From: Matt Johnston <matt@codeconstruct.com.au>
To: Jeremy Kerr <jk@codeconstruct.com.au>,
"David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Jakub Kicinski <kuba@kernel.org>,
Paolo Abeni <pabeni@redhat.com>, Simon Horman <horms@kernel.org>
Cc: netdev@vger.kernel.org, Matt Johnston <matt@codeconstruct.com.au>
Subject: [PATCH net-next v4 5/8] net: mctp: Use hashtable for binds
Date: Thu, 10 Jul 2025 16:55:58 +0800 [thread overview]
Message-ID: <20250710-mctp-bind-v4-5-8ec2f6460c56@codeconstruct.com.au> (raw)
In-Reply-To: <20250710-mctp-bind-v4-0-8ec2f6460c56@codeconstruct.com.au>
Ensure that a specific EID (remote or local) bind will match in
preference to a MCTP_ADDR_ANY bind.
This adds infrastructure for binding a socket to receive messages from a
specific remote peer address, a future commit will expose an API for
this.
Signed-off-by: Matt Johnston <matt@codeconstruct.com.au>
---
v2:
- Use DECLARE_HASHTABLE
- Fix long lines
---
include/net/netns/mctp.h | 20 +++++++++---
net/mctp/af_mctp.c | 11 ++++---
net/mctp/route.c | 81 ++++++++++++++++++++++++++++++++++++++----------
3 files changed, 87 insertions(+), 25 deletions(-)
diff --git a/include/net/netns/mctp.h b/include/net/netns/mctp.h
index 1db8f9aaddb4b96f4803df9f30a762f5f88d7f7f..89555f90b97b297e50a571b26c5232b824909da7 100644
--- a/include/net/netns/mctp.h
+++ b/include/net/netns/mctp.h
@@ -6,19 +6,25 @@
#ifndef __NETNS_MCTP_H__
#define __NETNS_MCTP_H__
+#include <linux/hash.h>
+#include <linux/hashtable.h>
#include <linux/mutex.h>
#include <linux/types.h>
+#define MCTP_BINDS_BITS 7
+
struct netns_mctp {
/* Only updated under RTNL, entries freed via RCU */
struct list_head routes;
- /* Bound sockets: list of sockets bound by type.
- * This list is updated from non-atomic contexts (under bind_lock),
- * and read (under rcu) in packet rx
+ /* Bound sockets: hash table of sockets, keyed by
+ * (type, src_eid, dest_eid).
+ * Specific src_eid/dest_eid entries also have an entry for
+ * MCTP_ADDR_ANY. This list is updated from non-atomic contexts
+ * (under bind_lock), and read (under rcu) in packet rx.
*/
struct mutex bind_lock;
- struct hlist_head binds;
+ DECLARE_HASHTABLE(binds, MCTP_BINDS_BITS);
/* tag allocations. This list is read and updated from atomic contexts,
* but elements are free()ed after a RCU grace-period
@@ -34,4 +40,10 @@ struct netns_mctp {
struct list_head neighbours;
};
+static inline u32 mctp_bind_hash(u8 type, u8 local_addr, u8 peer_addr)
+{
+ return hash_32(type | (u32)local_addr << 8 | (u32)peer_addr << 16,
+ MCTP_BINDS_BITS);
+}
+
#endif /* __NETNS_MCTP_H__ */
diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c
index 20edaf840a607700c04b740708763fbd02a2df47..16341de5cf2893bbc04a8c05a038c30be6570296 100644
--- a/net/mctp/af_mctp.c
+++ b/net/mctp/af_mctp.c
@@ -626,17 +626,17 @@ static int mctp_sk_hash(struct sock *sk)
struct net *net = sock_net(sk);
struct sock *existing;
struct mctp_sock *msk;
+ u32 hash;
int rc;
msk = container_of(sk, struct mctp_sock, sk);
- /* Bind lookup runs under RCU, remain live during that. */
- sock_set_flag(sk, SOCK_RCU_FREE);
+ hash = mctp_bind_hash(msk->bind_type, msk->bind_addr, MCTP_ADDR_ANY);
mutex_lock(&net->mctp.bind_lock);
/* Prevent duplicate binds. */
- sk_for_each(existing, &net->mctp.binds) {
+ sk_for_each(existing, &net->mctp.binds[hash]) {
struct mctp_sock *mex =
container_of(existing, struct mctp_sock, sk);
@@ -648,7 +648,10 @@ static int mctp_sk_hash(struct sock *sk)
}
}
- sk_add_node_rcu(sk, &net->mctp.binds);
+ /* Bind lookup runs under RCU, remain live during that. */
+ sock_set_flag(sk, SOCK_RCU_FREE);
+
+ sk_add_node_rcu(sk, &net->mctp.binds[hash]);
rc = 0;
out:
diff --git a/net/mctp/route.c b/net/mctp/route.c
index a20d6b11d4186b55cab9d76e367169ea712553c7..69cfb0e6c545c2b44e5defdfac4e602c4f0265b1 100644
--- a/net/mctp/route.c
+++ b/net/mctp/route.c
@@ -40,14 +40,45 @@ static int mctp_dst_discard(struct mctp_dst *dst, struct sk_buff *skb)
return 0;
}
-static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
+static struct mctp_sock *mctp_lookup_bind_details(struct net *net,
+ struct sk_buff *skb,
+ u8 type, u8 dest,
+ u8 src, bool allow_net_any)
{
struct mctp_skb_cb *cb = mctp_cb(skb);
- struct mctp_hdr *mh;
struct sock *sk;
- u8 type;
+ u8 hash;
- WARN_ON(!rcu_read_lock_held());
+ WARN_ON_ONCE(!rcu_read_lock_held());
+
+ hash = mctp_bind_hash(type, dest, src);
+
+ sk_for_each_rcu(sk, &net->mctp.binds[hash]) {
+ struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
+
+ if (!allow_net_any && msk->bind_net == MCTP_NET_ANY)
+ continue;
+
+ if (msk->bind_net != MCTP_NET_ANY && msk->bind_net != cb->net)
+ continue;
+
+ if (msk->bind_type != type)
+ continue;
+
+ if (!mctp_address_matches(msk->bind_addr, dest))
+ continue;
+
+ return msk;
+ }
+
+ return NULL;
+}
+
+static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
+{
+ struct mctp_sock *msk;
+ struct mctp_hdr *mh;
+ u8 type;
/* TODO: look up in skb->cb? */
mh = mctp_hdr(skb);
@@ -57,20 +88,36 @@ static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
type = (*(u8 *)skb->data) & 0x7f;
- sk_for_each_rcu(sk, &net->mctp.binds) {
- struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
-
- if (msk->bind_net != MCTP_NET_ANY && msk->bind_net != cb->net)
- continue;
-
- if (msk->bind_type != type)
- continue;
-
- if (!mctp_address_matches(msk->bind_addr, mh->dest))
- continue;
+ /* Look for binds in order of widening scope. A given destination or
+ * source address also implies matching on a particular network.
+ *
+ * - Matching destination and source
+ * - Matching destination
+ * - Matching source
+ * - Matching network, any address
+ * - Any network or address
+ */
+ msk = mctp_lookup_bind_details(net, skb, type, mh->dest, mh->src,
+ false);
+ if (msk)
+ return msk;
+ msk = mctp_lookup_bind_details(net, skb, type, MCTP_ADDR_ANY, mh->src,
+ false);
+ if (msk)
+ return msk;
+ msk = mctp_lookup_bind_details(net, skb, type, mh->dest, MCTP_ADDR_ANY,
+ false);
+ if (msk)
+ return msk;
+ msk = mctp_lookup_bind_details(net, skb, type, MCTP_ADDR_ANY,
+ MCTP_ADDR_ANY, false);
+ if (msk)
+ return msk;
+ msk = mctp_lookup_bind_details(net, skb, type, MCTP_ADDR_ANY,
+ MCTP_ADDR_ANY, true);
+ if (msk)
return msk;
- }
return NULL;
}
@@ -1671,7 +1718,7 @@ static int __net_init mctp_routes_net_init(struct net *net)
struct netns_mctp *ns = &net->mctp;
INIT_LIST_HEAD(&ns->routes);
- INIT_HLIST_HEAD(&ns->binds);
+ hash_init(ns->binds);
mutex_init(&ns->bind_lock);
INIT_HLIST_HEAD(&ns->keys);
spin_lock_init(&ns->keys_lock);
--
2.43.0
next prev parent reply other threads:[~2025-07-10 8:56 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-10 8:55 [PATCH net-next v4 0/8] net: mctp: Improved bind handling Matt Johnston
2025-07-10 8:55 ` [PATCH net-next v4 1/8] net: mctp: mctp_test_route_extaddr_input cleanup Matt Johnston
2025-07-10 8:55 ` [PATCH net-next v4 2/8] net: mctp: Prevent duplicate binds Matt Johnston
2025-07-15 9:50 ` Paolo Abeni
2025-07-10 8:55 ` [PATCH net-next v4 3/8] net: mctp: Treat MCTP_NET_ANY specially in bind() Matt Johnston
2025-07-10 8:55 ` [PATCH net-next v4 4/8] net: mctp: Add test for conflicting bind()s Matt Johnston
2025-07-10 8:55 ` Matt Johnston [this message]
2025-07-15 10:05 ` [PATCH net-next v4 5/8] net: mctp: Use hashtable for binds Paolo Abeni
2025-07-15 11:08 ` Matt Johnston
2025-07-10 8:55 ` [PATCH net-next v4 6/8] net: mctp: Allow limiting binds to a peer address Matt Johnston
2025-07-10 8:56 ` [PATCH net-next v4 7/8] net: mctp: Test conflicts of connect() with bind() Matt Johnston
2025-07-10 8:56 ` [PATCH net-next v4 8/8] net: mctp: Add bind lookup test Matt Johnston
2025-07-15 10:20 ` [PATCH net-next v4 0/8] net: mctp: Improved bind handling patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250710-mctp-bind-v4-5-8ec2f6460c56@codeconstruct.com.au \
--to=matt@codeconstruct.com.au \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=horms@kernel.org \
--cc=jk@codeconstruct.com.au \
--cc=kuba@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).