace_mctp_key_release(key, reason);
skb = key->reasm_head;
key->reasm_head = NULL;
- key->reasm_dead = true;
- spin_unlock_irqrestore(&key->reasm_lock, flags);
- spin_lock_irqsave(&net->mctp.keys_lock, flags);
- hlist_del_rcu(&key->hlist);
- hlist_del_rcu(&key->sklist);
- spin_unlock_irqrestore(&net->mctp.keys_lock, flags);
- kfree_rcu(key, rcu);
+ if (!key->manual_alloc) {
+ key->reasm_dead = true;
+ key->valid = false;
+ mctp_dev_release_key(key->dev, key);
+ }
+ spin_unlock_irqrestore(&key->lock, flags);
- if (skb)
- kfree_skb(skb);
+ if (!key->manual_alloc) {
+ spin_lock_irqsave(&net->mctp.keys_lock, flags);
+ hlist_del(&key->hlist);
+ hlist_del(&key->sklist);
+ spin_unlock_irqrestore(&net->mctp.keys_lock, flags);
+
+ /* unref for the lists */
+ mctp_key_unref(key);
+ }
+
+ /* and one for the local reference */
+ mctp_key_unref(key);
+
+ kfree_skb(skb);
+}
+
+#ifdef CONFIG_MCTP_FLOWS
+static void mctp_skb_set_flow(struct sk_buff *skb, struct mctp_sk_key *key)
+{
+ struct mctp_flow *flow;
+
+ flow = skb_ext_add(skb, SKB_EXT_MCTP);
+ if (!flow)
+ return;
+
+ refcount_inc(&key->refs);
+ flow->key = key;
}
+static void mctp_flow_prepare_output(struct sk_buff *skb, struct mctp_dev *dev)
+{
+ struct mctp_sk_key *key;
+ struct mctp_flow *flow;
+
+ flow = skb_ext_find(skb, SKB_EXT_MCTP);
+ if (!flow)
+ return;
+
+ key = flow->key;
+
+ if (WARN_ON(key->dev && key->dev != dev))
+ return;
+
+ mctp_dev_set_key(dev, key);
+}
+#else
+static void mctp_skb_set_flow(struct sk_buff *skb, struct mctp_sk_key *key) {}
+static void mctp_flow_prepare_output(struct sk_buff *skb, struct mctp_dev *dev) {}
+#endif
+
static int mctp_frag_queue(struct mctp_sk_key *key, struct sk_buff *skb)
{
struct mctp_hdr *hdr = mctp_hdr(skb);
@@ -248,8 +347,10 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
rcu_read_lock();
- /* lookup socket / reasm context, exactly matching (src,dest,tag) */
- key = mctp_lookup_key(net, skb, mh->src);
+ /* lookup socket / reasm context, exactly matching (src,dest,tag).
+ * we hold a ref on the key, and key->lock held.
+ */
+ key = mctp_lookup_key(net, skb, mh->src, &f);
if (flags & MCTP_HDR_FLAG_SOM) {
if (key) {
@@ -260,10 +361,12 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
* key for reassembly - we'll create a more specific
* one for future packets if required (ie, !EOM).
*/
- key = mctp_lookup_key(net, skb, MCTP_ADDR_ANY);
+ key = mctp_lookup_key(net, skb, MCTP_ADDR_ANY, &f);
if (key) {
msk = container_of(key->sk,
struct mctp_sock, sk);
+ spin_unlock_irqrestore(&key->lock, f);
+ mctp_key_unref(key);
key = NULL;
}
}
@@ -282,11 +385,12 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
if (flags & MCTP_HDR_FLAG_EOM) {
sock_queue_rcv_skb(&msk->sk, skb);
if (key) {
- spin_lock_irqsave(&key->reasm_lock, f);
/* we've hit a pending reassembly; not much we
* can do but drop it
*/
- __mctp_key_unlock_drop(key, net, f);
+ __mctp_key_done_in(key, net, f,
+ MCTP_TRACE_KEY_REPLIED);
+ key = NULL;
}
rc = 0;
goto out_unlock;
@@ -303,7 +407,7 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
goto out_unlock;
}
- /* we can queue without the reasm lock here, as the
+ /* we can queue without the key lock here, as the
* key isn't observable yet
*/
mctp_frag_queue(key, skb);
@@ -315,20 +419,25 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
* this function.
*/
rc = mctp_key_add(key, msk);
- if (rc)
+ if (rc) {
kfree(key);
+ } else {
+ trace_mctp_key_acquire(key);
- } else {
- /* existing key: start reassembly */
- spin_lock_irqsave(&key->reasm_lock, f);
+ /* we don't need to release key->lock on exit */
+ mctp_key_unref(key);
+ }
+ key = NULL;
+ } else {
if (key->reasm_head || key->reasm_dead) {
/* duplicate start? drop everything */
- __mctp_key_unlock_drop(key, net, f);
+ __mctp_key_done_in(key, net, f,
+ MCTP_TRACE_KEY_INVALIDATED);
rc = -EEXIST;
+ key = NULL;
} else {
rc = mctp_frag_queue(key, skb);
- spin_unlock_irqrestore(&key->reasm_lock, f);
}
}
@@ -337,8 +446,6 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
* using the message-specific key
*/
- spin_lock_irqsave(&key->reasm_lock, f);
-
/* we need to be continuing an existing reassembly... */
if (!key->reasm_head)
rc = -EINVAL;
@@ -351,9 +458,8 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
if (!rc && flags & MCTP_HDR_FLAG_EOM) {
sock_queue_rcv_skb(key->sk, key->reasm_head);
key->reasm_head = NULL;
- __mctp_key_unlock_drop(key, net, f);
- } else {
- spin_unlock_irqrestore(&key->reasm_lock, f);
+ __mctp_key_done_in(key, net, f, MCTP_TRACE_KEY_REPLIED);
+ key = NULL;
}
} else {
@@ -363,6 +469,10 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
out_unlock:
rcu_read_unlock();
+ if (key) {
+ spin_unlock_irqrestore(&key->lock, f);
+ mctp_key_unref(key);
+ }
out:
if (rc)
kfree_skb(skb);
@@ -376,6 +486,7 @@ static unsigned int mctp_route_mtu(struct mctp_route *rt)
static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
{
+ struct mctp_skb_cb *cb = mctp_cb(skb);
struct mctp_hdr *hdr = mctp_hdr(skb);
char daddr_buf[MAX_ADDR_LEN];
char *daddr = NULL;
@@ -390,9 +501,14 @@ static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
return -EMSGSIZE;
}
- /* If lookup fails let the device handle daddr==NULL */
- if (mctp_neigh_lookup(route->dev, hdr->dest, daddr_buf) == 0)
- daddr = daddr_buf;
+ if (cb->ifindex) {
+ /* direct route; use the hwaddr we stashed in sendmsg */
+ daddr = cb->haddr;
+ } else {
+ /* If lookup fails let the device handle daddr==NULL */
+ if (mctp_neigh_lookup(route->dev, hdr->dest, daddr_buf) == 0)
+ daddr = daddr_buf;
+ }
rc = dev_hard_header(skb, skb->dev, ntohs(skb->protocol),
daddr, skb->dev->dev_addr, skb->len);
@@ -401,6 +517,8 @@ static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
return -EHOSTUNREACH;
}
+ mctp_flow_prepare_output(skb, route->dev);
+
rc = dev_queue_xmit(skb);
if (rc)
rc = net_xmit_errno(rc);
@@ -412,7 +530,7 @@ static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
static void mctp_route_release(struct mctp_route *rt)
{
if (refcount_dec_and_test(&rt->refs)) {
- dev_put(rt->dev->dev);
+ mctp_dev_put(rt->dev);
kfree_rcu(rt, rcu);
}
}
@@ -454,30 +572,38 @@ static void mctp_reserve_tag(struct net *net, struct mctp_sk_key *key,
lockdep_assert_held(&mns->keys_lock);
+ key->expiry = jiffies + mctp_key_lifetime;
+ timer_reduce(&msk->key_expiry, key->expiry);
+
/* we hold the net->key_lock here, allowing updates to both
* then net and sk
*/
hlist_add_head_rcu(&key->hlist, &mns->keys);
hlist_add_head_rcu(&key->sklist, &msk->keys);
+ refcount_inc(&key->refs);
}
/* Allocate a locally-owned tag value for (saddr, daddr), and reserve
* it for the socket msk
*/
-static int mctp_alloc_local_tag(struct mctp_sock *msk,
- mctp_eid_t saddr, mctp_eid_t daddr, u8 *tagp)
+struct mctp_sk_key *mctp_alloc_local_tag(struct mctp_sock *msk,
+ mctp_eid_t daddr, mctp_eid_t saddr,
+ bool manual, u8 *tagp)
{
struct net *net = sock_net(&msk->sk);
struct netns_mctp *mns = &net->mctp;
struct mctp_sk_key *key, *tmp;
unsigned long flags;
- int rc = -EAGAIN;
u8 tagbits;
+ /* for NULL destination EIDs, we may get a response from any peer */
+ if (daddr == MCTP_ADDR_NULL)
+ daddr = MCTP_ADDR_ANY;
+
/* be optimistic, alloc now */
key = mctp_key_alloc(msk, saddr, daddr, 0, GFP_KERNEL);
if (!key)
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
/* 8 possible tag values */
tagbits = 0xff;
@@ -488,14 +614,25 @@ static int mctp_alloc_local_tag(struct mctp_sock *msk,
* tags. If we find a conflict, clear that bit from tagbits
*/
hlist_for_each_entry(tmp, &mns->keys, hlist) {
+ /* We can check the lookup fields (*_addr, tag) without the
+ * lock held, they don't change over the lifetime of the key.
+ */
+
/* if we don't own the tag, it can't conflict */
if (tmp->tag & MCTP_HDR_FLAG_TO)
continue;
- if ((tmp->peer_addr == daddr ||
- tmp->peer_addr == MCTP_ADDR_ANY) &&
- tmp->local_addr == saddr)
+ if (!(mctp_address_matches(tmp->peer_addr, daddr) &&
+ mctp_address_matches(tmp->local_addr, saddr)))
+ continue;
+
+ spin_lock(&tmp->lock);
+ /* key must still be valid. If we find a match, clear the
+ * potential tag value
+ */
+ if (tmp->valid)
tagbits &= ~(1 << tmp->tag);
+ spin_unlock(&tmp->lock);
if (!tagbits)
break;
@@ -504,16 +641,64 @@ static int mctp_alloc_local_tag(struct mctp_sock *msk,
if (tagbits) {
key->tag = __ffs(tagbits);
mctp_reserve_tag(net, key, msk);
+ trace_mctp_key_acquire(key);
+
+ key->manual_alloc = manual;
*tagp = key->tag;
- rc = 0;
}
spin_unlock_irqrestore(&mns->keys_lock, flags);
- if (!tagbits)
+ if (!tagbits) {
kfree(key);
+ return ERR_PTR(-EBUSY);
+ }
- return rc;
+ return key;
+}
+
+static struct mctp_sk_key *mctp_lookup_prealloc_tag(struct mctp_sock *msk,
+ mctp_eid_t daddr,
+ u8 req_tag, u8 *tagp)
+{
+ struct net *net = sock_net(&msk->sk);
+ struct netns_mctp *mns = &net->mctp;
+ struct mctp_sk_key *key, *tmp;
+ unsigned long flags;
+
+ req_tag &= ~(MCTP_TAG_PREALLOC | MCTP_TAG_OWNER);
+ key = NULL;
+
+ spin_lock_irqsave(&mns->keys_lock, flags);
+
+ hlist_for_each_entry(tmp, &mns->keys, hlist) {
+ if (tmp->tag != req_tag)
+ continue;
+
+ if (!mctp_address_matches(tmp->peer_addr, daddr))
+ continue;
+
+ if (!tmp->manual_alloc)
+ continue;
+
+ spin_lock(&tmp->lock);
+ if (tmp->valid) {
+ key = tmp;
+ refcount_inc(&key->refs);
+ spin_unlock(&tmp->lock);
+ break;
+ }
+ spin_unlock(&tmp->lock);
+ }
+ spin_unlock_irqrestore(&mns->keys_lock, flags);
+
+ if (!key)
+ return ERR_PTR(-ENOENT);
+
+ if (tagp)
+ *tagp = key->tag;
+
+ return key;
}
/* routing lookups */
@@ -552,14 +737,18 @@ struct mctp_route *mctp_route_lookup(struct net *net, unsigned int dnet,
return rt;
}
-/* sends a skb to rt and releases the route. */
-int mctp_do_route(struct mctp_route *rt, struct sk_buff *skb)
+static struct mctp_route *mctp_route_lookup_null(struct net *net,
+ struct net_device *dev)
{
- int rc;
+ struct mctp_route *rt;
- rc = rt->output(rt, skb);
- mctp_route_release(rt);
- return rc;
+ list_for_each_entry_rcu(rt, &net->mctp.routes, list) {
+ if (rt->dev->dev == dev && rt->type == RTN_LOCAL &&
+ refcount_inc_not_zero(&rt->refs))
+ return rt;
+ }
+
+ return NULL;
}
static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
@@ -628,7 +817,7 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
/* copy message payload */
skb_copy_bits(skb, pos, skb_transport_header(skb2), size);
- /* do route, but don't drop the rt reference */
+ /* do route */
rc = rt->output(rt, skb2);
if (rc)
break;
@@ -637,7 +826,6 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
pos += size;
}
- mctp_route_release(rt);
consume_skb(skb);
return rc;
}
@@ -647,15 +835,50 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt,
{
struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
struct mctp_skb_cb *cb = mctp_cb(skb);
+ struct mctp_route tmp_rt = {0};
+ struct mctp_sk_key *key;
struct mctp_hdr *hdr;
unsigned long flags;
unsigned int mtu;
mctp_eid_t saddr;
+ bool ext_rt;
int rc;
u8 tag;
- if (WARN_ON(!rt->dev))
+ rc = -ENODEV;
+
+ if (rt) {
+ ext_rt = false;
+ if (WARN_ON(!rt->dev))
+ goto out_release;
+
+ } else if (cb->ifindex) {
+ struct net_device *dev;
+
+ ext_rt = true;
+ rt = &tmp_rt;
+
+ rcu_read_lock();
+ dev = dev_get_by_index_rcu(sock_net(sk), cb->ifindex);
+ if (!dev) {
+ rcu_read_unlock();
+ return rc;
+ }
+ rt->dev = __mctp_dev_get(dev);
+ rcu_read_unlock();
+
+ if (!rt->dev)
+ goto out_release;
+
+ /* establish temporary route - we set up enough to keep
+ * mctp_route_output happy
+ */
+ rt->output = mctp_route_output;
+ rt->mtu = 0;
+
+ } else {
return -EINVAL;
+ }
spin_lock_irqsave(&rt->dev->addrs_lock, flags);
if (rt->dev->num_addrs == 0) {
@@ -668,18 +891,29 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt,
spin_unlock_irqrestore(&rt->dev->addrs_lock, flags);
if (rc)
- return rc;
+ goto out_release;
- if (req_tag & MCTP_HDR_FLAG_TO) {
- rc = mctp_alloc_local_tag(msk, saddr, daddr, &tag);
- if (rc)
- return rc;
+ if (req_tag & MCTP_TAG_OWNER) {
+ if (req_tag & MCTP_TAG_PREALLOC)
+ key = mctp_lookup_prealloc_tag(msk, daddr,
+ req_tag, &tag);
+ else
+ key = mctp_alloc_local_tag(msk, daddr, saddr,
+ false, &tag);
+
+ if (IS_ERR(key)) {
+ rc = PTR_ERR(key);
+ goto out_release;
+ }
+ mctp_skb_set_flow(skb, key);
+ /* done with the key in this scope */
+ mctp_key_unref(key);
tag |= MCTP_HDR_FLAG_TO;
} else {
- tag = req_tag;
+ key = NULL;
+ tag = req_tag & MCTP_TAG_MASK;
}
-
skb->protocol = htons(ETH_P_MCTP);
skb->priority = 0;
skb_reset_transport_header(skb);
@@ -699,12 +933,20 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt,
mtu = mctp_route_mtu(rt);
if (skb->len + sizeof(struct mctp_hdr) <= mtu) {
- hdr->flags_seq_tag = MCTP_HDR_FLAG_SOM | MCTP_HDR_FLAG_EOM |
- tag;
- return mctp_do_route(rt, skb);
+ hdr->flags_seq_tag = MCTP_HDR_FLAG_SOM |
+ MCTP_HDR_FLAG_EOM | tag;
+ rc = rt->output(rt, skb);
} else {
- return mctp_do_fragment_route(rt, skb, mtu, tag);
+ rc = mctp_do_fragment_route(rt, skb, mtu, tag);
}
+
+out_release:
+ if (!ext_rt)
+ mctp_route_release(rt);
+
+ mctp_dev_put(tmp_rt.dev);
+
+ return rc;
}
/* route management */
@@ -716,7 +958,7 @@ static int mctp_route_add(struct mctp_dev *mdev, mctp_eid_t daddr_start,
struct net *net = dev_net(mdev->dev);
struct mctp_route *rt, *ert;
- if (!mctp_address_ok(daddr_start))
+ if (!mctp_address_unicast(daddr_start))
return -EINVAL;
if (daddr_extent > 0xff || daddr_start + daddr_extent >= 255)
@@ -741,7 +983,7 @@ static int mctp_route_add(struct mctp_dev *mdev, mctp_eid_t daddr_start,
rt->max = daddr_start + daddr_extent;
rt->mtu = mtu;
rt->dev = mdev;
- dev_hold(rt->dev->dev);
+ mctp_dev_hold(rt->dev);
rt->type = type;
rt->output = rtfn;
@@ -822,13 +1064,18 @@ static int mctp_pkttype_receive(struct sk_buff *skb, struct net_device *dev,
struct net_device *orig_dev)
{
struct net *net = dev_net(dev);
+ struct mctp_dev *mdev;
struct mctp_skb_cb *cb;
struct mctp_route *rt;
struct mctp_hdr *mh;
- /* basic non-data sanity checks */
- if (dev->type != ARPHRD_MCTP)
+ rcu_read_lock();
+ mdev = __mctp_dev_get(dev);
+ rcu_read_unlock();
+ if (!mdev) {
+ /* basic non-data sanity checks */
goto err_drop;
+ }
if (!pskb_may_pull(skb, sizeof(struct mctp_hdr)))
goto err_drop;
@@ -841,21 +1088,45 @@ static int mctp_pkttype_receive(struct sk_buff *skb, struct net_device *dev,
if (mh->ver < MCTP_VER_MIN || mh->ver > MCTP_VER_MAX)
goto err_drop;
- cb = __mctp_cb(skb);
- rcu_read_lock();
- cb->net = READ_ONCE(__mctp_dev_get(dev)->net);
- rcu_read_unlock();
+ /* source must be valid unicast or null; drop reserved ranges and
+ * broadcast
+ */
+ if (!(mctp_address_unicast(mh->src) || mctp_address_null(mh->src)))
+ goto err_drop;
+
+ /* dest address: as above, but allow broadcast */
+ if (!(mctp_address_unicast(mh->dest) || mctp_address_null(mh->dest) ||
+ mctp_address_broadcast(mh->dest)))
+ goto err_drop;
+
+ /* MCTP drivers must populate halen/haddr */
+ if (dev->type == ARPHRD_MCTP) {
+ cb = mctp_cb(skb);
+ } else {
+ cb = __mctp_cb(skb);
+ cb->halen = 0;
+ }
+ cb->net = READ_ONCE(mdev->net);
+ cb->ifindex = dev->ifindex;
rt = mctp_route_lookup(net, cb->net, mh->dest);
+
+ /* NULL EID, but addressed to our physical address */
+ if (!rt && mh->dest == MCTP_ADDR_NULL && skb->pkt_type == PACKET_HOST)
+ rt = mctp_route_lookup_null(net, dev);
+
if (!rt)
goto err_drop;
- mctp_do_route(rt, skb);
+ rt->output(rt, skb);
+ mctp_route_release(rt);
+ mctp_dev_put(mdev);
return NET_RX_SUCCESS;
err_drop:
kfree_skb(skb);
+ mctp_dev_put(mdev);
return NET_RX_DROP;
}
@@ -927,10 +1198,15 @@ static int mctp_route_nlparse(struct sk_buff *skb, struct nlmsghdr *nlh,
return 0;
}
+static const struct nla_policy rta_metrics_policy[RTAX_MAX + 1] = {
+ [RTAX_MTU] = { .type = NLA_U32 },
+};
+
static int mctp_newroute(struct sk_buff *skb, struct nlmsghdr *nlh,
struct netlink_ext_ack *extack)
{
struct nlattr *tb[RTA_MAX + 1];
+ struct nlattr *tbx[RTAX_MAX + 1];
mctp_eid_t daddr_start;
struct mctp_dev *mdev;
struct rtmsg *rtm;
@@ -947,8 +1223,15 @@ static int mctp_newroute(struct sk_buff *skb, struct nlmsghdr *nlh,
return -EINVAL;
}
- /* TODO: parse mtu from nlparse */
mtu = 0;
+ if (tb[RTA_METRICS]) {
+ rc = nla_parse_nested(tbx, RTAX_MAX, tb[RTA_METRICS],
+ rta_metrics_policy, NULL);
+ if (rc < 0)
+ return rc;
+ if (tbx[RTAX_MTU])
+ mtu = nla_get_u32(tbx[RTAX_MTU]);
+ }
if (rtm->rtm_type != RTN_UNICAST)
return -EINVAL;
@@ -1117,3 +1400,7 @@ void __exit mctp_routes_exit(void)
rtnl_unregister(PF_MCTP, RTM_GETROUTE);
dev_remove_pack(&mctp_packet_type);
}
+
+#if IS_ENABLED(CONFIG_MCTP_TEST)
+#include "test/route-test.c"
+#endif
diff --git a/net/mctp/test/route-test.c b/net/mctp/test/route-test.c
new file mode 100644
index 000000000000..61205cf40074
--- /dev/null
+++ b/net/mctp/test/route-test.c
@@ -0,0 +1,684 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <kunit/test.h>
+
+#include "utils.h"
+
+struct mctp_test_route {
+ struct mctp_route rt;
+ struct sk_buff_head pkts;
+};
+
+static int mctp_test_route_output(struct mctp_route *rt, struct sk_buff *skb)
+{
+ struct mctp_test_route *test_rt = container_of(rt, struct mctp_test_route, rt);
+
+ skb_queue_tail(&test_rt->pkts, skb);
+
+ return 0;
+}
+
+/* local version of mctp_route_alloc() */
+static struct mctp_test_route *mctp_route_test_alloc(void)
+{
+ struct mctp_test_route *rt;
+
+ rt = kzalloc(sizeof(*rt), GFP_KERNEL);
+ if (!rt)
+ return NULL;
+
+ INIT_LIST_HEAD(&rt->rt.list);
+ refcount_set(&rt->rt.refs, 1);
+ rt->rt.output = mctp_test_route_output;
+
+ skb_queue_head_init(&rt->pkts);
+
+ return rt;
+}
+
+static struct mctp_test_route *mctp_test_create_route(struct net *net,
+ struct mctp_dev *dev,
+ mctp_eid_t eid,
+ unsigned int mtu)
+{
+ struct mctp_test_route *rt;
+
+ rt = mctp_route_test_alloc();
+ if (!rt)
+ return NULL;
+
+ rt->rt.min = eid;
+ rt->rt.max = eid;
+ rt->rt.mtu = mtu;
+ rt->rt.type = RTN_UNSPEC;
+ if (dev)
+ mctp_dev_hold(dev);
+ rt->rt.dev = dev;
+
+ list_add_rcu(&rt->rt.list, &net->mctp.routes);
+
+ return rt;
+}
+
+static void mctp_test_route_destroy(struct kunit *test,
+ struct mctp_test_route *rt)
+{
+ unsigned int refs;
+
+ rtnl_lock();
+ list_del_rcu(&rt->rt.list);
+ rtnl_unlock();
+
+ skb_queue_purge(&rt->pkts);
+ if (rt->rt.dev)
+ mctp_dev_put(rt->rt.dev);
+
+ refs = refcount_read(&rt->rt.refs);
+ KUNIT_ASSERT_EQ_MSG(test, refs, 1, "route ref imbalance");
+
+ kfree_rcu(&rt->rt, rcu);
+}
+
+static struct sk_buff *mctp_test_create_skb(const struct mctp_hdr *hdr,
+ unsigned int data_len)
+{
+ size_t hdr_len = sizeof(*hdr);
+ struct sk_buff *skb;
+ unsigned int i;
+ u8 *buf;
+
+ skb = alloc_skb(hdr_len + data_len, GFP_KERNEL);
+ if (!skb)
+ return NULL;
+
+ memcpy(skb_put(skb, hdr_len), hdr, hdr_len);
+
+ buf = skb_put(skb, data_len);
+ for (i = 0; i < data_len; i++)
+ buf[i] = i & 0xff;
+
+ return skb;
+}
+
+static struct sk_buff *__mctp_test_create_skb_data(const struct mctp_hdr *hdr,
+ const void *data,
+ size_t data_len)
+{
+ size_t hdr_len = sizeof(*hdr);
+ struct sk_buff *skb;
+
+ skb = alloc_skb(hdr_len + data_len, GFP_KERNEL);
+ if (!skb)
+ return NULL;
+
+ memcpy(skb_put(skb, hdr_len), hdr, hdr_len);
+ memcpy(skb_put(skb, data_len), data, data_len);
+
+ return skb;
+}
+
+#define mctp_test_create_skb_data(h, d) \
+ __mctp_test_create_skb_data(h, d, sizeof(*d))
+
+struct mctp_frag_test {
+ unsigned int mtu;
+ unsigned int msgsize;
+ unsigned int n_frags;
+};
+
+static void mctp_test_fragment(struct kunit *test)
+{
+ const struct mctp_frag_test *params;
+ int rc, i, n, mtu, msgsize;
+ struct mctp_test_route *rt;
+ struct sk_buff *skb;
+ struct mctp_hdr hdr;
+ u8 seq;
+
+ params = test->param_value;
+ mtu = params->mtu;
+ msgsize = params->msgsize;
+
+ hdr.ver = 1;
+ hdr.src = 8;
+ hdr.dest = 10;
+ hdr.flags_seq_tag = MCTP_HDR_FLAG_TO;
+
+ skb = mctp_test_create_skb(&hdr, msgsize);
+ KUNIT_ASSERT_TRUE(test, skb);
+
+ rt = mctp_test_create_route(&init_net, NULL, 10, mtu);
+ KUNIT_ASSERT_TRUE(test, rt);
+
+ rc = mctp_do_fragment_route(&rt->rt, skb, mtu, MCTP_TAG_OWNER);
+ KUNIT_EXPECT_FALSE(test, rc);
+
+ n = rt->pkts.qlen;
+
+ KUNIT_EXPECT_EQ(test, n, params->n_frags);
+
+ for (i = 0;; i++) {
+ struct mctp_hdr *hdr2;
+ struct sk_buff *skb2;
+ u8 tag_mask, seq2;
+ bool first, last;
+
+ first = i == 0;
+ last = i == (n - 1);
+
+ skb2 = skb_dequeue(&rt->pkts);
+
+ if (!skb2)
+ break;
+
+ hdr2 = mctp_hdr(skb2);
+
+ tag_mask = MCTP_HDR_TAG_MASK | MCTP_HDR_FLAG_TO;
+
+ KUNIT_EXPECT_EQ(test, hdr2->ver, hdr.ver);
+ KUNIT_EXPECT_EQ(test, hdr2->src, hdr.src);
+ KUNIT_EXPECT_EQ(test, hdr2->dest, hdr.dest);
+ KUNIT_EXPECT_EQ(test, hdr2->flags_seq_tag & tag_mask,
+ hdr.flags_seq_tag & tag_mask);
+
+ KUNIT_EXPECT_EQ(test,
+ !!(hdr2->flags_seq_tag & MCTP_HDR_FLAG_SOM), first);
+ KUNIT_EXPECT_EQ(test,
+ !!(hdr2->flags_seq_tag & MCTP_HDR_FLAG_EOM), last);
+
+ seq2 = (hdr2->flags_seq_tag >> MCTP_HDR_SEQ_SHIFT) &
+ MCTP_HDR_SEQ_MASK;
+
+ if (first) {
+ seq = seq2;
+ } else {
+ seq++;
+ KUNIT_EXPECT_EQ(test, seq2, seq & MCTP_HDR_SEQ_MASK);
+ }
+
+ if (!last)
+ KUNIT_EXPECT_EQ(test, skb2->len, mtu);
+ else
+ KUNIT_EXPECT_LE(test, skb2->len, mtu);
+
+ kfree_skb(skb2);
+ }
+
+ mctp_test_route_destroy(test, rt);
+}
+
+static const struct mctp_frag_test mctp_frag_tests[] = {
+ {.mtu = 68, .msgsize = 63, .n_frags = 1},
+ {.mtu = 68, .msgsize = 64, .n_frags = 1},
+ {.mtu = 68, .msgsize = 65, .n_frags = 2},
+ {.mtu = 68, .msgsize = 66, .n_frags = 2},
+ {.mtu = 68, .msgsize = 127, .n_frags = 2},
+ {.mtu = 68, .msgsize = 128, .n_frags = 2},
+ {.mtu = 68, .msgsize = 129, .n_frags = 3},
+ {.mtu = 68, .msgsize = 130, .n_frags = 3},
+};
+
+static void mctp_frag_test_to_desc(const struct mctp_frag_test *t, char *desc)
+{
+ sprintf(desc, "mtu %d len %d -> %d frags",
+ t->msgsize, t->mtu, t->n_frags);
+}
+
+KUNIT_ARRAY_PARAM(mctp_frag, mctp_frag_tests, mctp_frag_test_to_desc);
+
+struct mctp_rx_input_test {
+ struct mctp_hdr hdr;
+ bool input;
+};
+
+static void mctp_test_rx_input(struct kunit *test)
+{
+ const struct mctp_rx_input_test *params;
+ struct mctp_test_route *rt;
+ struct mctp_test_dev *dev;
+ struct sk_buff *skb;
+
+ params = test->param_value;
+
+ dev = mctp_test_create_dev();
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev);
+
+ rt = mctp_test_create_route(&init_net, dev->mdev, 8, 68);
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, rt);
+
+ skb = mctp_test_create_skb(&params->hdr, 1);
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb);
+
+ __mctp_cb(skb);
+
+ mctp_pkttype_receive(skb, dev->ndev, &mctp_packet_type, NULL);
+
+ KUNIT_EXPECT_EQ(test, !!rt->pkts.qlen, params->input);
+
+ mctp_test_route_destroy(test, rt);
+ mctp_test_destroy_dev(dev);
+}
+
+#define RX_HDR(_ver, _src, _dest, _fst) \
+ { .ver = _ver, .src = _src, .dest = _dest, .flags_seq_tag = _fst }
+
+/* we have a route for EID 8 only */
+static const struct mctp_rx_input_test mctp_rx_input_tests[] = {
+ { .hdr = RX_HDR(1, 10, 8, 0), .input = true },
+ { .hdr = RX_HDR(1, 10, 9, 0), .input = false }, /* no input route */
+ { .hdr = RX_HDR(2, 10, 8, 0), .input = false }, /* invalid version */
+};
+
+static void mctp_rx_input_test_to_desc(const struct mctp_rx_input_test *t,
+ char *desc)
+{
+ sprintf(desc, "{%x,%x,%x,%x}", t->hdr.ver, t->hdr.src, t->hdr.dest,
+ t->hdr.flags_seq_tag);
+}
+
+KUNIT_ARRAY_PARAM(mctp_rx_input, mctp_rx_input_tests,
+ mctp_rx_input_test_to_desc);
+
+/* set up a local dev, route on EID 8, and a socket listening on type 0 */
+static void __mctp_route_test_init(struct kunit *test,
+ struct mctp_test_dev **devp,
+ struct mctp_test_route **rtp,
+ struct socket **sockp)
+{
+ struct sockaddr_mctp addr = {0};
+ struct mctp_test_route *rt;
+ struct mctp_test_dev *dev;
+ struct socket *sock;
+ int rc;
+
+ dev = mctp_test_create_dev();
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev);
+
+ rt = mctp_test_create_route(&init_net, dev->mdev, 8, 68);
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, rt);
+
+ rc = sock_create_kern(&init_net, AF_MCTP, SOCK_DGRAM, 0, &sock);
+ KUNIT_ASSERT_EQ(test, rc, 0);
+
+ addr.smctp_family = AF_MCTP;
+ addr.smctp_network = MCTP_NET_ANY;
+ addr.smctp_addr.s_addr = 8;
+ addr.smctp_type = 0;
+ rc = kernel_bind(sock, (struct sockaddr *)&addr, sizeof(addr));
+ KUNIT_ASSERT_EQ(test, rc, 0);
+
+ *rtp = rt;
+ *devp = dev;
+ *sockp = sock;
+}
+
+static void __mctp_route_test_fini(struct kunit *test,
+ struct mctp_test_dev *dev,
+ struct mctp_test_route *rt,
+ struct socket *sock)
+{
+ sock_release(sock);
+ mctp_test_route_destroy(test, rt);
+ mctp_test_destroy_dev(dev);
+}
+
+struct mctp_route_input_sk_test {
+ struct mctp_hdr hdr;
+ u8 type;
+ bool deliver;
+};
+
+static void mctp_test_route_input_sk(struct kunit *test)
+{
+ const struct mctp_route_input_sk_test *params;
+ struct sk_buff *skb, *skb2;
+ struct mctp_test_route *rt;
+ struct mctp_test_dev *dev;
+ struct socket *sock;
+ int rc;
+
+ params = test->param_value;
+
+ __mctp_route_test_init(test, &dev, &rt, &sock);
+
+ skb = mctp_test_create_skb_data(&params->hdr, &params->type);
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb);
+
+ skb->dev = dev->ndev;
+ __mctp_cb(skb);
+
+ rc = mctp_route_input(&rt->rt, skb);
+
+ if (params->deliver) {
+ KUNIT_EXPECT_EQ(test, rc, 0);
+
+ skb2 = skb_recv_datagram(sock->sk, 0, 1, &rc);
+ KUNIT_EXPECT_NOT_ERR_OR_NULL(test, skb2);
+ KUNIT_EXPECT_EQ(test, skb->len, 1);
+
+ skb_free_datagram(sock->sk, skb2);
+
+ } else {
+ KUNIT_EXPECT_NE(test, rc, 0);
+ skb2 = skb_recv_datagram(sock->sk, 0, 1, &rc);
+ KUNIT_EXPECT_PTR_EQ(test, skb2, NULL);
+ }
+
+ __mctp_route_test_fini(test, dev, rt, sock);
+}
+
+#define FL_S (MCTP_HDR_FLAG_SOM)
+#define FL_E (MCTP_HDR_FLAG_EOM)
+#define FL_TO (MCTP_HDR_FLAG_TO)
+#define FL_T(t) ((t) & MCTP_HDR_TAG_MASK)
+
+static const struct mctp_route_input_sk_test mctp_route_input_sk_tests[] = {
+ { .hdr = RX_HDR(1, 10, 8, FL_S | FL_E | FL_TO), .type = 0, .deliver = true },
+ { .hdr = RX_HDR(1, 10, 8, FL_S | FL_E | FL_TO), .type = 1, .deliver = false },
+ { .hdr = RX_HDR(1, 10, 8, FL_S | FL_E), .type = 0, .deliver = false },
+ { .hdr = RX_HDR(1, 10, 8, FL_E | FL_TO), .type = 0, .deliver = false },
+ { .hdr = RX_HDR(1, 10, 8, FL_TO), .type = 0, .deliver = false },
+ { .hdr = RX_HDR(1, 10, 8, 0), .type = 0, .deliver = false },
+};
+
+static void mctp_route_input_sk_to_desc(const struct mctp_route_input_sk_test *t,
+ char *desc)
+{
+ sprintf(desc, "{%x,%x,%x,%x} type %d", t->hdr.ver, t->hdr.src,
+ t->hdr.dest, t->hdr.flags_seq_tag, t->type);
+}
+
+KUNIT_ARRAY_PARAM(mctp_route_input_sk, mctp_route_input_sk_tests,
+ mctp_route_input_sk_to_desc);
+
+struct mctp_route_input_sk_reasm_test {
+ const char *name;
+ struct mctp_hdr hdrs[4];
+ int n_hdrs;
+ int rx_len;
+};
+
+static void mctp_test_route_input_sk_reasm(struct kunit *test)
+{
+ const struct mctp_route_input_sk_reasm_test *params;
+ struct sk_buff *skb, *skb2;
+ struct mctp_test_route *rt;
+ struct mctp_test_dev *dev;
+ struct socket *sock;
+ int i, rc;
+ u8 c;
+
+ params = test->param_value;
+
+ __mctp_route_test_init(test, &dev, &rt, &sock);
+
+ for (i = 0; i < params->n_hdrs; i++) {
+ c = i;
+ skb = mctp_test_create_skb_data(&params->hdrs[i], &c);
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb);
+
+ skb->dev = dev->ndev;
+ __mctp_cb(skb);
+
+ rc = mctp_route_input(&rt->rt, skb);
+ }
+
+ skb2 = skb_recv_datagram(sock->sk, 0, 1, &rc);
+
+ if (params->rx_len) {
+ KUNIT_EXPECT_NOT_ERR_OR_NULL(test, skb2);
+ KUNIT_EXPECT_EQ(test, skb2->len, params->rx_len);
+ skb_free_datagram(sock->sk, skb2);
+
+ } else {
+ KUNIT_EXPECT_PTR_EQ(test, skb2, NULL);
+ }
+
+ __mctp_route_test_fini(test, dev, rt, sock);
+}
+
+#define RX_FRAG(f, s) RX_HDR(1, 10, 8, FL_TO | (f) | ((s) << MCTP_HDR_SEQ_SHIFT))
+
+static const struct mctp_route_input_sk_reasm_test mctp_route_input_sk_reasm_tests[] = {
+ {
+ .name = "single packet",
+ .hdrs = {
+ RX_FRAG(FL_S | FL_E, 0),
+ },
+ .n_hdrs = 1,
+ .rx_len = 1,
+ },
+ {
+ .name = "single packet, offset seq",
+ .hdrs = {
+ RX_FRAG(FL_S | FL_E, 1),
+ },
+ .n_hdrs = 1,
+ .rx_len = 1,
+ },
+ {
+ .name = "start & end packets",
+ .hdrs = {
+ RX_FRAG(FL_S, 0),
+ RX_FRAG(FL_E, 1),
+ },
+ .n_hdrs = 2,
+ .rx_len = 2,
+ },
+ {
+ .name = "start & end packets, offset seq",
+ .hdrs = {
+ RX_FRAG(FL_S, 1),
+ RX_FRAG(FL_E, 2),
+ },
+ .n_hdrs = 2,
+ .rx_len = 2,
+ },
+ {
+ .name = "start & end packets, out of order",
+ .hdrs = {
+ RX_FRAG(FL_E, 1),
+ RX_FRAG(FL_S, 0),
+ },
+ .n_hdrs = 2,
+ .rx_len = 0,
+ },
+ {
+ .name = "start, middle & end packets",
+ .hdrs = {
+ RX_FRAG(FL_S, 0),
+ RX_FRAG(0, 1),
+ RX_FRAG(FL_E, 2),
+ },
+ .n_hdrs = 3,
+ .rx_len = 3,
+ },
+ {
+ .name = "missing seq",
+ .hdrs = {
+ RX_FRAG(FL_S, 0),
+ RX_FRAG(FL_E, 2),
+ },
+ .n_hdrs = 2,
+ .rx_len = 0,
+ },
+ {
+ .name = "seq wrap",
+ .hdrs = {
+ RX_FRAG(FL_S, 3),
+ RX_FRAG(FL_E, 0),
+ },
+ .n_hdrs = 2,
+ .rx_len = 2,
+ },
+};
+
+static void mctp_route_input_sk_reasm_to_desc(
+ const struct mctp_route_input_sk_reasm_test *t,
+ char *desc)
+{
+ sprintf(desc, "%s", t->name);
+}
+
+KUNIT_ARRAY_PARAM(mctp_route_input_sk_reasm, mctp_route_input_sk_reasm_tests,
+ mctp_route_input_sk_reasm_to_desc);
+
+struct mctp_route_input_sk_keys_test {
+ const char *name;
+ mctp_eid_t key_peer_addr;
+ mctp_eid_t key_local_addr;
+ u8 key_tag;
+ struct mctp_hdr hdr;
+ bool deliver;
+};
+
+/* test packet rx in the presence of various key configurations */
+static void mctp_test_route_input_sk_keys(struct kunit *test)
+{
+ const struct mctp_route_input_sk_keys_test *params;
+ struct mctp_test_route *rt;
+ struct sk_buff *skb, *skb2;
+ struct mctp_test_dev *dev;
+ struct mctp_sk_key *key;
+ struct netns_mctp *mns;
+ struct mctp_sock *msk;
+ struct socket *sock;
+ unsigned long flags;
+ int rc;
+ u8 c;
+
+ params = test->param_value;
+
+ dev = mctp_test_create_dev();
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev);
+
+ rt = mctp_test_create_route(&init_net, dev->mdev, 8, 68);
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, rt);
+
+ rc = sock_create_kern(&init_net, AF_MCTP, SOCK_DGRAM, 0, &sock);
+ KUNIT_ASSERT_EQ(test, rc, 0);
+
+ msk = container_of(sock->sk, struct mctp_sock, sk);
+ mns = &sock_net(sock->sk)->mctp;
+
+ /* set the incoming tag according to test params */
+ key = mctp_key_alloc(msk, params->key_local_addr, params->key_peer_addr,
+ params->key_tag, GFP_KERNEL);
+
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, key);
+
+ spin_lock_irqsave(&mns->keys_lock, flags);
+ mctp_reserve_tag(&init_net, key, msk);
+ spin_unlock_irqrestore(&mns->keys_lock, flags);
+
+ /* create packet and route */
+ c = 0;
+ skb = mctp_test_create_skb_data(&params->hdr, &c);
+ KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb);
+
+ skb->dev = dev->ndev;
+ __mctp_cb(skb);
+
+ rc = mctp_route_input(&rt->rt, skb);
+
+ /* (potentially) receive message */
+ skb2 = skb_recv_datagram(sock->sk, 0, 1, &rc);
+
+ if (params->deliver)
+ KUNIT_EXPECT_NOT_ERR_OR_NULL(test, skb2);
+ else
+ KUNIT_EXPECT_PTR_EQ(test, skb2, NULL);
+
+ if (skb2)
+ skb_free_datagram(sock->sk, skb2);
+
+ mctp_key_unref(key);
+ __mctp_route_test_fini(test, dev, rt, sock);
+}
+
+static const struct mctp_route_input_sk_keys_test mctp_route_input_sk_keys_tests[] = {
+ {
+ .name = "direct match",
+ .key_peer_addr = 9,
+ .key_local_addr = 8,
+ .key_tag = 1,
+ .hdr = RX_HDR(1, 9, 8, FL_S | FL_E | FL_T(1)),
+ .deliver = true,
+ },
+ {
+ .name = "flipped src/dest",
+ .key_peer_addr = 8,
+ .key_local_addr = 9,
+ .key_tag = 1,
+ .hdr = RX_HDR(1, 9, 8, FL_S | FL_E | FL_T(1)),
+ .deliver = false,
+ },
+ {
+ .name = "peer addr mismatch",
+ .key_peer_addr = 9,
+ .key_local_addr = 8,
+ .key_tag = 1,
+ .hdr = RX_HDR(1, 10, 8, FL_S | FL_E | FL_T(1)),
+ .deliver = false,
+ },
+ {
+ .name = "tag value mismatch",
+ .key_peer_addr = 9,
+ .key_local_addr = 8,
+ .key_tag = 1,
+ .hdr = RX_HDR(1, 9, 8, FL_S | FL_E | FL_T(2)),
+ .deliver = false,
+ },
+ {
+ .name = "TO mismatch",
+ .key_peer_addr = 9,
+ .key_local_addr = 8,
+ .key_tag = 1,
+ .hdr = RX_HDR(1, 9, 8, FL_S | FL_E | FL_T(1) | FL_TO),
+ .deliver = false,
+ },
+ {
+ .name = "broadcast response",
+ .key_peer_addr = MCTP_ADDR_ANY,
+ .key_local_addr = 8,
+ .key_tag = 1,
+ .hdr = RX_HDR(1, 11, 8, FL_S | FL_E | FL_T(1)),
+ .deliver = true,
+ },
+ {
+ .name = "any local match",
+ .key_peer_addr = 12,
+ .key_local_addr = MCTP_ADDR_ANY,
+ .key_tag = 1,
+ .hdr = RX_HDR(1, 12, 8, FL_S | FL_E | FL_T(1)),
+ .deliver = true,
+ },
+};
+
+static void mctp_route_input_sk_keys_to_desc(
+ const struct mctp_route_input_sk_keys_test *t,
+ char *desc)
+{
+ sprintf(desc, "%s", t->name);
+}
+
+KUNIT_ARRAY_PARAM(mctp_route_input_sk_keys, mctp_route_input_sk_keys_tests,
+ mctp_route_input_sk_keys_to_desc);
+
+static struct kunit_case mctp_test_cases[] = {
+ KUNIT_CASE_PARAM(mctp_test_fragment, mctp_frag_gen_params),
+ KUNIT_CASE_PARAM(mctp_test_rx_input, mctp_rx_input_gen_params),
+ KUNIT_CASE_PARAM(mctp_test_route_input_sk, mctp_route_input_sk_gen_params),
+ KUNIT_CASE_PARAM(mctp_test_route_input_sk_reasm,
+ mctp_route_input_sk_reasm_gen_params),
+ KUNIT_CASE_PARAM(mctp_test_route_input_sk_keys,
+ mctp_route_input_sk_keys_gen_params),
+ {}
+};
+
+static struct kunit_suite mctp_test_suite = {
+ .name = "mctp",
+ .test_cases = mctp_test_cases,
+};
+
+kunit_test_suite(mctp_test_suite);
diff --git a/net/mctp/test/utils.c b/net/mctp/test/utils.c
new file mode 100644
index 000000000000..e03ba66bbe18
--- /dev/null
+++ b/net/mctp/test/utils.c
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/netdevice.h>
+#include <linux/mctp.h>
+#include <linux/if_arp.h>
+
+#include <net/mctpdevice.h>
+#include <net/pkt_sched.h>
+
+#include "utils.h"
+
+static netdev_tx_t mctp_test_dev_tx(struct sk_buff *skb,
+ struct net_device *ndev)
+{
+ kfree_skb(skb);
+ return NETDEV_TX_OK;
+}
+
+static const struct net_device_ops mctp_test_netdev_ops = {
+ .ndo_start_xmit = mctp_test_dev_tx,
+};
+
+static void mctp_test_dev_setup(struct net_device *ndev)
+{
+ ndev->type = ARPHRD_MCTP;
+ ndev->mtu = MCTP_DEV_TEST_MTU;
+ ndev->hard_header_len = 0;
+ ndev->addr_len = 0;
+ ndev->tx_queue_len = DEFAULT_TX_QUEUE_LEN;
+ ndev->flags = IFF_NOARP;
+ ndev->netdev_ops = &mctp_test_netdev_ops;
+ ndev->needs_free_netdev = true;
+}
+
+struct mctp_test_dev *mctp_test_create_dev(void)
+{
+ struct mctp_test_dev *dev;
+ struct net_device *ndev;
+ int rc;
+
+ ndev = alloc_netdev(sizeof(*dev), "mctptest%d", NET_NAME_ENUM,
+ mctp_test_dev_setup);
+ if (!ndev)
+ return NULL;
+
+ dev = netdev_priv(ndev);
+ dev->ndev = ndev;
+
+ rc = register_netdev(ndev);
+ if (rc) {
+ free_netdev(ndev);
+ return NULL;
+ }
+
+ rcu_read_lock();
+ dev->mdev = __mctp_dev_get(ndev);
+ rcu_read_unlock();
+
+ return dev;
+}
+
+void mctp_test_destroy_dev(struct mctp_test_dev *dev)
+{
+ mctp_dev_put(dev->mdev);
+ unregister_netdev(dev->ndev);
+}
diff --git a/net/mctp/test/utils.h b/net/mctp/test/utils.h
new file mode 100644
index 000000000000..df6aa1c03440
--- /dev/null
+++ b/net/mctp/test/utils.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef __NET_MCTP_TEST_UTILS_H
+#define __NET_MCTP_TEST_UTILS_H
+
+#include <kunit/test.h>
+
+#define MCTP_DEV_TEST_MTU 68
+
+struct mctp_test_dev {
+ struct net_device *ndev;
+ struct mctp_dev *mdev;
+};
+
+struct mctp_test_dev;
+
+struct mctp_test_dev *mctp_test_create_dev(void);
+void mctp_test_destroy_dev(struct mctp_test_dev *dev);
+
+#endif /* __NET_MCTP_TEST_UTILS_H */
generated by cgit v1.2.3 (git 2.39.0) at 2024-06-12 06:33:19 +0300