[DCCP]: Support for partial checksums (RFC 4340, sec. 9.2)
This patch does the following:
a) introduces variable-length checksums as specified in [RFC 4340, sec. 9.2]
b) provides necessary socket options and documentation as to how to use them
c) basic support and infrastructure for the Minimum Checksum Coverage feature
[RFC 4340, sec. 9.2.1]: acceptability tests, user notification and user
interface
In addition, it
(1) fixes two bugs in the DCCPv4 checksum computation:
* pseudo-header used checksum_len instead of skb->len
* incorrect checksum coverage calculation based on dccph_x
(2) removes dccp_v4_verify_checksum() since it reduplicates code of the
checksum computation; code calling this function is updated accordingly.
(3) now uses skb_checksum(), which is safer than checksum_partial() if the
sk_buff has is a non-linear buffer (has pages attached to it).
(4) fixes an outstanding TODO item:
* If P.CsCov is too large for the packet size, drop packet and return.
The code has been tested with applications, the latest version of tcpdump now
comes with support for partial DCCP checksums.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt
index 74563b3..a8142a8 100644
--- a/Documentation/networking/dccp.txt
+++ b/Documentation/networking/dccp.txt
@@ -47,6 +47,22 @@
is present). Connecting sockets set at most one service option; for
listening sockets, multiple service codes can be specified.
+DCCP_SOCKOPT_SEND_CSCOV and DCCP_SOCKOPT_RECV_CSCOV are used for setting the
+partial checksum coverage (RFC 4340, sec. 9.2). The default is that checksums
+always cover the entire packet and that only fully covered application data is
+accepted by the receiver. Hence, when using this feature on the sender, it must
+be enabled at the receiver, too with suitable choice of CsCov.
+
+DCCP_SOCKOPT_SEND_CSCOV sets the sender checksum coverage. Values in the
+ range 0..15 are acceptable. The default setting is 0 (full coverage),
+ values between 1..15 indicate partial coverage.
+DCCP_SOCKOPT_SEND_CSCOV is for the receiver and has a different meaning: it
+ sets a threshold, where again values 0..15 are acceptable. The default
+ of 0 means that all packets with a partial coverage will be discarded.
+ Values in the range 1..15 indicate that packets with minimally such a
+ coverage value are also acceptable. The higher the number, the more
+ restrictive this setting (see [RFC 4340, sec. 9.2.1]).
+
Notes
=====
diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 90d04ff..0502dfa 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -183,6 +183,7 @@
DCCPF_ACK_RATIO = 5,
DCCPF_SEND_ACK_VECTOR = 6,
DCCPF_SEND_NDP_COUNT = 7,
+ DCCPF_MIN_CSUM_COVER = 8,
/* 10-127 reserved */
DCCPF_MIN_CCID_SPECIFIC = 128,
DCCPF_MAX_CCID_SPECIFIC = 255,
@@ -200,6 +201,8 @@
#define DCCP_SOCKOPT_SERVICE 2
#define DCCP_SOCKOPT_CHANGE_L 3
#define DCCP_SOCKOPT_CHANGE_R 4
+#define DCCP_SOCKOPT_SEND_CSCOV 10
+#define DCCP_SOCKOPT_RECV_CSCOV 11
#define DCCP_SOCKOPT_CCID_RX_INFO 128
#define DCCP_SOCKOPT_CCID_TX_INFO 192
@@ -450,6 +453,8 @@
* @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option
* @dccps_packet_size - Set thru setsockopt
* @dccps_role - Role of this sock, one of %dccp_role
+ * @dccps_pcslen - sender partial checksum coverage (via sockopt)
+ * @dccps_pcrlen - receiver partial checksum coverage (via sockopt)
* @dccps_ndp_count - number of Non Data Packets since last data packet
* @dccps_hc_rx_ackvec - rx half connection ack vector
* @dccps_xmit_timer - timer for when CCID is not ready to send
@@ -474,6 +479,8 @@
__u32 dccps_packet_size;
__u16 dccps_l_ack_ratio;
__u16 dccps_r_ack_ratio;
+ __u16 dccps_pcslen;
+ __u16 dccps_pcrlen;
unsigned long dccps_ndp_count;
__u32 dccps_mss_cache;
struct dccp_minisock dccps_minisock;
diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h
index 2990bfb..d5c414b 100644
--- a/net/dccp/dccp.h
+++ b/net/dccp/dccp.h
@@ -129,6 +129,30 @@
#define DCCP_ADD_STATS_USER(field, val) \
SNMP_ADD_STATS_USER(dccp_statistics, field, val)
+/*
+ * Checksumming routines
+ */
+static inline int dccp_csum_coverage(const struct sk_buff *skb)
+{
+ const struct dccp_hdr* dh = dccp_hdr(skb);
+
+ if (dh->dccph_cscov == 0)
+ return skb->len;
+ return (dh->dccph_doff + dh->dccph_cscov - 1) * sizeof(u32);
+}
+
+static inline void dccp_csum_outgoing(struct sk_buff *skb)
+{
+ int cov = dccp_csum_coverage(skb);
+
+ if (cov >= skb->len)
+ dccp_hdr(skb)->dccph_cscov = 0;
+
+ skb->csum = skb_checksum(skb, 0, (cov > skb->len)? skb->len : cov, 0);
+}
+
+extern void dccp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb);
+
extern int dccp_retransmit_skb(struct sock *sk, struct sk_buff *skb);
extern void dccp_send_ack(struct sock *sk);
@@ -214,14 +238,9 @@
extern int inet_dccp_listen(struct socket *sock, int backlog);
extern unsigned int dccp_poll(struct file *file, struct socket *sock,
poll_table *wait);
-extern void dccp_v4_send_check(struct sock *sk, int len,
- struct sk_buff *skb);
extern int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr,
int addr_len);
-extern int dccp_v4_checksum(const struct sk_buff *skb,
- const __be32 saddr, const __be32 daddr);
-
extern int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code);
extern void dccp_send_close(struct sock *sk, const int active);
extern int dccp_invalid_packet(struct sk_buff *skb);
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 91bffaa..4961120 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -349,13 +349,19 @@
sock_put(sk);
}
-/* This routine computes an IPv4 DCCP checksum. */
-void dccp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb)
+static inline u16 dccp_v4_csum_finish(struct sk_buff *skb,
+ __be32 src, __be32 dst)
+{
+ return csum_tcpudp_magic(src, dst, skb->len, IPPROTO_DCCP, skb->csum);
+}
+
+void dccp_v4_send_check(struct sock *sk, int unused, struct sk_buff *skb)
{
const struct inet_sock *inet = inet_sk(sk);
struct dccp_hdr *dh = dccp_hdr(skb);
- dh->dccph_checksum = dccp_v4_checksum(skb, inet->saddr, inet->daddr);
+ dccp_csum_outgoing(skb);
+ dh->dccph_checksum = dccp_v4_csum_finish(skb, inet->saddr, inet->daddr);
}
EXPORT_SYMBOL_GPL(dccp_v4_send_check);
@@ -454,47 +460,6 @@
return sk;
}
-int dccp_v4_checksum(const struct sk_buff *skb, const __be32 saddr,
- const __be32 daddr)
-{
- const struct dccp_hdr* dh = dccp_hdr(skb);
- int checksum_len;
- u32 tmp;
-
- if (dh->dccph_cscov == 0)
- checksum_len = skb->len;
- else {
- checksum_len = (dh->dccph_cscov + dh->dccph_x) * sizeof(u32);
- checksum_len = checksum_len < skb->len ? checksum_len :
- skb->len;
- }
-
- tmp = csum_partial((unsigned char *)dh, checksum_len, 0);
- return csum_tcpudp_magic(saddr, daddr, checksum_len,
- IPPROTO_DCCP, tmp);
-}
-
-EXPORT_SYMBOL_GPL(dccp_v4_checksum);
-
-static int dccp_v4_verify_checksum(struct sk_buff *skb,
- const __be32 saddr, const __be32 daddr)
-{
- struct dccp_hdr *dh = dccp_hdr(skb);
- int checksum_len;
- u32 tmp;
-
- if (dh->dccph_cscov == 0)
- checksum_len = skb->len;
- else {
- checksum_len = (dh->dccph_cscov + dh->dccph_x) * sizeof(u32);
- checksum_len = checksum_len < skb->len ? checksum_len :
- skb->len;
- }
- tmp = csum_partial((unsigned char *)dh, checksum_len, 0);
- return csum_tcpudp_magic(saddr, daddr, checksum_len,
- IPPROTO_DCCP, tmp) == 0 ? 0 : -1;
-}
-
static struct dst_entry* dccp_v4_route_skb(struct sock *sk,
struct sk_buff *skb)
{
@@ -536,8 +501,8 @@
const struct inet_request_sock *ireq = inet_rsk(req);
struct dccp_hdr *dh = dccp_hdr(skb);
- dh->dccph_checksum = dccp_v4_checksum(skb, ireq->loc_addr,
- ireq->rmt_addr);
+ dh->dccph_checksum = dccp_v4_csum_finish(skb, ireq->loc_addr,
+ ireq->rmt_addr);
memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr,
ireq->rmt_addr,
@@ -602,8 +567,9 @@
dccp_hdr_set_ack(dccp_hdr_ack_bits(skb),
DCCP_SKB_CB(rxskb)->dccpd_seq);
- dh->dccph_checksum = dccp_v4_checksum(skb, rxskb->nh.iph->saddr,
- rxskb->nh.iph->daddr);
+ dccp_csum_outgoing(skb);
+ dh->dccph_checksum = dccp_v4_csum_finish(skb, rxskb->nh.iph->saddr,
+ rxskb->nh.iph->daddr);
bh_lock_sock(dccp_v4_ctl_socket->sk);
err = ip_build_and_send_pkt(skb, dccp_v4_ctl_socket->sk,
@@ -779,6 +745,7 @@
int dccp_invalid_packet(struct sk_buff *skb)
{
const struct dccp_hdr *dh;
+ unsigned int cscov;
if (skb->pkt_type != PACKET_HOST)
return 1;
@@ -830,6 +797,22 @@
return 1;
}
+ /*
+ * If P.CsCov is too large for the packet size, drop packet and return.
+ * This must come _before_ checksumming (not as RFC 4340 suggests).
+ */
+ cscov = dccp_csum_coverage(skb);
+ if (cscov > skb->len) {
+ LIMIT_NETDEBUG(KERN_WARNING
+ "DCCP: P.CsCov %u exceeds packet length %d\n",
+ dh->dccph_cscov, skb->len);
+ return 1;
+ }
+
+ /* If header checksum is incorrect, drop packet and return.
+ * (This step is completed in the AF-dependent functions.) */
+ skb->csum = skb_checksum(skb, 0, cscov, 0);
+
return 0;
}
@@ -840,16 +823,17 @@
{
const struct dccp_hdr *dh;
struct sock *sk;
+ int min_cov;
- /* Step 1: Check header basics: */
+ /* Step 1: Check header basics */
if (dccp_invalid_packet(skb))
goto discard_it;
- /* If the header checksum is incorrect, drop packet and return */
- if (dccp_v4_verify_checksum(skb, skb->nh.iph->saddr,
- skb->nh.iph->daddr) < 0) {
- LIMIT_NETDEBUG(KERN_WARNING "%s: incorrect header checksum\n",
+ /* Step 1: If header checksum is incorrect, drop packet and return */
+ if (dccp_v4_csum_finish(skb, skb->nh.iph->saddr, skb->nh.iph->daddr)) {
+ LIMIT_NETDEBUG(KERN_WARNING
+ "%s: dropped packet with invalid checksum\n",
__FUNCTION__);
goto discard_it;
}
@@ -905,6 +889,21 @@
goto no_dccp_socket;
}
+ /*
+ * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage
+ * o if MinCsCov = 0, only packets with CsCov = 0 are accepted
+ * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov
+ */
+ min_cov = dccp_sk(sk)->dccps_pcrlen;
+ if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) {
+ dccp_pr_debug("Packet CsCov %d does not satisfy MinCsCov %d\n",
+ dh->dccph_cscov, min_cov);
+ /* FIXME: "Such packets SHOULD be reported using Data Dropped
+ * options (Section 11.7) with Drop Code 0, Protocol
+ * Constraints." */
+ goto discard_and_relse;
+ }
+
if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
goto discard_and_relse;
nf_reset(skb);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 201801e..193b946 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -58,12 +58,22 @@
}
}
-static inline u16 dccp_v6_check(struct dccp_hdr *dh, int len,
- struct in6_addr *saddr,
- struct in6_addr *daddr,
- unsigned long base)
+/* add pseudo-header to DCCP checksum stored in skb->csum */
+static inline u16 dccp_v6_csum_finish(struct sk_buff *skb,
+ struct in6_addr *saddr,
+ struct in6_addr *daddr)
{
- return csum_ipv6_magic(saddr, daddr, len, IPPROTO_DCCP, base);
+ return csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_DCCP, skb->csum);
+}
+
+static inline void dccp_v6_send_check(struct sock *sk, int unused_value,
+ struct sk_buff *skb)
+{
+ struct ipv6_pinfo *np = inet6_sk(sk);
+ struct dccp_hdr *dh = dccp_hdr(skb);
+
+ dccp_csum_outgoing(skb);
+ dh->dccph_checksum = dccp_v6_csum_finish(skb, &np->saddr, &np->daddr);
}
static __u32 dccp_v6_init_sequence(struct sock *sk, struct sk_buff *skb)
@@ -280,12 +290,9 @@
if (skb != NULL) {
struct dccp_hdr *dh = dccp_hdr(skb);
- dh->dccph_checksum = dccp_v6_check(dh, skb->len,
- &ireq6->loc_addr,
- &ireq6->rmt_addr,
- csum_partial((char *)dh,
- skb->len,
- skb->csum));
+ dh->dccph_checksum = dccp_v6_csum_finish(skb,
+ &ireq6->loc_addr,
+ &ireq6->rmt_addr);
ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr);
err = ip6_xmit(sk, skb, &fl, opt, 0);
if (err == NET_XMIT_CN)
@@ -305,18 +312,6 @@
kfree_skb(inet6_rsk(req)->pktopts);
}
-static void dccp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb)
-{
- struct ipv6_pinfo *np = inet6_sk(sk);
- struct dccp_hdr *dh = dccp_hdr(skb);
-
- dh->dccph_checksum = csum_ipv6_magic(&np->saddr, &np->daddr,
- len, IPPROTO_DCCP,
- csum_partial((char *)dh,
- dh->dccph_doff << 2,
- skb->csum));
-}
-
static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb)
{
struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh;
@@ -360,12 +355,14 @@
dccp_hdr_set_ack(dccp_hdr_ack_bits(skb),
DCCP_SKB_CB(rxskb)->dccpd_seq);
+ dccp_csum_outgoing(skb);
+ dh->dccph_checksum = dccp_v6_csum_finish(skb, &rxskb->nh.ipv6h->saddr,
+ &rxskb->nh.ipv6h->daddr);
+
memset(&fl, 0, sizeof(fl));
ipv6_addr_copy(&fl.fl6_dst, &rxskb->nh.ipv6h->saddr);
ipv6_addr_copy(&fl.fl6_src, &rxskb->nh.ipv6h->daddr);
- dh->dccph_checksum = csum_ipv6_magic(&fl.fl6_src, &fl.fl6_dst,
- sizeof(*dh), IPPROTO_DCCP,
- skb->csum);
+
fl.proto = IPPROTO_DCCP;
fl.oif = inet6_iif(rxskb);
fl.fl_ip_dport = dh->dccph_dport;
@@ -825,12 +822,22 @@
const struct dccp_hdr *dh;
struct sk_buff *skb = *pskb;
struct sock *sk;
+ int min_cov;
- /* Step 1: Check header basics: */
+ /* Step 1: Check header basics */
if (dccp_invalid_packet(skb))
goto discard_it;
+ /* Step 1: If header checksum is incorrect, drop packet and return. */
+ if (dccp_v6_csum_finish(skb, &skb->nh.ipv6h->saddr,
+ &skb->nh.ipv6h->daddr)) {
+ LIMIT_NETDEBUG(KERN_WARNING
+ "%s: dropped packet with invalid checksum\n",
+ __FUNCTION__);
+ goto discard_it;
+ }
+
dh = dccp_hdr(skb);
DCCP_SKB_CB(skb)->dccpd_seq = dccp_hdr_seq(skb);
@@ -869,6 +876,19 @@
goto no_dccp_socket;
}
+ /*
+ * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage
+ * o if MinCsCov = 0, only packets with CsCov = 0 are accepted
+ * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov
+ */
+ min_cov = dccp_sk(sk)->dccps_pcrlen;
+ if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) {
+ dccp_pr_debug("Packet CsCov %d does not satisfy MinCsCov %d\n",
+ dh->dccph_cscov, min_cov);
+ /* FIXME: send Data Dropped option (see also dccp_v4_rcv) */
+ goto discard_and_relse;
+ }
+
if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
goto discard_and_relse;
diff --git a/net/dccp/output.c b/net/dccp/output.c
index 5165497..992caed 100644
--- a/net/dccp/output.c
+++ b/net/dccp/output.c
@@ -96,6 +96,7 @@
dh->dccph_dport = inet->dport;
dh->dccph_doff = (dccp_header_size + dcb->dccpd_opt_len) / 4;
dh->dccph_ccval = dcb->dccpd_ccval;
+ dh->dccph_cscov = dp->dccps_pcslen;
/* XXX For now we're using only 48 bits sequence numbers */
dh->dccph_x = 1;
@@ -115,7 +116,7 @@
break;
}
- icsk->icsk_af_ops->send_check(sk, skb->len, skb);
+ icsk->icsk_af_ops->send_check(sk, 0, skb);
if (set_ack)
dccp_event_ack_sent(sk);
@@ -329,7 +330,6 @@
skb_reserve(skb, sk->sk_prot->max_header);
skb->dst = dst_clone(dst);
- skb->csum = 0;
dreq = dccp_rsk(req);
DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_RESPONSE;
@@ -352,6 +352,8 @@
dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dreq->dreq_isr);
dccp_hdr_response(skb)->dccph_resp_service = dreq->dreq_service;
+ dccp_csum_outgoing(skb);
+
DCCP_INC_STATS(DCCP_MIB_OUTSEGS);
return skb;
}
@@ -376,7 +378,6 @@
skb_reserve(skb, sk->sk_prot->max_header);
skb->dst = dst_clone(dst);
- skb->csum = 0;
dccp_inc_seqno(&dp->dccps_gss);
@@ -401,7 +402,7 @@
dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dp->dccps_gsr);
dccp_hdr_reset(skb)->dccph_reset_code = code;
- inet_csk(sk)->icsk_af_ops->send_check(sk, skb->len, skb);
+ inet_csk(sk)->icsk_af_ops->send_check(sk, 0, skb);
DCCP_INC_STATS(DCCP_MIB_OUTSEGS);
return skb;
@@ -475,7 +476,6 @@
skb_reserve(skb, sk->sk_prot->max_header);
DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_REQUEST;
- skb->csum = 0;
dccp_skb_entail(sk, skb);
dccp_transmit_skb(sk, skb_clone(skb, GFP_KERNEL));
@@ -507,7 +507,6 @@
/* Reserve space for headers */
skb_reserve(skb, sk->sk_prot->max_header);
- skb->csum = 0;
DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_ACK;
dccp_transmit_skb(sk, skb);
}
@@ -561,7 +560,6 @@
/* Reserve space for headers and prepare control bits. */
skb_reserve(skb, sk->sk_prot->max_header);
- skb->csum = 0;
DCCP_SKB_CB(skb)->dccpd_type = pkt_type;
DCCP_SKB_CB(skb)->dccpd_seq = seq;
@@ -587,7 +585,6 @@
/* Reserve space for headers and prepare control bits. */
skb_reserve(skb, sk->sk_prot->max_header);
- skb->csum = 0;
DCCP_SKB_CB(skb)->dccpd_type = dp->dccps_role == DCCP_ROLE_CLIENT ?
DCCP_PKT_CLOSE : DCCP_PKT_CLOSEREQ;
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 047d170..db54e55 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -472,7 +472,6 @@
case DCCP_SOCKOPT_PACKET_SIZE:
dp->dccps_packet_size = val;
break;
-
case DCCP_SOCKOPT_CHANGE_L:
if (optlen != sizeof(struct dccp_so_feat))
err = -EINVAL;
@@ -481,7 +480,6 @@
(struct dccp_so_feat __user *)
optval);
break;
-
case DCCP_SOCKOPT_CHANGE_R:
if (optlen != sizeof(struct dccp_so_feat))
err = -EINVAL;
@@ -490,12 +488,26 @@
(struct dccp_so_feat __user *)
optval);
break;
-
+ case DCCP_SOCKOPT_SEND_CSCOV: /* sender side, RFC 4340, sec. 9.2 */
+ if (val < 0 || val > 15)
+ err = -EINVAL;
+ else
+ dp->dccps_pcslen = val;
+ break;
+ case DCCP_SOCKOPT_RECV_CSCOV: /* receiver side, RFC 4340 sec. 9.2.1 */
+ if (val < 0 || val > 15)
+ err = -EINVAL;
+ else {
+ dp->dccps_pcrlen = val;
+ /* FIXME: add feature negotiation,
+ * ChangeL(MinimumChecksumCoverage, val) */
+ }
+ break;
default:
err = -ENOPROTOOPT;
break;
}
-
+
release_sock(sk);
return err;
}
@@ -575,6 +587,12 @@
case DCCP_SOCKOPT_SERVICE:
return dccp_getsockopt_service(sk, len,
(__be32 __user *)optval, optlen);
+ case DCCP_SOCKOPT_SEND_CSCOV:
+ val = dp->dccps_pcslen;
+ break;
+ case DCCP_SOCKOPT_RECV_CSCOV:
+ val = dp->dccps_pcrlen;
+ break;
case 128 ... 191:
return ccid_hc_rx_getsockopt(dp->dccps_hc_rx_ccid, sk, optname,
len, (u32 __user *)optval, optlen);