aboutsummaryrefslogtreecommitdiffstats
path: root/net
diff options
context:
space:
mode:
authorDavid S. Miller <davem@davemloft.net>2015-12-15 16:50:28 -0500
committerDavid S. Miller <davem@davemloft.net>2015-12-15 16:50:28 -0500
commit93d085d222de7b8e7c7794dbd800c6a39df2eae2 (patch)
tree32f07a8966330133dd357ab2a6212ac252e4344c /net
parentravb: Add fixed-link support (diff)
parentnet: Elaborate on checksum offload interface description (diff)
downloadlinux-dev-93d085d222de7b8e7c7794dbd800c6a39df2eae2.tar.xz
linux-dev-93d085d222de7b8e7c7794dbd800c6a39df2eae2.zip
Merge branch 'end-of-ip-csum'
Tom Herbert says: ==================== net: The beginning of the end for NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM Background: This patch set starts to address one front in the battle against protocol ossification. Protocol ossification describes the state that we have arrived at in the evolution of the Internet where we are materially limited to only using a very narrow range of protocols and protocol features. For instance, only TCP and UDP is sufficiently supported on the Internet so that deploying alternative protocols, such as SCTP and DCCP, are non-starters. Similarly, IP options and IPv6 extension headers are typically not considered feasible for wide deployment, so we have loss the extensibility of IP protocols. Protocol ossification is not only a problem on the Internet, but in the data center as well. A root cause of this seems to be narrow, protocol specific optimizations implemented in switches (for doing EMCP) and in NICs (NIC offloads). These tend to be performance optimization around TCP and UDP packets, and these have become requirements to implement performant network solutions at scale. Attempts to deal with protocol ossification in data center have yielded ad hoc, sub-optimal solutions. A main driver of foo-over-UDP (e.g. GRE/UDP, MPLS/UDP) is to leverage the existing EMCP and RSS support for UDP by setting the source port as an entropy value. This has seen some success, but the cost of additional overhead and layering limits its usefulness. An even more extreme solution is STT where non-TCP packets are spoofed as TCP to leverage NIC offloads. This patch set endeavours to address protocol ossification caused by techniques used in transmit checksum offload for NICs. Future work will address protocol ossification in the other primary NIC offloads-- namely receive checksum offload, LSO, LRO, and RSS. NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM: NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM exemplify the problem of protocol ossification. These features are relics from a simpler time in the Internet, before encapsulation, before GRE and IPIP. Many hardware vendors only saw the need to provide checksum offload for simple UDP and TCP packets over IPv4 (IPv6 support is an afterthought also). In today's Internet and data centers, checksum offload is well established as a valuable feature, but we can no longer afford to be contsrained to use a handful of protocols and features that are supported at the discretion of NIC vendors. Generic and protocol agnostic methods are needed. The actual interface that the stack uses with drivers for checksum offload is CHECKSUM_PARTIAL. This is a generic and protocol agnostic interface. A driver for a device that supports this generic interface advertises NETIF_F_HW_CSUM. Goals of this patch set: We propose that drivers advertise NETIF_F_HW_CSUM instead of protocol specific values of NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM. If the driver's device is constrained (for instance it can only offlaod simple IPv4 and IPv6 packets) then these constraints can be checked in the transmit path and skb_checksum_help would be called for packets that the driver is unable to offload. In order to facilitate this, we add some helper functions that takes a specification argument indicating the type of packets a device is able to offload. If a packet does not match the specification, the helper function calls skb_checksum_help. Benefits of this approach are: - Simplify the stack and clarify the interface for checksum offload - Encourage NIC vendors to implement the generic. protocol agnostic checksum offload methods in hardware - Encourage feature parity in NIC offloads for IPv4 and IPv6 Many drivers advertise NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM and it probably isn't feasible to convert them all in a given time frame (although if we could this would be a great simplification to the stack). A reasonable direction may be to declare that new drivers must use NETIF_F_HW_CSUM as NETIF_F_IP_CSUM and NETIF_F_IPV6_CSUM are considered deprecated. There is a class of drivers that should now be converted to advertise NETIF_F_HW_CSUM, namely those that support offload of ecapsulated checksums. These drivers have to date been using skb->encapsulation to infer that checksum offload is being performed for an encapsulated checksum. This is strictly not correct. skb->encapsulation indicates that the inner headers are valid in the skbuff, whereas the stack indicates checksum offload arguments exclusively in csum_start and csum_offset. At some point we may want to set the inner headers for an skbuff but offload the outer transport checksum, so this needs to be fixed. In this patch set: - Rename some of constants involved in checksum offload to be more reflective of their function - Eliminate NETIF_F_GEN_CSUM and NETIF_F_V[46]_CSUM entirely as unnecessary convolutions - Fix conditions in tcp_sendpage and tcp_sendmsg to take IP protocol into account when determining if checksum offload can be done - Add driver helper functions for determining if a checksum can be offloaded to a device. If not, the helper function can call skb_checksum_help - Document the checksum offload interface between the stack and drivers with detail and specifics Testing: Have been testing ixgbe and mlx4. No noticeable regressions seen yet. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net')
-rw-r--r--net/8021q/vlan_dev.c4
-rw-r--r--net/core/dev.c156
-rw-r--r--net/core/ethtool.c4
-rw-r--r--net/core/pktgen.c4
-rw-r--r--net/ipv4/ip_output.c2
-rw-r--r--net/ipv4/netfilter/nf_nat_l3proto_ipv4.c3
-rw-r--r--net/ipv4/tcp.c4
-rw-r--r--net/ipv4/udp.c3
-rw-r--r--net/ipv4/udp_offload.c5
-rw-r--r--net/ipv6/ip6_output.c2
-rw-r--r--net/ipv6/netfilter/nf_nat_l3proto_ipv6.c3
-rw-r--r--net/netfilter/ipvs/ip_vs_proto_sctp.c2
-rw-r--r--net/sctp/output.c2
13 files changed, 167 insertions, 27 deletions
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 9f4bd137e045..ad5e2fd1012c 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -543,9 +543,9 @@ static int vlan_dev_init(struct net_device *dev)
(1<<__LINK_STATE_DORMANT))) |
(1<<__LINK_STATE_PRESENT);
- dev->hw_features = NETIF_F_ALL_CSUM | NETIF_F_SG |
+ dev->hw_features = NETIF_F_HW_CSUM | NETIF_F_SG |
NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE |
- NETIF_F_HIGHDMA | NETIF_F_SCTP_CSUM |
+ NETIF_F_HIGHDMA | NETIF_F_SCTP_CRC |
NETIF_F_ALL_FCOE;
dev->features |= real_dev->vlan_features | NETIF_F_LLTX |
diff --git a/net/core/dev.c b/net/core/dev.c
index 8f705fcedb94..914b4a24c654 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -138,6 +138,7 @@
#include <linux/errqueue.h>
#include <linux/hrtimer.h>
#include <linux/netfilter_ingress.h>
+#include <linux/sctp.h>
#include "net-sysfs.h"
@@ -2471,6 +2472,141 @@ out:
}
EXPORT_SYMBOL(skb_checksum_help);
+/* skb_csum_offload_check - Driver helper function to determine if a device
+ * with limited checksum offload capabilities is able to offload the checksum
+ * for a given packet.
+ *
+ * Arguments:
+ * skb - sk_buff for the packet in question
+ * spec - contains the description of what device can offload
+ * csum_encapped - returns true if the checksum being offloaded is
+ * encpasulated. That is it is checksum for the transport header
+ * in the inner headers.
+ * checksum_help - when set indicates that helper function should
+ * call skb_checksum_help if offload checks fail
+ *
+ * Returns:
+ * true: Packet has passed the checksum checks and should be offloadable to
+ * the device (a driver may still need to check for additional
+ * restrictions of its device)
+ * false: Checksum is not offloadable. If checksum_help was set then
+ * skb_checksum_help was called to resolve checksum for non-GSO
+ * packets and when IP protocol is not SCTP
+ */
+bool __skb_csum_offload_chk(struct sk_buff *skb,
+ const struct skb_csum_offl_spec *spec,
+ bool *csum_encapped,
+ bool csum_help)
+{
+ struct iphdr *iph;
+ struct ipv6hdr *ipv6;
+ void *nhdr;
+ int protocol;
+ u8 ip_proto;
+
+ if (skb->protocol == htons(ETH_P_8021Q) ||
+ skb->protocol == htons(ETH_P_8021AD)) {
+ if (!spec->vlan_okay)
+ goto need_help;
+ }
+
+ /* We check whether the checksum refers to a transport layer checksum in
+ * the outermost header or an encapsulated transport layer checksum that
+ * corresponds to the inner headers of the skb. If the checksum is for
+ * something else in the packet we need help.
+ */
+ if (skb_checksum_start_offset(skb) == skb_transport_offset(skb)) {
+ /* Non-encapsulated checksum */
+ protocol = eproto_to_ipproto(vlan_get_protocol(skb));
+ nhdr = skb_network_header(skb);
+ *csum_encapped = false;
+ if (spec->no_not_encapped)
+ goto need_help;
+ } else if (skb->encapsulation && spec->encap_okay &&
+ skb_checksum_start_offset(skb) ==
+ skb_inner_transport_offset(skb)) {
+ /* Encapsulated checksum */
+ *csum_encapped = true;
+ switch (skb->inner_protocol_type) {
+ case ENCAP_TYPE_ETHER:
+ protocol = eproto_to_ipproto(skb->inner_protocol);
+ break;
+ case ENCAP_TYPE_IPPROTO:
+ protocol = skb->inner_protocol;
+ break;
+ }
+ nhdr = skb_inner_network_header(skb);
+ } else {
+ goto need_help;
+ }
+
+ switch (protocol) {
+ case IPPROTO_IP:
+ if (!spec->ipv4_okay)
+ goto need_help;
+ iph = nhdr;
+ ip_proto = iph->protocol;
+ if (iph->ihl != 5 && !spec->ip_options_okay)
+ goto need_help;
+ break;
+ case IPPROTO_IPV6:
+ if (!spec->ipv6_okay)
+ goto need_help;
+ if (spec->no_encapped_ipv6 && *csum_encapped)
+ goto need_help;
+ ipv6 = nhdr;
+ nhdr += sizeof(*ipv6);
+ ip_proto = ipv6->nexthdr;
+ break;
+ default:
+ goto need_help;
+ }
+
+ip_proto_again:
+ switch (ip_proto) {
+ case IPPROTO_TCP:
+ if (!spec->tcp_okay ||
+ skb->csum_offset != offsetof(struct tcphdr, check))
+ goto need_help;
+ break;
+ case IPPROTO_UDP:
+ if (!spec->udp_okay ||
+ skb->csum_offset != offsetof(struct udphdr, check))
+ goto need_help;
+ break;
+ case IPPROTO_SCTP:
+ if (!spec->sctp_okay ||
+ skb->csum_offset != offsetof(struct sctphdr, checksum))
+ goto cant_help;
+ break;
+ case NEXTHDR_HOP:
+ case NEXTHDR_ROUTING:
+ case NEXTHDR_DEST: {
+ u8 *opthdr = nhdr;
+
+ if (protocol != IPPROTO_IPV6 || !spec->ext_hdrs_okay)
+ goto need_help;
+
+ ip_proto = opthdr[0];
+ nhdr += (opthdr[1] + 1) << 3;
+
+ goto ip_proto_again;
+ }
+ default:
+ goto need_help;
+ }
+
+ /* Passed the tests for offloading checksum */
+ return true;
+
+need_help:
+ if (csum_help && !skb_shinfo(skb)->gso_size)
+ skb_checksum_help(skb);
+cant_help:
+ return false;
+}
+EXPORT_SYMBOL(__skb_csum_offload_chk);
+
__be16 skb_network_protocol(struct sk_buff *skb, int *depth)
{
__be16 type = skb->protocol;
@@ -2645,7 +2781,7 @@ static netdev_features_t harmonize_features(struct sk_buff *skb,
if (skb->ip_summed != CHECKSUM_NONE &&
!can_checksum_protocol(features, type)) {
- features &= ~NETIF_F_ALL_CSUM;
+ features &= ~NETIF_F_CSUM_MASK;
} else if (illegal_highdma(skb->dev, skb)) {
features &= ~NETIF_F_SG;
}
@@ -2792,7 +2928,7 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device
else
skb_set_transport_header(skb,
skb_checksum_start_offset(skb));
- if (!(features & NETIF_F_ALL_CSUM) &&
+ if (!(features & NETIF_F_CSUM_MASK) &&
skb_checksum_help(skb))
goto out_kfree_skb;
}
@@ -6467,9 +6603,9 @@ static netdev_features_t netdev_fix_features(struct net_device *dev,
/* UFO needs SG and checksumming */
if (features & NETIF_F_UFO) {
/* maybe split UFO into V4 and V6? */
- if (!((features & NETIF_F_GEN_CSUM) ||
- (features & (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))
- == (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))) {
+ if (!(features & NETIF_F_HW_CSUM) &&
+ ((features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) !=
+ (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))) {
netdev_dbg(dev,
"Dropping NETIF_F_UFO since no checksum offload features.\n");
features &= ~NETIF_F_UFO;
@@ -7571,16 +7707,16 @@ static int dev_cpu_callback(struct notifier_block *nfb,
netdev_features_t netdev_increment_features(netdev_features_t all,
netdev_features_t one, netdev_features_t mask)
{
- if (mask & NETIF_F_GEN_CSUM)
- mask |= NETIF_F_ALL_CSUM;
+ if (mask & NETIF_F_HW_CSUM)
+ mask |= NETIF_F_CSUM_MASK;
mask |= NETIF_F_VLAN_CHALLENGED;
- all |= one & (NETIF_F_ONE_FOR_ALL|NETIF_F_ALL_CSUM) & mask;
+ all |= one & (NETIF_F_ONE_FOR_ALL | NETIF_F_CSUM_MASK) & mask;
all &= one | ~NETIF_F_ALL_FOR_ALL;
/* If one device supports hw checksumming, set for all. */
- if (all & NETIF_F_GEN_CSUM)
- all &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_GEN_CSUM);
+ if (all & NETIF_F_HW_CSUM)
+ all &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_HW_CSUM);
return all;
}
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 29edf74846fc..09948a726347 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -87,7 +87,7 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN]
[NETIF_F_GSO_UDP_TUNNEL_BIT] = "tx-udp_tnl-segmentation",
[NETIF_F_FCOE_CRC_BIT] = "tx-checksum-fcoe-crc",
- [NETIF_F_SCTP_CSUM_BIT] = "tx-checksum-sctp",
+ [NETIF_F_SCTP_CRC_BIT] = "tx-checksum-sctp",
[NETIF_F_FCOE_MTU_BIT] = "fcoe-mtu",
[NETIF_F_NTUPLE_BIT] = "rx-ntuple-filter",
[NETIF_F_RXHASH_BIT] = "rx-hashing",
@@ -235,7 +235,7 @@ static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd)
switch (eth_cmd) {
case ETHTOOL_GTXCSUM:
case ETHTOOL_STXCSUM:
- return NETIF_F_ALL_CSUM | NETIF_F_SCTP_CSUM;
+ return NETIF_F_CSUM_MASK | NETIF_F_SCTP_CRC;
case ETHTOOL_GRXCSUM:
case ETHTOOL_SRXCSUM:
return NETIF_F_RXCSUM;
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index de8d5cc5eb24..2be144498bcf 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2898,7 +2898,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
if (!(pkt_dev->flags & F_UDPCSUM)) {
skb->ip_summed = CHECKSUM_NONE;
- } else if (odev->features & NETIF_F_V4_CSUM) {
+ } else if (odev->features & (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM)) {
skb->ip_summed = CHECKSUM_PARTIAL;
skb->csum = 0;
udp4_hwcsum(skb, iph->saddr, iph->daddr);
@@ -3032,7 +3032,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
if (!(pkt_dev->flags & F_UDPCSUM)) {
skb->ip_summed = CHECKSUM_NONE;
- } else if (odev->features & NETIF_F_V6_CSUM) {
+ } else if (odev->features & (NETIF_F_HW_CSUM | NETIF_F_IPV6_CSUM)) {
skb->ip_summed = CHECKSUM_PARTIAL;
skb->csum_start = skb_transport_header(skb) - skb->head;
skb->csum_offset = offsetof(struct udphdr, check);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index e0b94cd843d7..568e2bc0d93d 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -911,7 +911,7 @@ static int __ip_append_data(struct sock *sk,
*/
if (transhdrlen &&
length + fragheaderlen <= mtu &&
- rt->dst.dev->features & NETIF_F_V4_CSUM &&
+ rt->dst.dev->features & (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM) &&
!(flags & MSG_MORE) &&
!exthdrlen)
csummode = CHECKSUM_PARTIAL;
diff --git a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c
index 5075b7ecd26d..61c7cc22ea68 100644
--- a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c
@@ -132,7 +132,8 @@ static void nf_nat_ipv4_csum_recalc(struct sk_buff *skb,
if (skb->ip_summed != CHECKSUM_PARTIAL) {
if (!(rt->rt_flags & RTCF_LOCAL) &&
- (!skb->dev || skb->dev->features & NETIF_F_V4_CSUM)) {
+ (!skb->dev || skb->dev->features &
+ (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) {
skb->ip_summed = CHECKSUM_PARTIAL;
skb->csum_start = skb_headroom(skb) +
skb_network_offset(skb) +
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index c82cca18c90f..92b3e61b847d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1018,7 +1018,7 @@ int tcp_sendpage(struct sock *sk, struct page *page, int offset,
ssize_t res;
if (!(sk->sk_route_caps & NETIF_F_SG) ||
- !(sk->sk_route_caps & NETIF_F_ALL_CSUM))
+ !sk_check_csum_caps(sk))
return sock_no_sendpage(sk->sk_socket, page, offset, size,
flags);
@@ -1175,7 +1175,7 @@ new_segment:
/*
* Check whether we can use HW checksum.
*/
- if (sk->sk_route_caps & NETIF_F_ALL_CSUM)
+ if (sk_check_csum_caps(sk))
skb->ip_summed = CHECKSUM_PARTIAL;
skb_entail(sk, skb);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 0c7b0e61b917..8841e984f8bf 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -772,7 +772,8 @@ void udp_set_csum(bool nocheck, struct sk_buff *skb,
else if (skb_is_gso(skb))
uh->check = ~udp_v4_check(len, saddr, daddr, 0);
else if (skb_dst(skb) && skb_dst(skb)->dev &&
- (skb_dst(skb)->dev->features & NETIF_F_V4_CSUM)) {
+ (skb_dst(skb)->dev->features &
+ (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) {
BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index f9386160cbee..130042660181 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -60,8 +60,9 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
/* Try to offload checksum if possible */
offload_csum = !!(need_csum &&
- (skb->dev->features &
- (is_ipv6 ? NETIF_F_V6_CSUM : NETIF_F_V4_CSUM)));
+ ((skb->dev->features & NETIF_F_HW_CSUM) ||
+ (skb->dev->features & (is_ipv6 ?
+ NETIF_F_IPV6_CSUM : NETIF_F_IP_CSUM))));
/* segment inner packet. */
enc_features = skb->dev->hw_enc_features & features;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index e6a7bd15b9b7..2f748452b4aa 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1322,7 +1322,7 @@ emsgsize:
headersize == sizeof(struct ipv6hdr) &&
length < mtu - headersize &&
!(flags & MSG_MORE) &&
- rt->dst.dev->features & NETIF_F_V6_CSUM)
+ rt->dst.dev->features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM))
csummode = CHECKSUM_PARTIAL;
if (sk->sk_type == SOCK_DGRAM || sk->sk_type == SOCK_RAW) {
diff --git a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c
index 238e70c3f7b7..6ce309928841 100644
--- a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c
@@ -136,7 +136,8 @@ static void nf_nat_ipv6_csum_recalc(struct sk_buff *skb,
if (skb->ip_summed != CHECKSUM_PARTIAL) {
if (!(rt->rt6i_flags & RTF_LOCAL) &&
- (!skb->dev || skb->dev->features & NETIF_F_V6_CSUM)) {
+ (!skb->dev || skb->dev->features &
+ (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM))) {
skb->ip_summed = CHECKSUM_PARTIAL;
skb->csum_start = skb_headroom(skb) +
skb_network_offset(skb) +
diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c
index 010ddeec135f..d952d67f904d 100644
--- a/net/netfilter/ipvs/ip_vs_proto_sctp.c
+++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c
@@ -169,7 +169,7 @@ sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
/* Only update csum if we really have to */
if (sctph->dest != cp->dport || payload_csum ||
(skb->ip_summed == CHECKSUM_PARTIAL &&
- !(skb_dst(skb)->dev->features & NETIF_F_SCTP_CSUM))) {
+ !(skb_dst(skb)->dev->features & NETIF_F_SCTP_CRC))) {
sctph->dest = cp->dport;
sctp_nat_csum(skb, sctph, sctphoff);
} else if (skb->ip_summed != CHECKSUM_PARTIAL) {
diff --git a/net/sctp/output.c b/net/sctp/output.c
index abe7c2db2412..9d610eddd19e 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -534,7 +534,7 @@ int sctp_packet_transmit(struct sctp_packet *packet)
* by CRC32-C as described in <draft-ietf-tsvwg-sctpcsum-02.txt>.
*/
if (!sctp_checksum_disable) {
- if (!(dst->dev->features & NETIF_F_SCTP_CSUM) ||
+ if (!(dst->dev->features & NETIF_F_SCTP_CRC) ||
(dst_xfrm(dst) != NULL) || packet->ipfragok) {
sh->checksum = sctp_compute_cksum(nskb, 0);
} else {