From: Herbert Xu Subject: [RHEL5 PATCH] [PACKET]: Add PACKET_AUXDATA cmsg Date: Wed, 10 Jan 2007 20:17:36 +1100 Message-Id: <20070110091736.GA28579@gondor.apana.org.au> Changelog: xen: Add PACKET_AUXDATA cmsg Hi: RHEL5 BZ 219681 This patch forms part of the solution to #219681 where the DHCP server can't serve any requests from clients running (in different domains) on the same Xen host. I've sent a similar patch (without the Xen-specific code) upstream. This obsoletes the previous patch that computed the checksums in kernel-space. [PACKET]: Add PACKET_AUXDATA cmsg HCP servers/clients using AF_PACKET) to be able to serve another client on the same Xen host. The problem is that packets between different domains on the same Xen host only have partial checksums. Unfortunately this piece of information is not passed along in AF_PACKET unless you're using the mmap interface. Since dhcpd doesn't support packet-mmap, UDP packets from the same host come out with apparently bogus checksums. This patch adds a mechanism for AF_PACKET recvmsg(2) to return the status along with the packet. It does so by adding a new cmsg that contains this information along with some other relevant data such as the original packet length. I didn't include the time stamp information since there is already a cmsg for that. This patch also changes the mmap code to set the CSUMNOTREADY flag on all packets instead of just outoing packets on cooked sockets. Signed-off-by: Herbert Xu Index: linux-2.6.20.i386/include/linux/if_packet.h =================================================================== --- linux-2.6.20.i386.orig/include/linux/if_packet.h +++ linux-2.6.20.i386/include/linux/if_packet.h @@ -41,6 +41,7 @@ struct sockaddr_ll #define PACKET_RX_RING 5 #define PACKET_STATISTICS 6 #define PACKET_COPY_THRESH 7 +#define PACKET_AUXDATA 8 struct tpacket_stats { @@ -48,6 +49,15 @@ struct tpacket_stats unsigned int tp_drops; }; +struct tpacket_auxdata +{ + __u32 tp_status; + __u32 tp_len; + __u32 tp_snaplen; + __u16 tp_mac; + __u16 tp_net; +}; + struct tpacket_hdr { unsigned long tp_status; Index: linux-2.6.20.i386/net/packet/af_packet.c =================================================================== --- linux-2.6.20.i386.orig/net/packet/af_packet.c +++ linux-2.6.20.i386/net/packet/af_packet.c @@ -200,7 +200,8 @@ struct packet_sock { #endif struct packet_type prot_hook; spinlock_t bind_lock; - char running; /* prot_hook is attached*/ + unsigned int running:1, /* prot_hook is attached*/ + auxdata:1; int ifindex; /* bound device */ __be16 num; #ifdef CONFIG_PACKET_MULTICAST @@ -214,6 +215,10 @@ struct packet_sock { #endif }; +#define PACKET_SKB_CB(__skb) ((struct tpacket_auxdata *)((__skb)->cb)) + +extern int skb_checksum_setup(struct sk_buff *skb); + #ifdef CONFIG_PACKET_MMAP static inline char *packet_lookup_frame(struct packet_sock *po, unsigned int position) @@ -462,6 +467,7 @@ static int packet_rcv(struct sk_buff *sk u8 * skb_head = skb->data; int skb_len = skb->len; unsigned int snaplen, res; + struct tpacket_auxdata *aux; if (skb->pkt_type == PACKET_LOOPBACK) goto drop; @@ -523,6 +529,18 @@ static int packet_rcv(struct sk_buff *sk if (dev->hard_header_parse) sll->sll_halen = dev->hard_header_parse(skb, sll->sll_addr); + if (skb_checksum_setup(skb)) + goto drop_n_acct; + + aux = PACKET_SKB_CB(skb); + aux->tp_status = TP_STATUS_USER; + if (skb->ip_summed == CHECKSUM_PARTIAL) + aux->tp_status |= TP_STATUS_CSUMNOTREADY; + aux->tp_len = skb->len; + aux->tp_snaplen = snaplen; + aux->tp_mac = 0; + aux->tp_net = skb->nh.raw - skb->data; + if (pskb_trim(skb, snaplen)) goto drop_n_acct; @@ -582,8 +600,6 @@ static int tpacket_rcv(struct sk_buff *s else if (skb->pkt_type == PACKET_OUTGOING) { /* Special case: outgoing packets have ll header at head */ skb_pull(skb, skb->nh.raw - skb->data); - if (skb->ip_summed == CHECKSUM_PARTIAL) - status |= TP_STATUS_CSUMNOTREADY; } } @@ -595,6 +611,11 @@ static int tpacket_rcv(struct sk_buff *s if (snaplen > res) snaplen = res; + if (skb_checksum_setup(skb)) + goto drop; + if (skb->ip_summed == CHECKSUM_PARTIAL) + status |= TP_STATUS_CSUMNOTREADY; + if (sk->sk_type == SOCK_DGRAM) { macoff = netoff = TPACKET_ALIGN(TPACKET_HDRLEN) + 16; } else { @@ -1119,6 +1140,11 @@ static int packet_recvmsg(struct kiocb * if (msg->msg_name) memcpy(msg->msg_name, skb->cb, msg->msg_namelen); + if (pkt_sk(sk)->auxdata) { + struct tpacket_auxdata *aux = PACKET_SKB_CB(skb); + put_cmsg(msg, SOL_PACKET, PACKET_AUXDATA, sizeof(*aux), aux); + } + /* * Free or return the buffer as appropriate. Again this * hides all the races and re-entrancy issues from us. @@ -1317,6 +1343,7 @@ static int packet_setsockopt(struct socket *sock, int level, int optname, char __user *optval, int optlen) { struct sock *sk = sock->sk; + struct packet_sock *po = pkt_sk(sk); int ret; if (level != SOL_PACKET) @@ -1369,6 +1396,18 @@ packet_setsockopt(struct socket *sock, i return 0; } #endif + case PACKET_AUXDATA: + { + int val; + + if (optlen < sizeof(val)) + return -EINVAL; + if (copy_from_user(&val, optval, sizeof(val))) + return -EFAULT; + + po->auxdata = !!val; + return 0; + } default: return -ENOPROTOOPT; } @@ -1378,8 +1417,11 @@ static int packet_getsockopt(struct sock char __user *optval, int __user *optlen) { int len; + int val; struct sock *sk = sock->sk; struct packet_sock *po = pkt_sk(sk); + void *data; + struct tpacket_stats st; if (level != SOL_PACKET) return -ENOPROTOOPT; @@ -1392,9 +1434,6 @@ static int packet_getsockopt(struct sock switch(optname) { case PACKET_STATISTICS: - { - struct tpacket_stats st; - if (len > sizeof(struct tpacket_stats)) len = sizeof(struct tpacket_stats); spin_lock_bh(&sk->sk_receive_queue.lock); @@ -1403,16 +1442,23 @@ static int packet_getsockopt(struct sock spin_unlock_bh(&sk->sk_receive_queue.lock); st.tp_packets += st.tp_drops; - if (copy_to_user(optval, &st, len)) - return -EFAULT; + data = &st; + break; + case PACKET_AUXDATA: + if (len > sizeof(int)) + len = sizeof(int); + val = po->auxdata; + + data = &val; break; - } default: return -ENOPROTOOPT; } if (put_user(len, optlen)) return -EFAULT; + if (copy_to_user(optval, data, len)) + return -EFAULT; return 0; }