* [PATCH v5 net-next 1/7] ipeh: Create exthdrs_options.c and ipeh.h
2019-10-03 21:57 [PATCH v5 net-next 0/7] ipv6: Extension header infrastructure Tom Herbert
@ 2019-10-03 21:57 ` Tom Herbert
2019-10-06 13:02 ` Simon Horman
2019-10-03 21:57 ` [PATCH v5 net-next 2/7] ipeh: Move generic EH functions to exthdrs_common.c Tom Herbert
` (5 subsequent siblings)
6 siblings, 1 reply; 20+ messages in thread
From: Tom Herbert @ 2019-10-03 21:57 UTC (permalink / raw)
To: davem, netdev; +Cc: Tom Herbert, Tom Herbert
From: Tom Herbert <tom@quantonium.net>
Create exthdrs_options.c to hold code related to specific Hop-by-Hop
and Destination extension header options. Move related functions in
exthdrs.c to the new file.
Create include net/ipeh.h to contain common definitions for IP extension
headers.
Signed-off-by: Tom Herbert <tom@herbertland.com>
---
include/net/ipeh.h | 22 +++++
include/net/ipv6.h | 1 +
net/ipv6/Makefile | 2 +-
net/ipv6/exthdrs.c | 204 ---------------------------------------------
net/ipv6/exthdrs_options.c | 201 ++++++++++++++++++++++++++++++++++++++++++++
5 files changed, 225 insertions(+), 205 deletions(-)
create mode 100644 include/net/ipeh.h
create mode 100644 net/ipv6/exthdrs_options.c
diff --git a/include/net/ipeh.h b/include/net/ipeh.h
new file mode 100644
index 0000000..ec2d186
--- /dev/null
+++ b/include/net/ipeh.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _NET_IPEH_H
+#define _NET_IPEH_H
+
+#include <linux/skbuff.h>
+
+/*
+ * Parsing tlv encoded headers.
+ *
+ * Parsing function "func" returns true, if parsing succeed
+ * and false, if it failed.
+ * It MUST NOT touch skb->h.
+ */
+struct tlvtype_proc {
+ int type;
+ bool (*func)(struct sk_buff *skb, int offset);
+};
+
+extern const struct tlvtype_proc tlvprocdestopt_lst[];
+extern const struct tlvtype_proc tlvprochopopt_lst[];
+
+#endif /* _NET_IPEH_H */
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index d04b7ab..6fc584c 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -20,6 +20,7 @@
#include <net/flow_dissector.h>
#include <net/snmp.h>
#include <net/netns/hash.h>
+#include <net/ipeh.h>
#define SIN6_LEN_RFC2133 24
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index 8ccf355..df3919b 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -10,7 +10,7 @@ ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \
route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \
raw.o icmp.o mcast.o reassembly.o tcp_ipv6.o ping.o \
exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o \
- udp_offload.o seg6.o fib6_notifier.o
+ udp_offload.o seg6.o fib6_notifier.o exthdrs_options.o
ipv6-offload := ip6_offload.o tcpv6_offload.o exthdrs_offload.o
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index ab5add0..664491e 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -39,7 +39,6 @@
#include <net/ndisc.h>
#include <net/ip6_route.h>
#include <net/addrconf.h>
-#include <net/calipso.h>
#if IS_ENABLED(CONFIG_IPV6_MIP6)
#include <net/xfrm.h>
#endif
@@ -51,19 +50,6 @@
#include <linux/uaccess.h>
-/*
- * Parsing tlv encoded headers.
- *
- * Parsing function "func" returns true, if parsing succeed
- * and false, if it failed.
- * It MUST NOT touch skb->h.
- */
-
-struct tlvtype_proc {
- int type;
- bool (*func)(struct sk_buff *skb, int offset);
-};
-
/*********************
Generic functions
*********************/
@@ -200,80 +186,6 @@ static bool ip6_parse_tlv(const struct tlvtype_proc *procs,
return false;
}
-/*****************************
- Destination options header.
- *****************************/
-
-#if IS_ENABLED(CONFIG_IPV6_MIP6)
-static bool ipv6_dest_hao(struct sk_buff *skb, int optoff)
-{
- struct ipv6_destopt_hao *hao;
- struct inet6_skb_parm *opt = IP6CB(skb);
- struct ipv6hdr *ipv6h = ipv6_hdr(skb);
- int ret;
-
- if (opt->dsthao) {
- net_dbg_ratelimited("hao duplicated\n");
- goto discard;
- }
- opt->dsthao = opt->dst1;
- opt->dst1 = 0;
-
- hao = (struct ipv6_destopt_hao *)(skb_network_header(skb) + optoff);
-
- if (hao->length != 16) {
- net_dbg_ratelimited("hao invalid option length = %d\n",
- hao->length);
- goto discard;
- }
-
- if (!(ipv6_addr_type(&hao->addr) & IPV6_ADDR_UNICAST)) {
- net_dbg_ratelimited("hao is not an unicast addr: %pI6\n",
- &hao->addr);
- goto discard;
- }
-
- ret = xfrm6_input_addr(skb, (xfrm_address_t *)&ipv6h->daddr,
- (xfrm_address_t *)&hao->addr, IPPROTO_DSTOPTS);
- if (unlikely(ret < 0))
- goto discard;
-
- if (skb_cloned(skb)) {
- if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
- goto discard;
-
- /* update all variable using below by copied skbuff */
- hao = (struct ipv6_destopt_hao *)(skb_network_header(skb) +
- optoff);
- ipv6h = ipv6_hdr(skb);
- }
-
- if (skb->ip_summed == CHECKSUM_COMPLETE)
- skb->ip_summed = CHECKSUM_NONE;
-
- swap(ipv6h->saddr, hao->addr);
-
- if (skb->tstamp == 0)
- __net_timestamp(skb);
-
- return true;
-
- discard:
- kfree_skb(skb);
- return false;
-}
-#endif
-
-static const struct tlvtype_proc tlvprocdestopt_lst[] = {
-#if IS_ENABLED(CONFIG_IPV6_MIP6)
- {
- .type = IPV6_TLV_HAO,
- .func = ipv6_dest_hao,
- },
-#endif
- {-1, NULL}
-};
-
static int ipv6_destopt_rcv(struct sk_buff *skb)
{
struct inet6_dev *idev = __in6_dev_get(skb->dev);
@@ -702,122 +614,6 @@ void ipv6_exthdrs_exit(void)
inet6_del_protocol(&rthdr_protocol, IPPROTO_ROUTING);
}
-/**********************************
- Hop-by-hop options.
- **********************************/
-
-/*
- * Note: we cannot rely on skb_dst(skb) before we assign it in ip6_route_input().
- */
-static inline struct inet6_dev *ipv6_skb_idev(struct sk_buff *skb)
-{
- return skb_dst(skb) ? ip6_dst_idev(skb_dst(skb)) : __in6_dev_get(skb->dev);
-}
-
-static inline struct net *ipv6_skb_net(struct sk_buff *skb)
-{
- return skb_dst(skb) ? dev_net(skb_dst(skb)->dev) : dev_net(skb->dev);
-}
-
-/* Router Alert as of RFC 2711 */
-
-static bool ipv6_hop_ra(struct sk_buff *skb, int optoff)
-{
- const unsigned char *nh = skb_network_header(skb);
-
- if (nh[optoff + 1] == 2) {
- IP6CB(skb)->flags |= IP6SKB_ROUTERALERT;
- memcpy(&IP6CB(skb)->ra, nh + optoff + 2, sizeof(IP6CB(skb)->ra));
- return true;
- }
- net_dbg_ratelimited("ipv6_hop_ra: wrong RA length %d\n",
- nh[optoff + 1]);
- kfree_skb(skb);
- return false;
-}
-
-/* Jumbo payload */
-
-static bool ipv6_hop_jumbo(struct sk_buff *skb, int optoff)
-{
- const unsigned char *nh = skb_network_header(skb);
- struct inet6_dev *idev = __in6_dev_get_safely(skb->dev);
- struct net *net = ipv6_skb_net(skb);
- u32 pkt_len;
-
- if (nh[optoff + 1] != 4 || (optoff & 3) != 2) {
- net_dbg_ratelimited("ipv6_hop_jumbo: wrong jumbo opt length/alignment %d\n",
- nh[optoff+1]);
- __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
- goto drop;
- }
-
- pkt_len = ntohl(*(__be32 *)(nh + optoff + 2));
- if (pkt_len <= IPV6_MAXPLEN) {
- __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
- icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff+2);
- return false;
- }
- if (ipv6_hdr(skb)->payload_len) {
- __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
- icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff);
- return false;
- }
-
- if (pkt_len > skb->len - sizeof(struct ipv6hdr)) {
- __IP6_INC_STATS(net, idev, IPSTATS_MIB_INTRUNCATEDPKTS);
- goto drop;
- }
-
- if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr)))
- goto drop;
-
- IP6CB(skb)->flags |= IP6SKB_JUMBOGRAM;
- return true;
-
-drop:
- kfree_skb(skb);
- return false;
-}
-
-/* CALIPSO RFC 5570 */
-
-static bool ipv6_hop_calipso(struct sk_buff *skb, int optoff)
-{
- const unsigned char *nh = skb_network_header(skb);
-
- if (nh[optoff + 1] < 8)
- goto drop;
-
- if (nh[optoff + 6] * 4 + 8 > nh[optoff + 1])
- goto drop;
-
- if (!calipso_validate(skb, nh + optoff))
- goto drop;
-
- return true;
-
-drop:
- kfree_skb(skb);
- return false;
-}
-
-static const struct tlvtype_proc tlvprochopopt_lst[] = {
- {
- .type = IPV6_TLV_ROUTERALERT,
- .func = ipv6_hop_ra,
- },
- {
- .type = IPV6_TLV_JUMBO,
- .func = ipv6_hop_jumbo,
- },
- {
- .type = IPV6_TLV_CALIPSO,
- .func = ipv6_hop_calipso,
- },
- { -1, }
-};
-
int ipv6_parse_hopopts(struct sk_buff *skb)
{
struct inet6_skb_parm *opt = IP6CB(skb);
diff --git a/net/ipv6/exthdrs_options.c b/net/ipv6/exthdrs_options.c
new file mode 100644
index 0000000..032e072
--- /dev/null
+++ b/net/ipv6/exthdrs_options.c
@@ -0,0 +1,201 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/errno.h>
+#include <linux/in6.h>
+#include <linux/net.h>
+#include <linux/netdevice.h>
+#include <linux/socket.h>
+#include <linux/types.h>
+#include <net/calipso.h>
+#include <net/ipv6.h>
+#include <net/ip6_route.h>
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
+#include <net/xfrm.h>
+#endif
+
+/* Destination options header */
+
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
+static bool ipv6_dest_hao(struct sk_buff *skb, int optoff)
+{
+ struct ipv6_destopt_hao *hao;
+ struct inet6_skb_parm *opt = IP6CB(skb);
+ struct ipv6hdr *ipv6h = ipv6_hdr(skb);
+ int ret;
+
+ if (opt->dsthao) {
+ net_dbg_ratelimited("hao duplicated\n");
+ goto discard;
+ }
+ opt->dsthao = opt->dst1;
+ opt->dst1 = 0;
+
+ hao = (struct ipv6_destopt_hao *)(skb_network_header(skb) + optoff);
+
+ if (hao->length != 16) {
+ net_dbg_ratelimited("hao invalid option length = %d\n",
+ hao->length);
+ goto discard;
+ }
+
+ if (!(ipv6_addr_type(&hao->addr) & IPV6_ADDR_UNICAST)) {
+ net_dbg_ratelimited("hao is not an unicast addr: %pI6\n",
+ &hao->addr);
+ goto discard;
+ }
+
+ ret = xfrm6_input_addr(skb, (xfrm_address_t *)&ipv6h->daddr,
+ (xfrm_address_t *)&hao->addr, IPPROTO_DSTOPTS);
+ if (unlikely(ret < 0))
+ goto discard;
+
+ if (skb_cloned(skb)) {
+ if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
+ goto discard;
+
+ /* update all variable using below by copied skbuff */
+ hao = (struct ipv6_destopt_hao *)(skb_network_header(skb) +
+ optoff);
+ ipv6h = ipv6_hdr(skb);
+ }
+
+ if (skb->ip_summed == CHECKSUM_COMPLETE)
+ skb->ip_summed = CHECKSUM_NONE;
+
+ swap(ipv6h->saddr, hao->addr);
+
+ if (skb->tstamp == 0)
+ __net_timestamp(skb);
+
+ return true;
+
+ discard:
+ kfree_skb(skb);
+ return false;
+}
+#endif
+
+const struct tlvtype_proc tlvprocdestopt_lst[] = {
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
+ {
+ .type = IPV6_TLV_HAO,
+ .func = ipv6_dest_hao,
+ },
+#endif
+ {-1, NULL}
+};
+
+/* Hop-by-hop options */
+
+/* Note: we cannot rely on skb_dst(skb) before we assign it in
+ * ip6_route_input().
+ */
+static inline struct inet6_dev *ipv6_skb_idev(struct sk_buff *skb)
+{
+ return skb_dst(skb) ? ip6_dst_idev(skb_dst(skb)) :
+ __in6_dev_get(skb->dev);
+}
+
+static inline struct net *ipv6_skb_net(struct sk_buff *skb)
+{
+ return skb_dst(skb) ? dev_net(skb_dst(skb)->dev) : dev_net(skb->dev);
+}
+
+/* Router Alert as of RFC 2711 */
+
+static bool ipv6_hop_ra(struct sk_buff *skb, int optoff)
+{
+ const unsigned char *nh = skb_network_header(skb);
+
+ if (nh[optoff + 1] == 2) {
+ IP6CB(skb)->flags |= IP6SKB_ROUTERALERT;
+ memcpy(&IP6CB(skb)->ra, nh + optoff + 2,
+ sizeof(IP6CB(skb)->ra));
+ return true;
+ }
+ net_dbg_ratelimited("%s: wrong RA length %d\n",
+ __func__, nh[optoff + 1]);
+ kfree_skb(skb);
+ return false;
+}
+
+/* Jumbo payload */
+
+static bool ipv6_hop_jumbo(struct sk_buff *skb, int optoff)
+{
+ const unsigned char *nh = skb_network_header(skb);
+ struct inet6_dev *idev = __in6_dev_get_safely(skb->dev);
+ struct net *net = ipv6_skb_net(skb);
+ u32 pkt_len;
+
+ if (nh[optoff + 1] != 4 || (optoff & 3) != 2) {
+ net_dbg_ratelimited("%s: wrong jumbo opt length/alignment %d\n",
+ __func__, nh[optoff + 1]);
+ __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
+ goto drop;
+ }
+
+ pkt_len = ntohl(*(__be32 *)(nh + optoff + 2));
+ if (pkt_len <= IPV6_MAXPLEN) {
+ __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
+ icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff + 2);
+ return false;
+ }
+ if (ipv6_hdr(skb)->payload_len) {
+ __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
+ icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff);
+ return false;
+ }
+
+ if (pkt_len > skb->len - sizeof(struct ipv6hdr)) {
+ __IP6_INC_STATS(net, idev, IPSTATS_MIB_INTRUNCATEDPKTS);
+ goto drop;
+ }
+
+ if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr)))
+ goto drop;
+
+ IP6CB(skb)->flags |= IP6SKB_JUMBOGRAM;
+ return true;
+
+drop:
+ kfree_skb(skb);
+ return false;
+}
+
+/* CALIPSO RFC 5570 */
+
+static bool ipv6_hop_calipso(struct sk_buff *skb, int optoff)
+{
+ const unsigned char *nh = skb_network_header(skb);
+
+ if (nh[optoff + 1] < 8)
+ goto drop;
+
+ if (nh[optoff + 6] * 4 + 8 > nh[optoff + 1])
+ goto drop;
+
+ if (!calipso_validate(skb, nh + optoff))
+ goto drop;
+
+ return true;
+
+drop:
+ kfree_skb(skb);
+ return false;
+}
+
+const struct tlvtype_proc tlvprochopopt_lst[] = {
+ {
+ .type = IPV6_TLV_ROUTERALERT,
+ .func = ipv6_hop_ra,
+ },
+ {
+ .type = IPV6_TLV_JUMBO,
+ .func = ipv6_hop_jumbo,
+ },
+ {
+ .type = IPV6_TLV_CALIPSO,
+ .func = ipv6_hop_calipso,
+ },
+ { -1, }
+};
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 1/7] ipeh: Create exthdrs_options.c and ipeh.h
2019-10-03 21:57 ` [PATCH v5 net-next 1/7] ipeh: Create exthdrs_options.c and ipeh.h Tom Herbert
@ 2019-10-06 13:02 ` Simon Horman
2019-12-14 18:44 ` Tom Herbert
0 siblings, 1 reply; 20+ messages in thread
From: Simon Horman @ 2019-10-06 13:02 UTC (permalink / raw)
To: Tom Herbert; +Cc: davem, netdev, Tom Herbert
On Thu, Oct 03, 2019 at 02:57:58PM -0700, Tom Herbert wrote:
> From: Tom Herbert <tom@quantonium.net>
>
> Create exthdrs_options.c to hold code related to specific Hop-by-Hop
> and Destination extension header options. Move related functions in
> exthdrs.c to the new file.
>
> Create include net/ipeh.h to contain common definitions for IP extension
> headers.
>
> Signed-off-by: Tom Herbert <tom@herbertland.com>
Hi Tom,
I'm not entirely clear on the direction this patchset it going in -
I assume its part of a larger journey - but in isolation this
patch seems fine to me.
Reviewed-by: Simon Horman <simon.horman@netronome.com>
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH v5 net-next 1/7] ipeh: Create exthdrs_options.c and ipeh.h
2019-10-06 13:02 ` Simon Horman
@ 2019-12-14 18:44 ` Tom Herbert
0 siblings, 0 replies; 20+ messages in thread
From: Tom Herbert @ 2019-12-14 18:44 UTC (permalink / raw)
To: Simon Horman
Cc: David S. Miller, Linux Kernel Network Developers, Tom Herbert
On Sun, Oct 6, 2019 at 6:02 AM Simon Horman <simon.horman@netronome.com> wrote:
>
> On Thu, Oct 03, 2019 at 02:57:58PM -0700, Tom Herbert wrote:
> > From: Tom Herbert <tom@quantonium.net>
> >
> > Create exthdrs_options.c to hold code related to specific Hop-by-Hop
> > and Destination extension header options. Move related functions in
> > exthdrs.c to the new file.
> >
> > Create include net/ipeh.h to contain common definitions for IP extension
> > headers.
> >
> > Signed-off-by: Tom Herbert <tom@herbertland.com>
>
> Hi Tom,
>
> I'm not entirely clear on the direction this patchset it going in -
> I assume its part of a larger journey - but in isolation this
> patch seems fine to me.
Hi Simon,
Sorry for the delayed response.
The overall direction here is to make extension headers as well as
TLVs more usable and better performance. This patch isolates extension
headers and TLV processing as a common facility which will allow
multiple uses (SR TLVs can use this, IPv4 DestOpt and HBH options, and
potentially other use cases of TLVs). Also, this set allows
non-privileged applications to set DO and HBH options with correct
permissions per option type and also validates options are properly
formed per some rules (for instance the length of an option being set
by an application can be validated that it falls in a required range).
Future patch sets will allow setting specific HBH and DO options on a
socket (as opposed to all or nothing currently done). Experimental
options will be allowed following the same format used for TCP
experimental options. I'd also like to add ability to parse options in
flow dissector and probably a BPF helper for XDP.
>
> Reviewed-by: Simon Horman <simon.horman@netronome.com>
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH v5 net-next 2/7] ipeh: Move generic EH functions to exthdrs_common.c
2019-10-03 21:57 [PATCH v5 net-next 0/7] ipv6: Extension header infrastructure Tom Herbert
2019-10-03 21:57 ` [PATCH v5 net-next 1/7] ipeh: Create exthdrs_options.c and ipeh.h Tom Herbert
@ 2019-10-03 21:57 ` Tom Herbert
2019-10-06 13:00 ` Simon Horman
2019-10-03 21:58 ` [PATCH v5 net-next 3/7] ipeh: Generic TLV parser Tom Herbert
` (4 subsequent siblings)
6 siblings, 1 reply; 20+ messages in thread
From: Tom Herbert @ 2019-10-03 21:57 UTC (permalink / raw)
To: davem, netdev; +Cc: Tom Herbert, Tom Herbert
From: Tom Herbert <tom@quantonium.net>
Move generic functions in exthdrs.c to new exthdrs_common.c so that
exthdrs.c only contains functions that are specific to IPv6 processing,
and exthdrs_common.c contains functions that are generic. These
functions include those that will be used with IPv4 extension headers.
Generic extension header related functions are prefixed by ipeh_.
Signed-off-by: Tom Herbert <tom@herbertland.com>
---
include/net/ipeh.h | 12 ++++
include/net/ipv6.h | 9 ---
net/dccp/ipv6.c | 2 +-
net/ipv6/Kconfig | 4 ++
net/ipv6/Makefile | 1 +
net/ipv6/calipso.c | 6 +-
net/ipv6/exthdrs.c | 138 --------------------------------------------
net/ipv6/exthdrs_common.c | 144 ++++++++++++++++++++++++++++++++++++++++++++++
net/ipv6/ipv6_sockglue.c | 2 +-
net/ipv6/raw.c | 2 +-
net/ipv6/tcp_ipv6.c | 2 +-
net/ipv6/udp.c | 2 +-
net/l2tp/l2tp_ip6.c | 2 +-
net/sctp/ipv6.c | 2 +-
14 files changed, 171 insertions(+), 157 deletions(-)
create mode 100644 net/ipv6/exthdrs_common.c
diff --git a/include/net/ipeh.h b/include/net/ipeh.h
index ec2d186..3b24831 100644
--- a/include/net/ipeh.h
+++ b/include/net/ipeh.h
@@ -19,4 +19,16 @@ struct tlvtype_proc {
extern const struct tlvtype_proc tlvprocdestopt_lst[];
extern const struct tlvtype_proc tlvprochopopt_lst[];
+struct ipv6_txoptions;
+struct ipv6_opt_hdr;
+
+struct ipv6_txoptions *ipeh_dup_options(struct sock *sk,
+ struct ipv6_txoptions *opt);
+struct ipv6_txoptions *ipeh_renew_options(struct sock *sk,
+ struct ipv6_txoptions *opt,
+ int newtype,
+ struct ipv6_opt_hdr *newopt);
+struct ipv6_txoptions *ipeh_fixup_options(struct ipv6_txoptions *opt_space,
+ struct ipv6_txoptions *opt);
+
#endif /* _NET_IPEH_H */
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 6fc584c..d34c36c 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -429,15 +429,6 @@ int ip6_ra_control(struct sock *sk, int sel);
int ipv6_parse_hopopts(struct sk_buff *skb);
-struct ipv6_txoptions *ipv6_dup_options(struct sock *sk,
- struct ipv6_txoptions *opt);
-struct ipv6_txoptions *ipv6_renew_options(struct sock *sk,
- struct ipv6_txoptions *opt,
- int newtype,
- struct ipv6_opt_hdr *newopt);
-struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
- struct ipv6_txoptions *opt);
-
bool ipv6_opt_accepted(const struct sock *sk, const struct sk_buff *skb,
const struct inet6_skb_parm *opt);
struct ipv6_txoptions *ipv6_update_options(struct sock *sk,
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 25aab67..b8843c1 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -515,7 +515,7 @@ static struct sock *dccp_v6_request_recv_sock(const struct sock *sk,
if (!opt)
opt = rcu_dereference(np->opt);
if (opt) {
- opt = ipv6_dup_options(newsk, opt);
+ opt = ipeh_dup_options(newsk, opt);
RCU_INIT_POINTER(newnp->opt, opt);
}
inet_csk(newsk)->icsk_ext_hdr_len = 0;
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index ae1344e..700fcea 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -3,9 +3,13 @@
# IPv6 configuration
#
+config EXTHDRS
+ bool
+
# IPv6 as module will cause a CRASH if you try to unload it
menuconfig IPV6
tristate "The IPv6 protocol"
+ select EXTHDRS
default y
---help---
Support for IP version 6 (IPv6).
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index df3919b..0bcab81 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -44,6 +44,7 @@ obj-$(CONFIG_IPV6_SIT) += sit.o
obj-$(CONFIG_IPV6_TUNNEL) += ip6_tunnel.o
obj-$(CONFIG_IPV6_GRE) += ip6_gre.o
obj-$(CONFIG_IPV6_FOU) += fou6.o
+obj-$(CONFIG_EXTHDRS) += exthdrs_common.o
obj-y += addrconf_core.o exthdrs_core.o ip6_checksum.o ip6_icmp.o
obj-$(CONFIG_INET) += output_core.o protocol.o $(ipv6-offload)
diff --git a/net/ipv6/calipso.c b/net/ipv6/calipso.c
index 221c81f..9c84848 100644
--- a/net/ipv6/calipso.c
+++ b/net/ipv6/calipso.c
@@ -785,7 +785,7 @@ static int calipso_opt_update(struct sock *sk, struct ipv6_opt_hdr *hop)
{
struct ipv6_txoptions *old = txopt_get(inet6_sk(sk)), *txopts;
- txopts = ipv6_renew_options(sk, old, IPV6_HOPOPTS, hop);
+ txopts = ipeh_renew_options(sk, old, IPV6_HOPOPTS, hop);
txopt_put(old);
if (IS_ERR(txopts))
return PTR_ERR(txopts);
@@ -1207,7 +1207,7 @@ static int calipso_req_setattr(struct request_sock *req,
if (IS_ERR(new))
return PTR_ERR(new);
- txopts = ipv6_renew_options(sk, req_inet->ipv6_opt, IPV6_HOPOPTS, new);
+ txopts = ipeh_renew_options(sk, req_inet->ipv6_opt, IPV6_HOPOPTS, new);
kfree(new);
@@ -1244,7 +1244,7 @@ static void calipso_req_delattr(struct request_sock *req)
if (calipso_opt_del(req_inet->ipv6_opt->hopopt, &new))
return; /* Nothing to do */
- txopts = ipv6_renew_options(sk, req_inet->ipv6_opt, IPV6_HOPOPTS, new);
+ txopts = ipeh_renew_options(sk, req_inet->ipv6_opt, IPV6_HOPOPTS, new);
if (!IS_ERR(txopts)) {
txopts = xchg(&req_inet->ipv6_opt, txopts);
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 664491e..e12d3a5 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -784,144 +784,6 @@ void ipv6_push_frag_opts(struct sk_buff *skb, struct ipv6_txoptions *opt, u8 *pr
}
EXPORT_SYMBOL(ipv6_push_frag_opts);
-struct ipv6_txoptions *
-ipv6_dup_options(struct sock *sk, struct ipv6_txoptions *opt)
-{
- struct ipv6_txoptions *opt2;
-
- opt2 = sock_kmalloc(sk, opt->tot_len, GFP_ATOMIC);
- if (opt2) {
- long dif = (char *)opt2 - (char *)opt;
- memcpy(opt2, opt, opt->tot_len);
- if (opt2->hopopt)
- *((char **)&opt2->hopopt) += dif;
- if (opt2->dst0opt)
- *((char **)&opt2->dst0opt) += dif;
- if (opt2->dst1opt)
- *((char **)&opt2->dst1opt) += dif;
- if (opt2->srcrt)
- *((char **)&opt2->srcrt) += dif;
- refcount_set(&opt2->refcnt, 1);
- }
- return opt2;
-}
-EXPORT_SYMBOL_GPL(ipv6_dup_options);
-
-static void ipv6_renew_option(int renewtype,
- struct ipv6_opt_hdr **dest,
- struct ipv6_opt_hdr *old,
- struct ipv6_opt_hdr *new,
- int newtype, char **p)
-{
- struct ipv6_opt_hdr *src;
-
- src = (renewtype == newtype ? new : old);
- if (!src)
- return;
-
- memcpy(*p, src, ipv6_optlen(src));
- *dest = (struct ipv6_opt_hdr *)*p;
- *p += CMSG_ALIGN(ipv6_optlen(*dest));
-}
-
-/**
- * ipv6_renew_options - replace a specific ext hdr with a new one.
- *
- * @sk: sock from which to allocate memory
- * @opt: original options
- * @newtype: option type to replace in @opt
- * @newopt: new option of type @newtype to replace (user-mem)
- * @newoptlen: length of @newopt
- *
- * Returns a new set of options which is a copy of @opt with the
- * option type @newtype replaced with @newopt.
- *
- * @opt may be NULL, in which case a new set of options is returned
- * containing just @newopt.
- *
- * @newopt may be NULL, in which case the specified option type is
- * not copied into the new set of options.
- *
- * The new set of options is allocated from the socket option memory
- * buffer of @sk.
- */
-struct ipv6_txoptions *
-ipv6_renew_options(struct sock *sk, struct ipv6_txoptions *opt,
- int newtype, struct ipv6_opt_hdr *newopt)
-{
- int tot_len = 0;
- char *p;
- struct ipv6_txoptions *opt2;
-
- if (opt) {
- if (newtype != IPV6_HOPOPTS && opt->hopopt)
- tot_len += CMSG_ALIGN(ipv6_optlen(opt->hopopt));
- if (newtype != IPV6_RTHDRDSTOPTS && opt->dst0opt)
- tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst0opt));
- if (newtype != IPV6_RTHDR && opt->srcrt)
- tot_len += CMSG_ALIGN(ipv6_optlen(opt->srcrt));
- if (newtype != IPV6_DSTOPTS && opt->dst1opt)
- tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst1opt));
- }
-
- if (newopt)
- tot_len += CMSG_ALIGN(ipv6_optlen(newopt));
-
- if (!tot_len)
- return NULL;
-
- tot_len += sizeof(*opt2);
- opt2 = sock_kmalloc(sk, tot_len, GFP_ATOMIC);
- if (!opt2)
- return ERR_PTR(-ENOBUFS);
-
- memset(opt2, 0, tot_len);
- refcount_set(&opt2->refcnt, 1);
- opt2->tot_len = tot_len;
- p = (char *)(opt2 + 1);
-
- ipv6_renew_option(IPV6_HOPOPTS, &opt2->hopopt,
- (opt ? opt->hopopt : NULL),
- newopt, newtype, &p);
- ipv6_renew_option(IPV6_RTHDRDSTOPTS, &opt2->dst0opt,
- (opt ? opt->dst0opt : NULL),
- newopt, newtype, &p);
- ipv6_renew_option(IPV6_RTHDR,
- (struct ipv6_opt_hdr **)&opt2->srcrt,
- (opt ? (struct ipv6_opt_hdr *)opt->srcrt : NULL),
- newopt, newtype, &p);
- ipv6_renew_option(IPV6_DSTOPTS, &opt2->dst1opt,
- (opt ? opt->dst1opt : NULL),
- newopt, newtype, &p);
-
- opt2->opt_nflen = (opt2->hopopt ? ipv6_optlen(opt2->hopopt) : 0) +
- (opt2->dst0opt ? ipv6_optlen(opt2->dst0opt) : 0) +
- (opt2->srcrt ? ipv6_optlen(opt2->srcrt) : 0);
- opt2->opt_flen = (opt2->dst1opt ? ipv6_optlen(opt2->dst1opt) : 0);
-
- return opt2;
-}
-
-struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
- struct ipv6_txoptions *opt)
-{
- /*
- * ignore the dest before srcrt unless srcrt is being included.
- * --yoshfuji
- */
- if (opt && opt->dst0opt && !opt->srcrt) {
- if (opt_space != opt) {
- memcpy(opt_space, opt, sizeof(*opt_space));
- opt = opt_space;
- }
- opt->opt_nflen -= ipv6_optlen(opt->dst0opt);
- opt->dst0opt = NULL;
- }
-
- return opt;
-}
-EXPORT_SYMBOL_GPL(ipv6_fixup_options);
-
/**
* fl6_update_dst - update flowi destination address with info given
* by srcrt option, if any.
diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
new file mode 100644
index 0000000..2c68184
--- /dev/null
+++ b/net/ipv6/exthdrs_common.c
@@ -0,0 +1,144 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Extension header and TLV library code that is not specific to IPv6. */
+#include <linux/export.h>
+#include <net/ipv6.h>
+
+struct ipv6_txoptions *
+ipeh_dup_options(struct sock *sk, struct ipv6_txoptions *opt)
+{
+ struct ipv6_txoptions *opt2;
+
+ opt2 = sock_kmalloc(sk, opt->tot_len, GFP_ATOMIC);
+ if (opt2) {
+ long dif = (char *)opt2 - (char *)opt;
+
+ memcpy(opt2, opt, opt->tot_len);
+ if (opt2->hopopt)
+ *((char **)&opt2->hopopt) += dif;
+ if (opt2->dst0opt)
+ *((char **)&opt2->dst0opt) += dif;
+ if (opt2->dst1opt)
+ *((char **)&opt2->dst1opt) += dif;
+ if (opt2->srcrt)
+ *((char **)&opt2->srcrt) += dif;
+ refcount_set(&opt2->refcnt, 1);
+ }
+ return opt2;
+}
+EXPORT_SYMBOL_GPL(ipeh_dup_options);
+
+static void ipeh_renew_option(int renewtype,
+ struct ipv6_opt_hdr **dest,
+ struct ipv6_opt_hdr *old,
+ struct ipv6_opt_hdr *new,
+ int newtype, char **p)
+{
+ struct ipv6_opt_hdr *src;
+
+ src = (renewtype == newtype ? new : old);
+ if (!src)
+ return;
+
+ memcpy(*p, src, ipv6_optlen(src));
+ *dest = (struct ipv6_opt_hdr *)*p;
+ *p += CMSG_ALIGN(ipv6_optlen(*dest));
+}
+
+/**
+ * ipeh_renew_options - replace a specific ext hdr with a new one.
+ *
+ * @sk: sock from which to allocate memory
+ * @opt: original options
+ * @newtype: option type to replace in @opt
+ * @newopt: new option of type @newtype to replace (user-mem)
+ * @newoptlen: length of @newopt
+ *
+ * Returns a new set of options which is a copy of @opt with the
+ * option type @newtype replaced with @newopt.
+ *
+ * @opt may be NULL, in which case a new set of options is returned
+ * containing just @newopt.
+ *
+ * @newopt may be NULL, in which case the specified option type is
+ * not copied into the new set of options.
+ *
+ * The new set of options is allocated from the socket option memory
+ * buffer of @sk.
+ */
+struct ipv6_txoptions *
+ipeh_renew_options(struct sock *sk, struct ipv6_txoptions *opt,
+ int newtype, struct ipv6_opt_hdr *newopt)
+{
+ int tot_len = 0;
+ char *p;
+ struct ipv6_txoptions *opt2;
+
+ if (opt) {
+ if (newtype != IPV6_HOPOPTS && opt->hopopt)
+ tot_len += CMSG_ALIGN(ipv6_optlen(opt->hopopt));
+ if (newtype != IPV6_RTHDRDSTOPTS && opt->dst0opt)
+ tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst0opt));
+ if (newtype != IPV6_RTHDR && opt->srcrt)
+ tot_len += CMSG_ALIGN(ipv6_optlen(opt->srcrt));
+ if (newtype != IPV6_DSTOPTS && opt->dst1opt)
+ tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst1opt));
+ }
+
+ if (newopt)
+ tot_len += CMSG_ALIGN(ipv6_optlen(newopt));
+
+ if (!tot_len)
+ return NULL;
+
+ tot_len += sizeof(*opt2);
+ opt2 = sock_kmalloc(sk, tot_len, GFP_ATOMIC);
+ if (!opt2)
+ return ERR_PTR(-ENOBUFS);
+
+ memset(opt2, 0, tot_len);
+ refcount_set(&opt2->refcnt, 1);
+ opt2->tot_len = tot_len;
+ p = (char *)(opt2 + 1);
+
+ ipeh_renew_option(IPV6_HOPOPTS, &opt2->hopopt,
+ (opt ? opt->hopopt : NULL),
+ newopt, newtype, &p);
+ ipeh_renew_option(IPV6_RTHDRDSTOPTS, &opt2->dst0opt,
+ (opt ? opt->dst0opt : NULL),
+ newopt, newtype, &p);
+ ipeh_renew_option(IPV6_RTHDR,
+ (struct ipv6_opt_hdr **)&opt2->srcrt,
+ (opt ? (struct ipv6_opt_hdr *)opt->srcrt : NULL),
+ newopt, newtype, &p);
+ ipeh_renew_option(IPV6_DSTOPTS, &opt2->dst1opt,
+ (opt ? opt->dst1opt : NULL),
+ newopt, newtype, &p);
+
+ opt2->opt_nflen = (opt2->hopopt ? ipv6_optlen(opt2->hopopt) : 0) +
+ (opt2->dst0opt ? ipv6_optlen(opt2->dst0opt) : 0) +
+ (opt2->srcrt ? ipv6_optlen(opt2->srcrt) : 0);
+ opt2->opt_flen = (opt2->dst1opt ? ipv6_optlen(opt2->dst1opt) : 0);
+
+ return opt2;
+}
+EXPORT_SYMBOL(ipeh_renew_options);
+
+struct ipv6_txoptions *ipeh_fixup_options(struct ipv6_txoptions *opt_space,
+ struct ipv6_txoptions *opt)
+{
+ /* ignore the dest before srcrt unless srcrt is being included.
+ * --yoshfuji
+ */
+ if (opt && opt->dst0opt && !opt->srcrt) {
+ if (opt_space != opt) {
+ memcpy(opt_space, opt, sizeof(*opt_space));
+ opt = opt_space;
+ }
+ opt->opt_nflen -= ipv6_optlen(opt->dst0opt);
+ opt->dst0opt = NULL;
+ }
+
+ return opt;
+}
+EXPORT_SYMBOL_GPL(ipeh_fixup_options);
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 264c292..8755ecc 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -427,7 +427,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
opt = rcu_dereference_protected(np->opt,
lockdep_sock_is_held(sk));
- opt = ipv6_renew_options(sk, opt, optname, new);
+ opt = ipeh_renew_options(sk, opt, optname, new);
kfree(new);
if (IS_ERR(opt)) {
retv = PTR_ERR(opt);
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 6e1888e..ace11b7 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -889,7 +889,7 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
}
if (flowlabel)
opt = fl6_merge_options(&opt_space, flowlabel, opt);
- opt = ipv6_fixup_options(&opt_space, opt);
+ opt = ipeh_fixup_options(&opt_space, opt);
fl6.flowi6_proto = proto;
fl6.flowi6_mark = ipc6.sockc.mark;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index e3d9f45..9c761c7 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1249,7 +1249,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff *
if (!opt)
opt = rcu_dereference(np->opt);
if (opt) {
- opt = ipv6_dup_options(newsk, opt);
+ opt = ipeh_dup_options(newsk, opt);
RCU_INIT_POINTER(newnp->opt, opt);
}
inet_csk(newsk)->icsk_ext_hdr_len = 0;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index aae4938..a1fbd47 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1386,7 +1386,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
}
if (flowlabel)
opt = fl6_merge_options(&opt_space, flowlabel, opt);
- opt = ipv6_fixup_options(&opt_space, opt);
+ opt = ipeh_fixup_options(&opt_space, opt);
ipc6.opt = opt;
fl6.flowi6_proto = sk->sk_protocol;
diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c
index 687e23a..bc4925c 100644
--- a/net/l2tp/l2tp_ip6.c
+++ b/net/l2tp/l2tp_ip6.c
@@ -590,7 +590,7 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
}
if (flowlabel)
opt = fl6_merge_options(&opt_space, flowlabel, opt);
- opt = ipv6_fixup_options(&opt_space, opt);
+ opt = ipeh_fixup_options(&opt_space, opt);
ipc6.opt = opt;
fl6.flowi6_proto = sk->sk_protocol;
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index dd860fe..39d0eec 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -441,7 +441,7 @@ static void sctp_v6_copy_ip_options(struct sock *sk, struct sock *newsk)
rcu_read_lock();
opt = rcu_dereference(np->opt);
if (opt) {
- opt = ipv6_dup_options(newsk, opt);
+ opt = ipeh_dup_options(newsk, opt);
if (!opt)
pr_err("%s: Failed to copy ip options\n", __func__);
}
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 2/7] ipeh: Move generic EH functions to exthdrs_common.c
2019-10-03 21:57 ` [PATCH v5 net-next 2/7] ipeh: Move generic EH functions to exthdrs_common.c Tom Herbert
@ 2019-10-06 13:00 ` Simon Horman
2019-12-14 18:47 ` Tom Herbert
0 siblings, 1 reply; 20+ messages in thread
From: Simon Horman @ 2019-10-06 13:00 UTC (permalink / raw)
To: Tom Herbert; +Cc: davem, netdev, Tom Herbert
On Thu, Oct 03, 2019 at 02:57:59PM -0700, Tom Herbert wrote:
> From: Tom Herbert <tom@quantonium.net>
>
> Move generic functions in exthdrs.c to new exthdrs_common.c so that
> exthdrs.c only contains functions that are specific to IPv6 processing,
> and exthdrs_common.c contains functions that are generic. These
> functions include those that will be used with IPv4 extension headers.
> Generic extension header related functions are prefixed by ipeh_.
>
> Signed-off-by: Tom Herbert <tom@herbertland.com>
...
> diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
> index 25aab67..b8843c1 100644
> --- a/net/dccp/ipv6.c
> +++ b/net/dccp/ipv6.c
> @@ -515,7 +515,7 @@ static struct sock *dccp_v6_request_recv_sock(const struct sock *sk,
> if (!opt)
> opt = rcu_dereference(np->opt);
> if (opt) {
> - opt = ipv6_dup_options(newsk, opt);
> + opt = ipeh_dup_options(newsk, opt);
> RCU_INIT_POINTER(newnp->opt, opt);
> }
> inet_csk(newsk)->icsk_ext_hdr_len = 0;
> diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
> index ae1344e..700fcea 100644
> --- a/net/ipv6/Kconfig
> +++ b/net/ipv6/Kconfig
> @@ -3,9 +3,13 @@
> # IPv6 configuration
> #
>
> +config EXTHDRS
> + bool
> +
> # IPv6 as module will cause a CRASH if you try to unload it
> menuconfig IPV6
> tristate "The IPv6 protocol"
> + select EXTHDRS
> default y
> ---help---
> Support for IP version 6 (IPv6).
Hi Tom,
could you expand on the motivation for this new Kconfig symbol.
It seems that at this time exthdrs_common.o could simply depend on IPV6.
Otherwise this patch seems fine to me.
> diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
> index df3919b..0bcab81 100644
> --- a/net/ipv6/Makefile
> +++ b/net/ipv6/Makefile
> @@ -44,6 +44,7 @@ obj-$(CONFIG_IPV6_SIT) += sit.o
> obj-$(CONFIG_IPV6_TUNNEL) += ip6_tunnel.o
> obj-$(CONFIG_IPV6_GRE) += ip6_gre.o
> obj-$(CONFIG_IPV6_FOU) += fou6.o
> +obj-$(CONFIG_EXTHDRS) += exthdrs_common.o
>
> obj-y += addrconf_core.o exthdrs_core.o ip6_checksum.o ip6_icmp.o
> obj-$(CONFIG_INET) += output_core.o protocol.o $(ipv6-offload)
...
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 2/7] ipeh: Move generic EH functions to exthdrs_common.c
2019-10-06 13:00 ` Simon Horman
@ 2019-12-14 18:47 ` Tom Herbert
2019-12-18 10:56 ` Simon Horman
0 siblings, 1 reply; 20+ messages in thread
From: Tom Herbert @ 2019-12-14 18:47 UTC (permalink / raw)
To: Simon Horman
Cc: David S. Miller, Linux Kernel Network Developers, Tom Herbert
On Sun, Oct 6, 2019 at 6:00 AM Simon Horman <simon.horman@netronome.com> wrote:
>
> On Thu, Oct 03, 2019 at 02:57:59PM -0700, Tom Herbert wrote:
> > From: Tom Herbert <tom@quantonium.net>
> >
> > Move generic functions in exthdrs.c to new exthdrs_common.c so that
> > exthdrs.c only contains functions that are specific to IPv6 processing,
> > and exthdrs_common.c contains functions that are generic. These
> > functions include those that will be used with IPv4 extension headers.
> > Generic extension header related functions are prefixed by ipeh_.
> >
> > Signed-off-by: Tom Herbert <tom@herbertland.com>
>
> ...
>
> > diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
> > index 25aab67..b8843c1 100644
> > --- a/net/dccp/ipv6.c
> > +++ b/net/dccp/ipv6.c
> > @@ -515,7 +515,7 @@ static struct sock *dccp_v6_request_recv_sock(const struct sock *sk,
> > if (!opt)
> > opt = rcu_dereference(np->opt);
> > if (opt) {
> > - opt = ipv6_dup_options(newsk, opt);
> > + opt = ipeh_dup_options(newsk, opt);
> > RCU_INIT_POINTER(newnp->opt, opt);
> > }
> > inet_csk(newsk)->icsk_ext_hdr_len = 0;
> > diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
> > index ae1344e..700fcea 100644
> > --- a/net/ipv6/Kconfig
> > +++ b/net/ipv6/Kconfig
> > @@ -3,9 +3,13 @@
> > # IPv6 configuration
> > #
> >
> > +config EXTHDRS
> > + bool
> > +
> > # IPv6 as module will cause a CRASH if you try to unload it
> > menuconfig IPV6
> > tristate "The IPv6 protocol"
> > + select EXTHDRS
> > default y
> > ---help---
> > Support for IP version 6 (IPv6).
>
> Hi Tom,
>
> could you expand on the motivation for this new Kconfig symbol.
> It seems that at this time exthdrs_common.o could simply depend on IPV6.
>
It anticipates other uses cases of extension headers, in particular
IPv4 extension headers
(https://tools.ietf.org/html/draft-herbert-ipv4-hbh-destopt-00)
> Otherwise this patch seems fine to me.
>
> > diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
> > index df3919b..0bcab81 100644
> > --- a/net/ipv6/Makefile
> > +++ b/net/ipv6/Makefile
> > @@ -44,6 +44,7 @@ obj-$(CONFIG_IPV6_SIT) += sit.o
> > obj-$(CONFIG_IPV6_TUNNEL) += ip6_tunnel.o
> > obj-$(CONFIG_IPV6_GRE) += ip6_gre.o
> > obj-$(CONFIG_IPV6_FOU) += fou6.o
> > +obj-$(CONFIG_EXTHDRS) += exthdrs_common.o
> >
> > obj-y += addrconf_core.o exthdrs_core.o ip6_checksum.o ip6_icmp.o
> > obj-$(CONFIG_INET) += output_core.o protocol.o $(ipv6-offload)
>
> ...
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 2/7] ipeh: Move generic EH functions to exthdrs_common.c
2019-12-14 18:47 ` Tom Herbert
@ 2019-12-18 10:56 ` Simon Horman
0 siblings, 0 replies; 20+ messages in thread
From: Simon Horman @ 2019-12-18 10:56 UTC (permalink / raw)
To: Tom Herbert; +Cc: David S. Miller, Linux Kernel Network Developers, Tom Herbert
On Sat, Dec 14, 2019 at 10:47:17AM -0800, Tom Herbert wrote:
> On Sun, Oct 6, 2019 at 6:00 AM Simon Horman <simon.horman@netronome.com> wrote:
> >
> > On Thu, Oct 03, 2019 at 02:57:59PM -0700, Tom Herbert wrote:
> > > From: Tom Herbert <tom@quantonium.net>
> > >
> > > Move generic functions in exthdrs.c to new exthdrs_common.c so that
> > > exthdrs.c only contains functions that are specific to IPv6 processing,
> > > and exthdrs_common.c contains functions that are generic. These
> > > functions include those that will be used with IPv4 extension headers.
> > > Generic extension header related functions are prefixed by ipeh_.
> > >
> > > Signed-off-by: Tom Herbert <tom@herbertland.com>
> >
> > ...
> >
> > > diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
> > > index 25aab67..b8843c1 100644
> > > --- a/net/dccp/ipv6.c
> > > +++ b/net/dccp/ipv6.c
> > > @@ -515,7 +515,7 @@ static struct sock *dccp_v6_request_recv_sock(const struct sock *sk,
> > > if (!opt)
> > > opt = rcu_dereference(np->opt);
> > > if (opt) {
> > > - opt = ipv6_dup_options(newsk, opt);
> > > + opt = ipeh_dup_options(newsk, opt);
> > > RCU_INIT_POINTER(newnp->opt, opt);
> > > }
> > > inet_csk(newsk)->icsk_ext_hdr_len = 0;
> > > diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
> > > index ae1344e..700fcea 100644
> > > --- a/net/ipv6/Kconfig
> > > +++ b/net/ipv6/Kconfig
> > > @@ -3,9 +3,13 @@
> > > # IPv6 configuration
> > > #
> > >
> > > +config EXTHDRS
> > > + bool
> > > +
> > > # IPv6 as module will cause a CRASH if you try to unload it
> > > menuconfig IPV6
> > > tristate "The IPv6 protocol"
> > > + select EXTHDRS
> > > default y
> > > ---help---
> > > Support for IP version 6 (IPv6).
> >
> > Hi Tom,
> >
> > could you expand on the motivation for this new Kconfig symbol.
> > It seems that at this time exthdrs_common.o could simply depend on IPV6.
> >
> It anticipates other uses cases of extension headers, in particular
> IPv4 extension headers
> (https://tools.ietf.org/html/draft-herbert-ipv4-hbh-destopt-00)
Thanks Tom,
I would lean towards adding the new Kconfig option when it is needed,
but I don't feel strongly aout this.
> > Otherwise this patch seems fine to me.
> >
> > > diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
> > > index df3919b..0bcab81 100644
> > > --- a/net/ipv6/Makefile
> > > +++ b/net/ipv6/Makefile
> > > @@ -44,6 +44,7 @@ obj-$(CONFIG_IPV6_SIT) += sit.o
> > > obj-$(CONFIG_IPV6_TUNNEL) += ip6_tunnel.o
> > > obj-$(CONFIG_IPV6_GRE) += ip6_gre.o
> > > obj-$(CONFIG_IPV6_FOU) += fou6.o
> > > +obj-$(CONFIG_EXTHDRS) += exthdrs_common.o
> > >
> > > obj-y += addrconf_core.o exthdrs_core.o ip6_checksum.o ip6_icmp.o
> > > obj-$(CONFIG_INET) += output_core.o protocol.o $(ipv6-offload)
> >
> > ...
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH v5 net-next 3/7] ipeh: Generic TLV parser
2019-10-03 21:57 [PATCH v5 net-next 0/7] ipv6: Extension header infrastructure Tom Herbert
2019-10-03 21:57 ` [PATCH v5 net-next 1/7] ipeh: Create exthdrs_options.c and ipeh.h Tom Herbert
2019-10-03 21:57 ` [PATCH v5 net-next 2/7] ipeh: Move generic EH functions to exthdrs_common.c Tom Herbert
@ 2019-10-03 21:58 ` Tom Herbert
2019-10-06 13:05 ` Simon Horman
2019-10-03 21:58 ` [PATCH v5 net-next 4/7] ip6tlvs: Registration of TLV handlers and parameters Tom Herbert
` (3 subsequent siblings)
6 siblings, 1 reply; 20+ messages in thread
From: Tom Herbert @ 2019-10-03 21:58 UTC (permalink / raw)
To: davem, netdev; +Cc: Tom Herbert, Tom Herbert
From: Tom Herbert <tom@quantonium.net>
Create a generic TLV parser. This will be used with various
extension headers that carry options including Destination,
Hop-by-Hop, Segment Routing TLVs, and other cases of simple
stateless parsing.
Signed-off-by: Tom Herbert <tom@herbertland.com>
---
include/net/ipeh.h | 25 ++++++++
net/ipv6/exthdrs.c | 159 +++++++++++-----------------------------------
net/ipv6/exthdrs_common.c | 114 +++++++++++++++++++++++++++++++++
3 files changed, 177 insertions(+), 121 deletions(-)
diff --git a/include/net/ipeh.h b/include/net/ipeh.h
index 3b24831..c1aa7b6 100644
--- a/include/net/ipeh.h
+++ b/include/net/ipeh.h
@@ -31,4 +31,29 @@ struct ipv6_txoptions *ipeh_renew_options(struct sock *sk,
struct ipv6_txoptions *ipeh_fixup_options(struct ipv6_txoptions *opt_space,
struct ipv6_txoptions *opt);
+/* Generic extension header TLV parser */
+
+enum ipeh_parse_errors {
+ IPEH_PARSE_ERR_PAD1, /* Excessive PAD1 */
+ IPEH_PARSE_ERR_PADN, /* Excessive PADN */
+ IPEH_PARSE_ERR_PADNZ, /* Non-zero padding data */
+ IPEH_PARSE_ERR_EH_TOOBIG, /* Length of EH exceeds limit */
+ IPEH_PARSE_ERR_OPT_TOOBIG, /* Option size exceeds limit */
+ IPEH_PARSE_ERR_OPT_TOOMANY, /* Option count exceeds limit */
+ IPEH_PARSE_ERR_OPT_UNK_DISALW, /* Unknown option disallowed */
+ IPEH_PARSE_ERR_OPT_UNK, /* Unknown option */
+};
+
+/* The generic TLV parser assumes that the type value of PAD1 is 0, and PADN
+ * is 1. This is true for Destination, Hop-by-Hop and current definition
+ * of Segment Routing TLVs.
+ */
+#define IPEH_TLV_PAD1 0
+#define IPEH_TLV_PADN 1
+
+bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
+ int max_count, int off, int len,
+ bool (*parse_error)(struct sk_buff *skb,
+ int off, enum ipeh_parse_errors error));
+
#endif /* _NET_IPEH_H */
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index e12d3a5..939d27c 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -54,135 +54,50 @@
Generic functions
*********************/
-/* An unknown option is detected, decide what to do */
-
-static bool ip6_tlvopt_unknown(struct sk_buff *skb, int optoff,
- bool disallow_unknowns)
+/* Handle parse errors from ipeh generic TLV parser */
+static bool ipv6_parse_error(struct sk_buff *skb, int off,
+ enum ipeh_parse_errors error)
{
- if (disallow_unknowns) {
- /* If unknown TLVs are disallowed by configuration
- * then always silently drop packet. Note this also
- * means no ICMP parameter problem is sent which
- * could be a good property to mitigate a reflection DOS
- * attack.
- */
-
- goto drop;
- }
-
- switch ((skb_network_header(skb)[optoff] & 0xC0) >> 6) {
- case 0: /* ignore */
- return true;
-
- case 1: /* drop packet */
- break;
-
- case 3: /* Send ICMP if not a multicast address and drop packet */
- /* Actually, it is redundant check. icmp_send
- will recheck in any case.
- */
- if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr))
+ switch (error) {
+ case IPEH_PARSE_ERR_OPT_UNK_DISALW:
+ /* Disallow unknown skip */
+ if (((skb_network_header(skb)[off] & 0xC0) >> 6) == 0) {
+ /* Silent drop */
break;
+ }
/* fall through */
- case 2: /* send ICMP PARM PROB regardless and drop packet */
- icmpv6_param_prob(skb, ICMPV6_UNK_OPTION, optoff);
- return false;
- }
-
-drop:
- kfree_skb(skb);
- return false;
-}
+ case IPEH_PARSE_ERR_OPT_UNK:
+ switch ((skb_network_header(skb)[off] & 0xC0) >> 6) {
+ case 0: /* ignore */
+ return true;
-/* Parse tlv encoded option header (hop-by-hop or destination) */
-
-static bool ip6_parse_tlv(const struct tlvtype_proc *procs,
- struct sk_buff *skb,
- int max_count)
-{
- int len = (skb_transport_header(skb)[1] + 1) << 3;
- const unsigned char *nh = skb_network_header(skb);
- int off = skb_network_header_len(skb);
- const struct tlvtype_proc *curr;
- bool disallow_unknowns = false;
- int tlv_count = 0;
- int padlen = 0;
-
- if (unlikely(max_count < 0)) {
- disallow_unknowns = true;
- max_count = -max_count;
- }
-
- if (skb_transport_offset(skb) + len > skb_headlen(skb))
- goto bad;
-
- off += 2;
- len -= 2;
-
- while (len > 0) {
- int optlen = nh[off + 1] + 2;
- int i;
-
- switch (nh[off]) {
- case IPV6_TLV_PAD1:
- optlen = 1;
- padlen++;
- if (padlen > 7)
- goto bad;
+ case 1: /* drop packet */
break;
- case IPV6_TLV_PADN:
- /* RFC 2460 states that the purpose of PadN is
- * to align the containing header to multiples
- * of 8. 7 is therefore the highest valid value.
- * See also RFC 4942, Section 2.1.9.5.
- */
- padlen += optlen;
- if (padlen > 7)
- goto bad;
- /* RFC 4942 recommends receiving hosts to
- * actively check PadN payload to contain
- * only zeroes.
+ case 3: /* Send ICMP if not a multicast address and drop packet
+ *
+ * Actually, it is redundant check. icmp_send
+ * will recheck in any case.
*/
- for (i = 2; i < optlen; i++) {
- if (nh[off + i] != 0)
- goto bad;
- }
- break;
+ if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr))
+ break;
- default: /* Other TLV code so scan list */
- if (optlen > len)
- goto bad;
-
- tlv_count++;
- if (tlv_count > max_count)
- goto bad;
-
- for (curr = procs; curr->type >= 0; curr++) {
- if (curr->type == nh[off]) {
- /* type specific length/alignment
- checks will be performed in the
- func(). */
- if (curr->func(skb, off) == false)
- return false;
- break;
- }
- }
- if (curr->type < 0 &&
- !ip6_tlvopt_unknown(skb, off, disallow_unknowns))
- return false;
-
- padlen = 0;
+ /* fall through */
+ case 2: /* send ICMP PARM PROB regardless and drop packet */
+ icmpv6_send(skb, ICMPV6_PARAMPROB,
+ ICMPV6_UNK_OPTION, off);
break;
}
- off += optlen;
- len -= optlen;
+ break;
+ default:
+ break;
}
- if (len == 0)
- return true;
-bad:
- kfree_skb(skb);
+ /* Will be dropping packet */
+
+ __IP6_INC_STATS(dev_net(skb->dev), __in6_dev_get(skb->dev),
+ IPSTATS_MIB_INHDRERRORS);
+
return false;
}
@@ -216,8 +131,9 @@ static int ipv6_destopt_rcv(struct sk_buff *skb)
dstbuf = opt->dst1;
#endif
- if (ip6_parse_tlv(tlvprocdestopt_lst, skb,
- init_net.ipv6.sysctl.max_dst_opts_cnt)) {
+ if (ipeh_parse_tlv(tlvprocdestopt_lst, skb,
+ init_net.ipv6.sysctl.max_dst_opts_cnt,
+ 2, extlen - 2, ipv6_parse_error)) {
skb->transport_header += extlen;
opt = IP6CB(skb);
#if IS_ENABLED(CONFIG_IPV6_MIP6)
@@ -639,8 +555,9 @@ int ipv6_parse_hopopts(struct sk_buff *skb)
goto fail_and_free;
opt->flags |= IP6SKB_HOPBYHOP;
- if (ip6_parse_tlv(tlvprochopopt_lst, skb,
- init_net.ipv6.sysctl.max_hbh_opts_cnt)) {
+ if (ipeh_parse_tlv(tlvprochopopt_lst, skb,
+ init_net.ipv6.sysctl.max_hbh_opts_cnt,
+ 2, extlen - 2, ipv6_parse_error)) {
skb->transport_header += extlen;
opt = IP6CB(skb);
opt->nhoff = sizeof(struct ipv6hdr);
diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
index 2c68184..99a0911 100644
--- a/net/ipv6/exthdrs_common.c
+++ b/net/ipv6/exthdrs_common.c
@@ -142,3 +142,117 @@ struct ipv6_txoptions *ipeh_fixup_options(struct ipv6_txoptions *opt_space,
return opt;
}
EXPORT_SYMBOL_GPL(ipeh_fixup_options);
+
+/* Generic extension header TLV parser
+ *
+ * Arguments:
+ * - skb_transport_header points to the extension header containing options
+ * - off is offset from skb_transport_header where first TLV is
+ * - len is length of TLV block
+ */
+bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
+ int max_count, int off, int len,
+ bool (*parse_error)(struct sk_buff *skb,
+ int off, enum ipeh_parse_errors error))
+{
+ const unsigned char *nh = skb_network_header(skb);
+ const struct tlvtype_proc *curr;
+ bool disallow_unknowns = false;
+ int tlv_count = 0;
+ int padlen = 0;
+
+ if (unlikely(max_count < 0)) {
+ disallow_unknowns = true;
+ max_count = -max_count;
+ }
+
+ if (skb_transport_offset(skb) + off + len > skb_headlen(skb)) {
+ if (!parse_error(skb, skb_transport_offset(skb),
+ IPEH_PARSE_ERR_EH_TOOBIG))
+ goto bad;
+
+ len = skb_headlen(skb) - skb_transport_offset(skb) - off;
+ }
+
+ /* ops function based offset on network header */
+ off += skb_network_header_len(skb);
+
+ while (len > 0) {
+ int optlen = nh[off + 1] + 2;
+ int i;
+
+ switch (nh[off]) {
+ case IPEH_TLV_PAD1:
+ optlen = 1;
+ padlen++;
+ if (padlen > 7 &&
+ !parse_error(skb, off, IPEH_PARSE_ERR_PAD1))
+ goto bad;
+
+ break;
+
+ case IPEH_TLV_PADN:
+ /* RFC 2460 states that the purpose of PadN is
+ * to align the containing header to multiples
+ * of 8. 7 is therefore the highest valid value.
+ * See also RFC 4942, Section 2.1.9.5.
+ */
+ padlen += optlen;
+ if (padlen > 7 &&
+ !parse_error(skb, off, IPEH_PARSE_ERR_PADN))
+ goto bad;
+
+ /* RFC 4942 recommends receiving hosts to
+ * actively check PadN payload to contain
+ * only zeroes.
+ */
+ for (i = 2; i < optlen; i++) {
+ if (nh[off + i] != 0 &&
+ !parse_error(skb, off + i,
+ IPEH_PARSE_ERR_PADNZ))
+ goto bad;
+ }
+ break;
+
+ default: /* Other TLV code so scan list */
+ if (optlen > len &&
+ !parse_error(skb, off, IPEH_PARSE_ERR_OPT_TOOBIG))
+ goto bad;
+
+ tlv_count++;
+ if (tlv_count > max_count &&
+ parse_error(skb, off, IPEH_PARSE_ERR_OPT_TOOMANY))
+ goto bad;
+
+ for (curr = procs; curr->type >= 0; curr++) {
+ if (curr->type == nh[off]) {
+ /* type specific length/alignment
+ * checks will be performed in the
+ * func().
+ */
+ if (curr->func(skb, off) == false)
+ return false;
+ break;
+ }
+ }
+ if (curr->type < 0 &&
+ !parse_error(skb, off,
+ disallow_unknowns ?
+ IPEH_PARSE_ERR_OPT_UNK_DISALW :
+ IPEH_PARSE_ERR_OPT_UNK))
+ goto bad;
+
+ padlen = 0;
+ break;
+ }
+ off += optlen;
+ len -= optlen;
+ }
+
+ if (len == 0)
+ return true;
+bad:
+ kfree_skb(skb);
+ return false;
+}
+EXPORT_SYMBOL(ipeh_parse_tlv);
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 3/7] ipeh: Generic TLV parser
2019-10-03 21:58 ` [PATCH v5 net-next 3/7] ipeh: Generic TLV parser Tom Herbert
@ 2019-10-06 13:05 ` Simon Horman
2019-12-14 19:08 ` Tom Herbert
0 siblings, 1 reply; 20+ messages in thread
From: Simon Horman @ 2019-10-06 13:05 UTC (permalink / raw)
To: Tom Herbert; +Cc: davem, netdev, Tom Herbert
On Thu, Oct 03, 2019 at 02:58:00PM -0700, Tom Herbert wrote:
> From: Tom Herbert <tom@quantonium.net>
>
> Create a generic TLV parser. This will be used with various
> extension headers that carry options including Destination,
> Hop-by-Hop, Segment Routing TLVs, and other cases of simple
> stateless parsing.
>
> Signed-off-by: Tom Herbert <tom@herbertland.com>
> ---
> include/net/ipeh.h | 25 ++++++++
> net/ipv6/exthdrs.c | 159 +++++++++++-----------------------------------
> net/ipv6/exthdrs_common.c | 114 +++++++++++++++++++++++++++++++++
> 3 files changed, 177 insertions(+), 121 deletions(-)
>
> diff --git a/include/net/ipeh.h b/include/net/ipeh.h
> index 3b24831..c1aa7b6 100644
> --- a/include/net/ipeh.h
> +++ b/include/net/ipeh.h
> @@ -31,4 +31,29 @@ struct ipv6_txoptions *ipeh_renew_options(struct sock *sk,
> struct ipv6_txoptions *ipeh_fixup_options(struct ipv6_txoptions *opt_space,
> struct ipv6_txoptions *opt);
>
> +/* Generic extension header TLV parser */
> +
> +enum ipeh_parse_errors {
> + IPEH_PARSE_ERR_PAD1, /* Excessive PAD1 */
> + IPEH_PARSE_ERR_PADN, /* Excessive PADN */
> + IPEH_PARSE_ERR_PADNZ, /* Non-zero padding data */
> + IPEH_PARSE_ERR_EH_TOOBIG, /* Length of EH exceeds limit */
> + IPEH_PARSE_ERR_OPT_TOOBIG, /* Option size exceeds limit */
> + IPEH_PARSE_ERR_OPT_TOOMANY, /* Option count exceeds limit */
> + IPEH_PARSE_ERR_OPT_UNK_DISALW, /* Unknown option disallowed */
> + IPEH_PARSE_ERR_OPT_UNK, /* Unknown option */
> +};
> +
> +/* The generic TLV parser assumes that the type value of PAD1 is 0, and PADN
> + * is 1. This is true for Destination, Hop-by-Hop and current definition
> + * of Segment Routing TLVs.
> + */
> +#define IPEH_TLV_PAD1 0
> +#define IPEH_TLV_PADN 1
> +
> +bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
> + int max_count, int off, int len,
> + bool (*parse_error)(struct sk_buff *skb,
> + int off, enum ipeh_parse_errors error));
> +
> #endif /* _NET_IPEH_H */
Hi Tom,
Unless I misread things, which is entirely possible, it seems
as well as moving code around this patch changes behaviour under
some error conditions via the parse_error callback and
the ipv6_parse_error() implementation of it below.
I think such a change is worth of at lest calling out in the changelog
and perhaps braking out into a separate patch.
...
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 3/7] ipeh: Generic TLV parser
2019-10-06 13:05 ` Simon Horman
@ 2019-12-14 19:08 ` Tom Herbert
0 siblings, 0 replies; 20+ messages in thread
From: Tom Herbert @ 2019-12-14 19:08 UTC (permalink / raw)
To: Simon Horman
Cc: David S. Miller, Linux Kernel Network Developers, Tom Herbert
On Sun, Oct 6, 2019 at 6:05 AM Simon Horman <simon.horman@netronome.com> wrote:
>
> On Thu, Oct 03, 2019 at 02:58:00PM -0700, Tom Herbert wrote:
> > From: Tom Herbert <tom@quantonium.net>
> >
> > Create a generic TLV parser. This will be used with various
> > extension headers that carry options including Destination,
> > Hop-by-Hop, Segment Routing TLVs, and other cases of simple
> > stateless parsing.
> >
> > Signed-off-by: Tom Herbert <tom@herbertland.com>
> > ---
> > include/net/ipeh.h | 25 ++++++++
> > net/ipv6/exthdrs.c | 159 +++++++++++-----------------------------------
> > net/ipv6/exthdrs_common.c | 114 +++++++++++++++++++++++++++++++++
> > 3 files changed, 177 insertions(+), 121 deletions(-)
> >
> > diff --git a/include/net/ipeh.h b/include/net/ipeh.h
> > index 3b24831..c1aa7b6 100644
> > --- a/include/net/ipeh.h
> > +++ b/include/net/ipeh.h
> > @@ -31,4 +31,29 @@ struct ipv6_txoptions *ipeh_renew_options(struct sock *sk,
> > struct ipv6_txoptions *ipeh_fixup_options(struct ipv6_txoptions *opt_space,
> > struct ipv6_txoptions *opt);
> >
> > +/* Generic extension header TLV parser */
> > +
> > +enum ipeh_parse_errors {
> > + IPEH_PARSE_ERR_PAD1, /* Excessive PAD1 */
> > + IPEH_PARSE_ERR_PADN, /* Excessive PADN */
> > + IPEH_PARSE_ERR_PADNZ, /* Non-zero padding data */
> > + IPEH_PARSE_ERR_EH_TOOBIG, /* Length of EH exceeds limit */
> > + IPEH_PARSE_ERR_OPT_TOOBIG, /* Option size exceeds limit */
> > + IPEH_PARSE_ERR_OPT_TOOMANY, /* Option count exceeds limit */
> > + IPEH_PARSE_ERR_OPT_UNK_DISALW, /* Unknown option disallowed */
> > + IPEH_PARSE_ERR_OPT_UNK, /* Unknown option */
> > +};
> > +
> > +/* The generic TLV parser assumes that the type value of PAD1 is 0, and PADN
> > + * is 1. This is true for Destination, Hop-by-Hop and current definition
> > + * of Segment Routing TLVs.
> > + */
> > +#define IPEH_TLV_PAD1 0
> > +#define IPEH_TLV_PADN 1
> > +
> > +bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
> > + int max_count, int off, int len,
> > + bool (*parse_error)(struct sk_buff *skb,
> > + int off, enum ipeh_parse_errors error));
> > +
> > #endif /* _NET_IPEH_H */
>
> Hi Tom,
>
> Unless I misread things, which is entirely possible, it seems
> as well as moving code around this patch changes behaviour under
> some error conditions via the parse_error callback and
> the ipv6_parse_error() implementation of it below.
>
> I think such a change is worth of at lest calling out in the changelog
> and perhaps braking out into a separate patch.
>
Okay, makes sense to split out the parse_error code. I also noticed
that there's no counter being bumped when we drop a HBH option, I'll
fix that.
> ...
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH v5 net-next 4/7] ip6tlvs: Registration of TLV handlers and parameters
2019-10-03 21:57 [PATCH v5 net-next 0/7] ipv6: Extension header infrastructure Tom Herbert
` (2 preceding siblings ...)
2019-10-03 21:58 ` [PATCH v5 net-next 3/7] ipeh: Generic TLV parser Tom Herbert
@ 2019-10-03 21:58 ` Tom Herbert
2019-10-06 13:18 ` Simon Horman
2019-10-03 21:58 ` [PATCH v5 net-next 5/7] ip6tlvs: Add TX parameters Tom Herbert
` (2 subsequent siblings)
6 siblings, 1 reply; 20+ messages in thread
From: Tom Herbert @ 2019-10-03 21:58 UTC (permalink / raw)
To: davem, netdev; +Cc: Tom Herbert, Tom Herbert
From: Tom Herbert <tom@quantonium.net>
Create a single TLV parameter table that holds meta information for IPv6
Hop-by-Hop and Destination TLVs. The data structure is composed of a 256
element array of u8's (one entry for each TLV type to allow O(1)
lookup). Each entry provides an offset into an array of TLV proc data
structures which follows the array of u8s. The TLV proc data structure
contains parameters and handler functions for receiving and transmitting
TLVs. The zeroth element in the TLV proc array provides default
parameters for TLVs.
A class attribute indicates the type of extension header in which the
TLV may be used (e.g. Hop-by-Hop options, Destination options, or
Destination options before the routing header).
Functions are defined to manipulate entries in the TLV parameter table.
* tlv_{set|unset}_proc set a TLV proc entry (ops and parameters)
* tlv_{set|unset}_params set parameters only
Receive TLV lookup and processing is modified to be a lookup in the TLV
parameter table. An init table containing parameters for TLVs supported
by the kernel is used to initialize the TLV table.
Signed-off-by: Tom Herbert <tom@herbertland.com>
---
include/net/ipeh.h | 107 ++++++++++++++++--
include/net/ipv6.h | 3 +
include/uapi/linux/ipeh.h | 16 +++
net/ipv6/exthdrs.c | 14 ++-
net/ipv6/exthdrs_common.c | 271 +++++++++++++++++++++++++++++++++++++++++----
net/ipv6/exthdrs_options.c | 63 +++++++----
6 files changed, 421 insertions(+), 53 deletions(-)
create mode 100644 include/uapi/linux/ipeh.h
diff --git a/include/net/ipeh.h b/include/net/ipeh.h
index c1aa7b6..aaa2910 100644
--- a/include/net/ipeh.h
+++ b/include/net/ipeh.h
@@ -11,13 +11,105 @@
* and false, if it failed.
* It MUST NOT touch skb->h.
*/
-struct tlvtype_proc {
- int type;
- bool (*func)(struct sk_buff *skb, int offset);
+struct tlv_ops {
+ bool (*func)(unsigned int class, struct sk_buff *skb, int offset);
};
-extern const struct tlvtype_proc tlvprocdestopt_lst[];
-extern const struct tlvtype_proc tlvprochopopt_lst[];
+struct tlv_rx_params {
+ unsigned char class : 4;
+};
+
+struct tlv_tx_params {
+};
+
+struct tlv_params {
+ struct tlv_rx_params r;
+ struct tlv_tx_params t;
+};
+
+struct tlv_proc {
+ struct tlv_ops ops;
+ struct tlv_params params;
+};
+
+struct tlv_type {
+ struct tlv_proc proc;
+};
+
+struct tlv_proc_init {
+ int type;
+ struct tlv_proc proc;
+};
+
+struct tlv_param_table_data {
+ unsigned char entries[256];
+ unsigned char count;
+ struct rcu_head rcu;
+ struct tlv_type types[0];
+};
+
+struct tlv_param_table {
+ struct tlv_param_table_data __rcu *data;
+};
+
+extern struct tlv_param_table ipv6_tlv_param_table;
+
+int __ipeh_tlv_set(struct tlv_param_table *tlv_param_table,
+ unsigned char type, const struct tlv_params *params,
+ const struct tlv_ops *ops);
+
+static inline int ipeh_tlv_set_params(struct tlv_param_table *tlv_param_table,
+ unsigned char type,
+ const struct tlv_params *params)
+{
+ return __ipeh_tlv_set(tlv_param_table, type, params, NULL);
+}
+
+static inline int ipeh_tlv_set_proc(struct tlv_param_table *tlv_param_table,
+ unsigned char type,
+ const struct tlv_proc *proc)
+{
+ return __ipeh_tlv_set(tlv_param_table, type,
+ &proc->params, &proc->ops);
+}
+
+int __ipeh_tlv_unset(struct tlv_param_table *tlv_param_table,
+ unsigned char type, bool params_only);
+
+static inline int ipeh_tlv_unset_params(struct tlv_param_table *tlv_param_table,
+ unsigned char type)
+{
+ return __ipeh_tlv_unset(tlv_param_table, type, true);
+}
+
+static inline int ipeh_tlv_unset_proc(struct tlv_param_table *tlv_param_table,
+ unsigned char type)
+{
+ return __ipeh_tlv_unset(tlv_param_table, type, false);
+}
+
+/* ipeh_tlv_get_proc_by_type assumes rcu_read_lock is held */
+static inline struct tlv_proc *ipeh_tlv_get_proc_by_type(
+ struct tlv_param_table *tlv_param_table, unsigned char type)
+{
+ struct tlv_param_table_data *tpt =
+ rcu_dereference(tlv_param_table->data);
+
+ return &tpt->types[tpt->entries[type]].proc;
+}
+
+/* ipeh_tlv_get_proc assumes rcu_read_lock is held */
+static inline struct tlv_proc *ipeh_tlv_get_proc(
+ struct tlv_param_table *tlv_param_table,
+ const __u8 *tlv)
+{
+ return ipeh_tlv_get_proc_by_type(tlv_param_table, tlv[0]);
+}
+
+int ipeh_exthdrs_init(struct tlv_param_table *tlv_param_table,
+ const struct tlv_proc_init *init_params,
+ int num_init_params);
+void ipeh_exthdrs_fini(struct tlv_param_table *tlv_param_table);
struct ipv6_txoptions;
struct ipv6_opt_hdr;
@@ -51,8 +143,9 @@ enum ipeh_parse_errors {
#define IPEH_TLV_PAD1 0
#define IPEH_TLV_PADN 1
-bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
- int max_count, int off, int len,
+bool ipeh_parse_tlv(unsigned int class,
+ struct tlv_param_table *tlv_param_table,
+ struct sk_buff *skb, int max_count, int off, int len,
bool (*parse_error)(struct sk_buff *skb,
int off, enum ipeh_parse_errors error));
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index d34c36c..7cefdf8 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -429,6 +429,9 @@ int ip6_ra_control(struct sock *sk, int sel);
int ipv6_parse_hopopts(struct sk_buff *skb);
+int ipv6_exthdrs_options_init(void);
+void ipv6_exthdrs_options_exit(void);
+
bool ipv6_opt_accepted(const struct sock *sk, const struct sk_buff *skb,
const struct inet6_skb_parm *opt);
struct ipv6_txoptions *ipv6_update_options(struct sock *sk,
diff --git a/include/uapi/linux/ipeh.h b/include/uapi/linux/ipeh.h
new file mode 100644
index 0000000..c4302b7
--- /dev/null
+++ b/include/uapi/linux/ipeh.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/* ipeh.h - IP extension header TLV management */
+
+#ifndef _UAPI_LINUX_IPEH_H
+#define _UAPI_LINUX_IPEH_H
+
+/* Flags for EH type that can use a TLV option */
+#define IPEH_TLV_CLASS_FLAG_HOPOPT BIT(0)
+#define IPEH_TLV_CLASS_FLAG_RTRDSTOPT BIT(1)
+#define IPEH_TLV_CLASS_FLAG_DSTOPT BIT(2)
+
+#define IPEH_TLV_CLASS_FLAG_MASK (IPEH_TLV_CLASS_FLAG_HOPOPT | \
+ IPEH_TLV_CLASS_FLAG_RTRDSTOPT | \
+ IPEH_TLV_CLASS_FLAG_DSTOPT)
+
+#endif /* _UAPI_LINUX_IPEH_H */
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 939d27c..0847d49 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -47,6 +47,7 @@
#ifdef CONFIG_IPV6_SEG6_HMAC
#include <net/seg6_hmac.h>
#endif
+#include <uapi/linux/ipeh.h>
#include <linux/uaccess.h>
@@ -131,7 +132,8 @@ static int ipv6_destopt_rcv(struct sk_buff *skb)
dstbuf = opt->dst1;
#endif
- if (ipeh_parse_tlv(tlvprocdestopt_lst, skb,
+ if (ipeh_parse_tlv(IPEH_TLV_CLASS_FLAG_DSTOPT,
+ &ipv6_tlv_param_table, skb,
init_net.ipv6.sysctl.max_dst_opts_cnt,
2, extlen - 2, ipv6_parse_error)) {
skb->transport_header += extlen;
@@ -514,8 +516,13 @@ int __init ipv6_exthdrs_init(void)
if (ret)
goto out_destopt;
+ ret = ipv6_exthdrs_options_init();
+ if (ret)
+ goto out_nodata;
out:
return ret;
+out_nodata:
+ inet6_del_protocol(&nodata_protocol, IPPROTO_NONE);
out_destopt:
inet6_del_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
out_rthdr:
@@ -525,6 +532,7 @@ int __init ipv6_exthdrs_init(void)
void ipv6_exthdrs_exit(void)
{
+ ipv6_exthdrs_options_exit();
inet6_del_protocol(&nodata_protocol, IPPROTO_NONE);
inet6_del_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
inet6_del_protocol(&rthdr_protocol, IPPROTO_ROUTING);
@@ -555,8 +563,8 @@ int ipv6_parse_hopopts(struct sk_buff *skb)
goto fail_and_free;
opt->flags |= IP6SKB_HOPBYHOP;
- if (ipeh_parse_tlv(tlvprochopopt_lst, skb,
- init_net.ipv6.sysctl.max_hbh_opts_cnt,
+ if (ipeh_parse_tlv(IPEH_TLV_CLASS_FLAG_HOPOPT, &ipv6_tlv_param_table,
+ skb, init_net.ipv6.sysctl.max_hbh_opts_cnt,
2, extlen - 2, ipv6_parse_error)) {
skb->transport_header += extlen;
opt = IP6CB(skb);
diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
index 99a0911..43737fc 100644
--- a/net/ipv6/exthdrs_common.c
+++ b/net/ipv6/exthdrs_common.c
@@ -150,14 +150,15 @@ EXPORT_SYMBOL_GPL(ipeh_fixup_options);
* - off is offset from skb_transport_header where first TLV is
* - len is length of TLV block
*/
-bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
- int max_count, int off, int len,
+bool ipeh_parse_tlv(unsigned int class,
+ struct tlv_param_table *tlv_param_table,
+ struct sk_buff *skb, int max_count, int off, int len,
bool (*parse_error)(struct sk_buff *skb,
int off, enum ipeh_parse_errors error))
{
const unsigned char *nh = skb_network_header(skb);
- const struct tlvtype_proc *curr;
bool disallow_unknowns = false;
+ const struct tlv_proc *curr;
int tlv_count = 0;
int padlen = 0;
@@ -168,8 +169,10 @@ bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
if (skb_transport_offset(skb) + off + len > skb_headlen(skb)) {
if (!parse_error(skb, skb_transport_offset(skb),
- IPEH_PARSE_ERR_EH_TOOBIG))
- goto bad;
+ IPEH_PARSE_ERR_EH_TOOBIG)) {
+ kfree_skb(skb);
+ return false;
+ }
len = skb_headlen(skb) - skb_transport_offset(skb) - off;
}
@@ -177,6 +180,8 @@ bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
/* ops function based offset on network header */
off += skb_network_header_len(skb);
+ rcu_read_lock();
+
while (len > 0) {
int optlen = nh[off + 1] + 2;
int i;
@@ -221,26 +226,22 @@ bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
tlv_count++;
if (tlv_count > max_count &&
- parse_error(skb, off, IPEH_PARSE_ERR_OPT_TOOMANY))
+ !parse_error(skb, off, IPEH_PARSE_ERR_OPT_TOOMANY))
goto bad;
- for (curr = procs; curr->type >= 0; curr++) {
- if (curr->type == nh[off]) {
- /* type specific length/alignment
- * checks will be performed in the
- * func().
- */
- if (curr->func(skb, off) == false)
- return false;
- break;
- }
- }
- if (curr->type < 0 &&
- !parse_error(skb, off,
+ curr = ipeh_tlv_get_proc(tlv_param_table, &nh[off]);
+ if ((curr->params.r.class & class) && curr->ops.func) {
+ /* Handler will apply additional checks to
+ * the TLV
+ */
+ if (!curr->ops.func(class, skb, off))
+ return false;
+ } else if (!parse_error(skb, off,
disallow_unknowns ?
IPEH_PARSE_ERR_OPT_UNK_DISALW :
- IPEH_PARSE_ERR_OPT_UNK))
+ IPEH_PARSE_ERR_OPT_UNK)) {
goto bad;
+ }
padlen = 0;
break;
@@ -249,10 +250,238 @@ bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
len -= optlen;
}
- if (len == 0)
+ if (len == 0) {
+ rcu_read_unlock();
return true;
+ }
bad:
+ rcu_read_unlock();
kfree_skb(skb);
return false;
}
EXPORT_SYMBOL(ipeh_parse_tlv);
+
+/* TLV parameter table functions and structures */
+
+/* Default (unset) values for TLV parameters */
+static const struct tlv_proc tlv_default_proc = {
+};
+
+static DEFINE_MUTEX(tlv_mutex);
+
+static size_t tlv_param_table_size(unsigned char count)
+{
+ return sizeof(struct tlv_param_table_data) +
+ (count * sizeof(struct tlv_type));
+}
+
+static void tlv_param_table_release(struct rcu_head *rcu)
+{
+ struct tlv_param_table_data *tpt =
+ container_of(rcu, struct tlv_param_table_data, rcu);
+
+ kvfree(tpt);
+}
+
+/* mutex held */
+static int __tlv_set_one(struct tlv_param_table *tlv_param_table,
+ unsigned char type, const struct tlv_params *params,
+ const struct tlv_ops *ops)
+{
+ struct tlv_param_table_data *tpt, *told;
+ struct tlv_type *ttype;
+
+ told = rcu_dereference_protected(tlv_param_table->data,
+ lockdep_is_held(&tlv_mutex));
+
+ /* Create new TLV table. If there is no exsiting entry then we are
+ * adding a new one to the table, else we're modifying an entry.
+ */
+ tpt = kvmalloc(tlv_param_table_size(told->count + !told->entries[type]),
+ GFP_KERNEL);
+ if (!tpt)
+ return -ENOMEM;
+
+ memcpy(tpt, told, tlv_param_table_size(told->count));
+
+ if (!told->entries[type]) {
+ memset(&tpt->types[told->count], 0, sizeof(struct tlv_type));
+ tpt->entries[type] = told->count;
+ tpt->count = told->count + 1;
+ }
+
+ ttype = &tpt->types[tpt->entries[type]];
+
+ ttype->proc.params = *params;
+ ttype->proc.ops = ops ? *ops : tlv_default_proc.ops;
+
+ rcu_assign_pointer(tlv_param_table->data, tpt);
+ call_rcu(&told->rcu, tlv_param_table_release);
+
+ return 0;
+}
+
+int __ipeh_tlv_set(struct tlv_param_table *tlv_param_table, unsigned char type,
+ const struct tlv_params *params, const struct tlv_ops *ops)
+{
+ int retv;
+
+ if (type < 2)
+ return -EINVAL;
+
+ mutex_lock(&tlv_mutex);
+ retv = __tlv_set_one(tlv_param_table, type, params, ops);
+ mutex_unlock(&tlv_mutex);
+
+ return retv;
+}
+EXPORT_SYMBOL(__ipeh_tlv_set);
+
+/* mutex held */
+static int __tlv_unset_one(struct tlv_param_table *tlv_param_table,
+ unsigned char type)
+{
+ struct tlv_param_table_data *tpt, *told;
+ unsigned int i, pos;
+
+ told = rcu_dereference_protected(tlv_param_table->data,
+ lockdep_is_held(&tlv_mutex));
+
+ if (!told->entries[type])
+ return 0;
+
+ tpt = kvmalloc(tlv_param_table_size(told->count - 1),
+ GFP_KERNEL);
+ if (!tpt)
+ return -ENOMEM;
+
+ pos = told->entries[type];
+
+ memcpy(tpt->types, told->types, pos * sizeof(struct tlv_type));
+ memcpy(&tpt->types[pos], &told->types[pos + 1],
+ (told->count - pos - 1) * sizeof(struct tlv_type));
+
+ for (i = 0; i < 256; i++) {
+ if (told->entries[i] > pos)
+ tpt->entries[i] = told->entries[i] - 1;
+ else
+ tpt->entries[i] = told->entries[i];
+ }
+
+ /* Clear entry for type being unset (point to default params) */
+ tpt->entries[type] = 0;
+
+ tpt->count = told->count - 1;
+
+ rcu_assign_pointer(tlv_param_table->data, tpt);
+ call_rcu(&told->rcu, tlv_param_table_release);
+
+ return 0;
+}
+
+/* tlv_internal_proc_type is used to check it the TLV proc was set
+ * internally. This is deduced by checking if any operations are defined.
+ */
+static bool tlv_internal_proc_type(struct tlv_proc *proc)
+{
+ return !!proc->ops.func;
+}
+
+int __ipeh_tlv_unset(struct tlv_param_table *tlv_param_table,
+ unsigned char type, bool params_only)
+{
+ struct tlv_proc *tproc;
+ int retv;
+
+ if (type < 2)
+ return -EINVAL;
+
+ mutex_lock(&tlv_mutex);
+
+ tproc = ipeh_tlv_get_proc_by_type(tlv_param_table, type);
+
+ if (params_only && tlv_internal_proc_type(tproc)) {
+ /* TLV was set by internal source, so maintain the
+ * non-parameter fields (i.e. the operations).
+ */
+ retv = __tlv_set_one(tlv_param_table, type,
+ &tlv_default_proc.params,
+ &tproc->ops);
+ } else {
+ retv = __tlv_unset_one(tlv_param_table, type);
+ }
+
+ mutex_unlock(&tlv_mutex);
+
+ return retv;
+}
+EXPORT_SYMBOL(__ipeh_tlv_unset);
+
+int ipeh_exthdrs_init(struct tlv_param_table *tlv_param_table,
+ const struct tlv_proc_init *tlv_init_params,
+ int num_init_params)
+{
+ struct tlv_param_table_data *tpt;
+ int pos = 0, i;
+ size_t tsize;
+
+ tsize = tlv_param_table_size(num_init_params + 1);
+
+ tpt = kvmalloc(tsize, GFP_KERNEL);
+ if (!tpt)
+ return -ENOMEM;
+
+ memset(tpt, 0, tsize);
+
+ /* Zeroth TLV proc entry is default */
+ tpt->types[pos++].proc = tlv_default_proc;
+
+ for (i = 0; i < num_init_params; i++, pos++) {
+ const struct tlv_proc_init *tpi = &tlv_init_params[i];
+
+ if (WARN_ON(tpi->type < 2)) {
+ /* Padding TLV initialized? */
+ goto err_inval;
+ }
+ if (WARN_ON(tpt->entries[tpi->type])) {
+ /* TLV type already set */
+ goto err_inval;
+ }
+
+ tpt->types[pos].proc = tpi->proc;
+ tpt->entries[tpi->type] = pos;
+ }
+
+ tpt->count = pos;
+
+ RCU_INIT_POINTER(tlv_param_table->data, tpt);
+
+ return 0;
+
+err_inval:
+ kvfree(tpt);
+ return -EINVAL;
+}
+EXPORT_SYMBOL(ipeh_exthdrs_init);
+
+static void tlv_destroy_param_table(struct tlv_param_table *tlv_param_table)
+{
+ struct tlv_param_table_data *tpt;
+
+ mutex_lock(&tlv_mutex);
+
+ tpt = rcu_dereference_protected(tlv_param_table->data,
+ lockdep_is_held(&tlv_mutex));
+ if (tpt) {
+ rcu_assign_pointer(tlv_param_table->data, NULL);
+ call_rcu(&tpt->rcu, tlv_param_table_release);
+ }
+
+ mutex_unlock(&tlv_mutex);
+}
+
+void ipeh_exthdrs_fini(struct tlv_param_table *tlv_param_table)
+{
+ tlv_destroy_param_table(tlv_param_table);
+}
+EXPORT_SYMBOL(ipeh_exthdrs_fini);
diff --git a/net/ipv6/exthdrs_options.c b/net/ipv6/exthdrs_options.c
index 032e072..d4b373e 100644
--- a/net/ipv6/exthdrs_options.c
+++ b/net/ipv6/exthdrs_options.c
@@ -11,11 +11,12 @@
#if IS_ENABLED(CONFIG_IPV6_MIP6)
#include <net/xfrm.h>
#endif
+#include <uapi/linux/ipeh.h>
/* Destination options header */
#if IS_ENABLED(CONFIG_IPV6_MIP6)
-static bool ipv6_dest_hao(struct sk_buff *skb, int optoff)
+static bool ipv6_dest_hao(unsigned int class, struct sk_buff *skb, int optoff)
{
struct ipv6_destopt_hao *hao;
struct inet6_skb_parm *opt = IP6CB(skb);
@@ -74,16 +75,6 @@ static bool ipv6_dest_hao(struct sk_buff *skb, int optoff)
}
#endif
-const struct tlvtype_proc tlvprocdestopt_lst[] = {
-#if IS_ENABLED(CONFIG_IPV6_MIP6)
- {
- .type = IPV6_TLV_HAO,
- .func = ipv6_dest_hao,
- },
-#endif
- {-1, NULL}
-};
-
/* Hop-by-hop options */
/* Note: we cannot rely on skb_dst(skb) before we assign it in
@@ -102,7 +93,7 @@ static inline struct net *ipv6_skb_net(struct sk_buff *skb)
/* Router Alert as of RFC 2711 */
-static bool ipv6_hop_ra(struct sk_buff *skb, int optoff)
+static bool ipv6_hop_ra(unsigned int class, struct sk_buff *skb, int optoff)
{
const unsigned char *nh = skb_network_header(skb);
@@ -120,7 +111,7 @@ static bool ipv6_hop_ra(struct sk_buff *skb, int optoff)
/* Jumbo payload */
-static bool ipv6_hop_jumbo(struct sk_buff *skb, int optoff)
+static bool ipv6_hop_jumbo(unsigned int class, struct sk_buff *skb, int optoff)
{
const unsigned char *nh = skb_network_header(skb);
struct inet6_dev *idev = __in6_dev_get_safely(skb->dev);
@@ -164,7 +155,8 @@ static bool ipv6_hop_jumbo(struct sk_buff *skb, int optoff)
/* CALIPSO RFC 5570 */
-static bool ipv6_hop_calipso(struct sk_buff *skb, int optoff)
+static bool ipv6_hop_calipso(unsigned int class, struct sk_buff *skb,
+ int optoff)
{
const unsigned char *nh = skb_network_header(skb);
@@ -184,18 +176,45 @@ static bool ipv6_hop_calipso(struct sk_buff *skb, int optoff)
return false;
}
-const struct tlvtype_proc tlvprochopopt_lst[] = {
+static const struct tlv_proc_init tlv_ipv6_init_params[] __initconst = {
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
{
- .type = IPV6_TLV_ROUTERALERT,
- .func = ipv6_hop_ra,
+ .type = IPV6_TLV_HAO,
+
+ .proc.ops.func = ipv6_dest_hao,
+ .proc.params.r.class = IPEH_TLV_CLASS_FLAG_DSTOPT,
},
+#endif
{
- .type = IPV6_TLV_JUMBO,
- .func = ipv6_hop_jumbo,
+ .type = IPV6_TLV_ROUTERALERT,
+
+ .proc.ops.func = ipv6_hop_ra,
+ .proc.params.r.class = IPEH_TLV_CLASS_FLAG_HOPOPT,
},
{
- .type = IPV6_TLV_CALIPSO,
- .func = ipv6_hop_calipso,
+ .type = IPV6_TLV_JUMBO,
+
+ .proc.ops.func = ipv6_hop_jumbo,
+ .proc.params.r.class = IPEH_TLV_CLASS_FLAG_HOPOPT,
+ },
+ {
+ .type = IPV6_TLV_CALIPSO,
+
+ .proc.ops.func = ipv6_hop_calipso,
+ .proc.params.r.class = IPEH_TLV_CLASS_FLAG_HOPOPT,
},
- { -1, }
};
+
+struct tlv_param_table __rcu ipv6_tlv_param_table;
+EXPORT_SYMBOL(ipv6_tlv_param_table);
+
+int __init ipv6_exthdrs_options_init(void)
+{
+ return ipeh_exthdrs_init(&ipv6_tlv_param_table, tlv_ipv6_init_params,
+ ARRAY_SIZE(tlv_ipv6_init_params));
+}
+
+void ipv6_exthdrs_options_exit(void)
+{
+ ipeh_exthdrs_fini(&ipv6_tlv_param_table);
+}
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 4/7] ip6tlvs: Registration of TLV handlers and parameters
2019-10-03 21:58 ` [PATCH v5 net-next 4/7] ip6tlvs: Registration of TLV handlers and parameters Tom Herbert
@ 2019-10-06 13:18 ` Simon Horman
0 siblings, 0 replies; 20+ messages in thread
From: Simon Horman @ 2019-10-06 13:18 UTC (permalink / raw)
To: Tom Herbert; +Cc: davem, netdev, Tom Herbert
On Thu, Oct 03, 2019 at 02:58:01PM -0700, Tom Herbert wrote:
> From: Tom Herbert <tom@quantonium.net>
>
> Create a single TLV parameter table that holds meta information for IPv6
> Hop-by-Hop and Destination TLVs. The data structure is composed of a 256
> element array of u8's (one entry for each TLV type to allow O(1)
> lookup). Each entry provides an offset into an array of TLV proc data
> structures which follows the array of u8s. The TLV proc data structure
> contains parameters and handler functions for receiving and transmitting
> TLVs. The zeroth element in the TLV proc array provides default
> parameters for TLVs.
>
> A class attribute indicates the type of extension header in which the
> TLV may be used (e.g. Hop-by-Hop options, Destination options, or
> Destination options before the routing header).
>
> Functions are defined to manipulate entries in the TLV parameter table.
>
> * tlv_{set|unset}_proc set a TLV proc entry (ops and parameters)
> * tlv_{set|unset}_params set parameters only
Perhaps it would be worth mentioning that these will
be used by a following patch in the series.
> Receive TLV lookup and processing is modified to be a lookup in the TLV
> parameter table. An init table containing parameters for TLVs supported
> by the kernel is used to initialize the TLV table.
>
> Signed-off-by: Tom Herbert <tom@herbertland.com>
Minor nit about types below, but overall this looks good to me.
Reviewed-by: Simon Horman <simon.horman@netronome.com>
...
> diff --git a/include/net/ipeh.h b/include/net/ipeh.h
> index c1aa7b6..aaa2910 100644
> --- a/include/net/ipeh.h
> +++ b/include/net/ipeh.h
...
> +/* ipeh_tlv_get_proc_by_type assumes rcu_read_lock is held */
> +static inline struct tlv_proc *ipeh_tlv_get_proc_by_type(
> + struct tlv_param_table *tlv_param_table, unsigned char type)
> +{
> + struct tlv_param_table_data *tpt =
> + rcu_dereference(tlv_param_table->data);
> +
> + return &tpt->types[tpt->entries[type]].proc;
> +}
> +
> +/* ipeh_tlv_get_proc assumes rcu_read_lock is held */
> +static inline struct tlv_proc *ipeh_tlv_get_proc(
> + struct tlv_param_table *tlv_param_table,
> + const __u8 *tlv)
> +{
> + return ipeh_tlv_get_proc_by_type(tlv_param_table, tlv[0]);
The type of tlv is const __u8 *, however,
1. the type of the 'type' parameter of ipeh_tlv_get_proc_by_type is
unsigned char; and
2. this function is passed a value of type const unsigned char * by
ipeh_parse_tlv().
I understand these are all one byte unsigned integer types.
But perhaps this could be made more consistent somehow.
> +}
...
> diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
> index 99a0911..43737fc 100644
> --- a/net/ipv6/exthdrs_common.c
> +++ b/net/ipv6/exthdrs_common.c
..
> @@ -221,26 +226,22 @@ bool ipeh_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb,
>
> tlv_count++;
> if (tlv_count > max_count &&
> - parse_error(skb, off, IPEH_PARSE_ERR_OPT_TOOMANY))
> + !parse_error(skb, off, IPEH_PARSE_ERR_OPT_TOOMANY))
> goto bad;
>
> - for (curr = procs; curr->type >= 0; curr++) {
> - if (curr->type == nh[off]) {
> - /* type specific length/alignment
> - * checks will be performed in the
> - * func().
> - */
> - if (curr->func(skb, off) == false)
> - return false;
> - break;
> - }
> - }
> - if (curr->type < 0 &&
> - !parse_error(skb, off,
> + curr = ipeh_tlv_get_proc(tlv_param_table, &nh[off]);
> + if ((curr->params.r.class & class) && curr->ops.func) {
> + /* Handler will apply additional checks to
> + * the TLV
> + */
> + if (!curr->ops.func(class, skb, off))
> + return false;
> + } else if (!parse_error(skb, off,
> disallow_unknowns ?
> IPEH_PARSE_ERR_OPT_UNK_DISALW :
> - IPEH_PARSE_ERR_OPT_UNK))
> + IPEH_PARSE_ERR_OPT_UNK)) {
> goto bad;
> + }
>
> padlen = 0;
> break;
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH v5 net-next 5/7] ip6tlvs: Add TX parameters
2019-10-03 21:57 [PATCH v5 net-next 0/7] ipv6: Extension header infrastructure Tom Herbert
` (3 preceding siblings ...)
2019-10-03 21:58 ` [PATCH v5 net-next 4/7] ip6tlvs: Registration of TLV handlers and parameters Tom Herbert
@ 2019-10-03 21:58 ` Tom Herbert
2019-10-06 13:25 ` Simon Horman
2019-10-03 21:58 ` [PATCH v5 net-next 6/7] ip6tlvs: Add netlink interface Tom Herbert
2019-10-03 21:58 ` [PATCH v5 net-next 7/7] ip6tlvs: Validation of TX Destination and Hop-by-Hop options Tom Herbert
6 siblings, 1 reply; 20+ messages in thread
From: Tom Herbert @ 2019-10-03 21:58 UTC (permalink / raw)
To: davem, netdev; +Cc: Tom Herbert, Tom Herbert
From: Tom Herbert <tom@quantonium.net>
Define a number of transmit parameters for TLV Parameter table
definitions. These will be used for validating TLVs that are set
on a socket.
Signed-off-by: Tom Herbert <tom@herbertland.com>
---
include/net/ipeh.h | 18 ++++++++++++++++
include/uapi/linux/ipeh.h | 8 +++++++
net/ipv6/exthdrs_common.c | 53 +++++++++++++++++++++++++++++++++++++++++++++-
net/ipv6/exthdrs_options.c | 45 +++++++++++++++++++++++++++++++++++++++
4 files changed, 123 insertions(+), 1 deletion(-)
diff --git a/include/net/ipeh.h b/include/net/ipeh.h
index aaa2910..de6d9d0 100644
--- a/include/net/ipeh.h
+++ b/include/net/ipeh.h
@@ -20,6 +20,17 @@ struct tlv_rx_params {
};
struct tlv_tx_params {
+ unsigned char admin_perm : 2;
+ unsigned char user_perm : 2;
+ unsigned char class : 3;
+ unsigned char rsvd : 1;
+ unsigned char align_mult : 4;
+ unsigned char align_off : 4;
+ unsigned char data_len_mult : 4;
+ unsigned char data_len_off : 4;
+ unsigned char min_data_len;
+ unsigned char max_data_len;
+ unsigned short preferred_order;
};
struct tlv_params {
@@ -54,6 +65,13 @@ struct tlv_param_table {
extern struct tlv_param_table ipv6_tlv_param_table;
+/* Preferred TLV ordering for HBH and Dest options (placed by increasing order)
+ */
+#define IPEH_TLV_PREF_ORDER_HAO 10
+#define IPEH_TLV_PREF_ORDER_ROUTERALERT 20
+#define IPEH_TLV_PREF_ORDER_JUMBO 30
+#define IPEH_TLV_PREF_ORDER_CALIPSO 40
+
int __ipeh_tlv_set(struct tlv_param_table *tlv_param_table,
unsigned char type, const struct tlv_params *params,
const struct tlv_ops *ops);
diff --git a/include/uapi/linux/ipeh.h b/include/uapi/linux/ipeh.h
index c4302b7..dbf0728 100644
--- a/include/uapi/linux/ipeh.h
+++ b/include/uapi/linux/ipeh.h
@@ -13,4 +13,12 @@
IPEH_TLV_CLASS_FLAG_RTRDSTOPT | \
IPEH_TLV_CLASS_FLAG_DSTOPT)
+/* TLV permissions values */
+enum {
+ IPEH_TLV_PERM_NONE,
+ IPEH_TLV_PERM_WITH_CHECK,
+ IPEH_TLV_PERM_NO_CHECK,
+ IPEH_TLV_PERM_MAX = IPEH_TLV_PERM_NO_CHECK
+};
+
#endif /* _UAPI_LINUX_IPEH_H */
diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
index 43737fc..feaa4a6 100644
--- a/net/ipv6/exthdrs_common.c
+++ b/net/ipv6/exthdrs_common.c
@@ -3,6 +3,7 @@
/* Extension header and TLV library code that is not specific to IPv6. */
#include <linux/export.h>
#include <net/ipv6.h>
+#include <uapi/linux/ipeh.h>
struct ipv6_txoptions *
ipeh_dup_options(struct sock *sk, struct ipv6_txoptions *opt)
@@ -265,6 +266,13 @@ EXPORT_SYMBOL(ipeh_parse_tlv);
/* Default (unset) values for TLV parameters */
static const struct tlv_proc tlv_default_proc = {
+ .params.t = {
+ .admin_perm = IPEH_TLV_PERM_NO_CHECK,
+ .user_perm = IPEH_TLV_PERM_NONE,
+ .align_mult = (4 - 1), /* Default alignment: 4n + 2 */
+ .align_off = 2,
+ .max_data_len = 255,
+ },
};
static DEFINE_MUTEX(tlv_mutex);
@@ -284,16 +292,45 @@ static void tlv_param_table_release(struct rcu_head *rcu)
}
/* mutex held */
+static int check_order(struct tlv_param_table_data *tpt, unsigned char type,
+ unsigned short order)
+{
+ int i;
+
+ if (!order)
+ return -EINVAL;
+
+ for (i = 2; i < 256; i++) {
+ struct tlv_type *ttype = &tpt->types[tpt->entries[i]];
+
+ if (!tpt->entries[i])
+ continue;
+
+ if (order == ttype->proc.params.t.preferred_order &&
+ i != type)
+ return -EALREADY;
+ }
+
+ return 0;
+}
+
+/* mutex held */
static int __tlv_set_one(struct tlv_param_table *tlv_param_table,
unsigned char type, const struct tlv_params *params,
const struct tlv_ops *ops)
{
struct tlv_param_table_data *tpt, *told;
struct tlv_type *ttype;
+ int retv;
told = rcu_dereference_protected(tlv_param_table->data,
lockdep_is_held(&tlv_mutex));
+ /* Check preferred order */
+ retv = check_order(told, type, params->t.preferred_order);
+ if (retv)
+ return retv;
+
/* Create new TLV table. If there is no exsiting entry then we are
* adding a new one to the table, else we're modifying an entry.
*/
@@ -422,7 +459,7 @@ int ipeh_exthdrs_init(struct tlv_param_table *tlv_param_table,
int num_init_params)
{
struct tlv_param_table_data *tpt;
- int pos = 0, i;
+ int pos = 0, i, j;
size_t tsize;
tsize = tlv_param_table_size(num_init_params + 1);
@@ -448,6 +485,20 @@ int ipeh_exthdrs_init(struct tlv_param_table *tlv_param_table,
goto err_inval;
}
+ if (WARN_ON(!tpi->proc.params.t.preferred_order)) {
+ /* Preferred order must be non-zero */
+ goto err_inval;
+ }
+
+ for (j = 0; j < i; j++) {
+ const struct tlv_proc_init *tpix = &tlv_init_params[j];
+
+ if (WARN_ON(tpi->proc.params.t.preferred_order ==
+ tpix->proc.params.t.preferred_order)) {
+ /* Preferred order must be unique */
+ goto err_inval;
+ }
+ }
tpt->types[pos].proc = tpi->proc;
tpt->entries[tpi->type] = pos;
}
diff --git a/net/ipv6/exthdrs_options.c b/net/ipv6/exthdrs_options.c
index d4b373e..3b50b58 100644
--- a/net/ipv6/exthdrs_options.c
+++ b/net/ipv6/exthdrs_options.c
@@ -183,6 +183,17 @@ static const struct tlv_proc_init tlv_ipv6_init_params[] __initconst = {
.proc.ops.func = ipv6_dest_hao,
.proc.params.r.class = IPEH_TLV_CLASS_FLAG_DSTOPT,
+
+ .proc.params.t = {
+ .preferred_order = IPEH_TLV_PREF_ORDER_HAO,
+ .admin_perm = IPEH_TLV_PERM_NO_CHECK,
+ .user_perm = IPEH_TLV_PERM_NONE,
+ .class = IPEH_TLV_CLASS_FLAG_DSTOPT,
+ .align_mult = (8 - 1), /* Align to 8n + 6 */
+ .align_off = 6,
+ .min_data_len = 16,
+ .max_data_len = 16,
+ },
},
#endif
{
@@ -190,18 +201,52 @@ static const struct tlv_proc_init tlv_ipv6_init_params[] __initconst = {
.proc.ops.func = ipv6_hop_ra,
.proc.params.r.class = IPEH_TLV_CLASS_FLAG_HOPOPT,
+
+ .proc.params.t = {
+ .preferred_order = IPEH_TLV_PREF_ORDER_ROUTERALERT,
+ .admin_perm = IPEH_TLV_PERM_NO_CHECK,
+ .user_perm = IPEH_TLV_PERM_NONE,
+ .class = IPEH_TLV_CLASS_FLAG_HOPOPT,
+ .align_mult = (2 - 1), /* Align to 2n */
+ .min_data_len = 2,
+ .max_data_len = 2,
+ },
},
{
.type = IPV6_TLV_JUMBO,
.proc.ops.func = ipv6_hop_jumbo,
.proc.params.r.class = IPEH_TLV_CLASS_FLAG_HOPOPT,
+
+ .proc.params.t = {
+ .preferred_order = IPEH_TLV_PREF_ORDER_JUMBO,
+ .admin_perm = IPEH_TLV_PERM_NO_CHECK,
+ .user_perm = IPEH_TLV_PERM_NONE,
+ .class = IPEH_TLV_CLASS_FLAG_HOPOPT,
+ .align_mult = (4 - 1), /* Align to 4n + 2 */
+ .align_off = 2,
+ .min_data_len = 4,
+ .max_data_len = 4,
+ },
},
{
.type = IPV6_TLV_CALIPSO,
.proc.ops.func = ipv6_hop_calipso,
.proc.params.r.class = IPEH_TLV_CLASS_FLAG_HOPOPT,
+
+ .proc.params.t = {
+ .preferred_order = IPEH_TLV_PREF_ORDER_CALIPSO,
+ .admin_perm = IPEH_TLV_PERM_NO_CHECK,
+ .user_perm = IPEH_TLV_PERM_NONE,
+ .class = IPEH_TLV_CLASS_FLAG_HOPOPT,
+ .align_mult = (4 - 1), /* Align to 4n + 2 */
+ .align_off = 2,
+ .min_data_len = 8,
+ .max_data_len = 252,
+ .data_len_mult = (4 - 1),
+ /* Length is multiple of 4 */
+ },
},
};
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 5/7] ip6tlvs: Add TX parameters
2019-10-03 21:58 ` [PATCH v5 net-next 5/7] ip6tlvs: Add TX parameters Tom Herbert
@ 2019-10-06 13:25 ` Simon Horman
2019-12-14 19:19 ` Tom Herbert
0 siblings, 1 reply; 20+ messages in thread
From: Simon Horman @ 2019-10-06 13:25 UTC (permalink / raw)
To: Tom Herbert; +Cc: davem, netdev, Tom Herbert
On Thu, Oct 03, 2019 at 02:58:02PM -0700, Tom Herbert wrote:
> From: Tom Herbert <tom@quantonium.net>
>
> Define a number of transmit parameters for TLV Parameter table
> definitions. These will be used for validating TLVs that are set
> on a socket.
>
> Signed-off-by: Tom Herbert <tom@herbertland.com>
> ---
> include/net/ipeh.h | 18 ++++++++++++++++
> include/uapi/linux/ipeh.h | 8 +++++++
> net/ipv6/exthdrs_common.c | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> net/ipv6/exthdrs_options.c | 45 +++++++++++++++++++++++++++++++++++++++
> 4 files changed, 123 insertions(+), 1 deletion(-)
>
> diff --git a/include/net/ipeh.h b/include/net/ipeh.h
> index aaa2910..de6d9d0 100644
> --- a/include/net/ipeh.h
> +++ b/include/net/ipeh.h
...
> @@ -54,6 +65,13 @@ struct tlv_param_table {
>
> extern struct tlv_param_table ipv6_tlv_param_table;
>
> +/* Preferred TLV ordering for HBH and Dest options (placed by increasing order)
> + */
> +#define IPEH_TLV_PREF_ORDER_HAO 10
> +#define IPEH_TLV_PREF_ORDER_ROUTERALERT 20
> +#define IPEH_TLV_PREF_ORDER_JUMBO 30
> +#define IPEH_TLV_PREF_ORDER_CALIPSO 40
> +
Hi Tom,
Could you expand on why thse values were chosen?
I can see that this patch implements a specific use of
the 255 indexes available. But its not at all clear to me that
this use fits expected use-cases (because I don't know what they are).
...
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 5/7] ip6tlvs: Add TX parameters
2019-10-06 13:25 ` Simon Horman
@ 2019-12-14 19:19 ` Tom Herbert
0 siblings, 0 replies; 20+ messages in thread
From: Tom Herbert @ 2019-12-14 19:19 UTC (permalink / raw)
To: Simon Horman
Cc: David S. Miller, Linux Kernel Network Developers, Tom Herbert
On Sun, Oct 6, 2019 at 6:25 AM Simon Horman <simon.horman@netronome.com> wrote:
>
> On Thu, Oct 03, 2019 at 02:58:02PM -0700, Tom Herbert wrote:
> > From: Tom Herbert <tom@quantonium.net>
> >
> > Define a number of transmit parameters for TLV Parameter table
> > definitions. These will be used for validating TLVs that are set
> > on a socket.
> >
> > Signed-off-by: Tom Herbert <tom@herbertland.com>
> > ---
> > include/net/ipeh.h | 18 ++++++++++++++++
> > include/uapi/linux/ipeh.h | 8 +++++++
> > net/ipv6/exthdrs_common.c | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > net/ipv6/exthdrs_options.c | 45 +++++++++++++++++++++++++++++++++++++++
> > 4 files changed, 123 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/net/ipeh.h b/include/net/ipeh.h
> > index aaa2910..de6d9d0 100644
> > --- a/include/net/ipeh.h
> > +++ b/include/net/ipeh.h
>
> ...
>
> > @@ -54,6 +65,13 @@ struct tlv_param_table {
> >
> > extern struct tlv_param_table ipv6_tlv_param_table;
> >
> > +/* Preferred TLV ordering for HBH and Dest options (placed by increasing order)
> > + */
> > +#define IPEH_TLV_PREF_ORDER_HAO 10
> > +#define IPEH_TLV_PREF_ORDER_ROUTERALERT 20
> > +#define IPEH_TLV_PREF_ORDER_JUMBO 30
> > +#define IPEH_TLV_PREF_ORDER_CALIPSO 40
> > +
>
> Hi Tom,
>
> Could you expand on why thse values were chosen?
>
Pseudo random selection :-). The idea of having an ordering is to
constrain the use of TLVs (in some environments there may be TLV
ordering requirements or optimizations around specific ordering). Note
that the ordering only applies to validation of TLVs being set to
send, there are no ordering constraints in RX. Also, in the next patch
set where application can set individual HBH and DO options on a
socket, the ordering attribute is used to always produce the same
order on the wire regardless of the ordering that the application set
the options. For non-priviledeged applications especially, I believe
it's good to be conservative and apply reasonable constraints such as
ordering for TX (i.e. follow robustness principle).
> I can see that this patch implements a specific use of
> the 255 indexes available. But its not at all clear to me that
> this use fits expected use-cases (because I don't know what they are).
>
There are at more 253 non-padding option types. Fortunately the
protocol designers had the foresight to limit option type to a byte
and so it's reasonable to represent for lookup in simple arrays. Two
bytes for type would have been much more painful (compare lookup on
EtherType to IP protocol numbers for instance).
> ...
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH v5 net-next 6/7] ip6tlvs: Add netlink interface
2019-10-03 21:57 [PATCH v5 net-next 0/7] ipv6: Extension header infrastructure Tom Herbert
` (4 preceding siblings ...)
2019-10-03 21:58 ` [PATCH v5 net-next 5/7] ip6tlvs: Add TX parameters Tom Herbert
@ 2019-10-03 21:58 ` Tom Herbert
2019-10-07 12:39 ` Simon Horman
2019-10-03 21:58 ` [PATCH v5 net-next 7/7] ip6tlvs: Validation of TX Destination and Hop-by-Hop options Tom Herbert
6 siblings, 1 reply; 20+ messages in thread
From: Tom Herbert @ 2019-10-03 21:58 UTC (permalink / raw)
To: davem, netdev; +Cc: Tom Herbert, Tom Herbert
From: Tom Herbert <tom@quantonium.net>
Add a netlink interface to manage the TX TLV parameters. Managed
parameters include those for validating and sending TLVs being sent
such as alignment, TLV ordering, length limits, etc.
Signed-off-by: Tom Herbert <tom@herbertland.com>
---
include/net/ipeh.h | 16 +++
include/net/ipv6.h | 1 +
include/uapi/linux/in6.h | 6 ++
include/uapi/linux/ipeh.h | 29 ++++++
net/ipv6/exthdrs_common.c | 238 +++++++++++++++++++++++++++++++++++++++++++++
net/ipv6/exthdrs_options.c | 81 ++++++++++++++-
6 files changed, 369 insertions(+), 2 deletions(-)
diff --git a/include/net/ipeh.h b/include/net/ipeh.h
index de6d9d0..8474a43 100644
--- a/include/net/ipeh.h
+++ b/include/net/ipeh.h
@@ -3,6 +3,7 @@
#define _NET_IPEH_H
#include <linux/skbuff.h>
+#include <net/genetlink.h>
/*
* Parsing tlv encoded headers.
@@ -106,6 +107,21 @@ static inline int ipeh_tlv_unset_proc(struct tlv_param_table *tlv_param_table,
return __ipeh_tlv_unset(tlv_param_table, type, false);
}
+extern const struct nla_policy ipeh_tlv_nl_policy[];
+
+int ipeh_tlv_nl_cmd_set(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct genl_info *info);
+int ipeh_tlv_nl_cmd_unset(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct genl_info *info);
+int ipeh_tlv_nl_cmd_get(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct genl_info *info);
+int ipeh_tlv_nl_dump(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct netlink_callback *cb);
+
/* ipeh_tlv_get_proc_by_type assumes rcu_read_lock is held */
static inline struct tlv_proc *ipeh_tlv_get_proc_by_type(
struct tlv_param_table *tlv_param_table, unsigned char type)
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 7cefdf8..54b6e53 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -14,6 +14,7 @@
#include <linux/jhash.h>
#include <linux/refcount.h>
#include <linux/jump_label_ratelimit.h>
+#include <net/genetlink.h>
#include <net/if_inet6.h>
#include <net/ndisc.h>
#include <net/flow.h>
diff --git a/include/uapi/linux/in6.h b/include/uapi/linux/in6.h
index 9f2273a..d5fe3d9 100644
--- a/include/uapi/linux/in6.h
+++ b/include/uapi/linux/in6.h
@@ -297,4 +297,10 @@ struct in6_flowlabel_req {
* ...
* MRT6_MAX
*/
+
+ /* NETLINK_GENERIC related info for IPv6 TLVs */
+
+#define IPV6_TLV_GENL_NAME "ipv6-tlv"
+#define IPV6_TLV_GENL_VERSION 0x1
+
#endif /* _UAPI_LINUX_IN6_H */
diff --git a/include/uapi/linux/ipeh.h b/include/uapi/linux/ipeh.h
index dbf0728..bac36a7 100644
--- a/include/uapi/linux/ipeh.h
+++ b/include/uapi/linux/ipeh.h
@@ -21,4 +21,33 @@ enum {
IPEH_TLV_PERM_MAX = IPEH_TLV_PERM_NO_CHECK
};
+/* NETLINK_GENERIC related info for IP TLVs */
+
+enum {
+ IPEH_TLV_ATTR_UNSPEC,
+ IPEH_TLV_ATTR_TYPE, /* u8, > 1 */
+ IPEH_TLV_ATTR_ORDER, /* u16 */
+ IPEH_TLV_ATTR_ADMIN_PERM, /* u8, perm value */
+ IPEH_TLV_ATTR_USER_PERM, /* u8, perm value */
+ IPEH_TLV_ATTR_CLASS, /* u8, 3 bit flags */
+ IPEH_TLV_ATTR_ALIGN_MULT, /* u8, 1 to 16 */
+ IPEH_TLV_ATTR_ALIGN_OFF, /* u8, 0 to 15 */
+ IPEH_TLV_ATTR_MIN_DATA_LEN, /* u8 (option data length) */
+ IPEH_TLV_ATTR_MAX_DATA_LEN, /* u8 (option data length) */
+ IPEH_TLV_ATTR_DATA_LEN_MULT, /* u8, 1 to 16 */
+ IPEH_TLV_ATTR_DATA_LEN_OFF, /* u8, 0 to 15 */
+
+ __IPEH_TLV_ATTR_MAX,
+};
+
+#define IPEH_TLV_ATTR_MAX (__IPEH_TLV_ATTR_MAX - 1)
+
+enum {
+ IPEH_TLV_CMD_SET,
+ IPEH_TLV_CMD_UNSET,
+ IPEH_TLV_CMD_GET,
+
+ __IPEH_TLV_CMD_MAX,
+};
+
#endif /* _UAPI_LINUX_IPEH_H */
diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
index feaa4a6..e3c1f33 100644
--- a/net/ipv6/exthdrs_common.c
+++ b/net/ipv6/exthdrs_common.c
@@ -454,6 +454,244 @@ int __ipeh_tlv_unset(struct tlv_param_table *tlv_param_table,
}
EXPORT_SYMBOL(__ipeh_tlv_unset);
+const struct nla_policy ipeh_tlv_nl_policy[IPEH_TLV_ATTR_MAX + 1] = {
+ [IPEH_TLV_ATTR_TYPE] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_ORDER] = { .type = NLA_U16, },
+ [IPEH_TLV_ATTR_ADMIN_PERM] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_USER_PERM] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_CLASS] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_ALIGN_MULT] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_ALIGN_OFF] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_MIN_DATA_LEN] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_MAX_DATA_LEN] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_DATA_LEN_OFF] = { .type = NLA_U8, },
+ [IPEH_TLV_ATTR_DATA_LEN_MULT] = { .type = NLA_U8, },
+};
+EXPORT_SYMBOL(ipeh_tlv_nl_policy);
+
+int ipeh_tlv_nl_cmd_set(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct genl_info *info)
+{
+ struct tlv_params new_params;
+ struct tlv_proc *tproc;
+ unsigned char type;
+ unsigned int v;
+ int retv = -EINVAL;
+
+ if (!info->attrs[IPEH_TLV_ATTR_TYPE])
+ return -EINVAL;
+
+ type = nla_get_u8(info->attrs[IPEH_TLV_ATTR_TYPE]);
+ if (type < 2)
+ return -EINVAL;
+
+ rcu_read_lock();
+
+ /* Base new parameters on existing ones */
+ tproc = ipeh_tlv_get_proc_by_type(tlv_param_table, type);
+ new_params = tproc->params;
+
+ if (info->attrs[IPEH_TLV_ATTR_ORDER]) {
+ v = nla_get_u16(info->attrs[IPEH_TLV_ATTR_ORDER]);
+ new_params.t.preferred_order = v;
+ }
+
+ if (info->attrs[IPEH_TLV_ATTR_ADMIN_PERM]) {
+ v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ADMIN_PERM]);
+ if (v > IPEH_TLV_PERM_MAX)
+ goto out;
+ new_params.t.admin_perm = v;
+ }
+
+ if (info->attrs[IPEH_TLV_ATTR_USER_PERM]) {
+ v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_USER_PERM]);
+ if (v > IPEH_TLV_PERM_MAX)
+ goto out;
+ new_params.t.user_perm = v;
+ }
+
+ if (info->attrs[IPEH_TLV_ATTR_CLASS]) {
+ v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_CLASS]);
+ if (!v || (v & ~IPEH_TLV_CLASS_FLAG_MASK))
+ goto out;
+ new_params.t.class = v;
+ }
+
+ if (info->attrs[IPEH_TLV_ATTR_ALIGN_MULT]) {
+ v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ALIGN_MULT]);
+ if (v > 16 || v < 1)
+ goto out;
+ new_params.t.align_mult = v - 1;
+ }
+
+ if (info->attrs[IPEH_TLV_ATTR_ALIGN_OFF]) {
+ v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ALIGN_OFF]);
+ if (v > 15)
+ goto out;
+ new_params.t.align_off = v;
+ }
+
+ if (info->attrs[IPEH_TLV_ATTR_MAX_DATA_LEN])
+ new_params.t.max_data_len =
+ nla_get_u8(info->attrs[IPEH_TLV_ATTR_MAX_DATA_LEN]);
+
+ if (info->attrs[IPEH_TLV_ATTR_MIN_DATA_LEN])
+ new_params.t.min_data_len =
+ nla_get_u8(info->attrs[IPEH_TLV_ATTR_MIN_DATA_LEN]);
+
+ if (info->attrs[IPEH_TLV_ATTR_DATA_LEN_MULT]) {
+ v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_DATA_LEN_MULT]);
+ if (v > 16 || v < 1)
+ goto out;
+ new_params.t.data_len_mult = v - 1;
+ }
+
+ if (info->attrs[IPEH_TLV_ATTR_DATA_LEN_OFF]) {
+ v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_DATA_LEN_OFF]);
+ if (v > 15)
+ goto out;
+ new_params.t.data_len_off = v;
+ }
+
+ retv = ipeh_tlv_set_params(tlv_param_table, type, &new_params);
+
+out:
+ rcu_read_unlock();
+ return retv;
+}
+EXPORT_SYMBOL(ipeh_tlv_nl_cmd_set);
+
+int ipeh_tlv_nl_cmd_unset(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct genl_info *info)
+{
+ unsigned char type;
+
+ if (!info->attrs[IPEH_TLV_ATTR_TYPE])
+ return -EINVAL;
+
+ type = nla_get_u8(info->attrs[IPEH_TLV_ATTR_TYPE]);
+ if (type < 2)
+ return -EINVAL;
+
+ return ipeh_tlv_unset_params(tlv_param_table, type);
+}
+EXPORT_SYMBOL(ipeh_tlv_nl_cmd_unset);
+
+static int tlv_fill_info(struct tlv_proc *tproc, unsigned char type,
+ struct sk_buff *msg, bool admin)
+{
+ struct tlv_params *tp = &tproc->params;
+ int ret = 0;
+
+ if (nla_put_u8(msg, IPEH_TLV_ATTR_TYPE, type) ||
+ nla_put_u16(msg, IPEH_TLV_ATTR_ORDER, tp->t.preferred_order) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_USER_PERM, tp->t.user_perm) ||
+ (admin && nla_put_u8(msg, IPEH_TLV_ATTR_ADMIN_PERM,
+ tp->t.admin_perm)) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_CLASS, tp->t.class) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_ALIGN_MULT, tp->t.align_mult + 1) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_ALIGN_OFF, tp->t.align_off) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_MIN_DATA_LEN, tp->t.min_data_len) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_MAX_DATA_LEN, tp->t.max_data_len) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_DATA_LEN_MULT,
+ tp->t.data_len_mult + 1) ||
+ nla_put_u8(msg, IPEH_TLV_ATTR_DATA_LEN_OFF, tp->t.data_len_off))
+ ret = -1;
+
+ return ret;
+}
+
+static int tlv_dump_info(struct tlv_proc *tproc, unsigned char type,
+ struct genl_family *tlv_nl_family, u32 portid,
+ u32 seq, u32 flags, struct sk_buff *skb, u8 cmd,
+ bool admin)
+{
+ void *hdr;
+
+ hdr = genlmsg_put(skb, portid, seq, tlv_nl_family, flags, cmd);
+ if (!hdr)
+ return -ENOMEM;
+
+ if (tlv_fill_info(tproc, type, skb, admin) < 0) {
+ genlmsg_cancel(skb, hdr);
+ return -EMSGSIZE;
+ }
+
+ genlmsg_end(skb, hdr);
+
+ return 0;
+}
+
+int ipeh_tlv_nl_cmd_get(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct genl_info *info)
+{
+ struct tlv_proc *tproc;
+ struct sk_buff *msg;
+ unsigned char type;
+ int ret;
+
+ if (!info->attrs[IPEH_TLV_ATTR_TYPE])
+ return -EINVAL;
+
+ type = nla_get_u8(info->attrs[IPEH_TLV_ATTR_TYPE]);
+ if (type < 2)
+ return -EINVAL;
+
+ msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+ if (!msg)
+ return -ENOMEM;
+
+ rcu_read_lock();
+
+ tproc = ipeh_tlv_get_proc_by_type(tlv_param_table, type);
+ ret = tlv_dump_info(tproc, type, tlv_nl_family, info->snd_portid,
+ info->snd_seq, 0, msg, info->genlhdr->cmd,
+ netlink_capable(skb, CAP_NET_ADMIN));
+
+ rcu_read_unlock();
+
+ if (ret < 0) {
+ nlmsg_free(msg);
+ return ret;
+ }
+
+ return genlmsg_reply(msg, info);
+}
+EXPORT_SYMBOL(ipeh_tlv_nl_cmd_get);
+
+int ipeh_tlv_nl_dump(struct tlv_param_table *tlv_param_table,
+ struct genl_family *tlv_nl_family,
+ struct sk_buff *skb, struct netlink_callback *cb)
+{
+ struct tlv_proc *tproc;
+ int idx = 0, ret, i;
+
+ rcu_read_lock();
+
+ for (i = 2; i < 256; i++) {
+ if (idx++ < cb->args[0])
+ continue;
+
+ tproc = ipeh_tlv_get_proc_by_type(tlv_param_table, i);
+ ret = tlv_dump_info(tproc, i, tlv_nl_family,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq, NLM_F_MULTI,
+ skb, IPEH_TLV_CMD_GET,
+ netlink_capable(cb->skb, CAP_NET_ADMIN));
+ if (ret)
+ break;
+ }
+
+ rcu_read_unlock();
+
+ cb->args[0] = idx;
+ return skb->len;
+}
+EXPORT_SYMBOL(ipeh_tlv_nl_dump);
+
int ipeh_exthdrs_init(struct tlv_param_table *tlv_param_table,
const struct tlv_proc_init *tlv_init_params,
int num_init_params)
diff --git a/net/ipv6/exthdrs_options.c b/net/ipv6/exthdrs_options.c
index 3b50b58..c1889f6 100644
--- a/net/ipv6/exthdrs_options.c
+++ b/net/ipv6/exthdrs_options.c
@@ -6,6 +6,7 @@
#include <linux/socket.h>
#include <linux/types.h>
#include <net/calipso.h>
+#include <net/genetlink.h>
#include <net/ipv6.h>
#include <net/ip6_route.h>
#if IS_ENABLED(CONFIG_IPV6_MIP6)
@@ -253,13 +254,89 @@ static const struct tlv_proc_init tlv_ipv6_init_params[] __initconst = {
struct tlv_param_table __rcu ipv6_tlv_param_table;
EXPORT_SYMBOL(ipv6_tlv_param_table);
+static int ipv6_tlv_nl_cmd_set(struct sk_buff *skb, struct genl_info *info);
+static int ipv6_tlv_nl_cmd_unset(struct sk_buff *skb, struct genl_info *info);
+static int ipv6_tlv_nl_cmd_get(struct sk_buff *skb, struct genl_info *info);
+static int ipv6_tlv_nl_dump(struct sk_buff *skb, struct netlink_callback *cb);
+
+static const struct genl_ops ipv6_tlv_nl_ops[] = {
+{
+ .cmd = IPEH_TLV_CMD_SET,
+ .doit = ipv6_tlv_nl_cmd_set,
+ .flags = GENL_ADMIN_PERM,
+},
+{
+ .cmd = IPEH_TLV_CMD_UNSET,
+ .doit = ipv6_tlv_nl_cmd_unset,
+ .flags = GENL_ADMIN_PERM,
+},
+{
+ .cmd = IPEH_TLV_CMD_GET,
+ .doit = ipv6_tlv_nl_cmd_get,
+ .dumpit = ipv6_tlv_nl_dump,
+},
+};
+
+struct genl_family ipv6_tlv_nl_family __ro_after_init = {
+ .hdrsize = 0,
+ .name = IPV6_TLV_GENL_NAME,
+ .version = IPV6_TLV_GENL_VERSION,
+ .maxattr = IPEH_TLV_ATTR_MAX,
+ .policy = ipeh_tlv_nl_policy,
+ .netnsok = true,
+ .parallel_ops = true,
+ .ops = ipv6_tlv_nl_ops,
+ .n_ops = ARRAY_SIZE(ipv6_tlv_nl_ops),
+ .module = THIS_MODULE,
+};
+
+static int ipv6_tlv_nl_cmd_set(struct sk_buff *skb, struct genl_info *info)
+{
+ return ipeh_tlv_nl_cmd_set(&ipv6_tlv_param_table, &ipv6_tlv_nl_family,
+ skb, info);
+}
+
+static int ipv6_tlv_nl_cmd_unset(struct sk_buff *skb, struct genl_info *info)
+{
+ return ipeh_tlv_nl_cmd_unset(&ipv6_tlv_param_table, &ipv6_tlv_nl_family,
+ skb, info);
+}
+
+static int ipv6_tlv_nl_cmd_get(struct sk_buff *skb, struct genl_info *info)
+{
+ return ipeh_tlv_nl_cmd_get(&ipv6_tlv_param_table, &ipv6_tlv_nl_family,
+ skb, info);
+}
+
+static int ipv6_tlv_nl_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+ return ipeh_tlv_nl_dump(&ipv6_tlv_param_table, &ipv6_tlv_nl_family,
+ skb, cb);
+}
+
int __init ipv6_exthdrs_options_init(void)
{
- return ipeh_exthdrs_init(&ipv6_tlv_param_table, tlv_ipv6_init_params,
- ARRAY_SIZE(tlv_ipv6_init_params));
+ int err;
+
+ err = genl_register_family(&ipv6_tlv_nl_family);
+ if (err)
+ goto genl_fail;
+
+ ipeh_exthdrs_init(&ipv6_tlv_param_table, tlv_ipv6_init_params,
+ ARRAY_SIZE(tlv_ipv6_init_params));
+ if (err)
+ goto ipv6_fail;
+
+ return 0;
+
+ipv6_fail:
+ genl_unregister_family(&ipv6_tlv_nl_family);
+genl_fail:
+ return err;
}
void ipv6_exthdrs_options_exit(void)
{
ipeh_exthdrs_fini(&ipv6_tlv_param_table);
+ genl_unregister_family(&ipv6_tlv_nl_family);
}
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 6/7] ip6tlvs: Add netlink interface
2019-10-03 21:58 ` [PATCH v5 net-next 6/7] ip6tlvs: Add netlink interface Tom Herbert
@ 2019-10-07 12:39 ` Simon Horman
2019-12-14 19:25 ` Tom Herbert
0 siblings, 1 reply; 20+ messages in thread
From: Simon Horman @ 2019-10-07 12:39 UTC (permalink / raw)
To: Tom Herbert; +Cc: davem, netdev, Tom Herbert
On Thu, Oct 03, 2019 at 02:58:03PM -0700, Tom Herbert wrote:
> From: Tom Herbert <tom@quantonium.net>
>
> Add a netlink interface to manage the TX TLV parameters. Managed
> parameters include those for validating and sending TLVs being sent
> such as alignment, TLV ordering, length limits, etc.
>
> Signed-off-by: Tom Herbert <tom@herbertland.com>
Hi Tom,
I am wondering if you considered including extack support in this code.
...
> diff --git a/include/uapi/linux/ipeh.h b/include/uapi/linux/ipeh.h
> index dbf0728..bac36a7 100644
> --- a/include/uapi/linux/ipeh.h
> +++ b/include/uapi/linux/ipeh.h
> @@ -21,4 +21,33 @@ enum {
> IPEH_TLV_PERM_MAX = IPEH_TLV_PERM_NO_CHECK
> };
>
> +/* NETLINK_GENERIC related info for IP TLVs */
> +
> +enum {
> + IPEH_TLV_ATTR_UNSPEC,
> + IPEH_TLV_ATTR_TYPE, /* u8, > 1 */
> + IPEH_TLV_ATTR_ORDER, /* u16 */
> + IPEH_TLV_ATTR_ADMIN_PERM, /* u8, perm value */
> + IPEH_TLV_ATTR_USER_PERM, /* u8, perm value */
My reading of struct tlv_tx_params is that admin_perm and user_perm are
2-bit entities whose valid values are currently 0, 1 and 2. Perhaps that
would be worth noting here in keeping with restrictions noted for other
attributes.
> + IPEH_TLV_ATTR_CLASS, /* u8, 3 bit flags */
> + IPEH_TLV_ATTR_ALIGN_MULT, /* u8, 1 to 16 */
> + IPEH_TLV_ATTR_ALIGN_OFF, /* u8, 0 to 15 */
> + IPEH_TLV_ATTR_MIN_DATA_LEN, /* u8 (option data length) */
> + IPEH_TLV_ATTR_MAX_DATA_LEN, /* u8 (option data length) */
> + IPEH_TLV_ATTR_DATA_LEN_MULT, /* u8, 1 to 16 */
> + IPEH_TLV_ATTR_DATA_LEN_OFF, /* u8, 0 to 15 */
> +
> + __IPEH_TLV_ATTR_MAX,
> +};
> +
> +#define IPEH_TLV_ATTR_MAX (__IPEH_TLV_ATTR_MAX - 1)
...
> diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
> index feaa4a6..e3c1f33 100644
> --- a/net/ipv6/exthdrs_common.c
> +++ b/net/ipv6/exthdrs_common.c
> @@ -454,6 +454,244 @@ int __ipeh_tlv_unset(struct tlv_param_table *tlv_param_table,
...
> +int ipeh_tlv_nl_cmd_set(struct tlv_param_table *tlv_param_table,
> + struct genl_family *tlv_nl_family,
> + struct sk_buff *skb, struct genl_info *info)
> +{
> + struct tlv_params new_params;
> + struct tlv_proc *tproc;
> + unsigned char type;
> + unsigned int v;
> + int retv = -EINVAL;
> +
> + if (!info->attrs[IPEH_TLV_ATTR_TYPE])
> + return -EINVAL;
> +
> + type = nla_get_u8(info->attrs[IPEH_TLV_ATTR_TYPE]);
> + if (type < 2)
> + return -EINVAL;
> +
> + rcu_read_lock();
> +
> + /* Base new parameters on existing ones */
> + tproc = ipeh_tlv_get_proc_by_type(tlv_param_table, type);
> + new_params = tproc->params;
> +
> + if (info->attrs[IPEH_TLV_ATTR_ORDER]) {
> + v = nla_get_u16(info->attrs[IPEH_TLV_ATTR_ORDER]);
> + new_params.t.preferred_order = v;
> + }
> +
> + if (info->attrs[IPEH_TLV_ATTR_ADMIN_PERM]) {
> + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ADMIN_PERM]);
> + if (v > IPEH_TLV_PERM_MAX)
> + goto out;
> + new_params.t.admin_perm = v;
> + }
> +
> + if (info->attrs[IPEH_TLV_ATTR_USER_PERM]) {
> + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_USER_PERM]);
> + if (v > IPEH_TLV_PERM_MAX)
> + goto out;
> + new_params.t.user_perm = v;
> + }
> +
> + if (info->attrs[IPEH_TLV_ATTR_CLASS]) {
> + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_CLASS]);
> + if (!v || (v & ~IPEH_TLV_CLASS_FLAG_MASK))
> + goto out;
> + new_params.t.class = v;
> + }
> +
> + if (info->attrs[IPEH_TLV_ATTR_ALIGN_MULT]) {
> + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ALIGN_MULT]);
> + if (v > 16 || v < 1)
> + goto out;
> + new_params.t.align_mult = v - 1;
> + }
> +
> + if (info->attrs[IPEH_TLV_ATTR_ALIGN_OFF]) {
> + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ALIGN_OFF]);
> + if (v > 15)
> + goto out;
> + new_params.t.align_off = v;
> + }
> +
> + if (info->attrs[IPEH_TLV_ATTR_MAX_DATA_LEN])
> + new_params.t.max_data_len =
> + nla_get_u8(info->attrs[IPEH_TLV_ATTR_MAX_DATA_LEN]);
> +
> + if (info->attrs[IPEH_TLV_ATTR_MIN_DATA_LEN])
> + new_params.t.min_data_len =
> + nla_get_u8(info->attrs[IPEH_TLV_ATTR_MIN_DATA_LEN]);
> +
> + if (info->attrs[IPEH_TLV_ATTR_DATA_LEN_MULT]) {
> + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_DATA_LEN_MULT]);
> + if (v > 16 || v < 1)
> + goto out;
> + new_params.t.data_len_mult = v - 1;
> + }
Is some sanity checking warranted for the min/max data len values.
f.e. that min <= max ?
> +
> + if (info->attrs[IPEH_TLV_ATTR_DATA_LEN_OFF]) {
> + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_DATA_LEN_OFF]);
> + if (v > 15)
> + goto out;
> + new_params.t.data_len_off = v;
> + }
> +
> + retv = ipeh_tlv_set_params(tlv_param_table, type, &new_params);
> +
> +out:
> + rcu_read_unlock();
> + return retv;
> +}
> +EXPORT_SYMBOL(ipeh_tlv_nl_cmd_set);
...
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH v5 net-next 6/7] ip6tlvs: Add netlink interface
2019-10-07 12:39 ` Simon Horman
@ 2019-12-14 19:25 ` Tom Herbert
0 siblings, 0 replies; 20+ messages in thread
From: Tom Herbert @ 2019-12-14 19:25 UTC (permalink / raw)
To: Simon Horman
Cc: David S. Miller, Linux Kernel Network Developers, Tom Herbert
On Mon, Oct 7, 2019 at 5:39 AM Simon Horman <simon.horman@netronome.com> wrote:
>
> On Thu, Oct 03, 2019 at 02:58:03PM -0700, Tom Herbert wrote:
> > From: Tom Herbert <tom@quantonium.net>
> >
> > Add a netlink interface to manage the TX TLV parameters. Managed
> > parameters include those for validating and sending TLVs being sent
> > such as alignment, TLV ordering, length limits, etc.
> >
> > Signed-off-by: Tom Herbert <tom@herbertland.com>
>
> Hi Tom,
>
> I am wondering if you considered including extack support in this code.
>
Good point, will look into adding it.
> ...
>
> > diff --git a/include/uapi/linux/ipeh.h b/include/uapi/linux/ipeh.h
> > index dbf0728..bac36a7 100644
> > --- a/include/uapi/linux/ipeh.h
> > +++ b/include/uapi/linux/ipeh.h
> > @@ -21,4 +21,33 @@ enum {
> > IPEH_TLV_PERM_MAX = IPEH_TLV_PERM_NO_CHECK
> > };
> >
> > +/* NETLINK_GENERIC related info for IP TLVs */
> > +
> > +enum {
> > + IPEH_TLV_ATTR_UNSPEC,
> > + IPEH_TLV_ATTR_TYPE, /* u8, > 1 */
> > + IPEH_TLV_ATTR_ORDER, /* u16 */
> > + IPEH_TLV_ATTR_ADMIN_PERM, /* u8, perm value */
> > + IPEH_TLV_ATTR_USER_PERM, /* u8, perm value */
>
> My reading of struct tlv_tx_params is that admin_perm and user_perm are
> 2-bit entities whose valid values are currently 0, 1 and 2. Perhaps that
> would be worth noting here in keeping with restrictions noted for other
> attributes.
>
Okay.
> > + IPEH_TLV_ATTR_CLASS, /* u8, 3 bit flags */
> > + IPEH_TLV_ATTR_ALIGN_MULT, /* u8, 1 to 16 */
> > + IPEH_TLV_ATTR_ALIGN_OFF, /* u8, 0 to 15 */
> > + IPEH_TLV_ATTR_MIN_DATA_LEN, /* u8 (option data length) */
> > + IPEH_TLV_ATTR_MAX_DATA_LEN, /* u8 (option data length) */
> > + IPEH_TLV_ATTR_DATA_LEN_MULT, /* u8, 1 to 16 */
> > + IPEH_TLV_ATTR_DATA_LEN_OFF, /* u8, 0 to 15 */
> > +
> > + __IPEH_TLV_ATTR_MAX,
> > +};
> > +
> > +#define IPEH_TLV_ATTR_MAX (__IPEH_TLV_ATTR_MAX - 1)
>
> ...
>
> > diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
> > index feaa4a6..e3c1f33 100644
> > --- a/net/ipv6/exthdrs_common.c
> > +++ b/net/ipv6/exthdrs_common.c
> > @@ -454,6 +454,244 @@ int __ipeh_tlv_unset(struct tlv_param_table *tlv_param_table,
>
> ...
>
> > +int ipeh_tlv_nl_cmd_set(struct tlv_param_table *tlv_param_table,
> > + struct genl_family *tlv_nl_family,
> > + struct sk_buff *skb, struct genl_info *info)
> > +{
> > + struct tlv_params new_params;
> > + struct tlv_proc *tproc;
> > + unsigned char type;
> > + unsigned int v;
> > + int retv = -EINVAL;
> > +
> > + if (!info->attrs[IPEH_TLV_ATTR_TYPE])
> > + return -EINVAL;
> > +
> > + type = nla_get_u8(info->attrs[IPEH_TLV_ATTR_TYPE]);
> > + if (type < 2)
> > + return -EINVAL;
> > +
> > + rcu_read_lock();
> > +
> > + /* Base new parameters on existing ones */
> > + tproc = ipeh_tlv_get_proc_by_type(tlv_param_table, type);
> > + new_params = tproc->params;
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_ORDER]) {
> > + v = nla_get_u16(info->attrs[IPEH_TLV_ATTR_ORDER]);
> > + new_params.t.preferred_order = v;
> > + }
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_ADMIN_PERM]) {
> > + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ADMIN_PERM]);
> > + if (v > IPEH_TLV_PERM_MAX)
> > + goto out;
> > + new_params.t.admin_perm = v;
> > + }
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_USER_PERM]) {
> > + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_USER_PERM]);
> > + if (v > IPEH_TLV_PERM_MAX)
> > + goto out;
> > + new_params.t.user_perm = v;
> > + }
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_CLASS]) {
> > + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_CLASS]);
> > + if (!v || (v & ~IPEH_TLV_CLASS_FLAG_MASK))
> > + goto out;
> > + new_params.t.class = v;
> > + }
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_ALIGN_MULT]) {
> > + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ALIGN_MULT]);
> > + if (v > 16 || v < 1)
> > + goto out;
> > + new_params.t.align_mult = v - 1;
> > + }
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_ALIGN_OFF]) {
> > + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_ALIGN_OFF]);
> > + if (v > 15)
> > + goto out;
> > + new_params.t.align_off = v;
> > + }
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_MAX_DATA_LEN])
> > + new_params.t.max_data_len =
> > + nla_get_u8(info->attrs[IPEH_TLV_ATTR_MAX_DATA_LEN]);
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_MIN_DATA_LEN])
> > + new_params.t.min_data_len =
> > + nla_get_u8(info->attrs[IPEH_TLV_ATTR_MIN_DATA_LEN]);
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_DATA_LEN_MULT]) {
> > + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_DATA_LEN_MULT]);
> > + if (v > 16 || v < 1)
> > + goto out;
> > + new_params.t.data_len_mult = v - 1;
> > + }
>
> Is some sanity checking warranted for the min/max data len values.
> f.e. that min <= max ?
Okay, will add.
>
> > +
> > + if (info->attrs[IPEH_TLV_ATTR_DATA_LEN_OFF]) {
> > + v = nla_get_u8(info->attrs[IPEH_TLV_ATTR_DATA_LEN_OFF]);
> > + if (v > 15)
> > + goto out;
> > + new_params.t.data_len_off = v;
> > + }
> > +
> > + retv = ipeh_tlv_set_params(tlv_param_table, type, &new_params);
> > +
> > +out:
> > + rcu_read_unlock();
> > + return retv;
> > +}
> > +EXPORT_SYMBOL(ipeh_tlv_nl_cmd_set);
>
> ...
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH v5 net-next 7/7] ip6tlvs: Validation of TX Destination and Hop-by-Hop options
2019-10-03 21:57 [PATCH v5 net-next 0/7] ipv6: Extension header infrastructure Tom Herbert
` (5 preceding siblings ...)
2019-10-03 21:58 ` [PATCH v5 net-next 6/7] ip6tlvs: Add netlink interface Tom Herbert
@ 2019-10-03 21:58 ` Tom Herbert
6 siblings, 0 replies; 20+ messages in thread
From: Tom Herbert @ 2019-10-03 21:58 UTC (permalink / raw)
To: davem, netdev; +Cc: Tom Herbert, Tom Herbert
From: Tom Herbert <tom@quantonium.net>
Validate Destination and Hop-by-Hop options. This uses the information
in the TLV parameters table to validate various aspects of both
individual TLVs as well as a list of TLVs in an extension header.
There are two levels of validation that can be performed: simple checks
and deep checks. Simple checks validate only the most basic properties
such as that the TLV list fits into the EH. Deep checks do a fine
grained validation that includes perferred ordering, length limits,
and length alignment.
With proper permissions set in the TLV parameter table, this patch
allows non-privileged users to send TLVs. Given that TLVs are open
ended and potentially a source of DOS attack, deep checks are
performed to limit the format that a non-privileged user can send.
If deep checks are enabled, a canonical format for sending TLVs is
enforced (in adherence with the robustness principle). A TLV must
be well ordered with respect to the preferred order for the TLV.
Each TLV must be aligned as described in the parameter table. Minimal
padding (one padding TLV) is used to align TLVs. The length of the
extension header as well as the count of non-padding TLVs is checked
against max_*_opts_len and max_*_opts_cnt. For individual TLVs, length
limits and length alignment is checked.
Signed-off-by: Tom Herbert <tom@herbertland.com>
---
include/net/ipeh.h | 22 +++
net/ipv6/datagram.c | 51 +++++--
net/ipv6/exthdrs_common.c | 382 ++++++++++++++++++++++++++++++++++++++++++++++
net/ipv6/ipv6_sockglue.c | 39 ++---
4 files changed, 455 insertions(+), 39 deletions(-)
diff --git a/include/net/ipeh.h b/include/net/ipeh.h
index 8474a43..e58a0f4 100644
--- a/include/net/ipeh.h
+++ b/include/net/ipeh.h
@@ -157,6 +157,28 @@ struct ipv6_txoptions *ipeh_renew_options(struct sock *sk,
struct ipv6_txoptions *ipeh_fixup_options(struct ipv6_txoptions *opt_space,
struct ipv6_txoptions *opt);
+int ipeh_opt_validate_tlvs(struct net *net,
+ struct tlv_param_table *tlv_param_table,
+ struct ipv6_opt_hdr *opt,
+ unsigned int optname, bool admin,
+ unsigned int max_len, unsigned int max_cnt);
+int ipeh_opt_validate_single_tlv(struct net *net,
+ struct tlv_param_table *tlv_param_table,
+ unsigned int optname, const __u8 *tlv,
+ size_t len, bool deleting, bool admin);
+int ipeh_opt_check_perm(struct net *net,
+ struct tlv_param_table *tlv_param_table,
+ struct ipv6_txoptions *txopt, int optname, bool admin);
+
+struct ipv6_txoptions *ipeh_txopt_from_opt(struct sock *sk,
+ struct tlv_param_table
+ *tlv_param_table,
+ struct ipv6_txoptions *opt,
+ int optname, char __user *optval,
+ unsigned int optlen,
+ unsigned int max_len,
+ unsigned int max_cnt);
+
/* Generic extension header TLV parser */
enum ipeh_parse_errors {
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index 96f9392..f9856a7 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -839,7 +839,10 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
break;
case IPV6_2292HOPOPTS:
- case IPV6_HOPOPTS:
+ case IPV6_HOPOPTS: {
+ int max_len = net->ipv6.sysctl.max_hbh_opts_len;
+ int max_cnt = net->ipv6.sysctl.max_hbh_opts_cnt;
+
if (opt->hopopt || cmsg->cmsg_len < CMSG_LEN(sizeof(struct ipv6_opt_hdr))) {
err = -EINVAL;
goto exit_f;
@@ -851,15 +854,24 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
err = -EINVAL;
goto exit_f;
}
- if (!ns_capable(net->user_ns, CAP_NET_RAW)) {
- err = -EPERM;
+
+ err = ipeh_opt_validate_tlvs(net, &ipv6_tlv_param_table,
+ hdr, IPV6_HOPOPTS,
+ ns_capable(net->user_ns,
+ CAP_NET_RAW),
+ max_len, max_cnt);
+ if (err < 0)
goto exit_f;
- }
+
opt->opt_nflen += len;
opt->hopopt = hdr;
break;
+ }
+
+ case IPV6_2292DSTOPTS: {
+ int max_len = net->ipv6.sysctl.max_dst_opts_len;
+ int max_cnt = net->ipv6.sysctl.max_dst_opts_cnt;
- case IPV6_2292DSTOPTS:
if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct ipv6_opt_hdr))) {
err = -EINVAL;
goto exit_f;
@@ -871,10 +883,14 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
err = -EINVAL;
goto exit_f;
}
- if (!ns_capable(net->user_ns, CAP_NET_RAW)) {
- err = -EPERM;
+ err = ipeh_opt_validate_tlvs(net, &ipv6_tlv_param_table,
+ hdr, IPV6_DSTOPTS,
+ ns_capable(net->user_ns,
+ CAP_NET_RAW),
+ max_len, max_cnt);
+ if (err < 0)
goto exit_f;
- }
+
if (opt->dst1opt) {
err = -EINVAL;
goto exit_f;
@@ -882,9 +898,13 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
opt->opt_flen += len;
opt->dst1opt = hdr;
break;
+ }
case IPV6_DSTOPTS:
- case IPV6_RTHDRDSTOPTS:
+ case IPV6_RTHDRDSTOPTS: {
+ int max_len = net->ipv6.sysctl.max_dst_opts_len;
+ int max_cnt = net->ipv6.sysctl.max_dst_opts_cnt;
+
if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct ipv6_opt_hdr))) {
err = -EINVAL;
goto exit_f;
@@ -896,10 +916,15 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
err = -EINVAL;
goto exit_f;
}
- if (!ns_capable(net->user_ns, CAP_NET_RAW)) {
- err = -EPERM;
+
+ err = ipeh_opt_validate_tlvs(net, &ipv6_tlv_param_table,
+ hdr, IPV6_DSTOPTS,
+ ns_capable(net->user_ns,
+ CAP_NET_RAW),
+ max_len, max_cnt);
+ if (err < 0)
goto exit_f;
- }
+
if (cmsg->cmsg_type == IPV6_DSTOPTS) {
opt->opt_flen += len;
opt->dst1opt = hdr;
@@ -908,7 +933,7 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
opt->dst0opt = hdr;
}
break;
-
+ }
case IPV6_2292RTHDR:
case IPV6_RTHDR:
if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct ipv6_rt_hdr))) {
diff --git a/net/ipv6/exthdrs_common.c b/net/ipv6/exthdrs_common.c
index e3c1f33..c3047df 100644
--- a/net/ipv6/exthdrs_common.c
+++ b/net/ipv6/exthdrs_common.c
@@ -262,6 +262,318 @@ bool ipeh_parse_tlv(unsigned int class,
}
EXPORT_SYMBOL(ipeh_parse_tlv);
+/* TLV validation functions */
+
+/* Validate a single non-padding TLV */
+static int __ipeh_opt_validate_single_tlv(struct net *net, const __u8 *tlv,
+ struct tlv_proc *tproc,
+ unsigned int class, bool *deep_check,
+ bool deleting, bool admin)
+{
+ struct tlv_tx_params *tptx = &tproc->params.t;
+
+ if (tlv[0] < 2) /* Must be non-padding */
+ return -EINVAL;
+
+ /* Check permissions */
+ switch (admin ? tptx->admin_perm : tptx->user_perm) {
+ case IPEH_TLV_PERM_NO_CHECK:
+ /* Allowed with no deep checks */
+ *deep_check = false;
+ return 0;
+ case IPEH_TLV_PERM_WITH_CHECK:
+ /* Allowed with deep checks */
+ *deep_check = true;
+ break;
+ default:
+ /* No permission */
+ return -EPERM;
+ }
+
+ /* Perform deep checks on the TLV */
+
+ /* Check class */
+ if ((tptx->class & class) != class)
+ return -EINVAL;
+
+ /* Don't bother checking lengths when deleting, the TLV is only
+ * needed here for lookup
+ */
+ if (deleting) {
+ /* Don't bother with deep checks when deleting */
+ *deep_check = false;
+ } else {
+ /* Check length */
+ if (tlv[1] < tptx->min_data_len || tlv[1] > tptx->max_data_len)
+ return -EINVAL;
+
+ /* Check length alignment */
+ if ((tlv[1] % (tptx->data_len_mult + 1)) != tptx->data_len_off)
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static unsigned int optname_to_tlv_class(int optname)
+{
+ switch (optname) {
+ case IPV6_HOPOPTS:
+ return IPEH_TLV_CLASS_FLAG_HOPOPT;
+ case IPV6_RTHDRDSTOPTS:
+ return IPEH_TLV_CLASS_FLAG_RTRDSTOPT;
+ case IPV6_DSTOPTS:
+ return IPEH_TLV_CLASS_FLAG_DSTOPT;
+ default:
+ return -1U;
+ }
+}
+
+static int __ipeh_opt_validate_tlvs(struct net *net,
+ struct tlv_param_table *tlv_param_table,
+ struct ipv6_opt_hdr *opt,
+ unsigned int optname, bool deleting,
+ bool admin, unsigned int max_len,
+ unsigned int max_cnt)
+{
+ bool deep_check = !admin, did_deep_check = false;
+ unsigned int opt_len, tlv_len, offset;
+ unsigned int padding = 0, numpad = 0;
+ unsigned short prev_tlv_order = 0;
+ unsigned int class, cnt = 0;
+ struct tlv_tx_params *tptx;
+ int retc, ret = -EINVAL;
+ __u8 *tlv = (__u8 *)opt;
+ struct tlv_proc *tproc;
+
+ opt_len = ipv6_optlen(opt);
+ offset = sizeof(*opt);
+
+ class = optname_to_tlv_class(optname);
+
+ rcu_read_lock();
+
+ while (offset < opt_len) {
+ switch (tlv[offset]) {
+ case IPV6_TLV_PAD1:
+ tlv_len = 1;
+ padding++;
+ numpad++;
+ break;
+ case IPV6_TLV_PADN:
+ if (offset + 1 >= opt_len)
+ goto out;
+
+ tlv_len = tlv[offset + 1] + 2;
+
+ if (offset + tlv_len > opt_len)
+ goto out;
+
+ padding += tlv_len;
+ numpad++;
+ break;
+ default:
+ if (offset + 1 >= opt_len)
+ goto out;
+
+ tlv_len = tlv[offset + 1] + 2;
+
+ if (offset + tlv_len > opt_len)
+ goto out;
+
+ tproc = ipeh_tlv_get_proc(tlv_param_table,
+ &tlv[offset]);
+ tptx = &tproc->params.t;
+
+ retc = __ipeh_opt_validate_single_tlv(net, &tlv[offset],
+ tproc, class,
+ &deep_check,
+ deleting, admin);
+ if (retc < 0) {
+ ret = retc;
+ goto out;
+ }
+
+ if (deep_check) {
+ /* Check for too many options */
+ if (++cnt > max_cnt) {
+ ret = -E2BIG;
+ goto out;
+ }
+
+ /* Check order */
+ if (tptx->preferred_order < prev_tlv_order)
+ goto out;
+
+ /* Check alignment */
+ if ((offset % (tptx->align_mult + 1)) !=
+ tptx->align_off)
+ goto out;
+
+ /* Check for right amount of padding */
+ if (numpad > 1 || padding > tptx->align_mult)
+ goto out;
+
+ prev_tlv_order = tptx->preferred_order;
+
+ did_deep_check = true;
+ }
+ padding = 0;
+ numpad = 0;
+ }
+ offset += tlv_len;
+ }
+
+ /* Check trailing padding. Note this covers the case option list
+ * only contains padding.
+ */
+ if (deep_check && (numpad > 1 || padding > 7))
+ goto out;
+
+ /* If we did at least one deep check apply length limit */
+ if (did_deep_check && opt_len > max_len) {
+ ret = -EMSGSIZE;
+ goto out;
+ }
+
+ /* All good */
+ ret = 0;
+out:
+ rcu_read_unlock();
+
+ return ret;
+}
+
+/**
+ * ipeh_opt_validate_tlvs - Validate TLVs.
+ * @net: Current net
+ * @tlv_param_table: TLV parameter table
+ * @opt: The option header
+ * @optname: IPV6_HOPOPTS, IPV6_RTHDRDSTOPTS, or IPV6_DSTOPTS
+ * @admin: Set for privileged user
+ * @max_len: Maximum length for TLV
+ * @max_cnt: Maximum number of non-padding TLVs
+ *
+ * Description:
+ * Walks the TLVs in a list to verify that the TLV lengths and other
+ * parameters are in bounds for a Destination or Hop-by-Hop option.
+ * Return -EINVAL is there is a problem, zero otherwise.
+ */
+int ipeh_opt_validate_tlvs(struct net *net,
+ struct tlv_param_table *tlv_param_table,
+ struct ipv6_opt_hdr *opt, unsigned int optname,
+ bool admin, unsigned int max_len,
+ unsigned int max_cnt)
+{
+ return __ipeh_opt_validate_tlvs(net, tlv_param_table, opt, optname,
+ false, admin, max_len, max_cnt);
+}
+EXPORT_SYMBOL(ipeh_opt_validate_tlvs);
+
+/**
+ * ipeh_opt_validate_single_tlv - Check that a single TLV is valid.
+ * @net: Current net
+ * @tlv_param_table: TLV parameter table
+ * @optname: IPV6_HOPOPTS, IPV6_RTHDRDSTOPTS, or IPV6_DSTOPTS
+ * @tlv: The TLV as array of bytes
+ * @len: Length of buffer holding TLV
+ * @deleting: TLV is being deleted
+ * @admin: Set for privileged user
+ *
+ * Description:
+ * Validates a single TLV. The TLV must be non-padding type. The length
+ * of the TLV (as determined by the second byte that gives length of the
+ * option data) must match @len.
+ */
+int ipeh_opt_validate_single_tlv(struct net *net,
+ struct tlv_param_table *tlv_param_table,
+ unsigned int optname, const __u8 *tlv,
+ size_t len, bool deleting, bool admin)
+{
+ struct tlv_proc *tproc;
+ unsigned int class;
+ bool deep_check;
+ int ret = 0;
+
+ class = optname_to_tlv_class(optname);
+
+ if (tlv[0] < 2)
+ return -EINVAL;
+
+ if (len < 2)
+ return -EINVAL;
+
+ if (tlv[1] + 2 != len)
+ return -EINVAL;
+
+ rcu_read_lock();
+
+ tproc = ipeh_tlv_get_proc(tlv_param_table, tlv);
+
+ ret = __ipeh_opt_validate_single_tlv(net, tlv, tproc, class,
+ &deep_check, deleting, admin);
+
+ rcu_read_unlock();
+
+ return ret;
+}
+EXPORT_SYMBOL(ipeh_opt_validate_single_tlv);
+
+/**
+ * ipeh_opt_check_perm - Check that current capabilities allows modifying
+ * txopts.
+ * @net: Current net
+ * @tlv_param_table: TLV parameter table
+ * @txopts: TX options from the socket
+ * @optname: IPV6_HOPOPTS, IPV6_RTHDRDSTOPTS, or IPV6_DSTOPTS
+ * @admin: Set for privileged user
+ *
+ * Description:
+ *
+ * Checks whether the permissions of TLV that are set on a socket permit
+ * modificationr.
+ *
+ */
+int ipeh_opt_check_perm(struct net *net,
+ struct tlv_param_table *tlv_param_table,
+ struct ipv6_txoptions *txopt, int optname, bool admin)
+{
+ struct ipv6_opt_hdr *opt;
+ int retv = -EPERM;
+
+ if (!txopt)
+ return 0;
+
+ switch (optname) {
+ case IPV6_HOPOPTS:
+ opt = txopt->hopopt;
+ break;
+ case IPV6_RTHDRDSTOPTS:
+ opt = txopt->dst0opt;
+ break;
+ case IPV6_DSTOPTS:
+ opt = txopt->dst1opt;
+ break;
+ default:
+ goto out;
+ }
+
+ if (!opt) {
+ retv = 0;
+ goto out;
+ }
+
+ /* Just call the validate function on the options as being
+ * deleted.
+ */
+ retv = __ipeh_opt_validate_tlvs(net, tlv_param_table, opt, optname,
+ true, admin, -1U, -1U);
+
+out:
+ return retv;
+}
+EXPORT_SYMBOL(ipeh_opt_check_perm);
+
/* TLV parameter table functions and structures */
/* Default (unset) values for TLV parameters */
@@ -454,6 +766,76 @@ int __ipeh_tlv_unset(struct tlv_param_table *tlv_param_table,
}
EXPORT_SYMBOL(__ipeh_tlv_unset);
+/* Utility function tp create TX options from a setsockopt that is setting
+ * options on a socket.
+ */
+struct ipv6_txoptions *ipeh_txopt_from_opt(struct sock *sk,
+ struct tlv_param_table
+ *tlv_param_table,
+ struct ipv6_txoptions *opt,
+ int optname, char __user *optval,
+ unsigned int optlen,
+ unsigned int max_len,
+ unsigned int max_cnt)
+{
+ struct ipv6_opt_hdr *new = NULL;
+ struct net *net = sock_net(sk);
+ int retv;
+
+ /* remove any sticky options header with a zero option
+ * length, per RFC3542.
+ */
+ if (optlen == 0) {
+ optval = NULL;
+ } else if (!optval) {
+ return ERR_PTR(-EINVAL);
+ } else if (optlen < sizeof(struct ipv6_opt_hdr) ||
+ optlen & 0x7 || optlen > 8 * 255) {
+ return ERR_PTR(-EINVAL);
+ } else {
+ new = memdup_user(optval, optlen);
+ if (IS_ERR(new))
+ return (struct ipv6_txoptions *)new;
+ if (unlikely(ipv6_optlen(new) > optlen)) {
+ kfree(new);
+ return ERR_PTR(-EINVAL);
+ }
+ }
+
+ if (optname != IPV6_RTHDR) {
+ bool cap = ns_capable(net->user_ns, CAP_NET_RAW);
+
+ /* First check if we have permission to delete
+ * the existing options on the socket.
+ */
+ retv = ipeh_opt_check_perm(net, tlv_param_table,
+ opt, optname, cap);
+ if (retv < 0) {
+ kfree(new);
+ return ERR_PTR(retv);
+ }
+
+ /* Check permissions and other validations on new
+ * TLVs
+ */
+ if (new) {
+ retv = ipeh_opt_validate_tlvs(net, tlv_param_table,
+ new, optname, cap,
+ max_len, max_cnt);
+ if (retv < 0) {
+ kfree(new);
+ return ERR_PTR(retv);
+ }
+ }
+ }
+
+ opt = ipeh_renew_options(sk, opt, optname, new);
+ kfree(new);
+
+ return opt;
+}
+EXPORT_SYMBOL(ipeh_txopt_from_opt);
+
const struct nla_policy ipeh_tlv_nl_policy[IPEH_TLV_ATTR_MAX + 1] = {
[IPEH_TLV_ATTR_TYPE] = { .type = NLA_U8, },
[IPEH_TLV_ATTR_ORDER] = { .type = NLA_U16, },
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 8755ecc..b8e007ca 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -395,40 +395,27 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
case IPV6_RTHDR:
case IPV6_DSTOPTS:
{
+ unsigned int max_len = -1U, max_cnt = -1U;
struct ipv6_txoptions *opt;
- struct ipv6_opt_hdr *new = NULL;
- /* hop-by-hop / destination options are privileged option */
- retv = -EPERM;
- if (optname != IPV6_RTHDR && !ns_capable(net->user_ns, CAP_NET_RAW))
+ switch (optname) {
+ case IPV6_HOPOPTS:
+ max_len = net->ipv6.sysctl.max_hbh_opts_len;
+ max_cnt = net->ipv6.sysctl.max_hbh_opts_cnt;
break;
-
- /* remove any sticky options header with a zero option
- * length, per RFC3542.
- */
- if (optlen == 0)
- optval = NULL;
- else if (!optval)
- goto e_inval;
- else if (optlen < sizeof(struct ipv6_opt_hdr) ||
- optlen & 0x7 || optlen > 8 * 255)
- goto e_inval;
- else {
- new = memdup_user(optval, optlen);
- if (IS_ERR(new)) {
- retv = PTR_ERR(new);
+ case IPV6_RTHDRDSTOPTS:
+ case IPV6_DSTOPTS:
+ max_len = net->ipv6.sysctl.max_dst_opts_len;
+ max_cnt = net->ipv6.sysctl.max_dst_opts_cnt;
break;
- }
- if (unlikely(ipv6_optlen(new) > optlen)) {
- kfree(new);
- goto e_inval;
- }
}
opt = rcu_dereference_protected(np->opt,
lockdep_sock_is_held(sk));
- opt = ipeh_renew_options(sk, opt, optname, new);
- kfree(new);
+ opt = ipeh_txopt_from_opt(sk, &ipv6_tlv_param_table, opt,
+ optname, optval, optlen, max_len,
+ max_cnt);
+
if (IS_ERR(opt)) {
retv = PTR_ERR(opt);
break;
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread