SO_ORIGINAL_DST for getsockopt() in SCTP
Subsystems (2)
PATCH 1/1 SO_ORIGINAL_DST for getsockopt() in SCTP
Hi,
a tiny kernel patch to allow getsockopt() with SO_ORIGINAL_DST to work for SCTP:
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c 2009-07-02 17:07:09.000000000 -0700 renamed from net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c 2008-12-24 15:26:37.000000000 -0800 +4 -4
@@ -258,11 +258,11 @@ getorigdst(struct sock *sk, int optval, tuple.dst.u3.ip = inet->daddr; tuple.dst.u.tcp.port = inet->dport; tuple.src.l3num = PF_INET; - tuple.dst.protonum = IPPROTO_TCP; + tuple.dst.protonum = sk->sk_protocol; - /* We only do TCP at the moment: is there a better way? */ - if (strcmp(sk->sk_prot->name, "TCP")) { - pr_debug("SO_ORIGINAL_DST: Not a TCP socket\n"); + /* We only do TCP and SCTP at the moment: is there a better way? */ + if ((sk->sk_protocol != IPPROTO_TCP) && (sk->sk_protocol != IPPROTO_SCTP)) { + pr_debug("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n"); return -ENOPROTOOPT; }
Rafael
David Miller
maintainerThunderbird has severely mangled your patch beyond usability.
Please read Documentation/email-clients.txt in the kernel sources.
Quoted 3 lines · David Miller
Thunderbird has severely mangled your patch beyond usability.
Please read
Documentation/email-clients.txtin the kernel sources.
Trying again:
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +4 -4
@@ -258,11 +258,11 @@ getorigdst(struct sock *sk, int optval, tuple.dst.u3.ip = inet->daddr; tuple.dst.u.tcp.port = inet->dport; tuple.src.l3num =PF_INET ; - tuple.dst.protonum =IPPROTO_TCP ; + tuple.dst.protonum = sk->sk_protocol; -/* We only do TCP at the moment: is there a better way? */ -if (strcmp (sk->sk_prot->name,"TCP" )) { -pr_debug ("SO_ORIGINAL_DST: Not a TCP socket\n" ); +/* We only do TCP and SCTP at the moment: is there a better way? */ +if ((sk->sk_protocol !=IPPROTO_TCP ) && (sk->sk_protocol !=IPPROTO_SCTP )) { +pr_debug ("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n" );return -ENOPROTOOPT ; }
Rafael
David Miller
maintainerFrom: Rafael Laufer <[email protected]>
Date: Thu, 02 Jul 2009 19:26:57 -0700
Quoted 6 lines
Quoted 3 lines · David Miller
Thunderbird has severely mangled your patch beyond usability.
Please read
Documentation/email-clients.txtin the kernel sources.Trying again:
You should make your new posting with full changelog and signoffs so that whoever applies this doesn't have to go searching into the mailing list archives just to hunt that down.
Quoted 16 lines · Rafael Laufer
Quoted 3 lines · David Miller
Thunderbird has severely mangled your patch beyond usability.
Please read
Documentation/email-clients.txtin the kernel sources.Trying again:
diff -rup
linux-2.6.28.orig/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.clinux-2.6.28/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.cnet/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +0
-1@@ -258,11 +258,11 @@ getorigdst(struct sock *sk, int optval, tuple.dst.u3.ip = inet->daddr; tuple.dst.u.tcp.port = inet->dport; tuple.src.l3num =PF_INET ; - tuple.dst.protonum =IPPROTO_TCP ;
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c getorigdst(struct sock *sk, int optval, · line 267
Quoted 10 lines
net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +4
-3+ tuple.dst.protonum = sk->sk_protocol; -/* We only do TCP at the moment: is there a better way? */ -if (strcmp (sk->sk_prot->name,"TCP" )) { -pr_debug ("SO_ORIGINAL_DST: Not a TCP socket\n" ); +/* We only do TCP and SCTP at the moment: is there a better way? */ +if ((sk->sk_protocol !=IPPROTO_TCP ) && (sk->sk_protocol !=IPPROTO_SCTP )) { +pr_debug ("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n" );return -ENOPROTOOPT ; }
Applied with minor stylistic changes.