* [PATCH Round 3 2/2] Core network changes to support network event notification.
From: Steve Wise @ 2006-06-27 20:51 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <20060627205050.30723.25753.stgit@stevo-desktop>
This patch adds netevent and netlink calls for neighbour change, route
add/del, pmtu change, and routing redirect events.
Netlink Details:
Neighbour change events are broadcast as a new ndmsg type RTM_NEIGHUPD.
Path mtu change events are broadcast as a new rtmsg type RTM_ROUTEUPD.
Routing redirect events are broadcast as a pair of rtmsgs, RTM_DELROUTE
and RTM_NEWROUTE.
---
include/linux/rtnetlink.h | 4 ++
net/core/Makefile | 2 +
net/core/neighbour.c | 37 ++++++++++++++++---
net/ipv4/fib_semantics.c | 9 +++++
net/ipv4/route.c | 86 ++++++++++++++++++++++++++++++++++++++++++--
net/ipv6/route.c | 87 +++++++++++++++++++++++++++++++++++++++++++++
6 files changed, 213 insertions(+), 12 deletions(-)
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index facd9ee..340ca4f 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -35,6 +35,8 @@ #define RTM_NEWROUTE RTM_NEWROUTE
#define RTM_DELROUTE RTM_DELROUTE
RTM_GETROUTE,
#define RTM_GETROUTE RTM_GETROUTE
+ RTM_ROUTEUPD,
+#define RTM_ROUTEUPD RTM_ROUTEUPD
RTM_NEWNEIGH = 28,
#define RTM_NEWNEIGH RTM_NEWNEIGH
@@ -42,6 +44,8 @@ #define RTM_NEWNEIGH RTM_NEWNEIGH
#define RTM_DELNEIGH RTM_DELNEIGH
RTM_GETNEIGH,
#define RTM_GETNEIGH RTM_GETNEIGH
+ RTM_NEIGHUPD,
+#define RTM_NEIGHUPD RTM_NEIGHUPD
RTM_NEWRULE = 32,
#define RTM_NEWRULE RTM_NEWRULE
diff --git a/net/core/Makefile b/net/core/Makefile
index e9bd246..2645ba4 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -7,7 +7,7 @@ obj-y := sock.o request_sock.o skbuff.o
obj-$(CONFIG_SYSCTL) += sysctl_net_core.o
-obj-y += dev.o ethtool.o dev_mcast.o dst.o \
+obj-y += dev.o ethtool.o dev_mcast.o dst.o netevent.o \
neighbour.o rtnetlink.o utils.o link_watch.o filter.o
obj-$(CONFIG_XFRM) += flow.o
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 50a8c73..bf70981 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -30,9 +30,11 @@ #include <linux/times.h>
#include <net/neighbour.h>
#include <net/dst.h>
#include <net/sock.h>
+#include <net/netevent.h>
#include <linux/rtnetlink.h>
#include <linux/random.h>
#include <linux/string.h>
+#include <linux/notifier.h>
#define NEIGH_DEBUG 1
@@ -59,6 +61,7 @@ static void neigh_app_notify(struct neig
#endif
static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev);
void neigh_changeaddr(struct neigh_table *tbl, struct net_device *dev);
+static void rtm_neigh_change(struct neighbour *n);
static struct neigh_table *neigh_tables;
#ifdef CONFIG_PROC_FS
@@ -755,6 +758,7 @@ #endif
neigh->nud_state = NUD_STALE;
neigh->updated = jiffies;
neigh_suspect(neigh);
+ notify = 1;
}
} else if (state & NUD_DELAY) {
if (time_before_eq(now,
@@ -763,6 +767,7 @@ #endif
neigh->nud_state = NUD_REACHABLE;
neigh->updated = jiffies;
neigh_connect(neigh);
+ notify = 1;
next = neigh->confirmed + neigh->parms->reachable_time;
} else {
NEIGH_PRINTK2("neigh %p is probed.\n", neigh);
@@ -820,6 +825,8 @@ #endif
out:
write_unlock(&neigh->lock);
}
+ if (notify)
+ rtm_neigh_change(neigh);
#ifdef CONFIG_ARPD
if (notify && neigh->parms->app_probes)
@@ -927,9 +934,7 @@ int neigh_update(struct neighbour *neigh
{
u8 old;
int err;
-#ifdef CONFIG_ARPD
int notify = 0;
-#endif
struct net_device *dev;
int update_isrouter = 0;
@@ -949,9 +954,7 @@ #endif
neigh_suspect(neigh);
neigh->nud_state = new;
err = 0;
-#ifdef CONFIG_ARPD
notify = old & NUD_VALID;
-#endif
goto out;
}
@@ -1023,9 +1026,7 @@ #endif
if (!(new & NUD_CONNECTED))
neigh->confirmed = jiffies -
(neigh->parms->base_reachable_time << 1);
-#ifdef CONFIG_ARPD
notify = 1;
-#endif
}
if (new == old)
goto out;
@@ -1056,7 +1057,11 @@ out:
(neigh->flags | NTF_ROUTER) :
(neigh->flags & ~NTF_ROUTER);
}
+
write_unlock_bh(&neigh->lock);
+
+ if (notify)
+ rtm_neigh_change(neigh);
#ifdef CONFIG_ARPD
if (notify && neigh->parms->app_probes)
neigh_app_notify(neigh);
@@ -2370,9 +2375,27 @@ static void neigh_app_notify(struct neig
NETLINK_CB(skb).dst_group = RTNLGRP_NEIGH;
netlink_broadcast(rtnl, skb, 0, RTNLGRP_NEIGH, GFP_ATOMIC);
}
-
#endif /* CONFIG_ARPD */
+static void rtm_neigh_change(struct neighbour *n)
+{
+ struct nlmsghdr *nlh;
+ int size = NLMSG_SPACE(sizeof(struct ndmsg) + 256);
+ struct sk_buff *skb = alloc_skb(size, GFP_ATOMIC);
+
+ call_netevent_notifiers(NETEVENT_NEIGH_UPDATE, n);
+ if (!skb)
+ return;
+
+ if (neigh_fill_info(skb, n, 0, 0, RTM_NEIGHUPD, 0) < 0) {
+ kfree_skb(skb);
+ return;
+ }
+ nlh = (struct nlmsghdr *)skb->data;
+ NETLINK_CB(skb).dst_group = RTNLGRP_NEIGH;
+ netlink_broadcast(rtnl, skb, 0, RTNLGRP_NEIGH, GFP_ATOMIC);
+}
+
#ifdef CONFIG_SYSCTL
static struct neigh_sysctl_table {
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 0f4145b..197c365 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -45,6 +45,7 @@ #include <net/tcp.h>
#include <net/sock.h>
#include <net/ip_fib.h>
#include <net/ip_mp_alg.h>
+#include <net/netevent.h>
#include "fib_lookup.h"
@@ -280,6 +281,14 @@ void rtmsg_fib(int event, u32 key, struc
struct sk_buff *skb;
u32 pid = req ? req->pid : n->nlmsg_pid;
int size = NLMSG_SPACE(sizeof(struct rtmsg)+256);
+ struct netevent_route_info nri;
+ int netevent;
+
+ nri.family = AF_INET;
+ nri.data = &fa->fa_info;
+ netevent = event == RTM_NEWROUTE ? NETEVENT_ROUTE_ADD
+ : NETEVENT_ROUTE_DEL;
+ call_netevent_notifiers(netevent, &nri);
skb = alloc_skb(size, GFP_KERNEL);
if (!skb)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 60b11ae..cef7c6d 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -105,6 +105,7 @@ #include <net/tcp.h>
#include <net/icmp.h>
#include <net/xfrm.h>
#include <net/ip_mp_alg.h>
+#include <net/netevent.h>
#ifdef CONFIG_SYSCTL
#include <linux/sysctl.h>
#endif
@@ -152,6 +153,8 @@ static struct dst_entry *ipv4_negative_a
static void ipv4_link_failure(struct sk_buff *skb);
static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu);
static int rt_garbage_collect(void);
+static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
+ int nowait, unsigned int flags, unsigned int prot);
static struct dst_ops ipv4_dst_ops = {
@@ -1112,6 +1115,52 @@ static void rt_del(unsigned hash, struct
spin_unlock_bh(rt_hash_lock_addr(hash));
}
+static void rtm_redirect(struct rtable *old, struct rtable *new)
+{
+ struct netevent_redirect netevent;
+ struct sk_buff *skb;
+ int err;
+
+ netevent.old = &old->u.dst;
+ netevent.new = &new->u.dst;
+
+ /* notify netevent subscribers */
+ call_netevent_notifiers(NETEVENT_REDIRECT, &netevent);
+
+ /* Post NETLINK messages: RTM_DELROUTE for old route,
+ RTM_NEWROUTE for new route */
+ skb = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
+ if (!skb)
+ return;
+ skb->mac.raw = skb->nh.raw = skb->data;
+ skb->dst = &old->u.dst;
+ NETLINK_CB(skb).dst_pid = 0;
+
+ err = rt_fill_info(skb, 0, 0, RTM_DELROUTE, 1, 0, RTPROT_UNSPEC);
+ if (err <= 0)
+ goto out_free;
+
+ netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV4_ROUTE, GFP_ATOMIC);
+
+ skb = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
+ if (!skb)
+ return;
+ skb->mac.raw = skb->nh.raw = skb->data;
+ skb->dst = &new->u.dst;
+ NETLINK_CB(skb).dst_pid = 0;
+
+ err = rt_fill_info(skb, 0, 0, RTM_NEWROUTE, 1, 0, RTPROT_REDIRECT);
+ if (err <= 0)
+ goto out_free;
+
+ netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV4_ROUTE, GFP_ATOMIC);
+ return;
+
+out_free:
+ kfree_skb(skb);
+ return;
+}
+
void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
u32 saddr, struct net_device *dev)
{
@@ -1211,6 +1260,8 @@ void ip_rt_redirect(u32 old_gw, u32 dadd
rt_drop(rt);
goto do_next;
}
+
+ rtm_redirect(rth, rt);
rt_del(hash, rth);
if (!rt_intern_hash(hash, rt, &rt))
@@ -1437,6 +1488,32 @@ unsigned short ip_rt_frag_needed(struct
return est_mtu ? : new_mtu;
}
+static void rtm_pmtu_update(struct rtable *rt)
+{
+ struct sk_buff *skb;
+ int err;
+
+ call_netevent_notifiers(NETEVENT_PMTU_UPDATE, &rt->u.dst);
+
+ skb = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
+ if (!skb)
+ return;
+ skb->mac.raw = skb->nh.raw = skb->data;
+ skb->dst = &rt->u.dst;
+ NETLINK_CB(skb).dst_pid = 0;
+
+ err = rt_fill_info(skb, 0, 0, RTM_ROUTEUPD, 1, 0, RTPROT_UNSPEC);
+ if (err <= 0)
+ goto out_free;
+
+ netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV4_ROUTE, GFP_ATOMIC);
+ return;
+
+out_free:
+ kfree_skb(skb);
+ return;
+}
+
static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
{
if (dst->metrics[RTAX_MTU-1] > mtu && mtu >= 68 &&
@@ -1447,6 +1524,7 @@ static void ip_rt_update_pmtu(struct dst
}
dst->metrics[RTAX_MTU-1] = mtu;
dst_set_expires(dst, ip_rt_mtu_expires);
+ rtm_pmtu_update((struct rtable *)dst);
}
}
@@ -2622,7 +2700,7 @@ int ip_route_output_key(struct rtable **
}
static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
- int nowait, unsigned int flags)
+ int nowait, unsigned int flags, unsigned int prot)
{
struct rtable *rt = (struct rtable*)skb->dst;
struct rtmsg *r;
@@ -2641,7 +2719,7 @@ #endif
r->rtm_table = RT_TABLE_MAIN;
r->rtm_type = rt->rt_type;
r->rtm_scope = RT_SCOPE_UNIVERSE;
- r->rtm_protocol = RTPROT_UNSPEC;
+ r->rtm_protocol = prot;
r->rtm_flags = (rt->rt_flags & ~0xFFFF) | RTM_F_CLONED;
if (rt->rt_flags & RTCF_NOTIFY)
r->rtm_flags |= RTM_F_NOTIFY;
@@ -2787,7 +2865,7 @@ int inet_rtm_getroute(struct sk_buff *in
NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid;
err = rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq,
- RTM_NEWROUTE, 0, 0);
+ RTM_NEWROUTE, 0, 0, RTPROT_UNSPEC);
if (!err)
goto out_free;
if (err < 0) {
@@ -2825,7 +2903,7 @@ int ip_rt_dump(struct sk_buff *skb, str
skb->dst = dst_clone(&rt->u.dst);
if (rt_fill_info(skb, NETLINK_CB(cb->skb).pid,
cb->nlh->nlmsg_seq, RTM_NEWROUTE,
- 1, NLM_F_MULTI) <= 0) {
+ 1, NLM_F_MULTI, RTPROT_UNSPEC) <= 0) {
dst_release(xchg(&skb->dst, NULL));
rcu_read_unlock_bh();
goto done;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 8a77793..95f68bc 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -54,6 +54,7 @@ #include <net/tcp.h>
#include <linux/rtnetlink.h>
#include <net/dst.h>
#include <net/xfrm.h>
+#include <net/netevent.h>
#include <asm/uaccess.h>
@@ -97,6 +98,10 @@ static int ip6_pkt_discard(struct sk_bu
static int ip6_pkt_discard_out(struct sk_buff *skb);
static void ip6_link_failure(struct sk_buff *skb);
static void ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu);
+static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
+ struct in6_addr *dst, struct in6_addr *src,
+ int iif, int type, u32 pid, u32 seq,
+ int prefix, unsigned int flags);
#ifdef CONFIG_IPV6_ROUTE_INFO
static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixlen,
@@ -732,6 +737,32 @@ static void ip6_link_failure(struct sk_b
}
}
+static void rtm_pmtu_update(struct rt6_info *rt)
+{
+ struct sk_buff *skb;
+ int err;
+
+ call_netevent_notifiers(NETEVENT_PMTU_UPDATE, &rt->u.dst);
+
+ skb = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
+ if (!skb)
+ return;
+ skb->mac.raw = skb->nh.raw = skb->data;
+ skb->dst = &rt->u.dst;
+ NETLINK_CB(skb).dst_pid = 0;
+
+ err = rt6_fill_node(skb, rt, NULL, NULL, 0, RTM_ROUTEUPD, 0, 0, 0, 0);
+ if (err <= 0)
+ goto out_free;
+
+ netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV6_ROUTE, GFP_ATOMIC);
+ return;
+
+out_free:
+ kfree_skb(skb);
+ return;
+}
+
static void ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
{
struct rt6_info *rt6 = (struct rt6_info*)dst;
@@ -743,6 +774,7 @@ static void ip6_rt_update_pmtu(struct ds
dst->metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
}
dst->metrics[RTAX_MTU-1] = mtu;
+ rtm_pmtu_update(rt6);
}
}
@@ -908,6 +940,7 @@ int ip6_route_add(struct in6_rtmsg *rtms
struct net_device *dev = NULL;
struct inet6_dev *idev = NULL;
int addr_type;
+ struct netevent_route_info nri;
rta = (struct rtattr **) _rtattr;
@@ -1086,6 +1119,9 @@ install_route:
rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
rt->u.dst.dev = dev;
rt->rt6i_idev = idev;
+ nri.family = AF_INET6;
+ nri.data = rt;
+ call_netevent_notifiers(NETEVENT_ROUTE_ADD, &nri);
return ip6_ins_rt(rt, nlh, _rtattr, req);
out:
@@ -1117,6 +1153,7 @@ static int ip6_route_del(struct in6_rtms
struct fib6_node *fn;
struct rt6_info *rt;
int err = -ESRCH;
+ struct netevent_route_info nri;
read_lock_bh(&rt6_lock);
@@ -1138,6 +1175,10 @@ static int ip6_route_del(struct in6_rtms
continue;
dst_hold(&rt->u.dst);
read_unlock_bh(&rt6_lock);
+
+ nri.family = AF_INET6;
+ nri.data = rt;
+ call_netevent_notifiers(NETEVENT_ROUTE_DEL, &nri);
return ip6_del_rt(rt, nlh, _rtattr, req);
}
@@ -1147,6 +1188,50 @@ static int ip6_route_del(struct in6_rtms
return err;
}
+static void rtm_redirect(struct rt6_info *old, struct rt6_info *new)
+{
+ struct netevent_redirect netevent;
+ struct sk_buff *skb;
+ int err;
+
+ netevent.old = &old->u.dst;
+ netevent.new = &new->u.dst;
+ call_netevent_notifiers(NETEVENT_REDIRECT, &netevent);
+
+ /* Post NETLINK messages: RTM_DELROUTE for old route,
+ RTM_NEWROUTE for new route */
+ skb = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
+ if (!skb)
+ return;
+ skb->mac.raw = skb->nh.raw = skb->data;
+ NETLINK_CB(skb).dst_pid = 0;
+ NETLINK_CB(skb).dst_group = RTNLGRP_IPV6_ROUTE;
+
+ err = rt6_fill_node(skb, old, NULL, NULL, 0, RTM_DELROUTE, 0, 0, 0, 0);
+ if (err <= 0)
+ goto out_free;
+
+ netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV6_ROUTE, GFP_ATOMIC);
+
+ skb = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
+ if (!skb)
+ return;
+ skb->mac.raw = skb->nh.raw = skb->data;
+ NETLINK_CB(skb).dst_pid = 0;
+ NETLINK_CB(skb).dst_group = RTNLGRP_IPV6_ROUTE;
+
+ err = rt6_fill_node(skb, new, NULL, NULL, 0, RTM_NEWROUTE, 0, 0, 0, 0);
+ if (err <= 0)
+ goto out_free;
+
+ netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV6_ROUTE, GFP_ATOMIC);
+ return;
+
+out_free:
+ kfree_skb(skb);
+ return;
+}
+
/*
* Handle redirects
*/
@@ -1253,6 +1338,8 @@ restart:
if (ip6_ins_rt(nrt, NULL, NULL, NULL))
goto out;
+ rtm_redirect(rt, nrt);
+
if (rt->rt6i_flags&RTF_CACHE) {
ip6_del_rt(rt, NULL, NULL, NULL);
return;
^ permalink raw reply related
* Re: Network namespaces a path to mergable code.
From: Sam Vilain @ 2006-06-27 22:20 UTC (permalink / raw)
To: Andrey Savochkin
Cc: Eric W. Biederman, dlezcano, linux-kernel, netdev, serue,
haveblue, clg, Andrew Morton, dev, herbert, devel, viro,
Alexey Kuznetsov, Mark Huang
In-Reply-To: <20060627215859.A20679@castle.nmd.msu.ru>
Andrey Savochkin wrote:
> On Tue, Jun 27, 2006 at 11:20:40AM -0600, Eric W. Biederman wrote:
>
>> Thinking about this I am going to suggest a slightly different direction
>> for get a patchset we can merge.
>>
>> First we concentrate on the fundamentals.
>> - How we mark a device as belonging to a specific network namespace.
>> - How we mark a socket as belonging to a specific network namespace.
>>
>
> I agree with the direction of your thoughts.
> I was trying to do a similar thing, define clear steps in network
> namespace merging.
>
> My first patchset covers devices but not sockets.
> The only difference from what you're suggesting is ipv4 routing.
> For me, it is not less important than devices and sockets. May be even
> more important, since routing exposes design deficiencies less obvious at
> socket level.
>
It sounds then like it would be a good start to have general socket
namespaces, if it would merge more easily - perhaps then network device
namespaces would fall into place more easily.
AIUI socket namespaces are also necessary for situations where you want
containers to share IP addresses. AIUI, PlanetLab do something like this
with a module atop of VServer already (but read
http://openvz.org/pipermail/devel/2006-June/000666.html for a proper
explanation from Mark Huang)
>> As part of the fundamentals we add a patch to the generic socket code
>> that by default will disable it for protocol families that do not indicate
>> support for handling network namespaces, on a non-default network namespace.
>>
>
> Fine
>
> Can you summarize you objections against my way of handling devices, please?
>
There were many objections, the major one being the patch was too large for certainty of adequate review.
Quoting what I perceived as a summary from Eric:
> When I went through this, my patchset just added an explicit
> continue if the devices was not in the appropriate namespace.
> I actually prefer the multiple list implementation but at
> the same time I think it is harder to get a clean implementation
> out of it.
You offered to re-do the patch without separate lists - I suggest that
this go ahead. No-one should really care; splitting it out into separate
lists can then be considered a performance optimization for later.
> And what was the typo you referred to in your letter to Kirill Korotaev?
>
I think this is the comment he refers to:
> These hunks should use for_each_netdev(ifp);
Both quotes are from http://lkml.org/lkml/2006/6/26/147
Though, in Kirill's defense, it seems a bit strange to expect him to
raise a fault that was just raised by Eric, in a reply to the message
where he raised it.
Sam.
^ permalink raw reply
* Re: [NET]: Added GSO header verification
From: Herbert Xu @ 2006-06-27 22:31 UTC (permalink / raw)
To: Michael Chan; +Cc: David S. Miller, netdev
In-Reply-To: <1151441195.3543.3.camel@rh4>
On Tue, Jun 27, 2006 at 01:46:35PM -0700, Michael Chan wrote:
> On Tue, 2006-06-27 at 22:07 +1000, Herbert Xu wrote:
>
> > [NET]: Added GSO header verification
> >
> > @@ -2166,10 +2166,14 @@ struct sk_buff *tcp_tso_segment(struct s
> > if (!pskb_may_pull(skb, thlen))
> > goto out;
> >
> > + segs = NULL;
> > + if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST))
> > + goto out;
> > +
>
> This logic doesn't look right to me. Perhaps it's backwards and should
> be:
>
> if (!skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST))
Oops, you're absolutely right. Here is the fix.
[NET]: Fix logical error in skb_gso_ok
The test in skb_gso_ok is backwards. Noticed by Michael Chan
<mchan@broadcom.com>.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 84b0f0d..efd1e2a 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -994,12 +994,12 @@ static inline int skb_gso_ok(struct sk_b
{
int feature = skb_shinfo(skb)->gso_size ?
skb_shinfo(skb)->gso_type << NETIF_F_GSO_SHIFT : 0;
- return (features & feature) != feature;
+ return (features & feature) == feature;
}
static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
{
- return skb_gso_ok(skb, dev->features);
+ return !skb_gso_ok(skb, dev->features);
}
#endif /* __KERNEL__ */
^ permalink raw reply related
* Re: [PATCH REPOST 0/2][RFC] Network Event Notifier Mechanism
From: Herbert Xu @ 2006-06-27 22:32 UTC (permalink / raw)
To: Steve Wise; +Cc: davem, netdev
In-Reply-To: <1151418717.3207.21.camel@stevo-desktop>
On Tue, Jun 27, 2006 at 09:31:57AM -0500, Steve Wise wrote:
>
> > I'd like to know more about what the RDMA device is going to do with this
> > information. I thought RDMA was for receiving packets? Most of the info
> > here pertains to transmission.
>
> RDMA Ethernet devices adhere to a set of protocols defined by the IETF.
> See the RDDP WG (http://www.ietf.org/html.charters/rddp-charter.html)
> for the Internet Drafts that define the protocols.
Would it be possible for you to give us a quick summary of the relevant
points?
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: [PATCH 00/21] e1000: driver update to 7.1.9-k2
From: Auke Kok @ 2006-06-27 22:48 UTC (permalink / raw)
To: Garzik, Jeff
Cc: Kok, Auke, netdev, Brandeburg, Jesse, Kok, Auke, Ronciak, John
In-Reply-To: <20060622051815.25497.89192.stgit@gitlost.site>
Jeff,
after comments I've made some adjustments. I'll list them below against the
old summary. The changes are available from our git-server:
Please pull from:
git://lost.foo-projects.org/~ahkok/git/netdev-2.6 upstream
These patches are against
netdev-2.6#upstream 612eff0e3715a6faff5ba1b74873b99e036c59fe
(Brian Haley <brian.haley@hp.com> / [PATCH] s2io: netpoll support)
> Summary of patches:
>
> [01]: fix loopback ethtool test
> [02]: rework driver hardware reset locking
> [03]: Make PHY powerup/down a function
> [04]: fix CONFIG_PM blocks
> [05]: small performance tweak by removing double code
> [06]: add smart power down code
> [07]: change printk into DPRINTK
> [08]: recycle skb
> [09]: rework module param code with uninitialized values
> [10]: force register write flushes to circumvent broken platforms
Unmodified. See comments here:
http://marc.theaimsgroup.com/?l=linux-netdev&m=115142459725123&w=2 [1]
> [11]: disable CRC stripping workaround
Removed all references to SECRC (crc stripping) instead of leaving it commented.
> [12]: fix adapter led blinking inconsistency
> [13]: add E1000_BIG_ENDIAN symbol
Dropped this patch entirely
> [14]: M88 PHY workaround
> [15]: check return value of _get_speed_and_duplex
> [16]: disable ERT
> [17]: add ich8lan core functions
> [18]: integrate ich8 support into driver
> [19]: allow user to disable ich8 lock loss workaround
> [20]: add ich8lan device ID's
> [21]: increase version to 7.1.9-k2
[1] I can drop #11 in case someone throws a fit ;) - as everyone I'd really
like to see patches 17->20 queued for 2.6.18 for obvious reasons - this is the
most important section of these patches!
Cheers,
Auke
---
drivers/net/e1000/e1000.h | 10
drivers/net/e1000/e1000_ethtool.c | 143 +--
drivers/net/e1000/e1000_hw.c | 1770 +++++++++++++++++++++++++++++++++++---
drivers/net/e1000/e1000_hw.h | 398 ++++++++
drivers/net/e1000/e1000_main.c | 384 +++++---
drivers/net/e1000/e1000_osdep.h | 13
drivers/net/e1000/e1000_param.c | 213 ++--
7 files changed, 2530 insertions(+), 401 deletions(-)
^ permalink raw reply
* Re: [patch 2/6] [Network namespace] Network device sharing by view
From: Herbert Poetzl @ 2006-06-27 22:48 UTC (permalink / raw)
To: Ben Greear
Cc: Eric W. Biederman, Daniel Lezcano, Andrey Savochkin, linux-kernel,
netdev, serue, haveblue, clg, Andrew Morton, dev, devel, sam,
viro, Alexey Kuznetsov
In-Reply-To: <44A157CA.70204@candelatech.com>
On Tue, Jun 27, 2006 at 09:07:38AM -0700, Ben Greear wrote:
> Ben Greear wrote:
> >Herbert Poetzl wrote:
> >
> >>On Mon, Jun 26, 2006 at 03:13:17PM -0700, Ben Greear wrote:
> >
> >>yes, that sounds good to me, any numbers how that
> >>affects networking in general (performance wise and
> >>memory wise, i.e. caches and hashes) ...
> >
> >I'll run some tests later today. Based on my previous tests,
> >I don't remember any significant overhead.
>
> Here's a quick benchmark using my redirect devices (RDD). Each RDD
> comes in a pair...when you tx on one, the pkt is rx'd on the peer.
> The idea is that it is exactly like two physical ethernet interfaces
> connected by a cross-over cable.
>
> My test system is a 64-bit dual-core Intel system, 3.013 Ghz processor
> with 1GB RAM. Fairly standard stuff..it's one of the Shuttle XPC
> systems. Kernel is 2.6.16.16 (64-bit).
>
>
> Test setup is: rdd1 -- rdd2 [bridge] rdd3 -- rdd4
>
> I am using my proprietary module for the bridge logic...and the
> default bridge should be at least this fast. I am injecting 1514 byte
> packets on rdd1 and rdd4 with pktgen (bi-directional flow). My pktgen
> is also receiving the pkts and gathering stats.
>
> This setup sustains 1.7Gbps of generated and received traffic between
> rdd1 and rdd4.
>
> Running only the [bridge] between two 10/100/1000 ports on an Intel
> PCI-E NIC will sustain about 870Mbps (bi-directional) on this system,
> so the virtual devices are quite efficient, as suspected.
>
> I have not yet had time to benchmark the mac-vlans...hopefully later
> today.
hmm, maybe you could also benchmark loopback connections
(and their throughput) on your system?
my (not so fancy) PIII, 32bit, 2.6.17.1 seems to do
roughly 2Gbs on the loopback device (tested with dd
and netcat)
best,
Herbert
> Thanks,
> Ben
>
> --
> Ben Greear <greearb@candelatech.com>
> Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [patch 2/6] [Network namespace] Network device sharing by view
From: Herbert Poetzl @ 2006-06-27 22:52 UTC (permalink / raw)
To: Ben Greear
Cc: Eric W. Biederman, Daniel Lezcano, Andrey Savochkin, linux-kernel,
netdev, serue, haveblue, clg, Andrew Morton, dev, devel, sam,
viro, Alexey Kuznetsov
In-Reply-To: <44A1689B.7060809@candelatech.com>
On Tue, Jun 27, 2006 at 10:19:23AM -0700, Ben Greear wrote:
> Eric W. Biederman wrote:
> >Herbert Poetzl <herbert@13thfloor.at> writes:
> >
> >
> >>On Tue, Jun 27, 2006 at 05:52:52AM -0600, Eric W. Biederman wrote:
> >>
> >>>Inside the containers I want all network devices named eth0!
> >>
> >>huh? even if there are two of them? also tun?
> >>
> >>I think you meant, you want to be able to have eth0 in
> >>_more_ than one guest where eth0 in a guest can also
> >>be/use/relate to eth1 on the host, right?
> >
> >
> >Right I want to have an eth0 in each guest where eth0 is
> >it's own network device and need have no relationship to
> >eth0 on the host.
>
> How does that help anything? Do you envision programs
> that make special decisions on whether the interface is
> called eth0 v/s eth151?
well, those poor folks who do not have ethernet
devices for networking :)
seriously, what I think Eric meant was that it
might be nice (especially for migration purposes)
to keep the device namespace completely virtualized
and not just isolated ...
I'm fine with that, as long as it does not add
overhead or complicate handling, and as far as I
can tell, it should not do that ...
best,
Herbert
> Ben
>
>
> --
> Ben Greear <greearb@candelatech.com>
> Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* [RFC 1/8] NetLabel: documentation
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-documentation --]
[-- Type: text/plain, Size: 40056 bytes --]
Documentation for the NetLabel system, this includes a basic overview of how
NetLabel works and how LSM developers can integrate it into their favorite
LSM. Also, due to the difficulty of finding expired IETF drafts, I am
including the IETF CIPSO draft that is the basis of the NetLabel CIPSO
implementation.
---
CREDITS | 7
Documentation/00-INDEX | 2
Documentation/netlabel/00-INDEX | 10
Documentation/netlabel/cipso_ipv4.txt | 48
Documentation/netlabel/draft-ietf-cipso-ipsecurity-01.txt | 791 ++++++++++++++
Documentation/netlabel/introduction.txt | 53
Documentation/netlabel/lsm_interface.txt | 47
7 files changed, 958 insertions(+)
Index: linux-2.6.17.i686-quilt/CREDITS
===================================================================
--- linux-2.6.17.i686-quilt.orig/CREDITS
+++ linux-2.6.17.i686-quilt/CREDITS
@@ -2383,6 +2383,13 @@ N: Thomas Molina
E: tmolina@cablespeed.com
D: bug fixes, documentation, minor hackery
+N: Paul Moore
+E: paul.moore@hp.com
+D: NetLabel author
+S: Hewlett-Packard
+S: 110 Spit Brook Road
+S: Nashua, NH 03062
+
N: James Morris
E: jmorris@namei.org
W: http://namei.org/
Index: linux-2.6.17.i686-quilt/Documentation/00-INDEX
===================================================================
--- linux-2.6.17.i686-quilt.orig/Documentation/00-INDEX
+++ linux-2.6.17.i686-quilt/Documentation/00-INDEX
@@ -184,6 +184,8 @@ mtrr.txt
- how to use PPro Memory Type Range Registers to increase performance.
nbd.txt
- info on a TCP implementation of a network block device.
+netlabel/
+ - directory with information on the NetLabel subsystem.
networking/
- directory with info on various aspects of networking with Linux.
nfsroot.txt
Index: linux-2.6.17.i686-quilt/Documentation/netlabel/00-INDEX
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/Documentation/netlabel/00-INDEX
@@ -0,0 +1,10 @@
+00-INDEX
+ - this file.
+cipso_ipv4.txt
+ - documentation on the IPv4 CIPSO protocol engine.
+draft-ietf-cipso-ipsecurity-01.txt
+ - IETF draft of the CIPSO protocol, dated 16 July 1992.
+introduction.txt
+ - NetLabel introduction, READ THIS FIRST.
+lsm_interface.txt
+ - documentation on the NetLabel kernel security module API.
Index: linux-2.6.17.i686-quilt/Documentation/netlabel/cipso_ipv4.txt
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/Documentation/netlabel/cipso_ipv4.txt
@@ -0,0 +1,48 @@
+NetLabel CIPSO/IPv4 Protocol Engine
+==============================================================================
+Paul Moore, paul.moore@hp.com
+
+May 17, 2006
+
+ * Overview
+
+The NetLabel CIPSO/IPv4 protocol engine is based on the IETF Commercial IP
+Security Option (CIPSO) draft from July 16, 1992. A copy of this draft can be
+found in this directory, consult '00-INDEX' for the filename. While the IETF
+draft never made it to an RFC standard it has become a de-facto standard for
+labeled networking and is used in many trusted operating systems.
+
+ * Outbound Packet Processing
+
+The CIPSO/IPv4 protocol engine applies the CIPSO IP option to packets by
+adding the CIPSO label to the socket. This causes all packets leaving the
+system through the socket to have the CIPSO IP option applied. The socket's
+CIPSO label can be changed at any point in time, however, it is recommended
+that it is set upon the socket's creation. The LSM can set the socket's CIPSO
+label by using the NetLabel security module API; if the NetLabel "domain" is
+configured to use CIPSO for packet labeling then a CIPSO IP option will be
+generated and attached to the socket.
+
+ * Inbound Packet Processing
+
+The CIPSO/IPv4 protocol engine validates every CIPSO IP option it finds at the
+IP layer without any special handling required by the LSM. However, in order
+to decode and translate the CIPSO label on the packet the LSM must use the
+NetLabel security module API to extract the security attributes of the packet.
+This is typically done at the socket layer using the 'socket_sock_rcv_skb()'
+LSM hook.
+
+ * Label Translation
+
+The CIPSO/IPv4 protocol engine contains a mechanism to translate CIPSO security
+attributes such as sensitivity level and category to values which are
+appropriate for the host. These mappings are defined as part of a CIPSO
+Domain Of Interpretation (DOI) definition and are configured through the
+NetLabel user space communication layer. Each DOI definition can have a
+different security attribute mapping table.
+
+ * Label Translation Cache
+
+The NetLabel system provides a framework for caching security attribute
+mappings from the network labels to the corresponding LSM identifiers. The
+CIPSO/IPv4 protocol engine supports this caching mechanism.
Index: linux-2.6.17.i686-quilt/Documentation/netlabel/draft-ietf-cipso-ipsecurity-01.txt
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/Documentation/netlabel/draft-ietf-cipso-ipsecurity-01.txt
@@ -0,0 +1,791 @@
+IETF CIPSO Working Group
+16 July, 1992
+
+
+
+ COMMERCIAL IP SECURITY OPTION (CIPSO 2.2)
+
+
+
+1. Status
+
+This Internet Draft provides the high level specification for a Commercial
+IP Security Option (CIPSO). This draft reflects the version as approved by
+the CIPSO IETF Working Group. Distribution of this memo is unlimited.
+
+This document is an Internet Draft. Internet Drafts are working documents
+of the Internet Engineering Task Force (IETF), its Areas, and its Working
+Groups. Note that other groups may also distribute working documents as
+Internet Drafts.
+
+Internet Drafts are draft documents valid for a maximum of six months.
+Internet Drafts may be updated, replaced, or obsoleted by other documents
+at any time. It is not appropriate to use Internet Drafts as reference
+material or to cite them other than as a "working draft" or "work in
+progress."
+
+Please check the I-D abstract listing contained in each Internet Draft
+directory to learn the current status of this or any other Internet Draft.
+
+
+
+
+2. Background
+
+Currently the Internet Protocol includes two security options. One of
+these options is the DoD Basic Security Option (BSO) (Type 130) which allows
+IP datagrams to be labeled with security classifications. This option
+provides sixteen security classifications and a variable number of handling
+restrictions. To handle additional security information, such as security
+categories or compartments, another security option (Type 133) exists and
+is referred to as the DoD Extended Security Option (ESO). The values for
+the fixed fields within these two options are administered by the Defense
+Information Systems Agency (DISA).
+
+Computer vendors are now building commercial operating systems with
+mandatory access controls and multi-level security. These systems are
+no longer built specifically for a particular group in the defense or
+intelligence communities. They are generally available commercial systems
+for use in a variety of government and civil sector environments.
+
+The small number of ESO format codes can not support all the possible
+applications of a commercial security option. The BSO and ESO were
+designed to only support the United States DoD. CIPSO has been designed
+to support multiple security policies. This Internet Draft provides the
+format and procedures required to support a Mandatory Access Control
+security policy. Support for additional security policies shall be
+defined in future RFCs.
+
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 1]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+
+3. CIPSO Format
+
+Option type: 134 (Class 0, Number 6, Copy on Fragmentation)
+Option length: Variable
+
+This option permits security related information to be passed between
+systems within a single Domain of Interpretation (DOI). A DOI is a
+collection of systems which agree on the meaning of particular values
+in the security option. An authority that has been assigned a DOI
+identifier will define a mapping between appropriate CIPSO field values
+and their human readable equivalent. This authority will distribute that
+mapping to hosts within the authority's domain. These mappings may be
+sensitive, therefore a DOI authority is not required to make these
+mappings available to anyone other than the systems that are included in
+the DOI.
+
+This option MUST be copied on fragmentation. This option appears at most
+once in a datagram. All multi-octet fields in the option are defined to be
+transmitted in network byte order. The format of this option is as follows:
+
++----------+----------+------//------+-----------//---------+
+| 10000110 | LLLLLLLL | DDDDDDDDDDDD | TTTTTTTTTTTTTTTTTTTT |
++----------+----------+------//------+-----------//---------+
+
+ TYPE=134 OPTION DOMAIN OF TAGS
+ LENGTH INTERPRETATION
+
+
+ Figure 1. CIPSO Format
+
+
+3.1 Type
+
+This field is 1 octet in length. Its value is 134.
+
+
+3.2 Length
+
+This field is 1 octet in length. It is the total length of the option
+including the type and length fields. With the current IP header length
+restriction of 40 octets the value of this field MUST not exceed 40.
+
+
+3.3 Domain of Interpretation Identifier
+
+This field is an unsigned 32 bit integer. The value 0 is reserved and MUST
+not appear as the DOI identifier in any CIPSO option. Implementations
+should assume that the DOI identifier field is not aligned on any particular
+byte boundary.
+
+To conserve space in the protocol, security levels and categories are
+represented by numbers rather than their ASCII equivalent. This requires
+a mapping table within CIPSO hosts to map these numbers to their
+corresponding ASCII representations. Non-related groups of systems may
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 2]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+have their own unique mappings. For example, one group of systems may
+use the number 5 to represent Unclassified while another group may use the
+number 1 to represent that same security level. The DOI identifier is used
+to identify which mapping was used for the values within the option.
+
+
+3.4 Tag Types
+
+A common format for passing security related information is necessary
+for interoperability. CIPSO uses sets of "tags" to contain the security
+information relevant to the data in the IP packet. Each tag begins with
+a tag type identifier followed by the length of the tag and ends with the
+actual security information to be passed. All multi-octet fields in a tag
+are defined to be transmitted in network byte order. Like the DOI
+identifier field in the CIPSO header, implementations should assume that
+all tags, as well as fields within a tag, are not aligned on any particular
+octet boundary. The tag types defined in this document contain alignment
+bytes to assist alignment of some information, however alignment can not
+be guaranteed if CIPSO is not the first IP option.
+
+CIPSO tag types 0 through 127 are reserved for defining standard tag
+formats. Their definitions will be published in RFCs. Tag types whose
+identifiers are greater than 127 are defined by the DOI authority and may
+only be meaningful in certain Domains of Interpretation. For these tag
+types, implementations will require the DOI identifier as well as the tag
+number to determine the security policy and the format associated with the
+tag. Use of tag types above 127 are restricted to closed networks where
+interoperability with other networks will not be an issue. Implementations
+that support a tag type greater than 127 MUST support at least one DOI that
+requires only tag types 1 to 127.
+
+Tag type 0 is reserved. Tag types 1, 2, and 5 are defined in this
+Internet Draft. Types 3 and 4 are reserved for work in progress.
+The standard format for all current and future CIPSO tags is shown below:
+
++----------+----------+--------//--------+
+| TTTTTTTT | LLLLLLLL | IIIIIIIIIIIIIIII |
++----------+----------+--------//--------+
+ TAG TAG TAG
+ TYPE LENGTH INFORMATION
+
+ Figure 2: Standard Tag Format
+
+In the three tag types described in this document, the length and count
+restrictions are based on the current IP limitation of 40 octets for all
+IP options. If the IP header is later expanded, then the length and count
+restrictions specified in this document may increase to use the full area
+provided for IP options.
+
+
+3.4.1 Tag Type Classes
+
+Tag classes consist of tag types that have common processing requirements
+and support the same security policy. The three tags defined in this
+Internet Draft belong to the Mandatory Access Control (MAC) Sensitivity
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 3]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+class and support the MAC Sensitivity security policy.
+
+
+3.4.2 Tag Type 1
+
+This is referred to as the "bit-mapped" tag type. Tag type 1 is included
+in the MAC Sensitivity tag type class. The format of this tag type is as
+follows:
+
++----------+----------+----------+----------+--------//---------+
+| 00000001 | LLLLLLLL | 00000000 | LLLLLLLL | CCCCCCCCCCCCCCCCC |
++----------+----------+----------+----------+--------//---------+
+
+ TAG TAG ALIGNMENT SENSITIVITY BIT MAP OF
+ TYPE LENGTH OCTET LEVEL CATEGORIES
+
+ Figure 3. Tag Type 1 Format
+
+
+3.4.2.1 Tag Type
+
+This field is 1 octet in length and has a value of 1.
+
+
+3.4.2.2 Tag Length
+
+This field is 1 octet in length. It is the total length of the tag type
+including the type and length fields. With the current IP header length
+restriction of 40 bytes the value within this field is between 4 and 34.
+
+
+3.4.2.3 Alignment Octet
+
+This field is 1 octet in length and always has the value of 0. Its purpose
+is to align the category bitmap field on an even octet boundary. This will
+speed many implementations including router implementations.
+
+
+3.4.2.4 Sensitivity Level
+
+This field is 1 octet in length. Its value is from 0 to 255. The values
+are ordered with 0 being the minimum value and 255 representing the maximum
+value.
+
+
+3.4.2.5 Bit Map of Categories
+
+The length of this field is variable and ranges from 0 to 30 octets. This
+provides representation of categories 0 to 239. The ordering of the bits
+is left to right or MSB to LSB. For example category 0 is represented by
+the most significant bit of the first byte and category 15 is represented
+by the least significant bit of the second byte. Figure 4 graphically
+shows this ordering. Bit N is binary 1 if category N is part of the label
+for the datagram, and bit N is binary 0 if category N is not part of the
+label. Except for the optimized tag 1 format described in the next section,
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 4]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+minimal encoding SHOULD be used resulting in no trailing zero octets in the
+category bitmap.
+
+ octet 0 octet 1 octet 2 octet 3 octet 4 octet 5
+ XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX . . .
+bit 01234567 89111111 11112222 22222233 33333333 44444444
+number 012345 67890123 45678901 23456789 01234567
+
+ Figure 4. Ordering of Bits in Tag 1 Bit Map
+
+
+3.4.2.6 Optimized Tag 1 Format
+
+Routers work most efficiently when processing fixed length fields. To
+support these routers there is an optimized form of tag type 1. The format
+does not change. The only change is to the category bitmap which is set to
+a constant length of 10 octets. Trailing octets required to fill out the 10
+octets are zero filled. Ten octets, allowing for 80 categories, was chosen
+because it makes the total length of the CIPSO option 20 octets. If CIPSO
+is the only option then the option will be full word aligned and additional
+filler octets will not be required.
+
+
+3.4.3 Tag Type 2
+
+This is referred to as the "enumerated" tag type. It is used to describe
+large but sparsely populated sets of categories. Tag type 2 is in the MAC
+Sensitivity tag type class. The format of this tag type is as follows:
+
++----------+----------+----------+----------+-------------//-------------+
+| 00000010 | LLLLLLLL | 00000000 | LLLLLLLL | CCCCCCCCCCCCCCCCCCCCCCCCCC |
++----------+----------+----------+----------+-------------//-------------+
+
+ TAG TAG ALIGNMENT SENSITIVITY ENUMERATED
+ TYPE LENGTH OCTET LEVEL CATEGORIES
+
+ Figure 5. Tag Type 2 Format
+
+
+3.4.3.1 Tag Type
+
+This field is one octet in length and has a value of 2.
+
+
+3.4.3.2 Tag Length
+
+This field is 1 octet in length. It is the total length of the tag type
+including the type and length fields. With the current IP header length
+restriction of 40 bytes the value within this field is between 4 and 34.
+
+
+3.4.3.3 Alignment Octet
+
+This field is 1 octet in length and always has the value of 0. Its purpose
+is to align the category field on an even octet boundary. This will
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 5]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+speed many implementations including router implementations.
+
+
+3.4.3.4 Sensitivity Level
+
+This field is 1 octet in length. Its value is from 0 to 255. The values
+are ordered with 0 being the minimum value and 255 representing the
+maximum value.
+
+
+3.4.3.5 Enumerated Categories
+
+In this tag, categories are represented by their actual value rather than
+by their position within a bit field. The length of each category is 2
+octets. Up to 15 categories may be represented by this tag. Valid values
+for categories are 0 to 65534. Category 65535 is not a valid category
+value. The categories MUST be listed in ascending order within the tag.
+
+
+3.4.4 Tag Type 5
+
+This is referred to as the "range" tag type. It is used to represent
+labels where all categories in a range, or set of ranges, are included
+in the sensitivity label. Tag type 5 is in the MAC Sensitivity tag type
+class. The format of this tag type is as follows:
+
++----------+----------+----------+----------+------------//-------------+
+| 00000101 | LLLLLLLL | 00000000 | LLLLLLLL | Top/Bottom | Top/Bottom |
++----------+----------+----------+----------+------------//-------------+
+
+ TAG TAG ALIGNMENT SENSITIVITY CATEGORY RANGES
+ TYPE LENGTH OCTET LEVEL
+
+ Figure 6. Tag Type 5 Format
+
+
+3.4.4.1 Tag Type
+
+This field is one octet in length and has a value of 5.
+
+
+3.4.4.2 Tag Length
+
+This field is 1 octet in length. It is the total length of the tag type
+including the type and length fields. With the current IP header length
+restriction of 40 bytes the value within this field is between 4 and 34.
+
+
+3.4.4.3 Alignment Octet
+
+This field is 1 octet in length and always has the value of 0. Its purpose
+is to align the category range field on an even octet boundary. This will
+speed many implementations including router implementations.
+
+
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 6]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+3.4.4.4 Sensitivity Level
+
+This field is 1 octet in length. Its value is from 0 to 255. The values
+are ordered with 0 being the minimum value and 255 representing the maximum
+value.
+
+
+3.4.4.5 Category Ranges
+
+A category range is a 4 octet field comprised of the 2 octet index of the
+highest numbered category followed by the 2 octet index of the lowest
+numbered category. These range endpoints are inclusive within the range of
+categories. All categories within a range are included in the sensitivity
+label. This tag may contain a maximum of 7 category pairs. The bottom
+category endpoint for the last pair in the tag MAY be omitted and SHOULD be
+assumed to be 0. The ranges MUST be non-overlapping and be listed in
+descending order. Valid values for categories are 0 to 65534. Category
+65535 is not a valid category value.
+
+
+3.4.5 Minimum Requirements
+
+A CIPSO implementation MUST be capable of generating at least tag type 1 in
+the non-optimized form. In addition, a CIPSO implementation MUST be able
+to receive any valid tag type 1 even those using the optimized tag type 1
+format.
+
+
+4. Configuration Parameters
+
+The configuration parameters defined below are required for all CIPSO hosts,
+gateways, and routers that support multiple sensitivity labels. A CIPSO
+host is defined to be the origination or destination system for an IP
+datagram. A CIPSO gateway provides IP routing services between two or more
+IP networks and may be required to perform label translations between
+networks. A CIPSO gateway may be an enhanced CIPSO host or it may just
+provide gateway services with no end system CIPSO capabilities. A CIPSO
+router is a dedicated IP router that routes IP datagrams between two or more
+IP networks.
+
+An implementation of CIPSO on a host MUST have the capability to reject a
+datagram for reasons that the information contained can not be adequately
+protected by the receiving host or if acceptance may result in violation of
+the host or network security policy. In addition, a CIPSO gateway or router
+MUST be able to reject datagrams going to networks that can not provide
+adequate protection or may violate the network's security policy. To
+provide this capability the following minimal set of configuration
+parameters are required for CIPSO implementations:
+
+HOST_LABEL_MAX - This parameter contains the maximum sensitivity label that
+a CIPSO host is authorized to handle. All datagrams that have a label
+greater than this maximum MUST be rejected by the CIPSO host. This
+parameter does not apply to CIPSO gateways or routers. This parameter need
+not be defined explicitly as it can be implicitly derived from the
+PORT_LABEL_MAX parameters for the associated interfaces.
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 7]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+
+HOST_LABEL_MIN - This parameter contains the minimum sensitivity label that
+a CIPSO host is authorized to handle. All datagrams that have a label less
+than this minimum MUST be rejected by the CIPSO host. This parameter does
+not apply to CIPSO gateways or routers. This parameter need not be defined
+explicitly as it can be implicitly derived from the PORT_LABEL_MIN
+parameters for the associated interfaces.
+
+PORT_LABEL_MAX - This parameter contains the maximum sensitivity label for
+all datagrams that may exit a particular network interface port. All
+outgoing datagrams that have a label greater than this maximum MUST be
+rejected by the CIPSO system. The label within this parameter MUST be
+less than or equal to the label within the HOST_LABEL_MAX parameter. This
+parameter does not apply to CIPSO hosts that support only one network port.
+
+PORT_LABEL_MIN - This parameter contains the minimum sensitivity label for
+all datagrams that may exit a particular network interface port. All
+outgoing datagrams that have a label less than this minimum MUST be
+rejected by the CIPSO system. The label within this parameter MUST be
+greater than or equal to the label within the HOST_LABEL_MIN parameter.
+This parameter does not apply to CIPSO hosts that support only one network
+port.
+
+PORT_DOI - This parameter is used to assign a DOI identifier value to a
+particular network interface port. All CIPSO labels within datagrams
+going out this port MUST use the specified DOI identifier. All CIPSO
+hosts and gateways MUST support either this parameter, the NET_DOI
+parameter, or the HOST_DOI parameter.
+
+NET_DOI - This parameter is used to assign a DOI identifier value to a
+particular IP network address. All CIPSO labels within datagrams destined
+for the particular IP network MUST use the specified DOI identifier. All
+CIPSO hosts and gateways MUST support either this parameter, the PORT_DOI
+parameter, or the HOST_DOI parameter.
+
+HOST_DOI - This parameter is used to assign a DOI identifier value to a
+particular IP host address. All CIPSO labels within datagrams destined for
+the particular IP host will use the specified DOI identifier. All CIPSO
+hosts and gateways MUST support either this parameter, the PORT_DOI
+parameter, or the NET_DOI parameter.
+
+This list represents the minimal set of configuration parameters required
+to be compliant. Implementors are encouraged to add to this list to
+provide enhanced functionality and control. For example, many security
+policies may require both incoming and outgoing datagrams be checked against
+the port and host label ranges.
+
+
+4.1 Port Range Parameters
+
+The labels represented by the PORT_LABEL_MAX and PORT_LABEL_MIN parameters
+MAY be in CIPSO or local format. Some CIPSO systems, such as routers, may
+want to have the range parameters expressed in CIPSO format so that incoming
+labels do not have to be converted to a local format before being compared
+against the range. If multiple DOIs are supported by one of these CIPSO
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 8]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+systems then multiple port range parameters would be needed, one set for
+each DOI supported on a particular port.
+
+The port range will usually represent the total set of labels that may
+exist on the logical network accessed through the corresponding network
+interface. It may, however, represent a subset of these labels that are
+allowed to enter the CIPSO system.
+
+
+4.2 Single Label CIPSO Hosts
+
+CIPSO implementations that support only one label are not required to
+support the parameters described above. These limited implementations are
+only required to support a NET_LABEL parameter. This parameter contains
+the CIPSO label that may be inserted in datagrams that exit the host. In
+addition, the host MUST reject any incoming datagram that has a label which
+is not equivalent to the NET_LABEL parameter.
+
+
+5. Handling Procedures
+
+This section describes the processing requirements for incoming and
+outgoing IP datagrams. Just providing the correct CIPSO label format
+is not enough. Assumptions will be made by one system on how a
+receiving system will handle the CIPSO label. Wrong assumptions may
+lead to non-interoperability or even a security incident. The
+requirements described below represent the minimal set needed for
+interoperability and that provide users some level of confidence.
+Many other requirements could be added to increase user confidence,
+however at the risk of restricting creativity and limiting vendor
+participation.
+
+
+5.1 Input Procedures
+
+All datagrams received through a network port MUST have a security label
+associated with them, either contained in the datagram or assigned to the
+receiving port. Without this label the host, gateway, or router will not
+have the information it needs to make security decisions. This security
+label will be obtained from the CIPSO if the option is present in the
+datagram. See section 4.1.2 for handling procedures for unlabeled
+datagrams. This label will be compared against the PORT (if appropriate)
+and HOST configuration parameters defined in section 3.
+
+If any field within the CIPSO option, such as the DOI identifier, is not
+recognized the IP datagram is discarded and an ICMP "parameter problem"
+(type 12) is generated and returned. The ICMP code field is set to "bad
+parameter" (code 0) and the pointer is set to the start of the CIPSO field
+that is unrecognized.
+
+If the contents of the CIPSO are valid but the security label is
+outside of the configured host or port label range, the datagram is
+discarded and an ICMP "destination unreachable" (type 3) is generated
+and returned. The code field of the ICMP is set to "communication with
+destination network administratively prohibited" (code 9) or to
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 9]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+"communication with destination host administratively prohibited"
+(code 10). The value of the code field used is dependent upon whether
+the originator of the ICMP message is acting as a CIPSO host or a CIPSO
+gateway. The recipient of the ICMP message MUST be able to handle either
+value. The same procedure is performed if a CIPSO can not be added to an
+IP packet because it is too large to fit in the IP options area.
+
+If the error is triggered by receipt of an ICMP message, the message
+is discarded and no response is permitted (consistent with general ICMP
+processing rules).
+
+
+5.1.1 Unrecognized tag types
+
+The default condition for any CIPSO implementation is that an
+unrecognized tag type MUST be treated as a "parameter problem" and
+handled as described in section 4.1. A CIPSO implementation MAY allow
+the system administrator to identify tag types that may safely be
+ignored. This capability is an allowable enhancement, not a
+requirement.
+
+
+5.1.2 Unlabeled Packets
+
+A network port may be configured to not require a CIPSO label for all
+incoming datagrams. For this configuration a CIPSO label must be
+assigned to that network port and associated with all unlabeled IP
+datagrams. This capability might be used for single level networks or
+networks that have CIPSO and non-CIPSO hosts and the non-CIPSO hosts
+all operate at the same label.
+
+If a CIPSO option is required and none is found, the datagram is
+discarded and an ICMP "parameter problem" (type 12) is generated and
+returned to the originator of the datagram. The code field of the ICMP
+is set to "option missing" (code 1) and the ICMP pointer is set to 134
+(the value of the option type for the missing CIPSO option).
+
+
+5.2 Output Procedures
+
+A CIPSO option MUST appear only once in a datagram. Only one tag type
+from the MAC Sensitivity class MAY be included in a CIPSO option. Given
+the current set of defined tag types, this means that CIPSO labels at
+first will contain only one tag.
+
+All datagrams leaving a CIPSO system MUST meet the following condition:
+
+ PORT_LABEL_MIN <= CIPSO label <= PORT_LABEL_MAX
+
+If this condition is not satisfied the datagram MUST be discarded.
+If the CIPSO system only supports one port, the HOST_LABEL_MIN and the
+HOST_LABEL_MAX parameters MAY be substituted for the PORT parameters in
+the above condition.
+
+The DOI identifier to be used for all outgoing datagrams is configured by
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 10]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+the administrator. If port level DOI identifier assignment is used, then
+the PORT_DOI configuration parameter MUST contain the DOI identifier to
+use. If network level DOI assignment is used, then the NET_DOI parameter
+MUST contain the DOI identifier to use. And if host level DOI assignment
+is employed, then the HOST_DOI parameter MUST contain the DOI identifier
+to use. A CIPSO implementation need only support one level of DOI
+assignment.
+
+
+5.3 DOI Processing Requirements
+
+A CIPSO implementation MUST support at least one DOI and SHOULD support
+multiple DOIs. System and network administrators are cautioned to
+ensure that at least one DOI is common within an IP network to allow for
+broadcasting of IP datagrams.
+
+CIPSO gateways MUST be capable of translating a CIPSO option from one
+DOI to another when forwarding datagrams between networks. For
+efficiency purposes this capability is only a desired feature for CIPSO
+routers.
+
+
+5.4 Label of ICMP Messages
+
+The CIPSO label to be used on all outgoing ICMP messages MUST be equivalent
+to the label of the datagram that caused the ICMP message. If the ICMP was
+generated due to a problem associated with the original CIPSO label then the
+following responses are allowed:
+
+ a. Use the CIPSO label of the original IP datagram
+ b. Drop the original datagram with no return message generated
+
+In most cases these options will have the same effect. If you can not
+interpret the label or if it is outside the label range of your host or
+interface then an ICMP message with the same label will probably not be
+able to exit the system.
+
+
+6. Assignment of DOI Identifier Numbers =
+
+Requests for assignment of a DOI identifier number should be addressed to
+the Internet Assigned Numbers Authority (IANA).
+
+
+7. Acknowledgements
+
+Much of the material in this RFC is based on (and copied from) work
+done by Gary Winiger of Sun Microsystems and published as Commercial
+IP Security Option at the INTEROP 89, Commercial IPSO Workshop.
+
+
+8. Author's Address
+
+To submit mail for distribution to members of the IETF CIPSO Working
+Group, send mail to: cipso@wdl1.wdl.loral.com.
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 11]
+
+
+
+CIPSO INTERNET DRAFT 16 July, 1992
+
+
+
+
+To be added to or deleted from this distribution, send mail to:
+cipso-request@wdl1.wdl.loral.com.
+
+
+9. References
+
+RFC 1038, "Draft Revised IP Security Option", M. St. Johns, IETF, January
+1988.
+
+RFC 1108, "U.S. Department of Defense Security Options
+for the Internet Protocol", Stephen Kent, IAB, 1 March, 1991.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Internet Draft, Expires 15 Jan 93 [PAGE 12]
+
+
+
Index: linux-2.6.17.i686-quilt/Documentation/netlabel/introduction.txt
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/Documentation/netlabel/introduction.txt
@@ -0,0 +1,53 @@
+NetLabel Introduction
+==============================================================================
+Paul Moore, paul.moore@hp.com
+
+June 27, 2006
+
+ * Overview
+
+NetLabel is a mechanism which can be used by kernel security modules to attach
+security attributes to outgoing network packets generated from user space
+applications and read security attributes from incoming network packets. It
+is composed of four main components, the protocol engines, the communication
+layer, the kernel security module API, and the network stack operations.
+
+ * Protocol Engines
+
+The protocol engines are responsible for both applying and retrieving the
+network packet's security attributes. If any translation between the network
+security attributes and those on the host are required then the protocol
+engine will handle those tasks as well. Other kernel subsystems should
+refrain from calling the protocol engines directly, instead they should use
+the NetLabel kernel security module API described below.
+
+Detailed information about each NetLabel protocol engine can be found in this
+directory, consult '00-INDEX' for filenames.
+
+ * Communication Layer
+
+The communication layer exists to allow NetLabel configuration and monitoring
+from user space. The NetLabel communication layer uses a message based
+protocol built on top of the Generic NETLINK transport mechanism. The exact
+formatting of these NetLabel messages as well as the Generic NETLINK family
+names can be found in the the 'net/netlabel/' directory as comments in the
+header files as well as in 'include/net/netlabel.h'.
+
+ * Security Module API
+
+The purpose of the NetLabel security module API is to provide a protocol
+independent interface to the underlying NetLabel protocol engines. In addition
+to protocol independence, the security module API is designed to be completely
+LSM independent which should allow multiple LSMs to leverage the same code
+base.
+
+Detailed information about the NetLabel security module API can be found in the
+'include/net/netlabel.h' header file as well as the 'lsm_interface.txt' file
+found in this directory.
+
+ * Network Stack Operations
+
+The reason for the network stack operations are to ensure the correct handling
+of the security attributes for kernel generated packets. In general only the
+core network stack should call these functions.
+
Index: linux-2.6.17.i686-quilt/Documentation/netlabel/lsm_interface.txt
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/Documentation/netlabel/lsm_interface.txt
@@ -0,0 +1,47 @@
+NetLabel Linux Security Module Interface
+==============================================================================
+Paul Moore, paul.moore@hp.com
+
+May 17, 2006
+
+ * Overview
+
+NetLabel is a mechanism which can set and retrieve security attributes from
+network packets. It is intended to be used by LSM developers who want to make
+use of a common code base for several different packet labeling protocols.
+The NetLabel security module API is defined in 'include/net/netlabel.h' but a
+brief overview is given below.
+
+ * NetLabel Security Attributes
+
+Since NetLabel supports multiple different packet labeling protocols and LSMs
+it uses the concept of security attributes to refer to the packet's security
+labels. The NetLabel security attributes are defined by the
+'netlbl_lsm_secattr' structure in the NetLabel header file. Internally the
+NetLabel subsystem converts the security attributes to and from the correct
+low-level packet label depending on the NetLabel build time and run time
+configuration. It is up to the LSM developer to translate the NetLabel
+security attributes into whatever security identifiers are in use for their
+particular LSM.
+
+ * NetLabel LSM Protocol Operations
+
+These are the functions which allow the LSM developer to manipulate the labels
+on outgoing packets as well as read the labels on incoming packets. Functions
+exist to operate both on sockets as well as the sk_buffs directly. These high
+level functions are translated into low level protocol operations based on how
+the administrator has configured the NetLabel subsystem.
+
+ * NetLabel Label Mapping Cache Operations
+
+Depending on the exact configuration, translation between the network packet
+label and the internal LSM security identifier can be time consuming. The
+NetLabel label mapping cache is a caching mechanism which can be used to
+sidestep much of this overhead once a mapping has been established. Once the
+LSM has received a packet, used NetLabel to decode it's security attributes,
+and translated the security attributes into a LSM internal identifier the LSM
+can use the NetLabel caching functions to associate the LSM internal
+identifier with the network packet's label. This means that in the future
+when a incoming packet matches a cached value not only are the internal
+NetLabel translation mechanisms bypassed but the LSM translation mechanisms are
+bypassed as well which should result in a significant reduction in overhead.
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 3/8] NetLabel: CIPSOv4 engine
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-net_cipsov4-2.6.17 --]
[-- Type: text/plain, Size: 49862 bytes --]
Add support for the Commercial IP Security Option (CIPSO) to the IPv4 network
stack. CIPSO has become a de-facto standard for trusted/labeled networking
amongst existing Trusted Operating Systems such as Trusted Solaris, HP-UX CMW,
etc. This implementation is designed to be used with the NetLabel subsystem
to provide explicit packet labeling to LSM developers.
The CIPSO/IPv4 packet labeling works by the LSM calling a NetLabel API function
which attaches a CIPSO label (IPv4 option) to a given socket; this in turn
attaches the CIPSO label to every packet leaving the socket without any extra
processing on the outbound side. On the inbound side the individual packet's
sk_buff is examined through a call to a NetLabel API function to determine if a
CIPSO/IPv4 label is present and if so the security attributes of the CIPSO
label are returned to the caller of the NetLabel API function.
---
net/ipv4/cipso_ipv4.c | 1749 ++++++++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 1749 insertions(+)
Index: linux-2.6.17.i686-quilt/net/ipv4/cipso_ipv4.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/ipv4/cipso_ipv4.c
@@ -0,0 +1,1749 @@
+/*
+ * CIPSO - Commercial IP Security Option
+ *
+ * This is an implementation of the CIPSO 2.2 protocol as specified in
+ * draft-ietf-cipso-ipsecurity-01.txt with additional tag types as found in
+ * FIPS-188, copies of both documents can be found in the Documentation
+ * directory. While CIPSO never became a full IETF RFC standard many vendors
+ * have chosen to adopt the protocol and over the years it has become a
+ * de-facto standard for labeled networking.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <linux/module.h>
+#include <linux/jhash.h>
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/tcp.h>
+#include <net/netlabel.h>
+#include <net/cipso_ipv4.h>
+#include <asm/bug.h>
+
+struct cipso_v4_domhsh_entry {
+ char *domain;
+ u32 valid;
+ struct list_head list;
+ struct rcu_head rcu;
+};
+
+/* List of available DOI definitions */
+/* XXX - Updates should be minimal so having a single lock for the
+ cipso_v4_doi_list and the cipso_v4_doi_list->dom_list should be
+ okay. */
+/* XXX - This currently assumes a minimal number of different DOIs in use,
+ if in practice there are a lot of different DOIs this list should
+ probably be turned into a hash table or something similar so we
+ can do quick lookups. */
+DEFINE_SPINLOCK(cipso_v4_doi_list_lock);
+static struct list_head cipso_v4_doi_list = LIST_HEAD_INIT(cipso_v4_doi_list);
+
+/* Label mapping cache */
+#define CIPSO_V4_CACHE_BUCKETBITS 7
+#define CIPSO_V4_CACHE_BUCKETS (1 << CIPSO_V4_CACHE_BUCKETBITS)
+#define CIPSO_V4_CACHE_BUCKETSIZE 10
+#define CIPSO_V4_CACHE_REORDERLIMIT 10
+/* PM - the number of cache buckets should probably be a compile time option */
+struct cipso_v4_map_cache_bkt {
+ spinlock_t lock;
+ u32 size;
+ struct list_head list;
+};
+struct cipso_v4_map_cache_entry {
+ u32 hash;
+ unsigned char *key;
+ u32 key_len;
+
+ struct netlbl_lsm_cache lsm_data;
+
+ u32 activity;
+ struct list_head list;
+};
+static u32 cipso_v4_cache_size = 0;
+static struct cipso_v4_map_cache_bkt *cipso_v4_cache = NULL;
+#define CIPSO_V4_CACHE_ENABLED (cipso_v4_cache_size > 0)
+
+/*
+ * Helper Functions
+ */
+
+/**
+ * cipso_v4_bitmap_walk - Walk a bitmap looking for a bit
+ * @bitmap: the bitmap
+ * @bitmap_len: length in bits
+ * @offset: starting offset
+ * @state: if non-zero, look for a set (1) bit else look for a cleared (0) bit
+ *
+ * Description:
+ * Starting at @offset, walk the bitmap from left to right until either the
+ * desired bit is found or we reach the end. Return the bit offset, -1 if
+ * not found, or -2 if error.
+ */
+static int cipso_v4_bitmap_walk(const unsigned char *bitmap,
+ const u32 bitmap_len,
+ const u32 offset,
+ const u8 state)
+{
+ u32 bit_spot;
+ u32 byte_offset;
+ unsigned char bitmask;
+ unsigned char byte;
+
+ /* gcc always rounds to zero when doing integer division */
+ byte_offset = offset / 8;
+ byte = bitmap[byte_offset];
+ bit_spot = offset;
+ bitmask = 0x80 >> offset % 8;
+
+ /* XXX - probably should use include/asm/bitops.h if we can */
+ while (bit_spot < bitmap_len) {
+ if ((state && (byte & bitmask) == bitmask) ||
+ (state == 0 && (byte & bitmask) == 0))
+ return bit_spot;
+
+ bit_spot++;
+ bitmask >>= 1;
+ if (bitmask == 0) {
+ byte = bitmap[++byte_offset];
+ bitmask = 0x80;
+ }
+ }
+
+ return -1;
+}
+
+/**
+ * cipso_v4_bitmap_setbit - Sets a single bit in a bitmap
+ * @bitmap: the bitmap
+ * @bit: the bit
+ * @state: if non-zero, set the bit (1) else clear the bit (0)
+ *
+ * Description:
+ * Set a single bit in the bitmask. Returns zero on success, negative values
+ * on error.
+ */
+static void cipso_v4_bitmap_setbit(unsigned char *bitmap,
+ const u32 bit,
+ const u8 state)
+{
+ u32 byte_spot;
+ u8 bitmask;
+
+ /* gcc always rounds to zero when doing integer division */
+ byte_spot = bit / 8;
+ bitmask = 0x80 >> bit % 8;
+ if (state)
+ bitmap[byte_spot] |= bitmask;
+ else
+ bitmap[byte_spot] &= ~bitmask;
+}
+
+/**
+ * cipso_v4_doi_domhsh_free - Frees a domain list entry
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that the memory allocated to a domain list entry can be released
+ * safely.
+ *
+ */
+static void cipso_v4_doi_domhsh_free(struct rcu_head *entry)
+{
+ struct cipso_v4_domhsh_entry *ptr;
+
+ ptr = container_of(entry, struct cipso_v4_domhsh_entry, rcu);
+ if (ptr->domain)
+ kfree(ptr->domain);
+ kfree(ptr);
+}
+
+/**
+ * cipso_v4_cache_entry_free - Frees a cache entry
+ * @entry: the entry to free
+ *
+ * Description:
+ * This function frees the memory associated with a cache entry.
+ *
+ */
+static void cipso_v4_cache_entry_free(struct cipso_v4_map_cache_entry *entry)
+{
+ if (entry->lsm_data.free)
+ entry->lsm_data.free(entry->lsm_data.data);
+ if (entry->key)
+ kfree(entry->key);
+ kfree(entry);
+}
+
+/**
+ * cipso_v4_map_cache_hash - Hashing function for the CIPSO cache
+ * @key: the hash key
+ * @key_len: the length of the key in bytes
+ *
+ * Description:
+ * The CIPSO tag hashing function. Returns a 32-bit hash value.
+ *
+ */
+static u32 cipso_v4_map_cache_hash(const unsigned char *key, const u32 key_len)
+{
+ /* PM - not sure if this is a good hash for this particular use but
+ it's probably better than what I had here before */
+ return jhash(key, key_len, 0);
+}
+
+/*
+ * Label Mapping Cache Functions
+ */
+
+/**
+ * cipso_v4_cache_init - Initialize the CIPSO cache
+ * @bkt_size: the number of cache buckets
+ *
+ * Description:
+ * Initializes the CIPSO label mapping cache, this function should be called
+ * before any of the other functions defined in this file. Returns zero on
+ * success, negative values on error.
+ *
+ */
+static int cipso_v4_cache_init(const u32 bkt_size)
+{
+ struct cipso_v4_map_cache_bkt *cache;
+ u32 iter;
+
+ if (bkt_size == 0)
+ return -EINVAL;
+
+ cache = kcalloc(bkt_size,
+ sizeof(struct cipso_v4_map_cache_bkt), GFP_KERNEL);
+ if (cache == NULL)
+ return -ENOMEM;
+
+ for (iter = 0; iter < bkt_size; iter++) {
+ cache[iter].lock = SPIN_LOCK_UNLOCKED;
+ cache[iter].size = 0;
+ INIT_LIST_HEAD(&cache[iter].list);
+ }
+ cipso_v4_cache = cache;
+ cipso_v4_cache_size = bkt_size;
+
+ return 0;
+}
+
+/**
+ * cipso_v4_cache_destroy - Destroy the CIPSO cache
+ *
+ * Description:
+ * Clears the CIPSO cache and frees all the memory. This function does not
+ * hold any locks and should only be called when the module is being unloaded.
+ *
+ */
+static int cipso_v4_cache_destroy(void)
+{
+ struct cipso_v4_map_cache_bkt *cache;
+ struct cipso_v4_map_cache_entry *entry;
+ u32 cache_size;
+ u32 iter;
+
+ cache = cipso_v4_cache;
+ cache_size = cipso_v4_cache_size;
+ cipso_v4_cache_size = 0;
+ cipso_v4_cache = NULL;
+
+ for (iter = 0; iter < cache_size; iter++)
+ list_for_each_entry(entry, &cache[iter].list, list) {
+ list_del(&entry->list);
+ cipso_v4_cache_entry_free(entry);
+ }
+
+ return 0;
+}
+
+/**
+ * cipso_v4_cache_invalidate - Invalidates the current CIPSO cache
+ *
+ * Description:
+ * Invalidates and frees any entries in the CIPSO cache. Returns zero on
+ * success and negative values on failure.
+ *
+ */
+void cipso_v4_cache_invalidate(void)
+{
+ struct cipso_v4_map_cache_entry *entry, *tmp_entry;
+ u32 iter;
+
+ if (!CIPSO_V4_CACHE_ENABLED)
+ return;
+
+ for (iter = 0; iter < cipso_v4_cache_size; iter++) {
+ spin_lock(&cipso_v4_cache[iter].lock);
+ list_for_each_entry_safe(entry,
+ tmp_entry,
+ &cipso_v4_cache[iter].list, list) {
+ list_del(&entry->list);
+ cipso_v4_cache_entry_free(entry);
+ }
+ cipso_v4_cache[iter].size = 0;
+ spin_unlock(&cipso_v4_cache[iter].lock);
+ }
+
+ return;
+}
+
+/**
+ * cipso_v4_cache_check - Check the CIPSO cache for a label mapping
+ * @key: the buffer to check
+ * @key_len: buffer length in bytes
+ * @secattr: the security attribute struct to use
+ *
+ * Description:
+ * This function checks the cache to see if a label mapping already exists for
+ * the given key. If there is a match then the cache is adjusted and the
+ * @secattr struct is populated with the correct LSM security attributes. The
+ * cache is adjusted in the following manner if the entry is not already the
+ * first in the cache bucket:
+ *
+ * 1. The cache entry's activity counter is incremented
+ * 2. The previous (higher ranking) entry's activity counter is decremented
+ * 3. If the difference between the two activity counters is geater than
+ * CIPSO_V4_CACHE_REORDERLIMIT the two entries are swapped
+ *
+ * Returns zero on success, -ENOENT for a cache miss, and other negative values
+ * on error.
+ *
+ */
+static int cipso_v4_cache_check(const unsigned char *key,
+ const u32 key_len,
+ struct netlbl_lsm_secattr *secattr)
+{
+ u32 bkt;
+ struct cipso_v4_map_cache_entry *entry;
+ struct cipso_v4_map_cache_entry *prev_entry = NULL;
+ u32 hash;
+
+ if (!CIPSO_V4_CACHE_ENABLED)
+ return -ENOENT;
+
+ hash = cipso_v4_map_cache_hash(key, key_len);
+ bkt = hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
+ spin_lock(&cipso_v4_cache[bkt].lock);
+ list_for_each_entry(entry, &cipso_v4_cache[bkt].list, list) {
+ if (entry->hash == hash &&
+ entry->key_len == key_len &&
+ memcmp(entry->key, key, key_len) == 0) {
+ entry->activity += 1;
+
+ if (prev_entry != NULL) {
+ if (prev_entry->activity > 0)
+ prev_entry->activity -= 1;
+
+ if (entry->activity > prev_entry->activity &&
+ entry->activity - prev_entry->activity >
+ CIPSO_V4_CACHE_REORDERLIMIT) {
+ __list_del(entry->list.prev,
+ entry->list.next);
+ __list_add(&entry->list,
+ prev_entry->list.prev,
+ &prev_entry->list);
+ }
+ }
+
+ secattr->cache.free = entry->lsm_data.free;
+ secattr->cache.data = entry->lsm_data.data;
+ spin_unlock(&cipso_v4_cache[bkt].lock);
+
+ secattr->set_cache = 1;
+ return 0;
+ }
+ prev_entry = entry;
+ }
+ spin_unlock(&cipso_v4_cache[bkt].lock);
+
+ return -ENOENT;
+}
+
+/**
+ * cipso_v4_cache_add - Add an entry to the CIPSO cache
+ * @skb: the packet
+ * @secattr: the packet's security attributes
+ *
+ * Description:
+ * Add a new entry into the CIPSO label mapping cache. Add the new entry to
+ * head of the cache bucket's list, if the cache bucket is out of room remove
+ * the last entry in the list first. It is important to note that there is
+ * currently no checking for duplicate keys. Returns zero on success,
+ * negative values on failure.
+ *
+ */
+int cipso_v4_cache_add(const struct sk_buff *skb,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -EPERM;
+ u32 bkt;
+ struct cipso_v4_map_cache_entry *entry = NULL;
+ struct cipso_v4_map_cache_entry *old_entry = NULL;
+ unsigned char *cipso_ptr;
+ u32 cipso_ptr_len;
+
+ if (!CIPSO_V4_CACHE_ENABLED)
+ return 0;
+
+ cipso_ptr = CIPSO_V4_OPTPTR(skb);
+ cipso_ptr_len = cipso_ptr[1];
+
+ entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
+ if (entry == NULL)
+ return -ENOMEM;
+ entry->key = kmalloc(cipso_ptr_len, GFP_ATOMIC);
+ if (entry->key == NULL) {
+ ret_val = -ENOMEM;
+ goto cache_add_failure;
+ }
+ memcpy(entry->key, cipso_ptr, cipso_ptr_len);
+ entry->key_len = cipso_ptr_len;
+ entry->hash = cipso_v4_map_cache_hash(cipso_ptr, cipso_ptr_len);
+ entry->lsm_data.free = secattr->cache.free;
+ entry->lsm_data.data = secattr->cache.data;
+
+ bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
+ spin_lock(&cipso_v4_cache[bkt].lock);
+ if (cipso_v4_cache[bkt].size < CIPSO_V4_CACHE_BUCKETSIZE) {
+ list_add(&entry->list, &cipso_v4_cache[bkt].list);
+ cipso_v4_cache[bkt].size += 1;
+ } else {
+ old_entry = list_entry(cipso_v4_cache[bkt].list.prev,
+ struct cipso_v4_map_cache_entry, list);
+ list_del(&old_entry->list);
+ list_add(&entry->list, &cipso_v4_cache[bkt].list);
+ cipso_v4_cache_entry_free(old_entry);
+ }
+ spin_unlock(&cipso_v4_cache[bkt].lock);
+
+ return 0;
+
+cache_add_failure:
+ if (entry)
+ cipso_v4_cache_entry_free(entry);
+ return ret_val;
+}
+
+/*
+ * DOI List Functions
+ */
+
+/**
+ * cipso_v4_doi_search - Searches for a DOI definition
+ * @doi: the DOI to search for
+ *
+ * Description:
+ * Search the DOI definition list for a DOI definition with a DOI value that
+ * matches @doi. The caller is responsibile for calling rcu_read_[un]lock().
+ * Returns a pointer to the DOI definition on success and NULL on failure.
+ */
+static struct cipso_v4_doi *cipso_v4_doi_search(const u32 doi)
+{
+ struct cipso_v4_doi *iter;
+
+ list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
+ if (iter->doi == doi && iter->valid)
+ return iter;
+ return NULL;
+}
+
+/**
+ * cipso_v4_doi_add - Add a new DOI to the CIPSO protocol engine
+ * @doi_def: the DOI structure
+ *
+ * Description:
+ * The caller defines a new DOI for use by the CIPSO engine and calls this
+ * function to add it to the list of acceptable domains. The caller must
+ * ensure that the mapping table specified in @doi_def->map meets all of the
+ * requirements of the mapping type (see cipso_ipv4.h for details). Returns
+ * zero on success and non-zero on failure.
+ *
+ */
+int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
+{
+ if (doi_def == NULL || doi_def->doi == CIPSO_V4_DOI_UNKNOWN)
+ return -EINVAL;
+
+ doi_def->valid = 1;
+ INIT_RCU_HEAD(&doi_def->rcu);
+ INIT_LIST_HEAD(&doi_def->dom_list);
+
+ rcu_read_lock();
+ if (cipso_v4_doi_search(doi_def->doi) != NULL) {
+ rcu_read_unlock();
+ return -EEXIST;
+ }
+ spin_lock(&cipso_v4_doi_list_lock);
+ list_add_tail_rcu(&doi_def->list, &cipso_v4_doi_list);
+ spin_unlock(&cipso_v4_doi_list_lock);
+ rcu_read_unlock();
+
+ return 0;
+}
+
+/**
+ * cipso_v4_doi_remove - Remove an existing DOI from the CIPSO protocol engine
+ * @doi: the DOI value
+ * @callback: the DOI cleanup/free callback
+ *
+ * Description:
+ * Removes a DOI definition from the CIPSO engine, @callback is called to
+ * free any memory. The NetLabel routines will be called to release their own
+ * LSM domain mappings as well as our own domain list. Returns zero on
+ * success and negative values on failure.
+ *
+ */
+int cipso_v4_doi_remove(const u32 doi,
+ void (*callback) (struct rcu_head * head))
+{
+ struct cipso_v4_doi *doi_def;
+ struct cipso_v4_domhsh_entry *dom_iter;
+
+ rcu_read_lock();
+ doi_def = cipso_v4_doi_search(doi);
+ if (doi_def != NULL) {
+ spin_lock(&cipso_v4_doi_list_lock);
+ doi_def->valid = 0;
+ list_del_rcu(&doi_def->list);
+ spin_unlock(&cipso_v4_doi_list_lock);
+ list_for_each_entry_rcu(dom_iter, &doi_def->dom_list, list)
+ if (dom_iter->valid)
+ netlbl_domhsh_remove(dom_iter->domain);
+ cipso_v4_cache_invalidate();
+ rcu_read_unlock();
+
+ call_rcu(&doi_def->rcu, callback);
+ return 0;
+ }
+ rcu_read_unlock();
+
+ return -ENOENT;
+}
+
+/**
+ * cipso_v4_doi_getdef - Returns a pointer to a valid DOI definition
+ * @doi: the DOI value
+ *
+ * Description:
+ * Searches for a valid DOI definition and if one is found it is returned to
+ * the caller. Otherwise NULL is returned. The caller must ensure that
+ * rcu_read_lock() is held while accessing the returned definition.
+ *
+ */
+struct cipso_v4_doi *cipso_v4_doi_getdef(const u32 doi)
+{
+ return cipso_v4_doi_search(doi);
+}
+
+/**
+ * cipso_v4_doi_dump - Dump the CIPSO DOI definitions into a sk_buff
+ * @doi: the DOI value
+ * @headroom: the amount of headroom to allocate for the sk_buff
+ *
+ * Description:
+ * If the @doi value is zero then dump a list of all the configured DOI values
+ * into a sk_buff. If the @doi value is non-zero, lookup the match DOI
+ * definition and dump it's contents into a sk_buff. The returned sk_buff has
+ * room at the front of the sk_buff for @headroom bytes. See
+ * netlabel_cipso_v4.h for the LIST message format. This function may fail if
+ * another process is changing the DOI list at the same time. Returns a
+ * pointer to a sk_buff on success, NULL on error.
+ *
+ */
+struct sk_buff *cipso_v4_doi_dump(const u32 doi, const u32 headroom)
+{
+ struct sk_buff *skb;
+ unsigned char *buf;
+ struct cipso_v4_doi *iter;
+ u32 doi_cnt = 0;
+ u32 tag_cnt = 0;
+ u32 lvl_cnt = 0;
+ u32 cat_cnt = 0;
+ u32 buf_len;
+ u32 tmp_len;
+
+ /* XXX - In both cases, this is kinda ugly as we have to go through
+ the list once to determine how large of a buffer we need,
+ drop the locks, allocate the buffer, grab the locks, and
+ finally fill the buffer. The problem is that there is that
+ open window where the table could grow and we will end up
+ short on space. */
+
+ if (doi == 0) {
+ buf_len = 4;
+ rcu_read_lock();
+ list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
+ if (iter->valid) {
+ doi_cnt += 1;
+ buf_len += 8;
+ }
+ rcu_read_unlock();
+
+ skb = alloc_skb(NLMSG_ALIGN(headroom + buf_len), GFP_KERNEL);
+ if (skb == NULL)
+ return NULL;
+ skb_reserve(skb, headroom);
+ tmp_len = skb_tailroom(skb);
+ if (tmp_len < buf_len)
+ goto doi_dump_failure;
+ buf = skb_put(skb, buf_len);
+ buf_len -= 4;
+ netlbl_putinc_u32(&buf, doi_cnt);
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
+ if (iter->valid) {
+ if (buf_len < 8) {
+ rcu_read_unlock();
+ goto doi_dump_failure;
+ }
+ buf_len -= 8;
+ netlbl_putinc_u32(&buf, iter->doi);
+ netlbl_putinc_u32(&buf, iter->type);
+ }
+ rcu_read_unlock();
+ } else {
+ rcu_read_lock();
+ iter = cipso_v4_doi_getdef(doi);
+ if (iter == NULL) {
+ rcu_read_unlock();
+ return NULL;
+ }
+ buf_len = 4;
+ switch (iter->type) {
+ case CIPSO_V4_MAP_PASS:
+ buf_len += 4;
+ while (tag_cnt < CIPSO_V4_TAG_MAXCNT &&
+ iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
+ tag_cnt += 1;
+ buf_len += 1;
+ }
+ break;
+ case CIPSO_V4_MAP_STD:
+ buf_len += 12;
+ while (tag_cnt < CIPSO_V4_TAG_MAXCNT &&
+ iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
+ tag_cnt += 1;
+ buf_len += 1;
+ }
+ tmp_len = 0;
+ while (tmp_len < iter->map.std->lvl.local_size)
+ if (iter->map.std->lvl.local[tmp_len++] !=
+ CIPSO_V4_INV_LVL) {
+ lvl_cnt += 1;
+ buf_len += 8;
+ }
+ tmp_len = 0;
+ while (tmp_len < iter->map.std->cat.local_size)
+ if (iter->map.std->cat.local[tmp_len++] !=
+ CIPSO_V4_INV_CAT) {
+ cat_cnt += 1;
+ buf_len += 8;
+ }
+ break;
+ default:
+ rcu_read_unlock();
+ return NULL;
+ }
+ rcu_read_unlock();
+
+ skb = alloc_skb(NLMSG_ALIGN(headroom + buf_len), GFP_KERNEL);
+ if (skb == NULL)
+ return NULL;
+ skb_reserve(skb, headroom);
+ tmp_len = skb_tailroom(skb);
+ if (tmp_len < buf_len)
+ goto doi_dump_failure;
+ buf = skb_put(skb, buf_len);
+ buf_len -= 4;
+ netlbl_putinc_u32(&buf, iter->type);
+
+ rcu_read_lock();
+ if (iter != cipso_v4_doi_getdef(doi)) {
+ rcu_read_unlock();
+ goto doi_dump_failure;
+ }
+ switch (iter->type) {
+ case CIPSO_V4_MAP_PASS:
+ netlbl_putinc_u32(&buf, tag_cnt);
+ tmp_len = 0;
+ while (tmp_len < CIPSO_V4_TAG_MAXCNT &&
+ iter->tags[tmp_len] != CIPSO_V4_TAG_INVALID) {
+ if (buf_len < 1) {
+ rcu_read_unlock();
+ goto doi_dump_failure;
+ }
+ buf_len -= 1;
+ netlbl_putinc_u8(&buf, iter->tags[tmp_len++]);
+ }
+ break;
+ case CIPSO_V4_MAP_STD:
+ buf_len -= 12;
+ netlbl_putinc_u32(&buf, tag_cnt);
+ netlbl_putinc_u32(&buf, lvl_cnt);
+ netlbl_putinc_u32(&buf, cat_cnt);
+ tmp_len = 0;
+ while (tmp_len < CIPSO_V4_TAG_MAXCNT &&
+ iter->tags[tmp_len] != CIPSO_V4_TAG_INVALID) {
+ if (buf_len < 1) {
+ rcu_read_unlock();
+ goto doi_dump_failure;
+ }
+ buf_len -= 1;
+ netlbl_putinc_u8(&buf, iter->tags[tmp_len++]);
+ }
+ tmp_len = 0;
+ while (tmp_len < iter->map.std->lvl.local_size) {
+ if (iter->map.std->lvl.local[tmp_len] !=
+ CIPSO_V4_INV_LVL) {
+ if (buf_len < 8) {
+ rcu_read_unlock();
+ goto doi_dump_failure;
+ }
+ buf_len -= 8;
+ netlbl_putinc_u32(&buf, tmp_len);
+ netlbl_putinc_u32(&buf,
+ iter->map.std->lvl.
+ local[tmp_len]);
+ }
+ tmp_len += 1;
+ }
+ tmp_len = 0;
+ while (tmp_len < iter->map.std->cat.local_size) {
+ if (iter->map.std->cat.local[tmp_len] !=
+ CIPSO_V4_INV_CAT) {
+ if (buf_len < 8) {
+ rcu_read_unlock();
+ goto doi_dump_failure;
+ }
+ buf_len -= 8;
+ netlbl_putinc_u32(&buf, tmp_len);
+ netlbl_putinc_u32(&buf,
+ iter->map.std->cat.
+ local[tmp_len]);
+ }
+ tmp_len += 1;
+ }
+ break;
+ }
+ rcu_read_unlock();
+ }
+
+ return skb;
+
+doi_dump_failure:
+ kfree(skb);
+ return NULL;
+}
+
+/**
+ * cipso_v4_doi_domhsh_add - Adds a domain entry to a DOI definition
+ * @doi_def: the DOI definition
+ * @domain: the domain to add
+ *
+ * Description:
+ * Adds the @domain to the the DOI specified by @doi_def, this function
+ * should only be called by external functions (i.e. NetLabel). This function
+ * does allocate memory. Returns zero on success, negative values on failure.
+ *
+ */
+int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain)
+{
+ struct cipso_v4_domhsh_entry *iter;
+ struct cipso_v4_domhsh_entry *new_dom;
+
+ new_dom = kzalloc(sizeof(*new_dom), GFP_KERNEL);
+ if (new_dom == NULL)
+ return -ENOMEM;
+ if (domain) {
+ new_dom->domain = kstrdup(domain, GFP_KERNEL);
+ if (new_dom->domain == NULL) {
+ kfree(new_dom);
+ return -ENOMEM;
+ }
+ }
+ new_dom->valid = 1;
+ INIT_RCU_HEAD(&new_dom->rcu);
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(iter, &doi_def->dom_list, list)
+ if (iter->valid &&
+ ((domain != NULL && iter->domain != NULL &&
+ strcmp(iter->domain, domain) == 0) ||
+ (domain == NULL && iter->domain == NULL))) {
+ rcu_read_unlock();
+ if (new_dom->domain != NULL)
+ kfree(new_dom->domain);
+ kfree(new_dom);
+ return -EEXIST;
+ }
+ spin_lock(&cipso_v4_doi_list_lock);
+ list_add_tail_rcu(&new_dom->list, &doi_def->dom_list);
+ spin_unlock(&cipso_v4_doi_list_lock);
+ rcu_read_unlock();
+
+ return 0;
+}
+
+/**
+ * cipso_v4_doi_domhsh_remove - Removes a domain entry from a DOI definition
+ * @doi_def: the DOI definition
+ * @domain: the domain to remove
+ *
+ * Description:
+ * Removes the @domain from the DOI specified by @doi_def, this function
+ * should only be called by external functions (i.e. NetLabel). Returns zero
+ * on success and negative values on error.
+ *
+ */
+int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
+ const char *domain)
+{
+ struct cipso_v4_domhsh_entry *iter;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(iter, &doi_def->dom_list, list)
+ if (iter->valid &&
+ ((domain != NULL && iter->domain != NULL &&
+ strcmp(iter->domain, domain) == 0) ||
+ (domain == NULL && iter->domain == NULL))) {
+ spin_lock(&cipso_v4_doi_list_lock);
+ iter->valid = 0;
+ list_del_rcu(&iter->list);
+ spin_unlock(&cipso_v4_doi_list_lock);
+ rcu_read_unlock();
+
+ call_rcu(&iter->rcu, cipso_v4_doi_domhsh_free);
+ return 0;
+ }
+ rcu_read_unlock();
+
+ return -ENOENT;
+}
+
+/*
+ * Label Mapping Functions
+ */
+
+/**
+ * cipso_v4_map_lvl_valid - Checks to see if the given level is understood
+ * @doi_def: the DOI definition
+ * @level: the level to check
+ *
+ * Description:
+ * Checks the given level against the given DOI definition and returns a
+ * negative value if the level does not have a valid mapping and a zero value
+ * if the level is defined by the DOI.
+ *
+ */
+static int cipso_v4_map_lvl_valid(const struct cipso_v4_doi *doi_def,
+ const u8 level)
+{
+ switch (doi_def->type) {
+ case CIPSO_V4_MAP_PASS:
+ return 0;
+ case CIPSO_V4_MAP_STD:
+ if (doi_def->map.std->lvl.cipso[level] < CIPSO_V4_INV_LVL)
+ return 0;
+ break;
+ }
+
+ return -EFAULT;
+}
+
+/**
+ * cipso_v4_map_lvl_hton - Perform a level mapping from the host to the network
+ * @doi_def: the DOI definition
+ * @host_lvl: the host MLS level
+ * @net_lvl: the network/CIPSO MLS level
+ *
+ * Description:
+ * Perform a label mapping to translate a local MLS level to the correct
+ * CIPSO level using the given DOI definition. Returns zero on success,
+ * negative values otherwise.
+ *
+ */
+static int cipso_v4_map_lvl_hton(const struct cipso_v4_doi *doi_def,
+ const u32 host_lvl,
+ u32 *net_lvl)
+{
+ switch (doi_def->type) {
+ case CIPSO_V4_MAP_PASS:
+ *net_lvl = host_lvl;
+ return 0;
+ case CIPSO_V4_MAP_STD:
+ if (host_lvl < doi_def->map.std->lvl.local_size) {
+ *net_lvl = doi_def->map.std->lvl.local[host_lvl];
+ return 0;
+ }
+ break;
+ }
+
+ return -EINVAL;
+}
+
+/**
+ * cipso_v4_map_lvl_ntoh - Perform a level mapping from the network to the host
+ * @doi_def: the DOI definition
+ * @net_lvl: the network/CIPSO MLS level
+ * @host_lvl: the host MLS level
+ *
+ * Description:
+ * Perform a label mapping to translate a CIPSO level to the correct local MLS
+ * level using the given DOI definition. Returns zero on success, negative
+ * values otherwise.
+ *
+ */
+static int cipso_v4_map_lvl_ntoh(const struct cipso_v4_doi *doi_def,
+ const u32 net_lvl,
+ u32 *host_lvl)
+{
+ switch (doi_def->type) {
+ case CIPSO_V4_MAP_PASS:
+ *host_lvl = net_lvl;
+ return 0;
+ case CIPSO_V4_MAP_STD:
+ if (net_lvl < doi_def->map.std->lvl.cipso_size) {
+ *host_lvl = doi_def->map.std->lvl.cipso[net_lvl];
+ return 0;
+ }
+ break;
+ }
+
+ return -EINVAL;
+}
+
+/**
+ * cipso_v4_map_cat_rbm_valid - Checks to see if the category bitmap is valid
+ * @doi_def: the DOI definition
+ * @bitmap: category bitmap
+ * @bitmap_len: bitmap length in bytes
+ *
+ * Description:
+ * Checks the given category bitmap against the given DOI definition and
+ * returns a negative value if any of the categories in the bitmap do not have
+ * a valid mapping and a zero value if all of the categories are valid.
+ *
+ */
+static int cipso_v4_map_cat_rbm_valid(const struct cipso_v4_doi *doi_def,
+ const unsigned char *bitmap,
+ const u32 bitmap_len)
+{
+ int cat = -1;
+ u32 bitmap_len_bits = bitmap_len * 8;
+ u32 cipso_cat_size = doi_def->map.std->cat.cipso_size;
+ u32 *cipso_array = doi_def->map.std->cat.cipso;
+
+ switch (doi_def->type) {
+ case CIPSO_V4_MAP_PASS:
+ return 0;
+ case CIPSO_V4_MAP_STD:
+ for (;;) {
+ cat = cipso_v4_bitmap_walk(bitmap,
+ bitmap_len_bits,
+ cat + 1,
+ 1);
+ if (cat < 0)
+ break;
+ if (cat >= cipso_cat_size ||
+ cipso_array[cat] >= CIPSO_V4_INV_CAT)
+ return -EFAULT;
+ }
+
+ if (cat == -1)
+ return 0;
+ break;
+ }
+
+ return -EFAULT;
+}
+
+/**
+ * cipso_v4_map_cat_rbm_hton - Perform a category mapping from host to network
+ * @doi_def: the DOI definition
+ * @host_cat: the category bitmap in host format
+ * @host_cat_len: the length of the host's category bitmap in bytes
+ * @net_cat: the zero'd out category bitmap in network/CIPSO format
+ * @net_cat_len: the length of the CIPSO bitmap in bytes
+ *
+ * Description:
+ * Perform a label mapping to translate a local MLS category bitmap to the
+ * correct CIPSO bitmap using the given DOI definition. Returns the minimum
+ * size in bytes of the network bitmap on success, negative values otherwise.
+ *
+ */
+static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def,
+ const unsigned char *host_cat,
+ const u32 host_cat_len,
+ unsigned char *net_cat,
+ const u32 net_cat_len)
+{
+ int host_spot = -1;
+ u32 net_spot;
+ u32 net_spot_max = 0;
+ u32 host_clen_bits = host_cat_len * 8;
+ u32 net_clen_bits = net_cat_len * 8;
+ u32 host_cat_size = doi_def->map.std->cat.local_size;
+ u32 *host_cat_array = doi_def->map.std->cat.local;
+
+ switch (doi_def->type) {
+ case CIPSO_V4_MAP_PASS:
+ net_spot_max = host_cat_len - 1;
+ while (net_spot_max > 0 && host_cat[net_spot_max] == 0)
+ net_spot_max--;
+ if (net_spot_max > net_cat_len)
+ return -EINVAL;
+ memcpy(net_cat, host_cat, net_spot_max);
+ return net_spot_max;
+ case CIPSO_V4_MAP_STD:
+ for (;;) {
+ host_spot = cipso_v4_bitmap_walk(host_cat,
+ host_clen_bits,
+ host_spot + 1,
+ 1);
+ if (host_spot < 0)
+ break;
+ if (host_spot >= host_cat_size)
+ return -EPERM;
+
+ net_spot = host_cat_array[host_spot];
+ if (net_spot >= net_clen_bits)
+ return -ENOSPC;
+ cipso_v4_bitmap_setbit(net_cat, net_spot, 1);
+
+ if (net_spot > net_spot_max)
+ net_spot_max = net_spot;
+ }
+
+ if (host_spot == -2)
+ return -EFAULT;
+
+ if (++net_spot_max % 8)
+ return net_spot_max / 8 + 1;
+ return net_spot_max / 8;
+ }
+
+ return -EINVAL;
+}
+
+/**
+ * cipso_v4_map_cat_rbm_ntoh - Perform a category mapping from network to host
+ * @doi_def: the DOI definition
+ * @net_cat: the category bitmap in network/CIPSO format
+ * @net_cat_len: the length of the CIPSO bitmap in bytes
+ * @host_cat: the zero'd out category bitmap in host format
+ * @host_cat_len: the length of the host's category bitmap in bytes
+ *
+ * Description:
+ * Perform a label mapping to translate a CIPSO bitmap to the correct local
+ * MLS category bitmap using the given DOI definition. Returns the minimum
+ * size in bytes of the host bitmap on success, negative values otherwise.
+ *
+ */
+static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def,
+ const unsigned char *net_cat,
+ const u32 net_cat_len,
+ unsigned char *host_cat,
+ const u32 host_cat_len)
+{
+ u32 host_spot;
+ u32 host_spot_max = 0;
+ int net_spot = -1;
+ u32 net_clen_bits = net_cat_len * 8;
+ u32 host_clen_bits = host_cat_len * 8;
+ u32 net_cat_size = doi_def->map.std->cat.cipso_size;
+ u32 *net_cat_array = doi_def->map.std->cat.cipso;
+
+ switch (doi_def->type) {
+ case CIPSO_V4_MAP_PASS:
+ if (net_cat_len > host_cat_len)
+ return -EINVAL;
+ memcpy(host_cat, net_cat, net_cat_len);
+ return host_cat_len;
+ case CIPSO_V4_MAP_STD:
+ for (;;) {
+ net_spot = cipso_v4_bitmap_walk(net_cat,
+ net_clen_bits,
+ net_spot + 1,
+ 1);
+ if (net_spot < 0)
+ break;
+ if (net_spot >= net_cat_size)
+ return -EPERM;
+
+ host_spot = net_cat_array[net_spot];
+ if (host_spot >= host_clen_bits)
+ return -ENOSPC;
+ cipso_v4_bitmap_setbit(host_cat, host_spot, 1);
+
+ if (host_spot > host_spot_max)
+ host_spot_max = host_spot;
+ }
+
+ if (net_spot == -2)
+ return -EFAULT;
+
+ if (++host_spot_max % 8)
+ return host_spot_max / 8 + 1;
+ return host_spot_max / 8;
+ }
+
+ return -EINVAL;
+}
+
+/*
+ * Protocol Handling Functions
+ */
+
+#define CIPSO_V4_HDR_LEN 6
+
+/**
+ * cipso_v4_gentag_hdr - Generate a CIPSO option header
+ * @doi_def: the DOI definition
+ * @len: the total tag length in bytes
+ * @buf: the CIPSO option buffer
+ *
+ * Description:
+ * Write a CIPSO header into the beginning of @buffer. Return zero on success,
+ * negative values on failure.
+ *
+ */
+static int cipso_v4_gentag_hdr(const struct cipso_v4_doi *doi_def,
+ const u32 len,
+ unsigned char *buf)
+{
+ if (CIPSO_V4_HDR_LEN + len > 40)
+ return -ENOSPC;
+
+ buf[0] = IPOPT_CIPSO;
+ buf[1] = CIPSO_V4_HDR_LEN + len;
+ *(u32 *)&buf[2] = htonl(doi_def->doi);
+
+ return 0;
+}
+
+#define CIPSO_V4_TAG1_CAT_LEN 30
+
+/**
+ * cipso_v4_gentag_rbm - Generate a CIPSO restricted bitmap tag (type #1)
+ * @doi_def: the DOI definition
+ * @secattr: the security attributes
+ * @buffer: the option buffer
+ * @buffer_len: length of buffer in bytes
+ *
+ * Description:
+ * Generate a CIPSO option using the restricted bitmap tag, tag type #1. The
+ * actual buffer length may be larger than the indicated size due to
+ * translation between host and network category bitmaps. Returns zero on
+ * success, negative values on failure.
+ *
+ */
+static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def,
+ const struct netlbl_lsm_secattr *secattr,
+ unsigned char **buffer,
+ u32 *buffer_len)
+{
+ int ret_val = -EPERM;
+ unsigned char *buf = NULL;
+ u32 buf_len;
+ u32 level;
+
+ if (secattr->set_mls_cat) {
+ buf = kzalloc(CIPSO_V4_HDR_LEN + 4 + CIPSO_V4_TAG1_CAT_LEN,
+ GFP_ATOMIC);
+ if (buf == NULL)
+ return -ENOMEM;
+
+ ret_val = cipso_v4_map_cat_rbm_hton(doi_def,
+ secattr->mls_cat,
+ secattr->mls_cat_len,
+ &buf[CIPSO_V4_HDR_LEN + 4],
+ CIPSO_V4_TAG1_CAT_LEN);
+ if (ret_val < 0)
+ goto gentag_failure;
+#if 0
+ /* PM - this should be a sysctl adjustable setting */
+ /* XXX - this will send packets using the "optimized" format
+ when possibile as specified in section 3.4.2.6 of the
+ CIPSO draft */
+ if (ret_val > 0 && ret_val < 10)
+ ret_val = 10;
+#endif
+ buf_len = 4 + ret_val;
+ } else {
+ buf = kzalloc(CIPSO_V4_HDR_LEN + 4, GFP_ATOMIC);
+ if (buf == NULL)
+ return -ENOMEM;
+ buf_len = 4;
+ }
+
+ ret_val = cipso_v4_map_lvl_hton(doi_def, secattr->mls_lvl, &level);
+ if (ret_val != 0)
+ goto gentag_failure;
+
+ ret_val = cipso_v4_gentag_hdr(doi_def, buf_len, buf);
+ if (ret_val != 0)
+ goto gentag_failure;
+
+ buf[CIPSO_V4_HDR_LEN] = 0x01;
+ buf[CIPSO_V4_HDR_LEN + 1] = buf_len;
+ buf[CIPSO_V4_HDR_LEN + 3] = level;
+
+ *buffer = buf;
+ *buffer_len = CIPSO_V4_HDR_LEN + buf_len;
+
+ return 0;
+
+gentag_failure:
+ if (buf)
+ kfree(buf);
+ return ret_val;
+}
+
+/**
+ * cipso_v4_parsetag_rbm - Parse a CIPSO restricted bitmap tag
+ * @doi_def: the DOI definition
+ * @tag: the CIPSO tag
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Parse a CIPSO restricted bitmap tag (tag type #1) and return the security
+ * attributes in @secattr. Return zero on success, negatives values on
+ * failure.
+ *
+ */
+static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def,
+ const unsigned char *tag,
+ struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val;
+ u8 tag_len = tag[1];
+ u32 level;
+
+ ret_val = cipso_v4_map_lvl_ntoh(doi_def, tag[3], &level);
+ if (ret_val != 0)
+ return ret_val;
+ secattr->mls_lvl = level;
+ secattr->set_mls_lvl = 1;
+
+ if (tag_len > 4) {
+ switch (doi_def->type) {
+ case CIPSO_V4_MAP_PASS:
+ secattr->mls_cat_len = tag_len - 4;
+ break;
+ case CIPSO_V4_MAP_STD:
+ secattr->mls_cat_len =
+ doi_def->map.std->cat.local_size;
+ break;
+ }
+ secattr->mls_cat = kzalloc(secattr->mls_cat_len, GFP_ATOMIC);
+ if (secattr->mls_cat == NULL)
+ return -ENOMEM;
+
+ ret_val = cipso_v4_map_cat_rbm_ntoh(doi_def,
+ &tag[4],
+ tag_len - 4,
+ secattr->mls_cat,
+ secattr->mls_cat_len);
+ if (ret_val < 0) {
+ kfree(secattr->mls_cat);
+ return ret_val;
+ }
+ secattr->mls_cat_len = ret_val;
+ secattr->set_mls_cat = 1;
+ }
+
+ return 0;
+}
+
+/**
+ * cipso_v4_validate - Validate a CIPSO option
+ * @option: the start of the option, on error it is set to point to the error
+ *
+ * Description:
+ * This routine is called to validate a CIPSO option, it checks all of the
+ * fields to ensure that they are at least valid, see the draft snippet below
+ * for details. If the option is valid then a zero value is returned and
+ * the value of @option is unchanged. If the option is invalid then a
+ * non-zero value is returned and @option is adjusted to point to the
+ * offending portion of the option. From the IETF draft ...
+ *
+ * "If any field within the CIPSO options, such as the DOI identifier, is not
+ * recognized the IP datagram is discarded and an ICMP 'parameter problem'
+ * (type 12) is generated and returned. The ICMP code field is set to 'bad
+ * parameter' (code 0) and the pointer is set to the start of the CIPSO field
+ * that is unrecognized."
+ *
+ */
+int cipso_v4_validate(unsigned char **option)
+{
+ unsigned char *opt = *option;
+ unsigned char *tag;
+ unsigned char opt_iter;
+ unsigned char err_offset = 0;
+ unsigned char locked = 0;
+ u8 opt_len;
+ u8 tag_len;
+ struct cipso_v4_doi *doi_def = NULL;
+ u32 tag_iter;
+
+ /* XXX: caller already checks for length values that are too large */
+ opt_len = opt[1];
+ if (opt_len < 8) {
+ err_offset = 1;
+ goto validate_return;
+ }
+
+ rcu_read_lock();
+ locked = 1;
+ doi_def = cipso_v4_doi_getdef(ntohl(*((u32 *)&opt[2])));
+ if (doi_def == NULL) {
+ err_offset = 2;
+ goto validate_return;
+ }
+
+ opt_iter = 6;
+ tag = opt + opt_iter;
+ while (opt_iter < opt_len) {
+ for (tag_iter = 0; doi_def->tags[tag_iter] != tag[0];)
+ if (doi_def->tags[tag_iter] == CIPSO_V4_TAG_INVALID ||
+ ++tag_iter == CIPSO_V4_TAG_MAXCNT) {
+ err_offset = opt_iter;
+ goto validate_return;
+ }
+
+ tag_len = tag[1];
+ if (tag_len > (opt_len - opt_iter)) {
+ err_offset = opt_iter + 1;
+ goto validate_return;
+ }
+
+ switch (tag[0]) {
+ case CIPSO_V4_TAG_RBITMAP:
+ if (tag_len < 4) {
+ err_offset = opt_iter + 1;
+ goto validate_return;
+ }
+#if 0
+ /* PM - i think we can skip this check safely */
+ if (tag[2] != 0) {
+ err_offset = opt_iter + 2;
+ goto validate_return;
+ }
+#endif
+ /* PM - how detailed a check do we need to make, are
+ these last two checks (level and category)
+ neccessary at this level? if we do drop these
+ checks we need to add a bit of code to the *_ntoh()
+ functions to check for invalid levels/cats. */
+ if (cipso_v4_map_lvl_valid(doi_def, tag[3]) < 0) {
+ err_offset = opt_iter + 3;
+ goto validate_return;
+ }
+ if (tag_len > 4 &&
+ cipso_v4_map_cat_rbm_valid(doi_def,
+ &tag[4],
+ tag_len - 4) < 0) {
+ err_offset = opt_iter + 4;
+ goto validate_return;
+ }
+ break;
+ case CIPSO_V4_TAG_ENUM:
+ case CIPSO_V4_TAG_RANGE:
+ case CIPSO_V4_TAG_PBITMAP:
+ case CIPSO_V4_TAG_FREEFORM:
+ default:
+ err_offset = opt_iter;
+ goto validate_return;
+ }
+
+ tag += tag_len;
+ opt_iter += tag_len;
+ }
+
+validate_return:
+ if (locked)
+ rcu_read_unlock();
+ *option = opt + err_offset;
+ return err_offset;
+}
+
+/**
+ * cipso_v4_error - Send the correct reponse for a bad packet
+ * @skb: the packet
+ * @error: the error code
+ * @gateway: CIPSO gateway flag
+ *
+ * Description:
+ * Based on the error code given in @error, send an ICMP error message back to
+ * the originating host. Returns zero on success, negative values on failure.
+ * From the IETF draft ...
+ *
+ * "If the contents of the CIPSO [option] are valid but the security label is
+ * outside of the configured host or port label range, the datagram is
+ * discarded and an ICMP 'destination unreachable' (type 3) is generated and
+ * returned. The code field of the ICMP is set to 'communication with
+ * destination network administratively prohibited' (code 9) or to
+ * 'communication with destination host administratively prohibited'
+ * (code 10). The value of the code is dependent on whether the originator
+ * of the ICMP message is acting as a CIPSO host or a CIPSO gateway. The
+ * recipient of the ICMP message MUST be able to handle either value. The
+ * same procedure is performed if a CIPSO [option] can not be added to an
+ * IP packet because it is too large to fit in the IP options area."
+ *
+ * "If the error is triggered by receipt of an ICMP message, the message is
+ * discarded and no response is permitted (consistent with general ICMP
+ * processing rules)."
+ *
+ */
+int cipso_v4_error(struct sk_buff *skb,
+ const int error,
+ const u32 gateway)
+{
+ switch (error) {
+ case -EACCES:
+ break;
+ default:
+ return 0;
+ }
+
+ if (skb->nh.iph->protocol == IPPROTO_ICMP)
+ return 0;
+
+ if (gateway)
+ icmp_send(skb, ICMP_DEST_UNREACH, ICMP_NET_ANO, 0);
+ else
+ icmp_send(skb, ICMP_DEST_UNREACH, ICMP_HOST_ANO, 0);
+
+ return 0;
+}
+
+/**
+ * cipso_v4_socket_setopt - Set the CIPSO option string on a socket
+ * @sock: the socket
+ * @opt: the option string
+ * @opt_len: the option string length
+ *
+ * Description:
+ * Attempt to add the given CIPSO option string to the socket. The caller is
+ * responsibile for locking and releasing the socket. Returns zero on success,
+ * negative values on failure.
+ *
+ */
+int cipso_v4_socket_setopt(struct socket *sock,
+ unsigned char *opt,
+ u32 opt_len)
+{
+ int ret_val;
+ struct ip_options *ip_opt = NULL;
+ struct sock *sk = sock->sk;
+ struct inet_sock *sk_inet;
+ struct inet_connection_sock *sk_conn;
+
+ ret_val = ip_options_get(&ip_opt, opt, opt_len);
+ if (ret_val != 0)
+ goto socket_setopt_return;
+ ip_opt->is_setbyuser = 0;
+ sk_inet = inet_sk(sk);
+ if (sk_inet->is_icsk) {
+ sk_conn = inet_csk(sk);
+ if (sk_inet->opt)
+ sk_conn->icsk_ext_hdr_len -= sk_inet->opt->optlen;
+ sk_conn->icsk_ext_hdr_len += ip_opt->optlen;
+ sk_conn->icsk_sync_mss(sk, sk_conn->icsk_pmtu_cookie);
+ }
+ ip_opt = xchg(&sk_inet->opt, ip_opt);
+
+ ret_val = 0;
+
+socket_setopt_return:
+ if (ip_opt)
+ kfree(opt);
+ return ret_val;
+}
+
+/**
+ * cipso_v4_socket_setattr - Add a CIPSO option to a socket
+ * @sock: the socket
+ * @doi_def: the CIPSO DOI to use
+ * @secattr: the specific security attributes of the socket
+ *
+ * Description:
+ * Set the CIPSO option on the given socket using the DOI definition and
+ * security attributes passed to the function. Returns zero on success and
+ * negative values on failure.
+ *
+ */
+int cipso_v4_socket_setattr(const struct socket *sock,
+ const struct cipso_v4_doi *doi_def,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -EPERM;
+ u32 iter;
+ unsigned char *buf = NULL;
+ u32 buf_len = 0;
+ struct ip_options *opt = NULL;
+ struct sock *sk;
+ struct inet_sock *sk_inet;
+ struct inet_connection_sock *sk_conn;
+
+ /* In the case of sock_create_lite(), the sock->sk field is not
+ defined yet but it is not a problem as the only users of these
+ "lite" PF_INET sockets are functions which do an accept() call
+ afterwards so we will label the socket as part of the accept(). */
+ /* PM - test this, nfs/sunrpc is a "lite" socket user */
+ sk = sock->sk;
+ if (sk == NULL)
+ return 0;
+
+ /* XXX - this code assumes only one tag per CIPSO option which isnt
+ really a good assumption to make but since we only support the MAC
+ tags right now it is a safe assumption */
+ iter = 0;
+ do {
+ switch (doi_def->tags[iter]) {
+ case CIPSO_V4_TAG_RBITMAP:
+ ret_val = cipso_v4_gentag_rbm(doi_def,
+ secattr,
+ &buf,
+ &buf_len);
+ break;
+ case CIPSO_V4_TAG_ENUM:
+ case CIPSO_V4_TAG_RANGE:
+ case CIPSO_V4_TAG_PBITMAP:
+ case CIPSO_V4_TAG_FREEFORM:
+ default:
+ ret_val = -EPERM;
+ goto socket_setattr_failure;
+ }
+
+ iter++;
+ } while (ret_val != 0 &&
+ iter < CIPSO_V4_TAG_MAXCNT &&
+ doi_def->tags[iter] != CIPSO_V4_TAG_INVALID);
+ if (ret_val != 0)
+ goto socket_setattr_failure;
+
+ ret_val = ip_options_get(&opt, buf, buf_len);
+ if (ret_val != 0)
+ goto socket_setattr_failure;
+ opt->is_setbyuser = 0;
+
+ kfree(buf);
+ buf = NULL;
+
+ lock_sock(sk);
+ sk_inet = inet_sk(sk);
+ if (sk_inet->is_icsk) {
+ sk_conn = inet_csk(sk);
+ if (sk_inet->opt)
+ sk_conn->icsk_ext_hdr_len -= sk_inet->opt->optlen;
+ sk_conn->icsk_ext_hdr_len += opt->optlen;
+ sk_conn->icsk_sync_mss(sk, sk_conn->icsk_pmtu_cookie);
+ }
+ opt = xchg(&sk_inet->opt, opt);
+ release_sock(sk);
+ if (opt)
+ kfree(opt);
+
+ return 0;
+
+socket_setattr_failure:
+ if (buf)
+ kfree(buf);
+ if (opt)
+ kfree(opt);
+ return ret_val;
+}
+
+/**
+ * cipso_v4_socket_getopt - Get the CIPSO IP option string from the socket
+ * @sock: the socket
+ * @opt: the option string
+ * @opt_len: the option string length in bytes
+ *
+ * Description:
+ * If a CIPSO option is present in the socket's IP options then return a
+ * a pointer to the option string. The caller is responsibile for locking
+ * and releasing the socket. Returns zero on success, negative values on
+ * failure.
+ *
+ */
+int cipso_v4_socket_getopt(const struct socket *sock,
+ unsigned char **opt,
+ u32 *opt_len)
+{
+ struct ip_options *ip_opt = inet_sk(sock->sk)->opt;
+ unsigned char *cipso_ptr;
+
+ if (ip_opt == NULL || ip_opt->cipso == 0)
+ return -ENOMSG;
+ cipso_ptr = ip_opt->__data + ip_opt->cipso - sizeof(struct iphdr);
+
+ *opt = cipso_ptr;
+ *opt_len = cipso_ptr[1];
+
+ return 0;
+}
+
+/**
+ * cipso_v4_socket_getattr - Get the security attributes from a socket
+ * @sock: the socket
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Query @sock to see if there is a CIPSO option attached to the socket and if
+ * there is return the CIPSO security attributes in @secattr. Returns zero on
+ * success and negative values on failure.
+ *
+ */
+int cipso_v4_socket_getattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -ENOMSG;
+ struct sock *sk;
+ struct inet_sock *sk_inet;
+ unsigned char *cipso_ptr;
+ u32 doi;
+ struct cipso_v4_doi *doi_def;
+
+ sk = sock->sk;
+ lock_sock(sk);
+ sk_inet = inet_sk(sk);
+ if (sk_inet->opt == NULL || sk_inet->opt->cipso == 0)
+ goto socket_getattr_return;
+ cipso_ptr = sk_inet->opt->__data + sk_inet->opt->cipso -
+ sizeof(struct iphdr);
+ ret_val = cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr);
+ if (ret_val == 0)
+ goto socket_getattr_return;
+
+ doi = ntohl(*(u32 *)&cipso_ptr[2]);
+ rcu_read_lock();
+ doi_def = cipso_v4_doi_getdef(doi);
+ if (doi_def == NULL) {
+ rcu_read_unlock();
+ goto socket_getattr_return;
+ }
+ /* XXX - this code assumes only one tag per CIPSO option, the draft is
+ a little unclear about supporting multiple tags per option
+ but this seems safe for now */
+ switch (cipso_ptr[6]) {
+ case CIPSO_V4_TAG_RBITMAP:
+ ret_val = cipso_v4_parsetag_rbm(doi_def,
+ &cipso_ptr[6],
+ secattr);
+ break;
+ }
+ rcu_read_unlock();
+
+socket_getattr_return:
+ release_sock(sk);
+ return ret_val;
+}
+
+/**
+ * cipso_v4_skbuff_getattr - Get the security attributes from the CIPSO option
+ * @skb: the packet
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Parse the given packet's CIPSO option and return the security attributes.
+ * Returns zero on success and negative values on failure.
+ *
+ */
+int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
+ struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -ENOMSG;
+ unsigned char *cipso_ptr;
+ u32 doi;
+ struct cipso_v4_doi *doi_def;
+
+ if (!CIPSO_V4_OPTEXIST(skb))
+ return -ENOMSG;
+ cipso_ptr = CIPSO_V4_OPTPTR(skb);
+ if (cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr) == 0)
+ return 0;
+
+ doi = ntohl(*(u32 *)&cipso_ptr[2]);
+ rcu_read_lock();
+ doi_def = cipso_v4_doi_getdef(doi);
+ if (doi_def == NULL)
+ goto skbuff_getattr_return;
+ /* XXX - this code assumes only one tag per CIPSO option which isnt
+ really a good assumption to make but since we only support the MAC
+ tags right now it is a safe assumption */
+ switch (cipso_ptr[6]) {
+ case CIPSO_V4_TAG_RBITMAP:
+ ret_val = cipso_v4_parsetag_rbm(doi_def,
+ &cipso_ptr[6],
+ secattr);
+ break;
+ }
+
+skbuff_getattr_return:
+ rcu_read_unlock();
+ return ret_val;
+}
+
+/*
+ * Setup Functions
+ */
+
+/**
+ * cipso_v4_init - Initialize the CIPSO module
+ *
+ * Description:
+ * Initialize the CIPSO module and prepare it for use. Returns zero on success
+ * and negative values on failure.
+ *
+ */
+static int __init cipso_v4_init(void)
+{
+ return cipso_v4_cache_init(CIPSO_V4_CACHE_BUCKETS);
+}
+
+/* PM - not sure there is any point to this now, delete this func? */
+/**
+ * cipso_v4_exit - Cleanup after the CIPSO module
+ *
+ * Description:
+ * Perform any necessary cleanup tasks before the CIPSO module goes away.
+ *
+ */
+static void __exit cipso_v4_exit(void)
+{
+ cipso_v4_cache_destroy();
+}
+
+subsys_initcall(cipso_v4_init);
+
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 5/8] NetLabel: SELinux support
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-selinux-2.6.17 --]
[-- Type: text/plain, Size: 44745 bytes --]
Add NetLabel support to the SELinux LSM and modify the socket_post_create() LSM
hook to return an error code. The most significant part of this patch is the
addition of NetLabel hooks into the following SELinux LSM hooks:
* selinux_file_permission()
* selinux_socket_sendmsg()
* selinux_socket_post_create()
* selinux_socket_post_accept() [NEW]
* selinux_socket_sock_rcv_skb()
* selinux_socket_getpeersec_stream()
* selinux_socket_getpeersec_dgram()
The basic reasoning behind this patch is that outgoing packets are "NetLabel'd"
by labeling their socket and the NetLabel security attributes are checked via
the additional hook in selinux_socket_sock_rcv_skb(). NetLabel itself is only
a labeling mechanism, similar to filesystem extended attributes, it is up to
the SELinux enforcement mechanism to perform the actual access checks.
In addition to the changes outlined above this patch also includes some changes
to the extended bitmap (ebitmap) and multi-level security (mls) code to import
and export SELinux TE/MLS attributes into and out of NetLabel.
---
include/linux/security.h | 25 -
net/socket.c | 13
security/dummy.c | 6
security/selinux/hooks.c | 59 ++
security/selinux/include/objsec.h | 11
security/selinux/include/selinux_netlabel.h | 94 ++++
security/selinux/ss/Makefile | 1
security/selinux/ss/ebitmap.c | 155 +++++++
security/selinux/ss/ebitmap.h | 6
security/selinux/ss/mls.c | 160 +++++++
security/selinux/ss/mls.h | 25 +
security/selinux/ss/selinux_netlabel.c | 574 ++++++++++++++++++++++++++++
security/selinux/ss/services.c | 12
security/selinux/ss/services.h | 2
14 files changed, 1113 insertions(+), 30 deletions(-)
Index: linux-2.6.17.i686-quilt/include/linux/security.h
===================================================================
--- linux-2.6.17.i686-quilt.orig/include/linux/security.h
+++ linux-2.6.17.i686-quilt/include/linux/security.h
@@ -1267,8 +1267,8 @@ struct security_operations {
int (*unix_may_send) (struct socket * sock, struct socket * other);
int (*socket_create) (int family, int type, int protocol, int kern);
- void (*socket_post_create) (struct socket * sock, int family,
- int type, int protocol, int kern);
+ int (*socket_post_create) (struct socket * sock, int family,
+ int type, int protocol, int kern);
int (*socket_bind) (struct socket * sock,
struct sockaddr * address, int addrlen);
int (*socket_connect) (struct socket * sock,
@@ -2677,13 +2677,13 @@ static inline int security_socket_create
return security_ops->socket_create(family, type, protocol, kern);
}
-static inline void security_socket_post_create(struct socket * sock,
- int family,
- int type,
- int protocol, int kern)
+static inline int security_socket_post_create(struct socket * sock,
+ int family,
+ int type,
+ int protocol, int kern)
{
- security_ops->socket_post_create(sock, family, type,
- protocol, kern);
+ return security_ops->socket_post_create(sock, family, type,
+ protocol, kern);
}
static inline int security_socket_bind(struct socket * sock,
@@ -2809,11 +2809,12 @@ static inline int security_socket_create
return 0;
}
-static inline void security_socket_post_create(struct socket * sock,
- int family,
- int type,
- int protocol, int kern)
+static inline int security_socket_post_create(struct socket * sock,
+ int family,
+ int type,
+ int protocol, int kern)
{
+ return 0;
}
static inline int security_socket_bind(struct socket * sock,
Index: linux-2.6.17.i686-quilt/net/socket.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/net/socket.c
+++ linux-2.6.17.i686-quilt/net/socket.c
@@ -976,11 +976,18 @@ int sock_create_lite(int family, int typ
goto out;
}
- security_socket_post_create(sock, family, type, protocol, 1);
sock->type = type;
+ err = security_socket_post_create(sock, family, type, protocol, 1);
+ if (err)
+ goto out_release;
+
out:
*res = sock;
return err;
+out_release:
+ sock_release(sock);
+ sock = NULL;
+ goto out;
}
/* No kernel lock held - perfect */
@@ -1218,7 +1225,9 @@ static int __sock_create(int family, int
*/
module_put(net_families[family]->owner);
*res = sock;
- security_socket_post_create(sock, family, type, protocol, kern);
+ err = security_socket_post_create(sock, family, type, protocol, kern);
+ if (err)
+ goto out_release;
out:
net_family_read_unlock();
Index: linux-2.6.17.i686-quilt/security/dummy.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/dummy.c
+++ linux-2.6.17.i686-quilt/security/dummy.c
@@ -692,10 +692,10 @@ static int dummy_socket_create (int fami
return 0;
}
-static void dummy_socket_post_create (struct socket *sock, int family, int type,
- int protocol, int kern)
+static int dummy_socket_post_create (struct socket *sock, int family, int type,
+ int protocol, int kern)
{
- return;
+ return 0;
}
static int dummy_socket_bind (struct socket *sock, struct sockaddr *address,
Index: linux-2.6.17.i686-quilt/security/selinux/hooks.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/hooks.c
+++ linux-2.6.17.i686-quilt/security/selinux/hooks.c
@@ -12,6 +12,8 @@
* Copyright (C) 2003 Red Hat, Inc., James Morris <jmorris@redhat.com>
* Copyright (C) 2004-2005 Trusted Computer Solutions, Inc.
* <dgoeddel@trustedcs.com>
+ * Copyright (C) 2006 Hewlett-Packard Development Company, L.P.
+ * Paul Moore, <paul.moore@hp.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2,
@@ -74,6 +76,7 @@
#include "objsec.h"
#include "netif.h"
#include "xfrm.h"
+#include "selinux_netlabel.h"
#define XATTR_SELINUX_SUFFIX "selinux"
#define XATTR_NAME_SELINUX XATTR_SECURITY_PREFIX XATTR_SELINUX_SUFFIX
@@ -2293,6 +2296,7 @@ static int selinux_inode_listsecurity(st
static int selinux_file_permission(struct file *file, int mask)
{
+ int rc;
struct inode *inode = file->f_dentry->d_inode;
if (!mask) {
@@ -2304,8 +2308,12 @@ static int selinux_file_permission(struc
if ((file->f_flags & O_APPEND) && (mask & MAY_WRITE))
mask |= MAY_APPEND;
- return file_has_perm(current, file,
- file_mask_to_av(inode->i_mode, mask));
+ rc = file_has_perm(current, file,
+ file_mask_to_av(inode->i_mode, mask));
+ if (rc)
+ return rc;
+
+ return selinux_netlbl_inode_permission(inode, mask);
}
static int selinux_file_alloc_security(struct file *file)
@@ -2922,8 +2930,8 @@ out:
return err;
}
-static void selinux_socket_post_create(struct socket *sock, int family,
- int type, int protocol, int kern)
+static int selinux_socket_post_create(struct socket *sock, int family,
+ int type, int protocol, int kern)
{
struct inode_security_struct *isec;
struct task_security_struct *tsec;
@@ -2935,7 +2943,7 @@ static void selinux_socket_post_create(s
isec->sid = kern ? SECINITSID_KERNEL : tsec->sid;
isec->initialized = 1;
- return;
+ return selinux_netlbl_socket_create(sock, family, isec->sid);
}
/* Range of port numbers used to automatically bind.
@@ -3113,10 +3121,24 @@ static int selinux_socket_accept(struct
return 0;
}
+#ifdef CONFIG_NETLABEL
+static void selinux_socket_post_accept(struct socket *sock,
+ struct socket *newsock)
+{
+ selinux_netlbl_socket_accept(sock, newsock);
+}
+#endif /* CONFIG_NETLABEL */
+
static int selinux_socket_sendmsg(struct socket *sock, struct msghdr *msg,
int size)
{
- return socket_has_perm(current, sock, SOCKET__WRITE);
+ int rc;
+
+ rc = socket_has_perm(current, sock, SOCKET__WRITE);
+ if (rc)
+ return rc;
+
+ return selinux_netlbl_inode_permission(SOCK_INODE(sock), MAY_WRITE);
}
static int selinux_socket_recvmsg(struct socket *sock, struct msghdr *msg,
@@ -3306,10 +3328,15 @@ static int selinux_socket_sock_rcv_skb(s
err = avc_has_perm(sock_sid, port_sid,
sock_class, recv_perm, &ad);
+ if (err)
+ goto out;
}
- if (!err)
- err = selinux_xfrm_sock_rcv_skb(sock_sid, skb);
+ err = selinux_netlbl_sock_rcv_skb(sock_class, sock_sid, skb, &ad);
+ if (err)
+ goto out;
+
+ err = selinux_xfrm_sock_rcv_skb(sock_sid, skb);
out:
return err;
@@ -3333,8 +3360,9 @@ static int selinux_socket_getpeersec_str
peer_sid = ssec->peer_sid;
}
else if (isec->sclass == SECCLASS_TCP_SOCKET) {
- peer_sid = selinux_socket_getpeer_stream(sock->sk);
-
+ err = selinux_netlbl_socket_getpeersec_stream(sock, &peer_sid);
+ if (err || peer_sid == SECSID_NULL)
+ peer_sid = selinux_socket_getpeer_stream(sock->sk);
if (peer_sid == SECSID_NULL) {
err = -ENOPROTOOPT;
goto out;
@@ -3369,8 +3397,12 @@ out:
static int selinux_socket_getpeersec_dgram(struct sk_buff *skb, char **secdata, u32 *seclen)
{
- int err = 0;
- u32 peer_sid = selinux_socket_getpeer_dgram(skb);
+ int err;
+ u32 peer_sid;
+
+ err = selinux_netlbl_socket_getpeersec_dgram(skb, &peer_sid);
+ if (err || peer_sid == SECSID_NULL)
+ peer_sid = selinux_socket_getpeer_dgram(skb);
if (peer_sid == SECSID_NULL)
return -EINVAL;
@@ -4353,6 +4385,9 @@ static struct security_operations selinu
.socket_connect = selinux_socket_connect,
.socket_listen = selinux_socket_listen,
.socket_accept = selinux_socket_accept,
+#ifdef CONFIG_NETLABEL
+ .socket_post_accept = selinux_socket_post_accept,
+#endif
.socket_sendmsg = selinux_socket_sendmsg,
.socket_recvmsg = selinux_socket_recvmsg,
.socket_getsockname = selinux_socket_getsockname,
Index: linux-2.6.17.i686-quilt/security/selinux/include/objsec.h
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/include/objsec.h
+++ linux-2.6.17.i686-quilt/security/selinux/include/objsec.h
@@ -35,6 +35,14 @@ struct task_security_struct {
u32 ptrace_sid; /* SID of ptrace parent */
};
+struct netlbl_security_struct {
+ u32 netlbl_sid; /* SID used to set the NetLabel */
+ u32 peer_sid; /* SID of the connected peer */
+ u32 req_netlbl:1, /* socket requires a NetLabel label */
+ labeled:1, /* socket is labeled with NetLabel */
+ __unused:30;
+};
+
struct inode_security_struct {
struct inode *inode; /* back pointer to inode object */
struct list_head list; /* list of inode_security_struct */
@@ -44,6 +52,9 @@ struct inode_security_struct {
unsigned char initialized; /* initialization flag */
struct semaphore sem;
unsigned char inherit; /* inherit SID from parent entry */
+#ifdef CONFIG_NETLABEL
+ struct netlbl_security_struct netlbl;
+#endif
};
struct file_security_struct {
Index: linux-2.6.17.i686-quilt/security/selinux/include/selinux_netlabel.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/security/selinux/include/selinux_netlabel.h
@@ -0,0 +1,94 @@
+/*
+ * SELinux interface to the NetLabel subsystem
+ *
+ * Author : Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _SELINUX_NETLABEL_H_
+#define _SELINUX_NETLABEL_H_
+
+#ifdef CONFIG_NETLABEL
+void selinux_netlbl_cache_invalidate(void);
+
+int selinux_netlbl_inode_permission(struct inode *inode, int mask);
+
+int selinux_netlbl_socket_create(struct socket *sock,
+ const int sock_family,
+ u32 sid);
+void selinux_netlbl_socket_accept(struct socket *sock, struct socket *newsock);
+int selinux_netlbl_sock_rcv_skb(const u16 sock_class,
+ const u32 sock_sid,
+ struct sk_buff *skb,
+ struct avc_audit_data *ad);
+int selinux_netlbl_socket_getpeersec_stream(struct socket *sock,
+ u32 *peer_sid);
+int selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb, u32 *peer_sid);
+#else
+static inline void selinux_netlbl_cache_invalidate(void)
+{
+ return;
+}
+
+static inline int selinux_netlbl_inode_permission(struct inode *inode,
+ int mask)
+{
+ return 0;
+}
+
+static inline int selinux_netlbl_socket_create(struct socket *sock,
+ const int sock_family,
+ u32 sid)
+{
+ return 0;
+}
+
+static inline void selinux_netlbl_socket_accept(struct socket *sock,
+ struct socket *newsock)
+{
+ return;
+}
+
+static inline int selinux_netlbl_sock_rcv_skb(const u16 sock_class,
+ const u32 sock_sid,
+ struct sk_buff *skb,
+ struct avc_audit_data *ad)
+{
+ return 0;
+}
+
+static inline int selinux_netlbl_socket_getpeersec_stream(struct socket *sock,
+ u32 *peer_sid)
+{
+ *peer_sid = SECSID_NULL;
+ return 0;
+}
+
+static inline int selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb,
+ u32 *peer_sid)
+{
+ *peer_sid = SECSID_NULL;
+ return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+#endif
Index: linux-2.6.17.i686-quilt/security/selinux/ss/Makefile
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/ss/Makefile
+++ linux-2.6.17.i686-quilt/security/selinux/ss/Makefile
@@ -7,3 +7,4 @@ obj-y := ss.o
ss-y := ebitmap.o hashtab.o symtab.o sidtab.o avtab.o policydb.o services.o conditional.o mls.o
+obj-$(CONFIG_NETLABEL) += selinux_netlabel.o
Index: linux-2.6.17.i686-quilt/security/selinux/ss/ebitmap.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/ss/ebitmap.c
+++ linux-2.6.17.i686-quilt/security/selinux/ss/ebitmap.c
@@ -3,6 +3,14 @@
*
* Author : Stephen Smalley, <sds@epoch.ncsc.mil>
*/
+/*
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ * Added ebitmap_export() and ebitmap_import()
+ *
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ */
+
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/errno.h>
@@ -59,6 +67,153 @@ int ebitmap_cpy(struct ebitmap *dst, str
return 0;
}
+/**
+ * ebitmap_export - Export an ebitmap to a unsigned char bitmap string
+ * @src: the ebitmap to export
+ * @dst: the resulting bitmap string
+ * @dst_len: length of dst in bytes
+ *
+ * Description:
+ * Allocate a buffer at least src->highbit bits long and export the extensible
+ * bitmap into the buffer. The bitmap string will be in little endian format,
+ * i.e. LSB first. The value returned in dst_len may not the true size of the
+ * buffer as the length of the buffer is rounded up to a multiple of MAPTYPE.
+ * The caller must free the buffer when finished. Returns zero on success,
+ * negative values on failure.
+ *
+ */
+int ebitmap_export(const struct ebitmap *src,
+ unsigned char **dst,
+ u32 *dst_len)
+{
+ u32 bitmap_len;
+ unsigned char *bitmap;
+ struct ebitmap_node *iter_node;
+ MAPTYPE node_val;
+ u32 bitmap_byte;
+ unsigned char bitmask;
+
+ if (src == NULL || dst == NULL || dst_len == NULL)
+ return -EINVAL;
+
+ *dst = NULL;
+ *dst_len = 0;
+
+ bitmap_len = src->highbit / 8;
+ if (src->highbit % 8 > 0)
+ bitmap_len += 1;
+ if (bitmap_len == 0)
+ return -EINVAL;
+
+ bitmap = kzalloc(bitmap_len + sizeof(MAPTYPE) -
+ (bitmap_len % sizeof(MAPTYPE)),
+ GFP_ATOMIC);
+ if (bitmap == NULL)
+ return -ENOMEM;
+
+ /* PM - there _has_ to be a faster way to do this, work on this more */
+ iter_node = src->node;
+ do {
+ bitmap_byte = iter_node->startbit / 8;
+ bitmask = 0x80;
+ node_val = iter_node->map;
+ do {
+ if (bitmask == 0) {
+ bitmap_byte++;
+ bitmask = 0x80;
+ }
+ if (node_val & (MAPTYPE)0x01)
+ bitmap[bitmap_byte] |= bitmask;
+ node_val >>= 1;
+ bitmask >>= 1;
+ } while (node_val > 0);
+ iter_node = iter_node->next;
+ } while (iter_node);
+
+ *dst = bitmap;
+ *dst_len = bitmap_len;
+ return 0;
+}
+
+/**
+ * ebitmap_import - Import an unsigned char bitmap string into an ebitmap
+ * @src: the bitmap string
+ * @src_len: the bitmap length in bytes
+ * @dst: the empty ebitmap
+ *
+ * Description:
+ * This function takes a little endian bitmap string in src and imports it into
+ * the ebitmap pointed to by dst. Returns zero on success, negative values on
+ * failure.
+ *
+ */
+int ebitmap_import(const unsigned char *src,
+ const u32 src_len,
+ struct ebitmap *dst)
+{
+ u32 src_off = 0;
+ struct ebitmap_node *node_new;
+ struct ebitmap_node *node_last = NULL;
+ u32 src_rem = src_len;
+ MAPTYPE tmp_val;
+ u32 iter;
+ u32 iter_bit;
+ unsigned char src_byte;
+
+ if (src == NULL || dst == NULL || src_len == 0)
+ return -EINVAL;
+
+ do {
+ node_new = kzalloc(sizeof(*node_new), GFP_ATOMIC);
+ if (node_new == NULL) {
+ ebitmap_destroy(dst);
+ return -ENOMEM;
+ }
+
+ /* PM - there _has_ to be a faster way to do this,
+ work on this more */
+ if (src_rem >= sizeof(MAPTYPE))
+ iter = sizeof(MAPTYPE);
+ else
+ iter = src_rem;
+ tmp_val = 0;
+ while (iter > 0) {
+ src_byte = src[src_off + --iter];
+ if (src_byte > 0)
+ for (iter_bit = 0; iter_bit < 8; iter_bit++) {
+ tmp_val <<= 1;
+ tmp_val |= src_byte & 0x01;
+ src_byte >>= 1;
+ }
+ else
+ tmp_val <<= 8;
+ }
+ node_new->map = tmp_val;
+ node_new->startbit = src_off * 8;
+
+ if (node_last != NULL)
+ node_last->next = node_new;
+ else
+ dst->node = node_new;
+ node_last = node_new;
+
+ if (src_rem >= sizeof(MAPTYPE)) {
+ src_off += sizeof(MAPTYPE);
+ src_rem -= sizeof(MAPTYPE);
+ } else
+ src_off += src_rem;
+ } while (src_off < src_len);
+
+ tmp_val = node_last->map;
+ dst->highbit = node_last->startbit;
+ while (tmp_val >= 1) {
+ dst->highbit += 1;
+ tmp_val >>= 1;
+ }
+
+ return 0;
+}
+
int ebitmap_contains(struct ebitmap *e1, struct ebitmap *e2)
{
struct ebitmap_node *n1, *n2;
Index: linux-2.6.17.i686-quilt/security/selinux/ss/ebitmap.h
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/ss/ebitmap.h
+++ linux-2.6.17.i686-quilt/security/selinux/ss/ebitmap.h
@@ -69,6 +69,12 @@ static inline int ebitmap_node_get_bit(s
int ebitmap_cmp(struct ebitmap *e1, struct ebitmap *e2);
int ebitmap_cpy(struct ebitmap *dst, struct ebitmap *src);
+int ebitmap_export(const struct ebitmap *src,
+ unsigned char **dst,
+ u32 *dst_len);
+int ebitmap_import(const unsigned char *src,
+ const u32 src_len,
+ struct ebitmap *dst);
int ebitmap_contains(struct ebitmap *e1, struct ebitmap *e2);
int ebitmap_get_bit(struct ebitmap *e, unsigned long bit);
int ebitmap_set_bit(struct ebitmap *e, unsigned long bit, int value);
Index: linux-2.6.17.i686-quilt/security/selinux/ss/mls.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/ss/mls.c
+++ linux-2.6.17.i686-quilt/security/selinux/ss/mls.c
@@ -10,6 +10,13 @@
*
* Copyright (C) 2004-2006 Trusted Computer Solutions, Inc.
*/
+/*
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ * Added support to import/export the MLS label
+ *
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ */
#include <linux/kernel.h>
#include <linux/slab.h>
@@ -585,3 +592,156 @@ int mls_compute_sid(struct context *scon
return -EINVAL;
}
+/**
+ * mls_export_lvl - Export the MLS sensitivity levels
+ * @context: the security context
+ * @lvl_low: the low sensitivity level
+ * @lvl_high: the high sensitivity level
+ *
+ * Description:
+ * Given the security context copy the low MLS sensitivity level into lvl_low
+ * and the high sensitivity level in lvl_high. The MLS levels are only
+ * exported if the pointers are not NULL, if they are NULL then that level is
+ * not exported. Returns zero on success, negative values on failure.
+ *
+ */
+int mls_export_lvl(const struct context *context, u32 *lvl_low, u32 *lvl_high)
+{
+ if (!selinux_mls_enabled)
+ return 0;
+
+ if (lvl_low != NULL)
+ *lvl_low = context->range.level[0].sens - 1;
+ if (lvl_high != NULL)
+ *lvl_high = context->range.level[1].sens - 1;
+
+ return 0;
+}
+
+/**
+ * mls_import_lvl - Import the MLS sensitivity levels
+ * @context: the security context
+ * @lvl_low: the low sensitivity level
+ * @lvl_high: the high sensitivity level
+ *
+ * Description:
+ * Given the security context and the two sensitivty levels, set the MLS levels
+ * in the context according the two given as parameters. Returns zero on
+ * success, negative values on failure.
+ *
+ */
+int mls_import_lvl(struct context *context,
+ const u32 lvl_low,
+ const u32 lvl_high)
+{
+ if (!selinux_mls_enabled)
+ return 0;
+
+ context->range.level[0].sens = lvl_low + 1;
+ context->range.level[1].sens = lvl_high + 1;
+
+ return 0;
+}
+
+/**
+ * mls_export_cat - Export the MLS categories
+ * @context: the security context
+ * @cat_low: the low category
+ * @cat_low_len: length of the cat_low bitmap in bytes
+ * @cat_high: the high category
+ * @cat_high_len: length of the cat_high bitmap in bytes
+ *
+ * Description:
+ * Given the security context export the low MLS category bitmap into cat_low
+ * and the high category bitmap into cat_high. The MLS categories are only
+ * exported if the pointers are not NULL, if they are NULL then that level is
+ * not exported. The caller is responsibile for freeing the memory when
+ * finished. Returns zero on success, negative values on failure.
+ *
+ */
+int mls_export_cat(const struct context *context,
+ unsigned char **cat_low,
+ u32 *cat_low_len,
+ unsigned char **cat_high,
+ u32 *cat_high_len)
+{
+ int ret_val = -EPERM;
+
+ if (!selinux_mls_enabled)
+ return 0;
+
+ if (cat_low != NULL && cat_low_len != NULL) {
+ ret_val = ebitmap_export(&context->range.level[0].cat,
+ cat_low,
+ cat_low_len);
+ if (ret_val != 0)
+ goto export_cat_failure;
+ }
+ if (cat_high != NULL && cat_high_len != NULL) {
+ ret_val = ebitmap_export(&context->range.level[1].cat,
+ cat_high,
+ cat_high_len);
+ if (ret_val != 0)
+ goto export_cat_failure;
+ }
+
+ return 0;
+
+export_cat_failure:
+ if (cat_low != NULL && *cat_low != NULL)
+ kfree(*cat_low);
+ if (cat_high != NULL && *cat_high != NULL)
+ kfree(*cat_high);
+ return ret_val;
+}
+
+/**
+ * mls_import_cat - Import the MLS categories
+ * @context: the security context
+ * @cat_low: the low category
+ * @cat_low_len: length of the cat_low bitmap in bytes
+ * @cat_high: the high category
+ * @cat_high_len: length of the cat_high bitmap in bytes
+ *
+ * Description:
+ * Given the security context and the two category bitmap strings import the
+ * categories into the security context. The MLS categories are only imported
+ * if the pointers are not NULL, if they are NULL they are skipped. Returns
+ * zero on success, negative values on failure.
+ *
+ */
+int mls_import_cat(struct context *context,
+ const unsigned char *cat_low,
+ const u32 cat_low_len,
+ const unsigned char *cat_high,
+ const u32 cat_high_len)
+{
+ int ret_val = -EPERM;
+
+ if (!selinux_mls_enabled)
+ return 0;
+
+ if (cat_low != NULL && cat_low_len > 0) {
+ ret_val = ebitmap_import(cat_low,
+ cat_low_len,
+ &context->range.level[0].cat);
+ if (ret_val != 0)
+ goto import_cat_failure;
+ }
+ if (cat_high != NULL && cat_high_len > 0) {
+ ret_val = ebitmap_import(cat_high,
+ cat_high_len,
+ &context->range.level[1].cat);
+ if (ret_val != 0)
+ goto import_cat_failure;
+ }
+
+ return 0;
+
+import_cat_failure:
+ if (cat_low)
+ ebitmap_destroy(&context->range.level[0].cat);
+ if (cat_high)
+ ebitmap_destroy(&context->range.level[1].cat);
+ return ret_val;
+}
Index: linux-2.6.17.i686-quilt/security/selinux/ss/mls.h
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/ss/mls.h
+++ linux-2.6.17.i686-quilt/security/selinux/ss/mls.h
@@ -10,6 +10,13 @@
*
* Copyright (C) 2004-2006 Trusted Computer Solutions, Inc.
*/
+/*
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ * Added support to import/export the MLS label
+ *
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ */
#ifndef _SS_MLS_H_
#define _SS_MLS_H_
@@ -42,5 +49,23 @@ int mls_compute_sid(struct context *scon
int mls_setup_user_range(struct context *fromcon, struct user_datum *user,
struct context *usercon);
+int mls_export_lvl(const struct context *context,
+ u32 *lvl_low,
+ u32 *lvl_high);
+int mls_import_lvl(struct context *context,
+ const u32 lvl_low,
+ const u32 lvl_high);
+
+int mls_export_cat(const struct context *context,
+ unsigned char **cat_low,
+ u32 *cat_low_len,
+ unsigned char **cat_high,
+ u32 *cat_high_len);
+int mls_import_cat(struct context *context,
+ const unsigned char *cat_low,
+ const u32 cat_low_len,
+ const unsigned char *cat_high,
+ const u32 cat_high_len);
+
#endif /* _SS_MLS_H */
Index: linux-2.6.17.i686-quilt/security/selinux/ss/selinux_netlabel.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/security/selinux/ss/selinux_netlabel.c
@@ -0,0 +1,574 @@
+/*
+ * SELinux interface to the NetLabel subsystem
+ *
+ * Author : Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/audit.h>
+#include <net/sock.h>
+#include <net/netlabel.h>
+
+#include "avc.h"
+#include "context.h"
+#include "sidtab.h"
+#include "services.h"
+#include "mls.h"
+#include "objsec.h"
+
+/*
+ * This is the structure we store inside the NetLabel cache block.
+ */
+#define NETLBL_CACHE(x) ((struct netlbl_cache *)(x))
+#define NETLBL_CACHE_T_NONE 0
+#define NETLBL_CACHE_T_SID 1
+#define NETLBL_CACHE_T_MLSLBL 2
+struct netlbl_cache {
+ u32 type;
+ union {
+ u32 sid;
+ struct mls_level mls_label;
+ } data;
+};
+
+/**
+ * selinux_netlbl_cache_free - Free the NetLabel cached data
+ * @data: the data to free
+ *
+ * Description:
+ * This function is intended to be used as the free() callback inside the
+ * netlbl_lsm_cache structure.
+ *
+ */
+static void selinux_netlbl_cache_free(const void *data)
+{
+ struct netlbl_cache *cache = NETLBL_CACHE(data);
+ switch (cache->type) {
+ case NETLBL_CACHE_T_MLSLBL:
+ ebitmap_destroy(&cache->data.mls_label.cat);
+ break;
+ }
+ kfree(data);
+}
+
+/**
+ * selinux_netlbl_cache_add - Add an entry to the NetLabel cache
+ * @skb: the packet
+ * @ctx: the SELinux context
+ *
+ * Description:
+ * Attempt to cache the context in @ctx, which was derived from the packet in
+ * @skb, in the NetLabel subsystem cache.
+ *
+ */
+static void selinux_netlbl_cache_add(struct sk_buff *skb, struct context *ctx)
+{
+ struct netlbl_cache *cache = NULL;
+ struct netlbl_lsm_secattr secattr;
+
+ netlbl_secattr_init(&secattr);
+
+ cache = kzalloc(sizeof(*cache), GFP_ATOMIC);
+ if (cache == NULL)
+ goto netlbl_cache_add_return;
+ secattr.cache.free = selinux_netlbl_cache_free;
+ secattr.cache.data = (void *)cache;
+ secattr.set_cache = 1;
+
+ if (ebitmap_cpy(&cache->data.mls_label.cat,
+ &ctx->range.level[0].cat) != 0)
+ goto netlbl_cache_add_return;
+ cache->data.mls_label.sens = ctx->range.level[0].sens;
+ cache->type = NETLBL_CACHE_T_MLSLBL;
+
+ if (netlbl_cache_add(skb, &secattr) != 0)
+ goto netlbl_cache_add_return;
+ secattr.set_cache = 0;
+
+netlbl_cache_add_return:
+ netlbl_secattr_destroy(&secattr);
+}
+
+/**
+ * selinux_netlbl_cache_invalidate - Invalidate the NetLabel cache
+ *
+ * Description:
+ * Invalidate the NetLabel security attribute mapping cache.
+ *
+ */
+void selinux_netlbl_cache_invalidate(void)
+{
+ netlbl_cache_invalidate();
+}
+
+/**
+ * selinux_netlbl_domain_export - Exports the type of a given context
+ * @context: the security context
+ * @scontext: the resulting type string
+ * @scontext_len: the length of scontext including the NULL byte
+ *
+ * Description:
+ * Allocate a buffer for the type name specified in context and copy the type
+ * name into the buffer. The caller must free the buffer when finished.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_domain_export(const struct context *context,
+ char **scontext,
+ u32 *scontext_len)
+{
+ char *str;
+ u32 str_len;
+
+ *scontext = NULL;
+ if (scontext_len != NULL)
+ *scontext_len = 0;
+ str_len = strlen(policydb.p_type_val_to_name[context->type - 1]) + 1;
+ str = kmalloc(str_len, GFP_ATOMIC);
+ if (str == NULL)
+ return -ENOMEM;
+ strcpy(str, policydb.p_type_val_to_name[context->type - 1]);
+
+ *scontext = str;
+ if (scontext_len != NULL)
+ *scontext_len = str_len;
+ return 0;
+}
+
+/**
+ * selinux_netlbl_secattr_to_sid - Convert a NetLabel secattr to a SELinux SID
+ * @skb: the network packet
+ * @secattr: the NetLabel packet security attributes
+ * @base_sid: the SELinux SID to use as a context for MLS only attributes
+ * @sid: the SELinux SID
+ *
+ * Description:
+ * Convert the given NetLabel packet security attributes in @secattr into a
+ * SELinux SID. If the @secattr field does not contain a full SELinux
+ * SID/context then use the context in @base_sid as the foundation. If @skb
+ * is not NULL attempt to cache as much data as possibile. Returns zero on
+ * success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb,
+ struct netlbl_lsm_secattr *secattr,
+ const u32 base_sid,
+ u32 *sid)
+{
+ int ret_val = -EIDRM;
+ struct context *ctx;
+ struct context ctx_new;
+ struct netlbl_cache *cache;
+ u32 ctx_new_destroy = 0;
+
+ if (secattr->set_cache) {
+ cache = NETLBL_CACHE(secattr->cache.data);
+ switch (cache->type) {
+ case NETLBL_CACHE_T_SID:
+ *sid = cache->data.sid;
+ break;
+ case NETLBL_CACHE_T_MLSLBL:
+ ctx = sidtab_search(&sidtab, base_sid);
+ if (ctx == NULL)
+ goto netlbl_secattr_to_sid_failure;
+ ret_val = context_cpy(&ctx_new, ctx);
+ if (ret_val != 0)
+ goto netlbl_secattr_to_sid_failure;
+ ctx_new_destroy = 1;
+ mls_context_destroy(&ctx_new);
+
+ ctx_new.range.level[0].sens =
+ cache->data.mls_label.sens;
+ ret_val = ebitmap_cpy(&ctx_new.range.level[0].cat,
+ &cache->data.mls_label.cat);
+ if (ret_val != 0)
+ goto netlbl_secattr_to_sid_failure;
+ ctx_new.range.level[1].sens =
+ cache->data.mls_label.sens;
+ ret_val = ebitmap_cpy(&ctx_new.range.level[1].cat,
+ &cache->data.mls_label.cat);
+ if (ret_val != 0)
+ goto netlbl_secattr_to_sid_failure;
+
+ ret_val = sidtab_context_to_sid(&sidtab,
+ &ctx_new,
+ sid);
+ break;
+ default:
+ goto netlbl_secattr_to_sid_failure;
+ }
+ } else if (secattr->set_mls_lvl) {
+ ctx = sidtab_search(&sidtab, base_sid);
+ if (ctx == NULL)
+ goto netlbl_secattr_to_sid_failure;
+ ret_val = context_cpy(&ctx_new, ctx);
+ if (ret_val != 0)
+ goto netlbl_secattr_to_sid_failure;
+ ctx_new_destroy = 1;
+ mls_context_destroy(&ctx_new);
+
+ if (mls_import_lvl(&ctx_new,
+ secattr->mls_lvl,
+ secattr->mls_lvl) != 0)
+ goto netlbl_secattr_to_sid_failure;
+ if (secattr->set_mls_cat) {
+ if (mls_import_cat(&ctx_new,
+ secattr->mls_cat,
+ secattr->mls_cat_len,
+ secattr->mls_cat,
+ secattr->mls_cat_len) != 0)
+ goto netlbl_secattr_to_sid_failure;
+ }
+
+ ret_val = mls_context_isvalid(&policydb, &ctx_new);
+ if (ret_val != 1)
+ goto netlbl_secattr_to_sid_failure;
+
+ if (skb != NULL)
+ selinux_netlbl_cache_add(skb, &ctx_new);
+
+ ret_val = sidtab_context_to_sid(&sidtab, &ctx_new, sid);
+ } else
+ *sid = SECINITSID_UNLABELED;
+
+ ret_val = 0;
+
+netlbl_secattr_to_sid_failure:
+ if (ctx_new_destroy)
+ context_destroy(&ctx_new);
+ return ret_val;
+}
+
+/**
+ * selinux_netlbl_skbuff_getsid - Get the sid of a packet using NetLabel
+ * @skb: the packet
+ * @base_sid: the SELinux SID to use as a context for MLS only attributes
+ * @sid: the SID
+ *
+ * Description:
+ * Call the NetLabel mechanism to get the security attributes of the given
+ * packet and use those attributes to determine the correct context/SID to
+ * assign to the packet. Returns zero on success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
+ const u32 base_sid,
+ u32 *sid)
+{
+ int ret_val;
+ struct netlbl_lsm_secattr secattr;
+
+ netlbl_secattr_init(&secattr);
+ ret_val = netlbl_skbuff_getattr(skb, &secattr);
+ if (ret_val == 0)
+ ret_val = selinux_netlbl_secattr_to_sid(skb,
+ &secattr,
+ base_sid,
+ sid);
+
+ secattr.set_cache = 0;
+ netlbl_secattr_destroy(&secattr);
+
+ return ret_val;
+}
+
+/**
+ * selinux_netlbl_socket_setsid - Label a socket using the NetLabel mechanism
+ * @sock: the socket to label
+ * @sock_family: the socket family
+ * @sid: the SID to use
+ *
+ * Description:
+ * Attempt to label a socket using the NetLabel mechanism using the given
+ * SID. Returns zero values on success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_socket_setsid(struct socket *sock,
+ const int sock_family,
+ u32 sid)
+{
+ int ret_val;
+ struct inode_security_struct *isec = SOCK_INODE(sock)->i_security;
+ struct netlbl_lsm_secattr secattr;
+ struct context *ctx;
+
+ ctx = sidtab_search(&sidtab, sid);
+ if (ctx != NULL) {
+ netlbl_secattr_init(&secattr);
+
+ if (selinux_netlbl_domain_export(ctx,
+ &secattr.domain,
+ NULL) == 0)
+ secattr.set_domain = 1;
+ if (mls_export_lvl(ctx, &secattr.mls_lvl, NULL) == 0)
+ secattr.set_mls_lvl = 1;
+ if (mls_export_cat(ctx,
+ &secattr.mls_cat,
+ &secattr.mls_cat_len,
+ NULL,
+ NULL) == 0)
+ secattr.set_mls_cat = 1;
+
+ ret_val = netlbl_socket_setattr(sock, &secattr);
+ if (ret_val == 0) {
+ isec->netlbl.netlbl_sid = sid;
+ isec->netlbl.labeled = 1;
+ }
+
+ netlbl_secattr_destroy(&secattr);
+ } else
+ ret_val = -ENOENT;
+
+ return ret_val;
+}
+
+/**
+ * selinux_netlbl_socket_peeksid - Get the SID of the first queued packet
+ * @sock: the socket to query
+ * @sid: the packet's SID
+ *
+ * Description:
+ * Examine the first incoming packet in the socket's queue and determine the
+ * packet's SELinux SID. Return zero on success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_socket_peeksid(struct socket *sock, u32 *sid)
+{
+ int ret_val;
+ struct netlbl_lsm_secattr secattr;
+ struct inode_security_struct *isec = SOCK_INODE(sock)->i_security;
+
+ netlbl_secattr_init(&secattr);
+ ret_val = netlbl_socket_peekattr(sock, &secattr);
+ if (ret_val == 0)
+ ret_val = selinux_netlbl_secattr_to_sid(NULL,
+ &secattr,
+ isec->sid,
+ sid);
+
+ secattr.set_cache = 0;
+ netlbl_secattr_destroy(&secattr);
+
+ return ret_val;
+}
+
+/**
+ * selinux_netlbl_socket_create - Label a socket using the NetLabel mechanism
+ * @sock: the socket to label
+ * @sock_family: the socket family
+ * @sid: the SID to use
+ *
+ * Description:
+ * Attempt to label a socket using the NetLabel mechanism using the given
+ * SID. Returns zero values on success, negative values on failure.
+ *
+ */
+int selinux_netlbl_socket_create(struct socket *sock,
+ const int sock_family,
+ u32 sid)
+{
+ struct inode_security_struct *isec;
+
+ if (!ss_initialized || sock_family != PF_INET)
+ return 0;
+
+ isec = SOCK_INODE(sock)->i_security;
+ isec->netlbl.req_netlbl = 1;
+
+ /* PM - should we audit failures, or at the very least do a printk()
+ so users are not left wondering? */
+ return selinux_netlbl_socket_setsid(sock, sock_family, sid);
+}
+
+/**
+ * selinux_netlbl_socket_accept - Handle the labeling of an accept()ed socket
+ * @sock: the original socket
+ * @newsock: the new accept()ed socket
+ *
+ * Description:
+ * Attempt to label a socket using the NetLabel mechanism based on the packets
+ * in the queue and the original socket's SID.
+ *
+ */
+void selinux_netlbl_socket_accept(struct socket *sock, struct socket *newsock)
+{
+ int ret_val;
+ struct inode_security_struct *newisec;
+ struct netlbl_lsm_secattr secattr;
+
+ if (!ss_initialized ||
+ newsock->sk == NULL || newsock->sk->sk_family != PF_INET)
+ return;
+
+ newisec = SOCK_INODE(newsock)->i_security;
+ newisec->netlbl.req_netlbl = 1;
+
+ netlbl_secattr_init(&secattr);
+ ret_val = netlbl_socket_getattr(newsock, &secattr);
+ if (ret_val == 0)
+ ret_val = selinux_netlbl_secattr_to_sid(NULL,
+ &secattr,
+ newisec->sid,
+ &newisec->netlbl.peer_sid);
+ if (ret_val != 0)
+ newisec->netlbl.peer_sid = SECSID_NULL;
+ secattr.set_cache = 0;
+ netlbl_secattr_destroy(&secattr);
+}
+
+/**
+ * selinux_netlbl_file_permission - Very the file/socket is NetLabel labeled
+ * @inode: the file descriptor's inode
+ * @mask: the permission mask
+ *
+ * Description:
+ * Looks at a file's inode and if it is marked as a socket protected by
+ * NetLabel then verify that the socket has been labeled, if not try to label
+ * the socket now with the task's SID. Returns zero on success, negative
+ * values on failure.
+ *
+ */
+int selinux_netlbl_inode_permission(struct inode *inode, int mask)
+{
+ int ret_val = 0;
+ struct inode_security_struct *isec = inode->i_security;
+ struct task_security_struct *tsec;
+ struct netlbl_security_struct *netlbl_sec = &isec->netlbl;
+ struct socket *sock;
+
+ if (netlbl_sec->req_netlbl && (mask & (MAY_WRITE | MAY_APPEND))) {
+ tsec = current->security;
+ if (netlbl_sec->labeled == 0) {
+ sock = SOCKET_I(inode);
+ ret_val = selinux_netlbl_socket_setsid(sock,
+ sock->sk->sk_family,
+ tsec->sid);
+ } else if (netlbl_sec->netlbl_sid != tsec->sid &&
+ netlbl_sec->netlbl_sid != SECINITSID_KERNEL)
+ ret_val = -EACCES;
+ }
+
+ return ret_val;
+}
+
+/**
+ * selinux_netlbl_sock_rcv_skb - Do an inbound access check using NetLabel
+ * @sock_class: the socket class
+ * @sock_sid: the socket's SID
+ * @skb: the incoming packet
+ * @ad: the audit data
+ *
+ * Description:
+ * Fetch the NetLabel security attributes from @skb and perform an access check
+ * against the receiving socket. Returns zero on success, negative values on
+ * error.
+ *
+ */
+int selinux_netlbl_sock_rcv_skb(const u16 sock_class,
+ const u32 sock_sid,
+ struct sk_buff *skb,
+ struct avc_audit_data *ad)
+{
+ int ret_val;
+ u32 netlbl_sid;
+ u32 recv_perm;
+
+ ret_val = selinux_netlbl_skbuff_getsid(skb, sock_sid, &netlbl_sid);
+ if (ret_val)
+ goto netlbl_sock_rcv_skb;
+
+ /* PM - at some point we want to do an unlabeled check here too but
+ we need a way to enable/disable NetLabel checks at runtime first,
+ see RH BZ #195238 for more details */
+ if (netlbl_sid != SECINITSID_UNLABELED) {
+ switch (sock_class) {
+ case SECCLASS_UDP_SOCKET:
+ recv_perm = UDP_SOCKET__RECV_MSG;
+ break;
+ case SECCLASS_TCP_SOCKET:
+ recv_perm = TCP_SOCKET__RECV_MSG;
+ break;
+ default:
+ recv_perm = RAWIP_SOCKET__RECV_MSG;
+ break;
+ }
+
+ ret_val = avc_has_perm(sock_sid,
+ netlbl_sid,
+ sock_class,
+ recv_perm,
+ ad);
+ if (ret_val)
+ netlbl_skbuff_err(skb, ret_val);
+ }
+
+netlbl_sock_rcv_skb:
+ return ret_val;
+}
+
+/**
+ * selinux_netlbl_socket_peersid - Return the peer SID of a connected socket
+ * @sock: the socket
+ * @peer_sid: the peer SID
+ *
+ * Description:
+ * Examine @sock to find the connected peer's SID, if that is not possibile
+ * check the socket's receive queue and take the SID from the incoming
+ * packets. Returns zero on success, negative values on error.
+ *
+ */
+int selinux_netlbl_socket_getpeersec_stream(struct socket *sock, u32 *peer_sid)
+{
+ int ret_val = 0;
+ struct inode_security_struct *isec = SOCK_INODE(sock)->i_security;
+
+ if (isec->netlbl.peer_sid == 0)
+ ret_val = selinux_netlbl_socket_peeksid(sock, peer_sid);
+ else
+ *peer_sid = isec->netlbl.peer_sid;
+
+ return ret_val;
+}
+
+/**
+ * selinux_netlbl_socket_getpeersec_dgram - Return the SID of a NetLabel packet
+ * @skb: the packet
+ * @peer_sid: the packet's SID
+ *
+ * Description:
+ * Examine @skb to find the SID assigned to it by NetLabel. Returns zero on
+ * success, negative values on error.
+ *
+ */
+int selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb, u32 *peer_sid)
+{
+ struct inode_security_struct *isec;
+
+ if (skb->sk == NULL || skb->sk->sk_socket == NULL)
+ return -ENOPROTOOPT;
+
+ isec = SOCK_INODE(skb->sk->sk_socket)->i_security;
+
+ return selinux_netlbl_skbuff_getsid(skb, isec->sid, peer_sid);
+}
Index: linux-2.6.17.i686-quilt/security/selinux/ss/services.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/ss/services.c
+++ linux-2.6.17.i686-quilt/security/selinux/ss/services.c
@@ -13,6 +13,11 @@
*
* Added conditional policy language extensions
*
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ * Added support for NetLabel
+ *
+ * Copyright (C) 2006 Hewlett-Packard Development Company, L.P.
* Copyright (C) 2004-2006 Trusted Computer Solutions, Inc.
* Copyright (C) 2003 - 2004 Tresys Technology, LLC
* Copyright (C) 2003 Red Hat, Inc., James Morris <jmorris@redhat.com>
@@ -29,6 +34,7 @@
#include <linux/sched.h>
#include <linux/audit.h>
#include <linux/mutex.h>
+#include <net/sock.h>
#include "flask.h"
#include "avc.h"
@@ -40,6 +46,8 @@
#include "services.h"
#include "conditional.h"
#include "mls.h"
+#include "objsec.h"
+#include "selinux_netlabel.h"
extern void selnl_notify_policyload(u32 seqno);
unsigned int policydb_loaded_version;
@@ -54,7 +62,7 @@ static DEFINE_MUTEX(load_mutex);
#define LOAD_LOCK mutex_lock(&load_mutex)
#define LOAD_UNLOCK mutex_unlock(&load_mutex)
-static struct sidtab sidtab;
+struct sidtab sidtab;
struct policydb policydb;
int ss_initialized = 0;
@@ -1241,6 +1249,7 @@ int security_load_policy(void *data, siz
selinux_complete_init();
avc_ss_reset(seqno);
selnl_notify_policyload(seqno);
+ selinux_netlbl_cache_invalidate();
return 0;
}
@@ -1295,6 +1304,7 @@ int security_load_policy(void *data, siz
avc_ss_reset(seqno);
selnl_notify_policyload(seqno);
+ selinux_netlbl_cache_invalidate();
return 0;
Index: linux-2.6.17.i686-quilt/security/selinux/ss/services.h
===================================================================
--- linux-2.6.17.i686-quilt.orig/security/selinux/ss/services.h
+++ linux-2.6.17.i686-quilt/security/selinux/ss/services.h
@@ -9,7 +9,9 @@
#include "policydb.h"
#include "sidtab.h"
+extern struct sidtab sidtab;
extern struct policydb policydb;
+extern int ss_initialized;
#endif /* _SS_SERVICES_H_ */
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 6/8] NetLabel: CIPSOv4 integration
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-cipsov4 --]
[-- Type: text/plain, Size: 17517 bytes --]
Add CIPSO/IPv4 support and management to the NetLabel subsystem. These changes
integrate the CIPSO/IPv4 configuration into the existing NetLabel code and
enable the use of CIPSO/IPv4 within the overall NetLabel framework.
---
net/netlabel/netlabel_cipso_v4.c | 634 +++++++++++++++++++++++++++++++++++++++
1 files changed, 634 insertions(+)
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_cipso_v4.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_cipso_v4.c
@@ -0,0 +1,634 @@
+/*
+ * NetLabel CIPSO/IPv4 Support
+ *
+ * This file defines the CIPSO/IPv4 functions for the NetLabel system. The
+ * NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/string.h>
+#include <linux/skbuff.h>
+#include <net/sock.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+#include <net/netlabel.h>
+#include <net/cipso_ipv4.h>
+
+#include "netlabel_user.h"
+#include "netlabel_cipso_v4.h"
+
+/* NetLabel Generic NETLINK CIPSOv4 family */
+static struct genl_family netlbl_cipsov4_gnl_family = {
+ .id = GENL_ID_GENERATE,
+ .hdrsize = 0,
+ .name = NETLBL_NLTYPE_CIPSOV4_NAME,
+ .version = NETLBL_PROTO_VERSION,
+ .maxattr = 0,
+};
+
+
+/*
+ * Local Prototypes
+ */
+
+static void netlbl_cipsov4_send_ack(const struct genl_info *info,
+ const u32 ret_code);
+
+
+/*
+ * Helper Functions
+ */
+
+/**
+ * netlbl_cipsov4_doi_free - Frees a CIPSO V4 DOI definition
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that the memory allocated to the DOI definition can be released
+ * safely.
+ *
+ */
+static void netlbl_cipsov4_doi_free(struct rcu_head *entry)
+{
+ struct cipso_v4_doi *ptr;
+
+ ptr = container_of(entry, struct cipso_v4_doi, rcu);
+ switch (ptr->type) {
+ case CIPSO_V4_MAP_STD:
+ if (ptr->map.std->lvl.cipso_size > 0)
+ kfree(ptr->map.std->lvl.cipso);
+ if (ptr->map.std->lvl.local_size > 0)
+ kfree(ptr->map.std->lvl.local);
+ if (ptr->map.std->cat.cipso_size > 0)
+ kfree(ptr->map.std->cat.cipso);
+ if (ptr->map.std->cat.local_size > 0)
+ kfree(ptr->map.std->cat.local);
+ break;
+ }
+ kfree(ptr);
+}
+
+
+/*
+ * NetLabel Command Handlers
+ */
+
+/**
+ * netlbl_cipsov4_add_std - Adds a CIPSO V4 DOI definition
+ * @doi: the DOI value
+ * @msg: the ADD message data
+ * @msg_size: the size of the ADD message buffer
+ *
+ * Description:
+ * Create a new CIPSO_V4_MAP_STD DOI definition based on the given ADD message
+ * and add it to the CIPSO V4 engine. Return zero on success and non-zero on
+ * error.
+ *
+ */
+static int netlbl_cipsov4_add_std(const u32 doi,
+ const unsigned char *msg,
+ const u32 msg_size)
+{
+ int ret_val = -EPERM;
+ unsigned char *msg_ptr = (unsigned char *)msg;
+ u32 msg_len = msg_size;
+ u32 num_tags;
+ u32 num_lvls;
+ u32 num_cats;
+ struct cipso_v4_doi *doi_def = NULL;
+ u32 iter;
+ u32 tmp_val_a;
+ u32 tmp_val_b;
+
+ if (msg_len < 4)
+ goto add_std_failure;
+ num_tags = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (num_tags == 0 || num_tags > CIPSO_V4_TAG_MAXCNT)
+ goto add_std_failure;
+
+ doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL);
+ if (doi_def == NULL) {
+ ret_val = -ENOMEM;
+ goto add_std_failure;
+ }
+ doi_def->map.std = kzalloc(sizeof(*doi_def->map.std),
+ GFP_KERNEL);
+ if (doi_def->map.std == NULL) {
+ ret_val = -ENOMEM;
+ goto add_std_failure;
+ }
+ doi_def->type = CIPSO_V4_MAP_STD;
+
+ if (msg_len < num_tags)
+ goto add_std_failure;
+ msg_len -= num_tags;
+ for (iter = 0; iter < num_tags; iter++) {
+ doi_def->tags[iter] = netlbl_getinc_u8(&msg_ptr);
+ switch (doi_def->tags[iter]) {
+ case CIPSO_V4_TAG_RBITMAP:
+ break;
+ default:
+ goto add_std_failure;
+ }
+ }
+ if (iter < CIPSO_V4_TAG_MAXCNT)
+ doi_def->tags[iter] = CIPSO_V4_TAG_INVALID;
+
+ if (msg_len < 24)
+ goto add_std_failure;
+
+ num_lvls = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (num_lvls == 0)
+ goto add_std_failure;
+ doi_def->map.std->lvl.local_size = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (doi_def->map.std->lvl.local_size > CIPSO_V4_MAX_LOC_LVLS)
+ goto add_std_failure;
+ doi_def->map.std->lvl.local = kcalloc(doi_def->map.std->lvl.local_size,
+ sizeof(u32), GFP_KERNEL);
+ if (doi_def->map.std->lvl.local == NULL) {
+ ret_val = -ENOMEM;
+ goto add_std_failure;
+ }
+ doi_def->map.std->lvl.cipso_size = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (doi_def->map.std->lvl.cipso_size > CIPSO_V4_MAX_REM_LVLS)
+ goto add_std_failure;
+ doi_def->map.std->lvl.cipso = kcalloc(doi_def->map.std->lvl.cipso_size,
+ sizeof(u32), GFP_KERNEL);
+ if (doi_def->map.std->lvl.cipso == NULL) {
+ ret_val = -ENOMEM;
+ goto add_std_failure;
+ }
+
+ num_cats = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ doi_def->map.std->cat.local_size = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (doi_def->map.std->cat.local_size > CIPSO_V4_MAX_LOC_CATS)
+ goto add_std_failure;
+ doi_def->map.std->cat.local = kcalloc(doi_def->map.std->cat.local_size,
+ sizeof(u32), GFP_KERNEL);
+ if (doi_def->map.std->cat.local == NULL) {
+ ret_val = -ENOMEM;
+ goto add_std_failure;
+ }
+ doi_def->map.std->cat.cipso_size = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (doi_def->map.std->cat.cipso_size > CIPSO_V4_MAX_REM_CATS)
+ goto add_std_failure;
+ doi_def->map.std->cat.cipso = kcalloc(doi_def->map.std->cat.cipso_size,
+ sizeof(u32), GFP_KERNEL);
+ if (doi_def->map.std->cat.cipso == NULL) {
+ ret_val = -ENOMEM;
+ goto add_std_failure;
+ }
+
+ if (msg_len < num_lvls * 8 + num_cats * 8)
+ goto add_std_failure;
+ msg_len -= num_lvls * 8 + num_cats * 8;
+
+ for (iter = 0; iter < doi_def->map.std->lvl.cipso_size; iter++)
+ doi_def->map.std->lvl.cipso[iter] = CIPSO_V4_INV_LVL;
+ for (iter = 0; iter < doi_def->map.std->lvl.local_size; iter++)
+ doi_def->map.std->lvl.local[iter] = CIPSO_V4_INV_LVL;
+ for (iter = 0; iter < doi_def->map.std->cat.cipso_size; iter++)
+ doi_def->map.std->cat.cipso[iter] = CIPSO_V4_INV_CAT;
+ for (iter = 0; iter < doi_def->map.std->cat.local_size; iter++)
+ doi_def->map.std->cat.local[iter] = CIPSO_V4_INV_CAT;
+
+ for (iter = 0; iter < num_lvls; iter++) {
+ tmp_val_a = netlbl_getinc_u32(&msg_ptr);
+ tmp_val_b = netlbl_getinc_u32(&msg_ptr);
+
+ if (tmp_val_a >= doi_def->map.std->lvl.local_size ||
+ tmp_val_b >= doi_def->map.std->lvl.cipso_size)
+ goto add_std_failure;
+
+ doi_def->map.std->lvl.cipso[tmp_val_b] = tmp_val_a;
+ doi_def->map.std->lvl.local[tmp_val_a] = tmp_val_b;
+ }
+
+ for (iter = 0; iter < num_cats; iter++) {
+ tmp_val_a = netlbl_getinc_u32(&msg_ptr);
+ tmp_val_b = netlbl_getinc_u32(&msg_ptr);
+
+ if (tmp_val_a >= doi_def->map.std->cat.local_size ||
+ tmp_val_b >= doi_def->map.std->cat.cipso_size)
+ goto add_std_failure;
+
+ doi_def->map.std->cat.cipso[tmp_val_b] = tmp_val_a;
+ doi_def->map.std->cat.local[tmp_val_a] = tmp_val_b;
+ }
+
+ doi_def->doi = doi;
+ ret_val = cipso_v4_doi_add(doi_def);
+ if (ret_val != 0)
+ goto add_std_failure;
+ return 0;
+
+add_std_failure:
+ if (doi_def)
+ netlbl_cipsov4_doi_free(&doi_def->rcu);
+ return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_add_pass - Adds a CIPSO V4 DOI definition
+ * @doi: the DOI value
+ * @msg: the ADD message data
+ * @msg_size: the size of the ADD message buffer
+ *
+ * Description:
+ * Create a new CIPSO_V4_MAP_PASS DOI definition based on the given ADD message
+ * and add it to the CIPSO V4 engine. Return zero on success and non-zero on
+ * error.
+ *
+ */
+static int netlbl_cipsov4_add_pass(const u32 doi,
+ const unsigned char *msg,
+ const u32 msg_size)
+{
+ int ret_val = -EPERM;
+ unsigned char *msg_ptr = (unsigned char *)msg;
+ u32 msg_len = msg_size;
+ u32 num_tags;
+ struct cipso_v4_doi *doi_def = NULL;
+ u32 iter;
+
+ if (msg_len < 4)
+ goto add_pass_failure;
+ num_tags = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (num_tags == 0 || num_tags > CIPSO_V4_TAG_MAXCNT)
+ goto add_pass_failure;
+
+ doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL);
+ if (doi_def == NULL) {
+ ret_val = -ENOMEM;
+ goto add_pass_failure;
+ }
+ doi_def->type = CIPSO_V4_MAP_PASS;
+
+ if (msg_len < num_tags)
+ goto add_pass_failure;
+ msg_len -= num_tags;
+ for (iter = 0; iter < num_tags; iter++) {
+ doi_def->tags[iter] = netlbl_getinc_u8(&msg_ptr);
+ switch (doi_def->tags[iter]) {
+ case CIPSO_V4_TAG_RBITMAP:
+ break;
+ default:
+ goto add_pass_failure;
+ }
+ }
+ if (iter < CIPSO_V4_TAG_MAXCNT)
+ doi_def->tags[iter] = CIPSO_V4_TAG_INVALID;
+
+ doi_def->doi = doi;
+ ret_val = cipso_v4_doi_add(doi_def);
+ if (ret_val != 0)
+ goto add_pass_failure;
+ return 0;
+
+add_pass_failure:
+ if (doi_def)
+ netlbl_cipsov4_doi_free(&doi_def->rcu);
+ return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_add - Handle an ADD message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Create a new DOI definition based on the given ADD message and add it to the
+ * CIPSO V4 engine. Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info)
+
+{
+ int ret_val = -EINVAL;
+ u32 doi;
+ u32 map_type;
+ u32 msg_len = netlbl_netlink_payload_len(skb);
+ unsigned char *msg = netlbl_netlink_payload_data(skb);
+
+ ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
+ if (ret_val != 0)
+ goto add_return;
+
+ if (msg_len < 8)
+ goto add_return;
+
+ doi = netlbl_getinc_u32(&msg);
+ map_type = netlbl_getinc_u32(&msg);
+ msg_len -= 8;
+ switch (map_type) {
+ case CIPSO_V4_MAP_STD:
+ ret_val = netlbl_cipsov4_add_std(doi, msg, msg_len);
+ break;
+ case CIPSO_V4_MAP_PASS:
+ ret_val = netlbl_cipsov4_add_pass(doi, msg, msg_len);
+ break;
+ }
+
+add_return:
+ netlbl_cipsov4_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_list - Handle a LIST message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated LIST message and respond accordingly. Returns
+ * zero on success and negative values on error.
+ *
+ */
+static int netlbl_cipsov4_list(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val = -EINVAL;
+ u32 doi;
+ unsigned char *msg = netlbl_netlink_payload_data(skb);
+ struct sk_buff *ans_skb;
+ unsigned char *buf_ptr;
+
+ if (netlbl_netlink_payload_len(skb) != 4)
+ goto list_failure;
+
+ doi = netlbl_getinc_u32(&msg);
+ ans_skb = cipso_v4_doi_dump(doi, NLMSG_SPACE(GENL_HDRLEN));
+ if (ans_skb == NULL) {
+ ret_val = -ENOMEM;
+ goto list_failure;
+ }
+ buf_ptr = skb_push(ans_skb, NLMSG_SPACE(GENL_HDRLEN));
+ __netlbl_netlink_hdr(buf_ptr,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_cipsov4_gnl_family.id,
+ NLBL_CIPSOV4_C_LIST,
+ ans_skb->len);
+
+ ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+ if (ret_val != 0)
+ goto list_failure;
+
+ return 0;
+
+list_failure:
+ netlbl_cipsov4_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_listall - Handle a LISTALL message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated LISTALL message and respond accordingly. Returns
+ * zero on success and negative values on error.
+ *
+ */
+static int netlbl_cipsov4_listall(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val = -EINVAL;
+ struct sk_buff *ans_skb;
+ unsigned char *buf_ptr;
+
+ ans_skb = cipso_v4_doi_dump(0, NLMSG_SPACE(GENL_HDRLEN));
+ if (ans_skb == NULL) {
+ ret_val = -ENOMEM;
+ goto list_failure;
+ }
+ buf_ptr = skb_push(ans_skb, NLMSG_SPACE(GENL_HDRLEN));
+ __netlbl_netlink_hdr(buf_ptr,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_cipsov4_gnl_family.id,
+ NLBL_CIPSOV4_C_LISTALL,
+ ans_skb->len);
+
+ ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+ if (ret_val != 0)
+ goto list_failure;
+
+ return 0;
+
+list_failure:
+ netlbl_cipsov4_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_remove - Handle a REMOVE message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated REMOVE message and respond accordingly. Returns
+ * zero on success, negative values on failure.
+ *
+ */
+static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val;
+ u32 doi;
+ unsigned char *msg = netlbl_netlink_payload_data(skb);
+
+ ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
+ if (ret_val != 0)
+ goto remove_return;
+
+ if (netlbl_netlink_payload_len(skb) != 4) {
+ ret_val = -EINVAL;
+ goto remove_return;
+ }
+
+ doi = netlbl_get_u32(msg);
+ ret_val = cipso_v4_doi_remove(doi, netlbl_cipsov4_doi_free);
+
+remove_return:
+ netlbl_cipsov4_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/*
+ * NetLabel Generic NETLINK Command Definitions
+ */
+
+static struct genl_ops netlbl_cipsov4_genl_c_add = {
+ .cmd = NLBL_CIPSOV4_C_ADD,
+ .flags = 0,
+ .doit = netlbl_cipsov4_add,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_cipsov4_genl_c_remove = {
+ .cmd = NLBL_CIPSOV4_C_REMOVE,
+ .flags = 0,
+ .doit = netlbl_cipsov4_remove,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_cipsov4_genl_c_list = {
+ .cmd = NLBL_CIPSOV4_C_LIST,
+ .flags = 0,
+ .doit = netlbl_cipsov4_list,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_cipsov4_genl_c_listall = {
+ .cmd = NLBL_CIPSOV4_C_LISTALL,
+ .flags = 0,
+ .doit = netlbl_cipsov4_listall,
+ .dumpit = NULL,
+};
+
+/*
+ * NetLabel Generic NETLINK Protocol Functions
+ */
+
+/**
+ * netlbl_cipsov4_send_ack - Send an ACK message
+ * @info: the generic NETLINK information
+ * @ret_code: return code to use
+ *
+ * Description:
+ * This function sends an ACK message to the sender of the NETLINK message
+ * specified by @req_skb.
+ *
+ */
+static void netlbl_cipsov4_send_ack(const struct genl_info *info,
+ const u32 ret_code)
+{
+ size_t msg_size;
+ size_t data_size;
+ struct sk_buff *skb;
+ unsigned char *data;
+
+ data_size = GENL_HDRLEN + 8;
+ msg_size = NLMSG_SPACE(data_size);
+
+ skb = alloc_skb(msg_size, GFP_KERNEL);
+ if (skb == NULL)
+ return;
+
+ data = netlbl_netlink_hdr_put(skb,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_cipsov4_gnl_family.id,
+ NLBL_CIPSOV4_C_ACK,
+ data_size);
+ if (data == NULL)
+ goto send_ack_failure;
+
+ netlbl_putinc_u32(&data, info->snd_seq);
+ netlbl_putinc_u32(&data, ret_code);
+
+ netlbl_netlink_snd(skb, info->snd_pid);
+ return;
+
+send_ack_failure:
+ kfree_skb(skb);
+}
+
+/**
+ * netlbl_cipsov4_genl_init - Register the CIPSOv4 NetLabel component
+ *
+ * Description:
+ * Register the CIPSOv4 packet NetLabel component with the Generic NETLINK
+ * mechanism. Returns zero on success, negative values on failure.
+ *
+ */
+int netlbl_cipsov4_genl_init(void)
+{
+ int ret_val;
+
+ ret_val = genl_register_family(&netlbl_cipsov4_gnl_family);
+ if (ret_val != 0)
+ return ret_val;
+
+ ret_val = genl_register_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_add);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_remove);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_list);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_listall);
+ if (ret_val != 0)
+ goto genl_init_failure;
+
+ return 0;
+
+genl_init_failure:
+ netlbl_cipsov4_genl_exit();
+ return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_genl_exit - Unregister the CIPSOv4 NetLabel component
+ *
+ * Description:
+ * Unregister the CIPSOv4 packet NetLabel component with the Generic NETLINK
+ * mechanism. Returns zero on success, negative values on failure.
+ *
+ */
+void netlbl_cipsov4_genl_exit(void)
+{
+ genl_unregister_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_listall);
+ genl_unregister_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_list);
+ genl_unregister_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_remove);
+ genl_unregister_ops(&netlbl_cipsov4_gnl_family,
+ &netlbl_cipsov4_genl_c_add);
+
+ genl_unregister_family(&netlbl_cipsov4_gnl_family);
+ netlbl_cipsov4_gnl_family.id = GENL_ID_GENERATE;
+}
+
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 8/8] NetLabel: tie NetLabel into the Kconfig system
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-net_kconfig-2.6.17 --]
[-- Type: text/plain, Size: 521 bytes --]
Modify the net/Kconfig file to enable selecting the NetLabel Kconfig options.
---
net/Kconfig | 2 ++
1 files changed, 2 insertions(+)
Index: linux-2.6.17.i686-quilt/net/Kconfig
===================================================================
--- linux-2.6.17.i686-quilt.orig/net/Kconfig
+++ linux-2.6.17.i686-quilt/net/Kconfig
@@ -228,6 +228,8 @@ source "net/tux/Kconfig"
config WIRELESS_EXT
bool
+source "net/netlabel/Kconfig"
+
endif # if NET
endmenu # Networking
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 0/8] NetLabel: updated to use generic netlink
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
An updated patch set with some small changes as well as one big one - NetLabel
now uses the generic netlink interface for it's kernel-userland communication
as opposed to it's own dedicated netlink type. Needless to say this requires
an updated userland configuration tool, so for those of you running this patch
please grab version 0.14 (or later) of the netlabel_tools which can be found
here:
* http://free.linux.hp.com/~pmoore/projects/linux_cipso
Thanks.
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 2/8] NetLabel: core network changes
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-net_core-2.6.17 --]
[-- Type: text/plain, Size: 24948 bytes --]
Changes to the core network stack to support the NetLabel subsystem. This
includes changes to the IPv4 option handling to support CIPSO labels, and a new
NetLabel hook in inet_accept() to handle NetLabel attributes across a
accept()s done by in-kernel daemons.
---
include/linux/ip.h | 1
include/net/cipso_ipv4.h | 251 ++++++++++++++++++++++++
include/net/inet_sock.h | 2
include/net/netlabel.h | 488 +++++++++++++++++++++++++++++++++++++++++++++++
net/ipv4/Makefile | 1
net/ipv4/af_inet.c | 3
net/ipv4/ah4.c | 2
net/ipv4/ip_options.c | 19 +
8 files changed, 765 insertions(+), 2 deletions(-)
Index: linux-2.6.17.i686-quilt/include/linux/ip.h
===================================================================
--- linux-2.6.17.i686-quilt.orig/include/linux/ip.h
+++ linux-2.6.17.i686-quilt/include/linux/ip.h
@@ -57,6 +57,7 @@
#define IPOPT_SEC (2 |IPOPT_CONTROL|IPOPT_COPY)
#define IPOPT_LSRR (3 |IPOPT_CONTROL|IPOPT_COPY)
#define IPOPT_TIMESTAMP (4 |IPOPT_MEASUREMENT)
+#define IPOPT_CIPSO (6 |IPOPT_CONTROL|IPOPT_COPY)
#define IPOPT_RR (7 |IPOPT_CONTROL)
#define IPOPT_SID (8 |IPOPT_CONTROL|IPOPT_COPY)
#define IPOPT_SSRR (9 |IPOPT_CONTROL|IPOPT_COPY)
Index: linux-2.6.17.i686-quilt/include/net/cipso_ipv4.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/include/net/cipso_ipv4.h
@@ -0,0 +1,251 @@
+/*
+ * CIPSO - Commercial IP Security Option
+ *
+ * This is an implementation of the CIPSO 2.2 protocol as specified in
+ * draft-ietf-cipso-ipsecurity-01.txt with additional tag types as found in
+ * FIPS-188, copies of both documents can be found in the Documentation
+ * directory. While CIPSO never became a full IETF RFC standard many vendors
+ * have chosen to adopt the protocol and over the years it has become a
+ * de-facto standard for labeled networking.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _CIPSO_IPV4_H
+#define _CIPSO_IPV4_H
+
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <net/netlabel.h>
+
+/* known doi values */
+#define CIPSO_V4_DOI_UNKNOWN 0x00000000
+
+/* tag types */
+#define CIPSO_V4_TAG_INVALID 0
+#define CIPSO_V4_TAG_RBITMAP 1
+#define CIPSO_V4_TAG_ENUM 2
+#define CIPSO_V4_TAG_RANGE 5
+#define CIPSO_V4_TAG_PBITMAP 6
+#define CIPSO_V4_TAG_FREEFORM 7
+
+/* doi mapping types */
+#define CIPSO_V4_MAP_UNKNOWN 0
+#define CIPSO_V4_MAP_STD 1
+#define CIPSO_V4_MAP_PASS 2
+
+/* limits */
+#define CIPSO_V4_MAX_REM_LVLS 256
+#define CIPSO_V4_INV_LVL 0x80000000
+#define CIPSO_V4_MAX_LOC_LVLS (CIPSO_V4_INV_LVL - 1)
+#define CIPSO_V4_MAX_REM_CATS 65536
+#define CIPSO_V4_INV_CAT 0x80000000
+#define CIPSO_V4_MAX_LOC_CATS (CIPSO_V4_INV_CAT - 1)
+
+/*
+ * CIPSO DOI definitions
+ */
+
+/* DOI definition struct */
+#define CIPSO_V4_TAG_MAXCNT 5
+struct cipso_v4_doi {
+ u32 doi;
+ u32 type;
+ union {
+ struct cipso_v4_std_map_tbl *std;
+ } map;
+ u8 tags[CIPSO_V4_TAG_MAXCNT];
+
+ u32 valid;
+ struct list_head list;
+ struct rcu_head rcu;
+ struct list_head dom_list;
+};
+
+/* Standard CIPSO mapping table */
+/* NOTE: the highest order bit (i.e. 0x80000000) is an 'invalid' flag, if the
+ * bit is set then consider that value as unspecified, meaning the
+ * mapping for that particular level/category is invalid */
+struct cipso_v4_std_map_tbl {
+ struct {
+ u32 *cipso;
+ u32 *local;
+ u32 cipso_size;
+ u32 local_size;
+ } lvl;
+ struct {
+ u32 *cipso;
+ u32 *local;
+ u32 cipso_size;
+ u32 local_size;
+ } cat;
+};
+
+/*
+ * Helper Functions
+ */
+
+#define CIPSO_V4_OPTEXIST(x) (IPCB(x)->opt.cipso != 0)
+#define CIPSO_V4_OPTPTR(x) ((x)->nh.raw + IPCB(x)->opt.cipso)
+
+/*
+ * DOI List Functions
+ */
+
+#ifdef CONFIG_NETLABEL
+int cipso_v4_doi_add(struct cipso_v4_doi *doi_def);
+int cipso_v4_doi_remove(const u32 doi,
+ void (*callback) (struct rcu_head * head));
+struct cipso_v4_doi *cipso_v4_doi_getdef(const u32 doi);
+struct sk_buff *cipso_v4_doi_dump(const u32 doi, const u32 headroom);
+int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain);
+int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
+ const char *domain);
+#else
+static inline int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
+{
+ return -ENOSYS;
+}
+
+static inline int cipso_v4_doi_remove(const u32 doi,
+ void (*callback) (struct rcu_head * head))
+{
+ return 0;
+}
+
+static inline struct cipso_v4_doi *cipso_v4_doi_getdef(const u32 doi)
+{
+ return NULL;
+}
+
+static inline struct sk_buff *cipso_v4_doi_dump(const u32 doi,
+ const u32 headroom)
+{
+ return NULL;
+}
+
+static inline int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def,
+ const char *domain)
+{
+ return -ENOSYS;
+}
+
+static inline int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
+ const char *domain)
+{
+ return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+/*
+ * Label Mapping Cache Functions
+ */
+
+#ifdef CONFIG_NETLABEL
+void cipso_v4_cache_invalidate(void);
+int cipso_v4_cache_add(const struct sk_buff *skb,
+ const struct netlbl_lsm_secattr *secattr);
+#else
+static inline void cipso_v4_cache_invalidate(void)
+{
+ return;
+}
+
+static inline int cipso_v4_cache_add(const struct sk_buff *skb,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+/*
+ * Protocol Handling Functions
+ */
+
+#ifdef CONFIG_NETLABEL
+int cipso_v4_error(struct sk_buff *skb,
+ const int error,
+ const u32 gateway);
+int cipso_v4_socket_setopt(struct socket *sock,
+ unsigned char *opt,
+ u32 opt_len);
+int cipso_v4_socket_setattr(const struct socket *sock,
+ const struct cipso_v4_doi *doi_def,
+ const struct netlbl_lsm_secattr *secattr);
+int cipso_v4_socket_getopt(const struct socket *sock,
+ unsigned char **opt,
+ u32 *opt_len);
+int cipso_v4_socket_getattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr);
+int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
+ struct netlbl_lsm_secattr *secattr);
+int cipso_v4_validate(unsigned char **option);
+#else
+static inline int cipso_v4_error(struct sk_buff *skb,
+ const int error,
+ const u32 gateway)
+{
+ return 0;
+}
+
+static inline int cipso_v4_socket_setopt(struct socket *sock,
+ unsigned char *opt,
+ u32 opt_len)
+{
+ return -ENOSYS;
+}
+
+static inline int cipso_v4_socket_setattr(const struct socket *sock,
+ const struct cipso_v4_doi *doi_def,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ return -ENOSYS;
+}
+
+static inline int cipso_v4_socket_getopt(const struct socket *sock,
+ unsigned char **opt,
+ u32 *opt_len)
+{
+ return -ENOSYS;
+}
+
+static inline int cipso_v4_socket_getattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr)
+{
+ return -ENOSYS;
+}
+
+static inline int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
+ struct netlbl_lsm_secattr *secattr)
+{
+ return -ENOSYS;
+}
+
+static inline int cipso_v4_validate(unsigned char **option)
+{
+ return -ENOSYS;
+}
+#endif /* CONFIG_NETLABEL */
+
+#endif /* _CIPSO_IPV4_H */
Index: linux-2.6.17.i686-quilt/include/net/inet_sock.h
===================================================================
--- linux-2.6.17.i686-quilt.orig/include/net/inet_sock.h
+++ linux-2.6.17.i686-quilt/include/net/inet_sock.h
@@ -52,7 +52,7 @@ struct ip_options {
ts_needtime:1,
ts_needaddr:1;
unsigned char router_alert;
- unsigned char __pad1;
+ unsigned char cipso;
unsigned char __pad2;
unsigned char __data[0];
};
Index: linux-2.6.17.i686-quilt/include/net/netlabel.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/include/net/netlabel.h
@@ -0,0 +1,488 @@
+/*
+ * NetLabel System
+ *
+ * The NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_H
+#define _NETLABEL_H
+
+#include <linux/types.h>
+#include <linux/skbuff.h>
+#include <net/netlink.h>
+
+/*
+ * NetLabel - A management interface for maintaining network packet label
+ * mapping tables for explicit packet labling protocols.
+ *
+ * Network protocols such as CIPSO and RIPSO require a label translation layer
+ * to convert the label on the packet into something meaningful on the host
+ * machine. In the current Linux implementation these mapping tables live
+ * inside the kernel; NetLabel provides a mechanism for user space applications
+ * to manage these mapping tables.
+ *
+ * NetLabel makes use of the Generic NETLINK mechanism as a transport layer to
+ * send messages between kernel and user space. The general format of a
+ * NetLabel message is shown below:
+ *
+ * +-----------------+-------------------+--------- --- -- -
+ * | struct nlmsghdr | struct genlmsghdr | payload
+ * +-----------------+-------------------+--------- --- -- -
+ *
+ * The 'nlmsghdr' and 'genlmsghdr' structs should be dealt with like normal.
+ * The payload is dependent on the subsystem specified in the
+ * 'nlmsghdr->nlmsg_type' and should be defined below, supporting functions
+ * should be defined in the corresponding net/netlabel/netlabel_<subsys>.h|c
+ * file.
+ *
+ */
+
+/*
+ * NetLabel NETLINK protocol
+ */
+
+#define NETLBL_PROTO_VERSION 1
+
+/* NetLabel NETLINK types/families */
+#define NETLBL_NLTYPE_NONE 0
+#define NETLBL_NLTYPE_MGMT 1
+#define NETLBL_NLTYPE_MGMT_NAME "NLBL_MGMT"
+#define NETLBL_NLTYPE_RIPSO 2
+#define NETLBL_NLTYPE_RIPSO_NAME "NLBL_RIPSO"
+#define NETLBL_NLTYPE_CIPSOV4 3
+#define NETLBL_NLTYPE_CIPSOV4_NAME "NLBL_CIPSOv4"
+#define NETLBL_NLTYPE_CIPSOV6 4
+#define NETLBL_NLTYPE_CIPSOV6_NAME "NLBL_CIPSOv6"
+#define NETLBL_NLTYPE_UNLABELED 5
+#define NETLBL_NLTYPE_UNLABELED_NAME "NLBL_UNLBL"
+
+/* NetLabel return codes */
+#define NETLBL_E_OK 0
+
+/*
+ * Helper functions
+ */
+
+/**
+ * netlbl_put_u8 - Write a u8 value into a buffer
+ * @buffer: the buffer
+ * @val: the value
+ *
+ * Description:
+ * Write the value specified in @val into the buffer specified by @buffer.
+ *
+ */
+static inline void netlbl_put_u8(unsigned char *buffer, const u8 val)
+{
+ *(u8 *)buffer = val;
+}
+
+/**
+ * netlbl_put_u32 - Write a u32 value into a buffer
+ * @buffer: the buffer
+ * @val: the value
+ *
+ * Description:
+ * Write the value specified in @val into the buffer specified by @buffer.
+ *
+ */
+static inline void netlbl_put_u32(unsigned char *buffer, const u32 val)
+{
+ *(u32 *)buffer = val;
+}
+
+/**
+ * netlbl_put_str - Write a string into a buffer
+ * @buffer: the buffer
+ * @val: the string
+ *
+ * Description:
+ * Write the string specified in @val into the buffer specified by @buffer.
+ *
+ */
+static inline void netlbl_put_str(unsigned char *buffer, const char *val)
+{
+ strcpy((char *)buffer, val);
+}
+
+/**
+ * netlbl_put_hdr - Write a NETLINK header into a buffer
+ * @buffer: the buffer
+ * @msg_type: the NETLINK message type
+ * @msg_len: the NETLINK message length
+ * @msg_flags: the NETLINK message flags
+ * @msg_pid: the NETLINK message PID
+ * @msg_seq: the NETLINK message sequence number
+ *
+ * Description:
+ * Use the given values to write a NETLINK header into the given buffer.
+ *
+ */
+static inline void netlbl_put_hdr(unsigned char *buffer,
+ const u32 msg_type,
+ const u16 msg_len,
+ const u16 msg_flags,
+ const u32 msg_pid,
+ const u32 msg_seq)
+{
+ struct nlmsghdr *hdr = (struct nlmsghdr *)buffer;
+ hdr->nlmsg_len = msg_len;
+ hdr->nlmsg_type = msg_type;
+ hdr->nlmsg_flags = msg_flags;
+ hdr->nlmsg_seq = msg_seq;
+ hdr->nlmsg_pid = msg_pid;
+}
+
+/**
+ * netlbl_put_u8 - Write a u8 value into a buffer and increment the buffer
+ * @buffer: the buffer
+ * @val: the value
+ *
+ * Description:
+ * Write the value specified in @val into the buffer specified by @buffer
+ * and advance the buffer pointer past the newly written value.
+ *
+ */
+static inline void netlbl_putinc_u8(unsigned char **buffer, const u8 val)
+{
+ netlbl_put_u8(*buffer, val);
+ *buffer += sizeof(u8);
+}
+
+/**
+ * netlbl_put_u32 - Write a u32 value into a buffer and increment the buffer
+ * @buffer: the buffer
+ * @val: the value
+ *
+ * Description:
+ * Write the value specified in @val into the buffer specified by @buffer
+ * and advance the buffer pointer past the newly written value.
+ *
+ */
+static inline void netlbl_putinc_u32(unsigned char **buffer, const u32 val)
+{
+ netlbl_put_u32(*buffer, val);
+ *buffer += sizeof(u32);
+}
+
+/**
+ * netlbl_put_u8 - Write a string into a buffer and increment the buffer
+ * @buffer: the buffer
+ * @val: the value
+ *
+ * Description:
+ * Write the string specified in @val into the buffer specified by @buffer
+ * and advance the buffer pointer past the newly written value.
+ *
+ */
+static inline void netlbl_putinc_str(unsigned char **buffer, const char *val)
+{
+ netlbl_put_str(*buffer, val);
+ *buffer += strlen(val) + 1;
+}
+
+/**
+ * netlbl_put_hdr - Write a NETLINK header into a buffer and increment the ptr
+ * @buffer: the buffer
+ * @msg_type: the NETLINK message type
+ * @msg_len: the NETLINK message length
+ * @msg_flags: the NETLINK message flags
+ * @msg_pid: the NETLINK message PID
+ * @msg_seq: the NETLINK message sequence number
+ *
+ * Description:
+ * Use the given values to write a NETLINK header into the given buffer and
+ * then increment the buffer pointer past the header.
+ *
+ */
+static inline void netlbl_putinc_hdr(unsigned char **buffer,
+ const u32 msg_type,
+ const u16 msg_len,
+ const u16 msg_flags,
+ const u32 msg_pid,
+ const u32 msg_seq)
+{
+ netlbl_put_hdr(*buffer,
+ msg_type,
+ msg_len,
+ msg_flags,
+ msg_pid,
+ msg_seq);
+ *buffer += NLMSG_HDRLEN;
+}
+
+/**
+ * netlbl_get_u8 - Read a u8 value from a buffer
+ * @buffer: the buffer
+ *
+ * Description:
+ * Return a u8 value pointed to by @buffer.
+ *
+ */
+static inline u8 netlbl_get_u8(const unsigned char *buffer)
+{
+ return *(u8 *)buffer;
+}
+
+/**
+ * netlbl_get_u32 - Read a u32 value from a buffer
+ * @buffer: the buffer
+ *
+ * Description:
+ * Return a u8 value pointed to by @buffer.
+ *
+ */
+static inline u32 netlbl_get_u32(const unsigned char *buffer)
+{
+ return *(u32 *)buffer;
+}
+
+/**
+ * netlbl_getinc_u8 - Read a u8 value from a buffer and increment the buffer
+ * @buffer: the buffer
+ *
+ * Description:
+ * Return a u8 value pointed to by @buffer and increment the buffer pointer
+ * past the value.
+ *
+ */
+static inline u8 netlbl_getinc_u8(unsigned char **buffer)
+{
+ u8 val = netlbl_get_u8(*buffer);
+ *buffer += sizeof(u8);
+ return val;
+}
+
+/**
+ * netlbl_getinc_u32 - Read a u32 value from a buffer and increment the buffer
+ * @buffer: the buffer
+ *
+ * Description:
+ * Return a u32 value pointed to by @buffer and increment the buffer pointer
+ * past the value.
+ *
+ */
+static inline u32 netlbl_getinc_u32(unsigned char **buffer)
+{
+ u32 val = netlbl_get_u32(*buffer);
+ *buffer += sizeof(u32);
+ return val;
+}
+
+/*
+ * NetLabel - Kernel API for accessing the network packet label mappings.
+ *
+ * The following functions are provided for use by other kernel modules,
+ * specifically kernel LSM modules, to provide a consistent, transparent API
+ * for dealing with explicit packet labeling protocols such as CIPSO and
+ * RIPSO. The functions defined here are implemented in the
+ * net/netlabel/netlabel_kapi.c file.
+ *
+ */
+
+/* Domain mapping definition struct */
+struct netlbl_dom_map;
+
+/* Domain mapping operations */
+int netlbl_domhsh_remove(const char *domain);
+
+/* LSM security attributes */
+struct netlbl_lsm_cache {
+ void (*free) (const void *data);
+ void *data;
+};
+struct netlbl_lsm_secattr {
+ char *domain;
+
+ u32 mls_lvl;
+ unsigned char *mls_cat;
+ u32 mls_cat_len;
+
+ struct netlbl_lsm_cache cache;
+
+ u32 set_domain:1,
+ set_mls_lvl:1,
+ set_mls_cat:1,
+ set_cache:1,
+ __unused:28;
+};
+
+/*
+ * LSM security attribute operations
+ */
+
+/**
+ * netlbl_secattr_alloc - Allocate and initialize a netlbl_lsm_secattr struct
+ * @flags: the memory allocation flags
+ *
+ * Description:
+ * Allocate and initialize a netlbl_lsm_secattr struct. Returns a valid
+ * pointer on success, or NULL on failure.
+ *
+ */
+static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc(const int flags)
+{
+ return kzalloc(sizeof(struct netlbl_lsm_secattr), flags);
+}
+
+/**
+ * netlbl_secattr_free - Frees a netlbl_lsm_secattr struct
+ * @secattr: the struct to free
+ *
+ * Description:
+ * Frees @secattr including all of the internal buffers.
+ *
+ */
+static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr)
+{
+ BUG_ON(secattr == NULL);
+ if (secattr->set_domain)
+ kfree(secattr->domain);
+ if (secattr->set_mls_cat)
+ kfree(secattr->mls_cat);
+ if (secattr->set_cache && secattr->cache.free)
+ secattr->cache.free(secattr->cache.data);
+ kfree(secattr);
+}
+
+/**
+ * netlbl_secattr_init - Initialize a netlbl_lsm_secattr struct
+ * @secattr: the struct to initialize
+ *
+ * Description:
+ * Initialize an already allocated netlbl_lsm_secattr struct. Returns zero on
+ * success, negative values on error.
+ *
+ */
+static inline int netlbl_secattr_init(struct netlbl_lsm_secattr *secattr)
+{
+ BUG_ON(secattr == NULL);
+ memset(secattr, 0, sizeof(*secattr));
+ return 0;
+}
+
+/**
+ * netlbl_secattr_destroy - Clears a netlbl_lsm_secattr struct
+ * @secattr: the struct to clear
+ *
+ * Description:
+ * Destroys the @secattr struct, including freeing all of the internal buffers.
+ * On return the struct is suitable for reuse.
+ *
+ */
+static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr)
+{
+ BUG_ON(secattr == NULL);
+ if (secattr->set_domain)
+ kfree(secattr->domain);
+ if (secattr->set_mls_cat)
+ kfree(secattr->mls_cat);
+ if (secattr->set_cache && secattr->cache.free)
+ secattr->cache.free(secattr->cache.data);
+ memset(secattr, 0, sizeof(*secattr));
+}
+
+/*
+ * LSM protocol operations
+ */
+
+#ifdef CONFIG_NETLABEL
+int netlbl_socket_setattr(const struct socket *sock,
+ const struct netlbl_lsm_secattr *secattr);
+int netlbl_socket_peekattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr);
+int netlbl_socket_getattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr);
+int netlbl_skbuff_getattr(const struct sk_buff *skb,
+ struct netlbl_lsm_secattr *secattr);
+int netlbl_skbuff_err(struct sk_buff *skb, int error);
+#else
+static inline int netlbl_socket_setattr(const struct socket *sock,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ return -ENOSYS;
+}
+
+static inline int netlbl_socket_peekattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr)
+{
+ return -ENOSYS;
+}
+
+static inline int netlbl_socket_getattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr)
+{
+ return -ENOSYS;
+}
+
+static inline int netlbl_skbuff_getattr(const struct sk_buff *skb,
+ struct netlbl_lsm_secattr *secattr)
+{
+ return -ENOSYS;
+}
+
+static inline int netlbl_skbuff_err(struct sk_buff *skb, int error)
+{
+ return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+/*
+ * LSM label mapping cache operations
+ */
+
+#ifdef CONFIG_NETLABEL
+void netlbl_cache_invalidate(void);
+int netlbl_cache_add(const struct sk_buff *skb,
+ const struct netlbl_lsm_secattr *secattr);
+#else
+static inline void netlbl_cache_invalidate(void)
+{
+ return;
+}
+
+static inline int netlbl_cache_add(const struct sk_buff *skb,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+/*
+ * Network stack operations
+ */
+
+#ifdef CONFIG_NETLABEL
+void netlbl_socket_inet_accept(struct socket *sock, struct socket *newsock);
+#else
+static inline void netlbl_socket_inet_accept(struct socket *sock,
+ struct socket *newsock)
+{
+ return;
+}
+#endif /* CONFIG_NETLABEL */
+
+#endif /* _NETLABEL_H */
Index: linux-2.6.17.i686-quilt/net/ipv4/Makefile
===================================================================
--- linux-2.6.17.i686-quilt.orig/net/ipv4/Makefile
+++ linux-2.6.17.i686-quilt/net/ipv4/Makefile
@@ -42,6 +42,7 @@ obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybl
obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o
obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o
obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
+obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
xfrm4_output.o
Index: linux-2.6.17.i686-quilt/net/ipv4/af_inet.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/net/ipv4/af_inet.c
+++ linux-2.6.17.i686-quilt/net/ipv4/af_inet.c
@@ -114,6 +114,7 @@
#ifdef CONFIG_IP_MROUTE
#include <linux/mroute.h>
#endif
+#include <net/netlabel.h>
DEFINE_SNMP_STAT(struct linux_mib, net_statistics) __read_mostly;
@@ -616,6 +617,8 @@ int inet_accept(struct socket *sock, str
sock_graft(sk2, newsock);
+ netlbl_socket_inet_accept(sock, newsock);
+
newsock->state = SS_CONNECTED;
err = 0;
release_sock(sk2);
Index: linux-2.6.17.i686-quilt/net/ipv4/ah4.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/net/ipv4/ah4.c
+++ linux-2.6.17.i686-quilt/net/ipv4/ah4.c
@@ -35,7 +35,7 @@ static int ip_clear_mutable_options(stru
switch (*optptr) {
case IPOPT_SEC:
case 0x85: /* Some "Extended Security" crap. */
- case 0x86: /* Another "Commercial Security" crap. */
+ case IPOPT_CIPSO:
case IPOPT_RA:
case 0x80|21: /* RFC1770 */
break;
Index: linux-2.6.17.i686-quilt/net/ipv4/ip_options.c
===================================================================
--- linux-2.6.17.i686-quilt.orig/net/ipv4/ip_options.c
+++ linux-2.6.17.i686-quilt/net/ipv4/ip_options.c
@@ -24,6 +24,7 @@
#include <net/ip.h>
#include <net/icmp.h>
#include <net/route.h>
+#include <net/cipso_ipv4.h>
/*
* Write options to IP header, record destination address to
@@ -194,6 +195,13 @@ int ip_options_echo(struct ip_options *
dopt->is_strictroute = sopt->is_strictroute;
}
}
+ if (sopt->cipso) {
+ optlen = sptr[sopt->cipso+1];
+ dopt->cipso = dopt->optlen+sizeof(struct iphdr);
+ memcpy(dptr, sptr+sopt->cipso, optlen);
+ dptr += optlen;
+ dopt->optlen += optlen;
+ }
while (dopt->optlen & 3) {
*dptr++ = IPOPT_END;
dopt->optlen++;
@@ -435,6 +443,17 @@ int ip_options_compile(struct ip_options
if (optptr[2] == 0 && optptr[3] == 0)
opt->router_alert = optptr - iph;
break;
+ case IPOPT_CIPSO:
+ if (opt->cipso) {
+ pp_ptr = optptr;
+ goto error;
+ }
+ opt->cipso = optptr - iph;
+ if (cipso_v4_validate(&optptr)) {
+ pp_ptr = optptr;
+ goto error;
+ }
+ break;
case IPOPT_SEC:
case IPOPT_SID:
default:
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 4/8] NetLabel: core NetLabel subsystem
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-core --]
[-- Type: text/plain, Size: 78406 bytes --]
Add a new kernel subsystem, NetLabel, to provide explicit packet labeling
services (CIPSO, RIPSO, etc.) to LSM developers. NetLabel is designed to work
in conjunction with a LSM to intercept and decode security labels on incoming
network packets as well as ensure that outgoing network packets are labeled
according to the security mechanism employed by the LSM. The NetLabel
subsystem is configured through a Generic NETLINK interface described in the
header files included in this patch.
---
net/Makefile | 1
net/netlabel/Kconfig | 14
net/netlabel/Makefile | 16
net/netlabel/netlabel_cipso_v4.h | 218 +++++++++++
net/netlabel/netlabel_domainhash.c | 597 +++++++++++++++++++++++++++++++
net/netlabel/netlabel_domainhash.h | 64 +++
net/netlabel/netlabel_kapi.c | 363 ++++++++++++++++++
net/netlabel/netlabel_mgmt.c | 707 +++++++++++++++++++++++++++++++++++++
net/netlabel/netlabel_mgmt.h | 250 +++++++++++++
net/netlabel/netlabel_unlabeled.h | 84 ++++
net/netlabel/netlabel_user.c | 131 ++++++
net/netlabel/netlabel_user.h | 175 +++++++++
12 files changed, 2620 insertions(+)
Index: linux-2.6.17.i686-quilt/net/Makefile
===================================================================
--- linux-2.6.17.i686-quilt.orig/net/Makefile
+++ linux-2.6.17.i686-quilt/net/Makefile
@@ -47,6 +47,7 @@ obj-$(CONFIG_IP_DCCP) += dccp/
obj-$(CONFIG_IP_SCTP) += sctp/
obj-$(CONFIG_IEEE80211) += ieee80211/
obj-$(CONFIG_TIPC) += tipc/
+obj-$(CONFIG_NETLABEL) += netlabel/
ifeq ($(CONFIG_NET),y)
obj-$(CONFIG_SYSCTL) += sysctl_net.o
Index: linux-2.6.17.i686-quilt/net/netlabel/Kconfig
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/Kconfig
@@ -0,0 +1,14 @@
+#
+# NetLabel configuration
+#
+
+config NETLABEL
+ bool "NetLabel subsystem support"
+ depends on NET && SECURITY
+ default n
+ ---help---
+ NetLabel provides support for explicit network packet labeling
+ protocols such as CIPSO and RIPSO. For more information see
+ Documentation/netlabel.
+
+ If you are unsure, say N.
Index: linux-2.6.17.i686-quilt/net/netlabel/Makefile
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/Makefile
@@ -0,0 +1,16 @@
+#
+# Makefile for the NetLabel subsystem.
+#
+# Feb 9, 2006, Paul Moore <paul.moore@hp.com>
+#
+
+# base objects
+obj-y := netlabel_user.o netlabel_kapi.o netlabel_domainhash.o
+
+# management objects
+obj-y += netlabel_mgmt.o
+
+# protocol modules
+obj-y += netlabel_unlabeled.o
+obj-y += netlabel_cipso_v4.o
+
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_cipso_v4.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_cipso_v4.h
@@ -0,0 +1,218 @@
+/*
+ * NetLabel CIPSO/IPv4 Support
+ *
+ * This file defines the CIPSO/IPv4 functions for the NetLabel system. The
+ * NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_CIPSO_V4
+#define _NETLABEL_CIPSO_V4
+
+#include <net/netlabel.h>
+
+/*
+ * The following NetLabel payloads are supported by the CIPSO subsystem, all
+ * of which are preceeded by the nlmsghdr struct.
+ *
+ * o ACK:
+ * Sent by the kernel in response to an applications message, applications
+ * should never send this message.
+ *
+ * +----------------------+-----------------------+
+ * | seq number (32 bits) | return code (32 bits) |
+ * +----------------------+-----------------------+
+ *
+ * seq number: the sequence number of the original message, taken from the
+ * nlmsghdr structure
+ * return code: return value, based on errno values
+ *
+ * o ADD:
+ * Sent by an application to add a new DOI mapping table, after completion
+ * of the task the kernel should ACK this message.
+ *
+ * +-----+--------------------+---------------------+
+ * | DOI | map type (32 bits) | tag count (32 bits) | ...
+ * +-----+--------------------+---------------------+
+ *
+ * +-----------------+
+ * | tag #X (8 bits) | ... repeated
+ * +-----------------+
+ *
+ * +-------------- ---- --- -- -
+ * | mapping data
+ * +-------------- ---- --- -- -
+ *
+ * DOI: the DOI value
+ * map type: the mapping table type (defined in the cipso_ipv4.h header
+ * as CIPSO_V4_MAP_*)
+ * tag count: the number of tags, must be greater than zero
+ * tag: the CIPSO tag for the DOI, tags listed first are given
+ * higher priorirty when sending packets
+ * mapping data: specific to the map type (see below)
+ *
+ * CIPSO_V4_MAP_STD
+ *
+ * +------------------+-----------------------+-----------------------+
+ * | levels (32 bits) | max l level (32 bits) | max r level (32 bits) | ...
+ * +------------------+-----------------------+-----------------------+
+ *
+ * +----------------------+---------------------+---------------------+
+ * | categories (32 bits) | max l cat (32 bits) | max r cat (32 bits) | ...
+ * +----------------------+---------------------+---------------------+
+ *
+ * +--------------------------+--------------------------+
+ * | local level #X (32 bits) | CIPSO level #X (32 bits) | ... repeated
+ * +--------------------------+--------------------------+
+ *
+ * +-----------------------------+-----------------------------+
+ * | local category #X (32 bits) | CIPSO category #X (32 bits) | ... repeated
+ * +-----------------------------+-----------------------------+
+ *
+ * levels: the number of level mappings
+ * max l level: the highest local level
+ * max r level: the highest remote/CIPSO level
+ * categories: the number of category mappings
+ * max l cat: the highest local category
+ * max r cat: the highest remote/CIPSO category
+ * local level: the local part of a level mapping
+ * CIPSO level: the remote/CIPSO part of a level mapping
+ * local category: the local part of a category mapping
+ * CIPSO category: the remote/CIPSO part of a category mapping
+ *
+ * CIPSO_V4_MAP_PASS
+ *
+ * No mapping data is needed for this map type.
+ *
+ * o REMOVE:
+ * Sent by an application to remove a specific DOI mapping table from the
+ * CIPSO V4 system. The kernel should ACK this message.
+ *
+ * +-----+
+ * | DOI |
+ * +-----+
+ *
+ * DOI: the DOI value
+ *
+ * o LIST:
+ * Sent by an application to list the details of a DOI definition. The
+ * kernel should send an ACK on error or a response as indicated below. The
+ * application generated message format is shown below.
+ *
+ * +-----+
+ * | DOI |
+ * +-----+
+ *
+ * DOI: the DOI value
+ *
+ * The valid response message format depends on the type of the DOI mapping,
+ * the known formats are shown below.
+ *
+ * +--------------------+
+ * | map type (32 bits) | ...
+ * +--------------------+
+ *
+ * map type: the DOI mapping table type (defined in the cipso_ipv4.h
+ * header as CIPSO_V4_MAP_*)
+ *
+ * (map type == CIPSO_V4_MAP_STD)
+ *
+ * +----------------+------------------+----------------------+
+ * | tags (32 bits) | levels (32 bits) | categories (32 bits) | ...
+ * +----------------+------------------+----------------------+
+ *
+ * +-----------------+
+ * | tag #X (8 bits) | ... repeated
+ * +-----------------+
+ *
+ * +--------------------------+--------------------------+
+ * | local level #X (32 bits) | CIPSO level #X (32 bits) | ... repeated
+ * +--------------------------+--------------------------+
+ *
+ * +-----------------------------+-----------------------------+
+ * | local category #X (32 bits) | CIPSO category #X (32 bits) | ... repeated
+ * +-----------------------------+-----------------------------+
+ *
+ * tags: the number of CIPSO tag types
+ * levels: the number of level mappings
+ * categories: the number of category mappings
+ * tag: the tag number, tags listed first are given higher
+ * priority when sending packets
+ * local level: the local part of a level mapping
+ * CIPSO level: the remote/CIPSO part of a level mapping
+ * local category: the local part of a category mapping
+ * CIPSO category: the remote/CIPSO part of a category mapping
+ *
+ * (map type == CIPSO_V4_MAP_PASS)
+ *
+ * +----------------+
+ * | tags (32 bits) | ...
+ * +----------------+
+ *
+ * +-----------------+
+ * | tag #X (8 bits) | ... repeated
+ * +-----------------+
+ *
+ * tags: the number of CIPSO tag types
+ * tag: the tag number, tags listed first are given higher
+ * priority when sending packets
+ *
+ * o LISTALL:
+ * This message is sent by an application to list the valid DOIs on the
+ * system. There is no payload and the kernel should respond with an ACK
+ * or the following message.
+ *
+ * +---------------------+------------------+-----------------------+
+ * | DOI count (32 bits) | DOI #X (32 bits) | map type #X (32 bits) |
+ * +---------------------+------------------+-----------------------+
+ *
+ * +-----------------------+
+ * | map type #X (32 bits) | ...
+ * +-----------------------+
+ *
+ * DOI count: the number of DOIs
+ * DOI: the DOI value
+ * map type: the DOI mapping table type (defined in the cipso_ipv4.h
+ * header as CIPSO_V4_MAP_*)
+ *
+ */
+
+/* NetLabel CIPSOv4 commands */
+enum {
+ NLBL_CIPSOV4_C_UNSPEC,
+ NLBL_CIPSOV4_C_ACK,
+ NLBL_CIPSOV4_C_ADD,
+ NLBL_CIPSOV4_C_REMOVE,
+ NLBL_CIPSOV4_C_LIST,
+ NLBL_CIPSOV4_C_LISTALL,
+ __NLBL_CIPSOV4_C_MAX,
+};
+#define NLBL_CIPSOV4_C_MAX (__NLBL_CIPSOV4_C_MAX - 1)
+
+/* NetLabel protocol functions */
+int netlbl_cipsov4_genl_init(void);
+void netlbl_cipsov4_genl_exit(void);
+
+#endif
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_domainhash.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_domainhash.c
@@ -0,0 +1,597 @@
+/*
+ * NetLabel Domain Hash Table
+ *
+ * This file manages the domain hash table that NetLabel uses to determine
+ * which network labeling protocol to use for a given domain. The NetLabel
+ * system manages static and dynamic label mappings for network protocols such
+ * as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <linux/skbuff.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <net/netlabel.h>
+#include <net/cipso_ipv4.h>
+#include <asm/bug.h>
+
+#include "netlabel_mgmt.h"
+#include "netlabel_domainhash.h"
+
+struct netlbl_domhsh_tbl {
+ struct list_head *tbl;
+ u32 size;
+};
+
+/* Domain hash table */
+/* XXX - updates should be so rare that having one spinlock for the entire
+ hash table should be okay */
+DEFINE_SPINLOCK(netlbl_domhsh_lock);
+static struct netlbl_domhsh_tbl *netlbl_domhsh = NULL;
+
+/* Default domain mapping */
+DEFINE_SPINLOCK(netlbl_domhsh_def_lock);
+static struct netlbl_dom_map *netlbl_domhsh_def = NULL;
+
+/*
+ * Domain Hash Table Helper Functions
+ */
+
+/**
+ * netlbl_domhsh_free_entry - Frees a domain hash table entry
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that the memory allocated to a hash table entry can be released
+ * safely.
+ *
+ */
+static void netlbl_domhsh_free_entry(struct rcu_head *entry)
+{
+ struct netlbl_dom_map *ptr;
+
+ ptr = container_of(entry, struct netlbl_dom_map, rcu);
+ if (ptr->domain)
+ kfree(ptr->domain);
+ kfree(ptr);
+}
+
+/**
+ * netlbl_domhsh_hash - Hashing function for the domain hash table
+ * @domain: the domain name to hash
+ *
+ * Description:
+ * This is the hashing function for the domain hash table, it returns the
+ * correct bucket number for the domain. The caller is responsibile for
+ * calling the rcu_read_[un]lock() functions.
+ *
+ */
+static u32 netlbl_domhsh_hash(const char *domain)
+{
+ char *p;
+ char *keyp = (char *)domain;
+ u32 size;
+ u32 val = 0;
+
+ /* This is taken (with slight modification) from
+ security/selinux/ss/symtab.c:symhash() */
+
+ size = strlen(keyp);
+ for (p = keyp; (p - keyp) < size; p++)
+ val = (val << 4 | (val >> (8 * sizeof(u32) - 4))) ^ (*p);
+ return val & (rcu_dereference(netlbl_domhsh)->size - 1);
+}
+
+/**
+ * netlbl_domhsh_search - Search for a domain entry
+ * @domain: the domain
+ * @def: return default if no match is found
+ *
+ * Description:
+ * Searches the domain hash table and returns a pointer to the hash table
+ * entry if found, otherwise NULL is returned. If @def is non-zero and a
+ * match is not found in the domain hash table the default mapping is returned
+ * if it exists. The caller is responsibile for the rcu hash table locks
+ * (i.e. the caller much call rcu_read_[un]lock()).
+ *
+ */
+static struct netlbl_dom_map *netlbl_domhsh_search(const char *domain,
+ const u32 def)
+{
+ u32 bkt;
+ struct netlbl_dom_map *iter;
+
+ if (domain != NULL) {
+ bkt = netlbl_domhsh_hash(domain);
+ list_for_each_entry_rcu(iter, &netlbl_domhsh->tbl[bkt], list)
+ if (iter->valid && strcmp(iter->domain, domain) == 0)
+ return iter;
+ }
+
+ if (def != 0) {
+ iter = rcu_dereference(netlbl_domhsh_def);
+ if (iter != NULL && iter->valid)
+ return iter;
+ }
+
+ return NULL;
+}
+
+/*
+ * Domain Hash Table Functions
+ */
+
+/**
+ * netlbl_domhsh_init - Init for the domain hash
+ * @size: the number of bits to use for the hash buckets
+ *
+ * Description:
+ * Initializes the domain hash table, should be called only by
+ * netlbl_user_init() during module load. Returns zero on success, non-zero
+ * values on error.
+ *
+ */
+int netlbl_domhsh_init(const u32 size)
+{
+ u32 iter;
+ struct netlbl_domhsh_tbl *hsh_tbl;
+
+ if (size <= 0)
+ return -EINVAL;
+
+ hsh_tbl = kmalloc(sizeof(*hsh_tbl), GFP_KERNEL);
+ if (hsh_tbl == NULL)
+ return -ENOMEM;
+ hsh_tbl->size = 1 << size;
+ hsh_tbl->tbl = kcalloc(hsh_tbl->size,
+ sizeof(struct list_head),
+ GFP_KERNEL);
+ if (hsh_tbl->tbl == NULL) {
+ kfree(hsh_tbl);
+ return -ENOMEM;
+ }
+ for (iter = 0; iter < hsh_tbl->size; iter++)
+ INIT_LIST_HEAD(&hsh_tbl->tbl[iter]);
+
+ rcu_read_lock();
+ spin_lock(&netlbl_domhsh_lock);
+ rcu_assign_pointer(netlbl_domhsh, hsh_tbl);
+ spin_unlock(&netlbl_domhsh_lock);
+ rcu_read_unlock();
+
+ return 0;
+}
+
+/**
+ * netlbl_domhsh_exit - Cleanup for the domain hash
+ *
+ * Description:
+ * Clears and frees the domain hash table. No real serious effort is made
+ * towards locking but that shouldn't be an issue since we should only be
+ * called when the module is being unloaded. Returns zero on success, non-zero
+ * values on error.
+ *
+ */
+void netlbl_domhsh_exit(void)
+{
+ u32 iter_bkt;
+ struct netlbl_domhsh_tbl *hsh_tbl;
+ struct netlbl_dom_map *iter_list;
+
+ rcu_read_lock();
+ if (rcu_dereference(netlbl_domhsh) == NULL) {
+ rcu_read_unlock();
+ return;
+ }
+ iter_list = rcu_dereference(netlbl_domhsh_def);
+ spin_lock(&netlbl_domhsh_def_lock);
+ rcu_assign_pointer(netlbl_domhsh_def, NULL);
+ spin_unlock(&netlbl_domhsh_def_lock);
+ rcu_read_unlock();
+
+ if (iter_list) {
+ synchronize_rcu();
+ netlbl_domhsh_free_entry(&iter_list->rcu);
+ }
+
+ rcu_read_lock();
+ hsh_tbl = rcu_dereference(netlbl_domhsh);
+ spin_lock(&netlbl_domhsh_lock);
+ rcu_assign_pointer(netlbl_domhsh, NULL);
+ spin_unlock(&netlbl_domhsh_lock);
+ rcu_read_unlock();
+
+ synchronize_rcu();
+
+ /* PM - it probably doesn't matter, but do we need to use the rcu list
+ functions here since we are the only one who has access to the
+ list now? */
+ for (iter_bkt = 0; iter_bkt < hsh_tbl->size; iter_bkt++)
+ list_for_each_entry_rcu(iter_list,
+ &hsh_tbl->tbl[iter_bkt], list) {
+ iter_list->valid = 0;
+ list_del_rcu(&iter_list->list);
+ netlbl_domhsh_free_entry(&iter_list->rcu);
+ }
+ kfree(hsh_tbl->tbl);
+ kfree(hsh_tbl);
+
+ return;
+}
+
+/**
+ * netlbl_domhsh_add - Adds a entry to the domain hash table
+ * @entry: the entry to add
+ *
+ * Description:
+ * Adds a new entry to the domain hash table and handles any updates to the
+ * lower level protocol handler (i.e. CIPSO). Returns zero on success,
+ * negative on failure.
+ *
+ */
+int netlbl_domhsh_add(struct netlbl_dom_map *entry)
+{
+ int ret_val = -EPERM;
+ u32 bkt;
+
+ BUG_ON(entry == NULL);
+
+ switch (entry->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ ret_val = 0;
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ ret_val = cipso_v4_doi_domhsh_add(entry->type_def.cipsov4,
+ entry->domain);
+ break;
+ default:
+ BUG();
+ }
+ if (ret_val != 0)
+ return ret_val;
+
+ entry->valid = 1;
+ INIT_RCU_HEAD(&entry->rcu);
+
+ rcu_read_lock();
+ if (entry->domain != NULL &&
+ netlbl_domhsh_search(entry->domain, 0) == NULL) {
+ bkt = netlbl_domhsh_hash(entry->domain);
+ spin_lock(&netlbl_domhsh_lock);
+ list_add_tail_rcu(&entry->list, &netlbl_domhsh->tbl[bkt]);
+ spin_unlock(&netlbl_domhsh_lock);
+ } else if (entry->domain == NULL &&
+ rcu_dereference(netlbl_domhsh_def) == NULL) {
+ INIT_LIST_HEAD(&entry->list);
+ spin_lock(&netlbl_domhsh_def_lock);
+ rcu_assign_pointer(netlbl_domhsh_def, entry);
+ spin_unlock(&netlbl_domhsh_def_lock);
+ } else {
+ rcu_read_unlock();
+ switch (entry->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ if (cipso_v4_doi_domhsh_remove(entry->type_def.cipsov4,
+ entry->domain) != 0)
+ BUG();
+ break;
+ default:
+ BUG();
+ }
+ return -EPERM;
+ }
+ rcu_read_unlock();
+
+ return 0;
+}
+
+/**
+ * netlbl_domhsh_add_default - Adds the default entry to the domain hash table
+ * @entry: the entry to add
+ *
+ * Description:
+ * Adds a new default entry to the domain hash table and handles any updates
+ * to the lower level protocol handler (i.e. CIPSO). Returns zero on success,
+ * negative on failure.
+ *
+ */
+int netlbl_domhsh_add_default(struct netlbl_dom_map *entry)
+{
+ return netlbl_domhsh_add(entry);
+}
+
+/**
+ * netlbl_domhsh_remove - Removes an entry from the domain hash table
+ * @domain: the domain to remove
+ *
+ * Description:
+ * Removes an entry from the domain hash table and handles any updates to the
+ * lower level protocol handler (i.e. CIPSO). Returns zero on success,
+ * negative on failure.
+ *
+ */
+int netlbl_domhsh_remove(const char *domain)
+{
+ int ret_val = -EPERM;
+ struct netlbl_dom_map *entry;
+
+ rcu_read_lock();
+ if (domain != NULL)
+ entry = netlbl_domhsh_search(domain, 0);
+ else
+ entry = netlbl_domhsh_search(domain, 1);
+ if (entry == NULL)
+ goto remove_return;
+ switch (entry->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ if (cipso_v4_doi_domhsh_remove(entry->type_def.cipsov4,
+ entry->domain) != 0)
+ goto remove_return;
+ break;
+ default:
+ BUG();
+ }
+ if (entry != rcu_dereference(netlbl_domhsh_def)) {
+ spin_lock(&netlbl_domhsh_lock);
+ entry->valid = 0;
+ list_del_rcu(&entry->list);
+ spin_unlock(&netlbl_domhsh_lock);
+ } else {
+ spin_lock(&netlbl_domhsh_def_lock);
+ entry->valid = 0;
+ rcu_assign_pointer(netlbl_domhsh_def, NULL);
+ spin_unlock(&netlbl_domhsh_def_lock);
+ }
+ call_rcu(&entry->rcu, netlbl_domhsh_free_entry);
+
+ ret_val = 0;
+
+remove_return:
+ rcu_read_unlock();
+ return ret_val;
+}
+
+/**
+ * netlbl_domhsh_remove_default - Removes the default entry from the table
+ *
+ * Description:
+ * Removes/resets the default entry for the domain hash table and handles any
+ * updates to the lower level protocol handler (i.e. CIPSO). Returns zero on
+ * success, non-zero on failure.
+ *
+ */
+int netlbl_domhsh_remove_default(void)
+{
+ return netlbl_domhsh_remove(NULL);
+}
+
+/**
+ * netlbl_domhsh_getentry - Get an entry from the domain hash table
+ * @domain: the domain name to search for
+ *
+ * Description:
+ * Look through the domain hash table searching for an entry to match @domain,
+ * return a pointer to a copy of the entry or NULL. The caller is responsibile
+ * for ensuring that rcu_read_[un]lock() is called.
+ *
+ */
+struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain)
+{
+ return netlbl_domhsh_search(domain, 1);
+}
+
+/**
+ * netlbl_domhsh_dump - Dump the domain hash table into a sk_buff
+ *
+ * Description:
+ * Dump the domain hash table into a buffer suitable for returning to an
+ * application in response to a NetLabel management DOMAIN message. This
+ * function may fail if another process is growing the hash table at the same
+ * time. The returned sk_buff has room at the front of the sk_buff for
+ * @headroom bytes. See netlabel.h for the DOMAIN message format. Returns a
+ * pointer to a sk_buff on success, NULL on error.
+ *
+ */
+struct sk_buff *netlbl_domhsh_dump(const u32 headroom)
+{
+ struct sk_buff *skb;
+ unsigned char *buf;
+ u32 buf_len;
+ u32 bkt_iter;
+ u32 dom_cnt = 0;
+ struct netlbl_domhsh_tbl *hsh_tbl;
+ struct netlbl_dom_map *list_iter;
+ u32 tmp_len;
+
+ /* XXX - This is kinda ugly as we have to go through the table once to
+ determine how large of a buffer we need, drop the locks, allocate
+ the buffer, grab the locks, and finally fill the buffer. The
+ problem is that there is that open window where the table could
+ grow and we will end up short on space. */
+
+ buf_len = 4;
+ rcu_read_lock();
+ hsh_tbl = rcu_dereference(netlbl_domhsh);
+ for (bkt_iter = 0; bkt_iter < hsh_tbl->size; bkt_iter++)
+ list_for_each_entry_rcu(list_iter,
+ &hsh_tbl->tbl[bkt_iter], list) {
+ buf_len += 9 + strlen(list_iter->domain);
+ switch (list_iter->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ buf_len += 8;
+ break;
+ default:
+ rcu_read_unlock();
+ BUG();
+ }
+ dom_cnt++;
+ }
+ rcu_read_unlock();
+
+ skb = alloc_skb(NLMSG_ALIGN(headroom + buf_len), GFP_KERNEL);
+ if (skb == NULL)
+ return NULL;
+ skb_reserve(skb, headroom);
+ tmp_len = skb_tailroom(skb);
+ if (tmp_len < buf_len)
+ goto dump_failure;
+ buf = skb_put(skb, buf_len);
+ buf_len -= 4;
+ netlbl_putinc_u32(&buf, dom_cnt);
+
+ rcu_read_lock();
+ hsh_tbl = rcu_dereference(netlbl_domhsh);
+ for (bkt_iter = 0; bkt_iter < hsh_tbl->size; bkt_iter++)
+ list_for_each_entry_rcu(list_iter,
+ &hsh_tbl->tbl[bkt_iter], list) {
+ tmp_len = strlen(list_iter->domain) + 1;
+ if (buf_len < tmp_len + 8) {
+ rcu_read_unlock();
+ goto dump_failure;
+ }
+ buf_len -= tmp_len + 8;
+ netlbl_putinc_u32(&buf, tmp_len);
+ netlbl_putinc_str(&buf, list_iter->domain);
+ netlbl_putinc_u32(&buf, list_iter->type);
+ switch (list_iter->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ if (buf_len < 8) {
+ rcu_read_unlock();
+ goto dump_failure;
+ }
+ buf_len -= 8;
+ netlbl_putinc_u32(&buf,
+ list_iter->type_def.cipsov4->type);
+ netlbl_putinc_u32(&buf,
+ list_iter->type_def.cipsov4->doi);
+ break;
+ default:
+ rcu_read_unlock();
+ BUG();
+ }
+ }
+ rcu_read_unlock();
+
+ return skb;
+
+dump_failure:
+ kfree_skb(skb);
+ return NULL;
+}
+
+/**
+ * netlbl_domhsh_dump_default - Dump the default domain mapping into a sk_buff
+ *
+ * Description:
+ * Dump the default domain mapping into a buffer suitable for returning to an
+ * application in response to a NetLabel management DEFDOMAIN message. This
+ * function may fail if another process is changing the default domain mapping
+ * at the same time. The returned sk_buff has room at the front of the
+ * skb_buff for @headroom bytes. See netlabel.h for the DEFDOMAIN message
+ * format. Returns a pointer to a sk_buff on success, NULL on error.
+ *
+ */
+struct sk_buff *netlbl_domhsh_dump_default(const u32 headroom)
+{
+ struct sk_buff *skb;
+ unsigned char *buf;
+ u32 buf_len;
+ u32 tmp_len;
+ struct netlbl_dom_map *entry;
+
+ /* XXX - This is kinda ugly as we have to go look at the default
+ mapping once to determine how large of a buffer we need, drop the
+ locks, allocate the buffer, grab the locks, and finally fill the
+ buffer. The problem is that there is that open window where the
+ default mapping could change on us, if that is the case we fail. */
+
+ buf_len = 4;
+ rcu_read_lock();
+ entry = rcu_dereference(netlbl_domhsh_def);
+ if (entry != NULL)
+ switch (entry->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ buf_len += 8;
+ break;
+ default:
+ BUG();
+ }
+ rcu_read_unlock();
+
+ skb = alloc_skb(NLMSG_ALIGN(headroom + buf_len), GFP_KERNEL);
+ if (skb == NULL)
+ return NULL;
+ skb_reserve(skb, headroom);
+ tmp_len = skb_tailroom(skb);
+ if (tmp_len < buf_len)
+ goto dump_default_failure;
+ buf = skb_put(skb, buf_len);
+
+ rcu_read_lock();
+ if (entry != rcu_dereference(netlbl_domhsh_def)) {
+ rcu_read_unlock();
+ goto dump_default_failure;
+ }
+ buf_len -= 4;
+ if (entry != NULL) {
+ netlbl_putinc_u32(&buf, entry->type);
+ switch (entry->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ if (buf_len < 8) {
+ rcu_read_unlock();
+ goto dump_default_failure;
+ }
+ buf_len -= 8;
+ netlbl_putinc_u32(&buf, entry->type_def.cipsov4->type);
+ netlbl_putinc_u32(&buf, entry->type_def.cipsov4->doi);
+ break;
+ default:
+ rcu_read_unlock();
+ BUG();
+ }
+ } else
+ netlbl_putinc_u32(&buf, NETLBL_NLTYPE_NONE);
+ rcu_read_unlock();
+
+ return skb;
+
+dump_default_failure:
+ kfree_skb(skb);
+ return NULL;
+}
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_domainhash.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_domainhash.h
@@ -0,0 +1,64 @@
+/*
+ * NetLabel Domain Hash Table
+ *
+ * This file manages the domain hash table that NetLabel uses to determine
+ * which network labeling protocol to use for a given domain. The NetLabel
+ * system manages static and dynamic label mappings for network protocols such
+ * as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_DOMAINHASH_H
+#define _NETLABEL_DOMAINHASH_H
+
+/* Domain hash table size */
+/* XXX - currently this number is an uneducated guess */
+#define NETLBL_DOMHSH_BITSIZE 7
+
+/* Domain mapping definition struct */
+struct netlbl_dom_map {
+ char *domain;
+ u32 type;
+ union {
+ struct cipso_v4_doi *cipsov4;
+ } type_def;
+
+ u32 valid;
+ struct list_head list;
+ struct rcu_head rcu;
+};
+
+/* init/exit functions */
+int netlbl_domhsh_init(const u32 size);
+void netlbl_domhsh_exit(void);
+
+/* Manipulate the domain hash table */
+int netlbl_domhsh_add(struct netlbl_dom_map *entry);
+int netlbl_domhsh_add_default(struct netlbl_dom_map *entry);
+int netlbl_domhsh_remove_default(void);
+struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain);
+struct sk_buff *netlbl_domhsh_dump(const u32 headroom);
+struct sk_buff *netlbl_domhsh_dump_default(const u32 headroom);
+
+#endif
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_kapi.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_kapi.c
@@ -0,0 +1,363 @@
+/*
+ * NetLabel Kernel API
+ *
+ * This file defines the kernel API for the NetLabel system. The NetLabel
+ * system manages static and dynamic label mappings for network protocols such
+ * as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <net/ip.h>
+#include <net/netlabel.h>
+#include <net/cipso_ipv4.h>
+#include <asm/bug.h>
+
+#include "netlabel_domainhash.h"
+#include "netlabel_unlabeled.h"
+#include "netlabel_user.h"
+
+/* PM - remove a lot of the input verification BUG()s once this gets farther
+ along */
+
+/*
+ * LSM Functions
+ */
+
+/**
+ * netlbl_socket_setattr - Label a socket using the correct protocol
+ * @sock: the socket to label
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Attach the correct label to the given socket using the security attributes
+ * specified in @secattr. Returns zero on success, negative values on failure.
+ *
+ */
+int netlbl_socket_setattr(const struct socket *sock,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -ENOENT;
+ struct netlbl_dom_map *dom_entry;
+
+ /* PM - split these into individual BUGs so we can pinpoint problems
+ easier */
+ BUG_ON(sock == NULL);
+ BUG_ON(secattr == NULL);
+ BUG_ON(secattr->set_domain == 0);
+
+ rcu_read_lock();
+ dom_entry = netlbl_domhsh_getentry(secattr->domain);
+ if (dom_entry == NULL)
+ goto socket_setattr_return;
+ switch (dom_entry->type) {
+ case NETLBL_NLTYPE_CIPSOV4:
+ ret_val = cipso_v4_socket_setattr(sock,
+ dom_entry->type_def.cipsov4,
+ secattr);
+ break;
+ case NETLBL_NLTYPE_UNLABELED:
+ ret_val = 0;
+ break;
+ default:
+ ret_val = -ENOENT;
+ }
+
+socket_setattr_return:
+ rcu_read_unlock();
+ return ret_val;
+}
+
+/**
+ * netlbl_socket_peekattr - Get the security attributes on a queued packet
+ * @sock: the socket
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Peek at the incoming packet queue of @sock and return the security
+ * attributes of the first packet in the queue. If there are no packets in
+ * the queue or no packets with security attributes return -ENOMSG. Otherwise
+ * return zero on success, negative values on failure.
+ *
+ */
+int netlbl_socket_peekattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -ENOMSG;
+ struct sk_buff_head *sock_queue;
+ struct sk_buff *skb;
+ struct sk_buff *skb_tmp = NULL;
+
+ /* PM - split these into individual BUGs so we can pinpoint problems
+ easier */
+ BUG_ON(sock == NULL);
+ BUG_ON(sock->sk == NULL);
+ BUG_ON(sock->sk->sk_family != PF_INET);
+ BUG_ON(secattr == NULL);
+
+ sock_queue = &sock->sk->sk_receive_queue;
+ /* XXX - We could just take the lock here and do a skb_peek() but we
+ might end up holding the lock for a long time so grab the
+ lock, do a peek, and clone the result */
+ spin_lock(&sock_queue->lock);
+ skb = skb_peek(sock_queue);
+ if (skb != NULL)
+ skb_tmp = skb_clone(skb, GFP_ATOMIC);
+ spin_unlock(&sock_queue->lock);
+ if (skb_tmp != NULL) {
+ ret_val = netlbl_skbuff_getattr(skb_tmp, secattr);
+ kfree_skb(skb_tmp);
+ } else
+ ret_val = -ENOMEM;
+
+ return ret_val;
+}
+
+/**
+ * netlbl_socket_getattr - Determine the security attributes of a socket
+ * @sock: the socket
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Examines the given socket to see any NetLabel style labeling has been
+ * applied to the socket, if so it parses the socket label and returns the
+ * security attributes in @secattr. Returns zero on success, negative values
+ * on failure.
+ *
+ */
+int netlbl_socket_getattr(const struct socket *sock,
+ struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -ENOMSG;
+
+ /* PM - split these into individual BUGs so we can pinpoint problems
+ easier */
+ BUG_ON(sock == NULL);
+ BUG_ON(sock->sk == NULL);
+ BUG_ON(sock->sk->sk_family != PF_INET);
+ BUG_ON(secattr == NULL);
+
+ ret_val = cipso_v4_socket_getattr(sock, secattr);
+ if (ret_val == 0)
+ return 0;
+
+ ret_val = netlbl_unlabel_getattr(secattr);
+
+ return ret_val;
+}
+
+/**
+ * netlbl_skbuff_getattr - Determine the security attributes of a packet
+ * @skb: the packet
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Examines the given packet to see if a recognized form of packet labeling
+ * is present, if so it parses the packet label and returns the security
+ * attributes in @secattr. Returns zero on success, negative values on
+ * failure.
+ *
+ */
+int netlbl_skbuff_getattr(const struct sk_buff *skb,
+ struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -ENOMSG;
+
+ /* PM - split these into individual BUGs so we can pinpoint problems
+ easier */
+ BUG_ON(skb == NULL);
+ BUG_ON(secattr == NULL);
+
+ ret_val = cipso_v4_skbuff_getattr(skb, secattr);
+ if (ret_val == 0)
+ return 0;
+
+ ret_val = netlbl_unlabel_getattr(secattr);
+
+ return ret_val;
+}
+
+/**
+ * netlbl_skbuff_err - Handle a LSM error on a sk_buff
+ * @skb: the packet
+ * @error: the error code
+ *
+ * Description:
+ * Deal with a LSM problem when handling the packet in @skb, typically this is
+ * a permission denied problem (-EACCES). The correct action is determined
+ * according to the packet's labeling protocol. Returns zero on success,
+ * negative values on failure.
+ *
+ */
+int netlbl_skbuff_err(struct sk_buff *skb, int error)
+{
+ BUG_ON(skb == NULL);
+
+ if (CIPSO_V4_OPTEXIST(skb))
+ return cipso_v4_error(skb, error, 0);
+
+ return -ENOMSG;
+}
+
+/**
+ * netlbl_cache_invalidate - Invalidate all of the NetLabel protocol caches
+ *
+ * Description:
+ * For all of the NetLabel protocols that support some form of label mapping
+ * cache, invalidate the cache. Returns zero on success, negative values on
+ * error.
+ *
+ */
+void netlbl_cache_invalidate(void)
+{
+ cipso_v4_cache_invalidate();
+}
+
+/**
+ * netlbl_cache_add - Add an entry to a NetLabel protocol cache
+ * @skb: the packet
+ * @secattr: the packet's security attributes
+ *
+ * Description:
+ * Add the LSM security attributes for the given packet to the underlying
+ * NetLabel protocol's label mapping cache. Returns zero on success, negative
+ * values on error.
+ *
+ */
+int netlbl_cache_add(const struct sk_buff *skb,
+ const struct netlbl_lsm_secattr *secattr)
+{
+ int ret_val = -ENOMSG;
+
+ /* PM - split these into individual BUGs so we can pinpoint problems
+ easier */
+ BUG_ON(skb == NULL);
+ BUG_ON(secattr == NULL);
+
+ if (secattr->set_cache != 1)
+ return -ENOMSG;
+
+ if (CIPSO_V4_OPTEXIST(skb))
+ ret_val = cipso_v4_cache_add(skb, secattr);
+
+ return ret_val;
+}
+
+/*
+ * Network Stack Functions
+ */
+
+/**
+ * netlbl_socket_inet_accept - Preserve the NetLabel across an accept()
+ * @sock: the old socket
+ * @newsock: the new socket
+ *
+ * Description:
+ * For kernel threads only, clear the NetLabel from @newsock and copy the
+ * NetLabel from @sock. The caller is responsibile for locking and releasing
+ * the sockets.
+ *
+ */
+void netlbl_socket_inet_accept(struct socket *sock, struct socket *newsock)
+{
+ unsigned char opt_null = IPOPT_NOOP, *opt = NULL;
+ u32 opt_len = 0;
+
+ /* We are only interested in preserving the NetLabel on kernel
+ threads across an accept, the LSM should handle user level
+ accept calls. */
+ if (current->mm)
+ return;
+
+ /* PM - we can't return a error here (no way to un-accept a socket)
+ and i'm pretty sure i would be lynched if i put a BUG_ON() here
+ but we should probably do a printk() at the very least ... */
+
+ if (cipso_v4_socket_getopt(sock, &opt, &opt_len) == 0)
+ cipso_v4_socket_setopt(newsock, opt, opt_len);
+ else
+ cipso_v4_socket_setopt(newsock, &opt_null, 1);
+
+ return;
+}
+
+/*
+ * Setup Functions
+ */
+
+/**
+ * netlbl_init - Init for the NetLabel module
+ *
+ * Description:
+ * Perform the initialization of the NetLabel module before first use.
+ *
+ */
+static int __init netlbl_init(void)
+{
+ int ret_val = 0;
+
+ printk(KERN_INFO "NetLabel: Initializing\n");
+ printk(KERN_INFO "NetLabel: domain hash size = %u\n",
+ (1 << NETLBL_DOMHSH_BITSIZE));
+ printk(KERN_INFO "NetLabel: protocols ="
+ " UNLABELED"
+ " CIPSOv4"
+ "\n");
+
+ ret_val = netlbl_domhsh_init(NETLBL_DOMHSH_BITSIZE);
+ if (ret_val != 0)
+ goto init_failure;
+
+ ret_val = netlbl_netlink_init();
+ if (ret_val != 0)
+ goto init_failure;
+
+ ret_val = netlbl_unlabel_defconf();
+ if (ret_val != 0)
+ goto init_failure;
+ printk(KERN_INFO "NetLabel: unlabeled traffic allowed by default\n");
+
+ return 0;
+
+init_failure:
+ panic("NetLabel: failed to initialize properly (%d)\n", ret_val);
+}
+
+/* PM - not sure there is any point to this now, delete this func? */
+/**
+ * netlbl_exit - Cleanup for the NetLabel module
+ *
+ * Description:
+ * Perform any cleanup required before removing the NetLabel module.
+ *
+ */
+static void __exit netlbl_exit(void)
+{
+ netlbl_netlink_exit();
+ netlbl_domhsh_exit();
+}
+
+subsys_initcall(netlbl_init);
+
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_mgmt.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_mgmt.c
@@ -0,0 +1,707 @@
+/*
+ * NetLabel Management Support
+ *
+ * This file defines the management functions for the NetLabel system. The
+ * NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/string.h>
+#include <linux/skbuff.h>
+#include <net/sock.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+#include <net/netlabel.h>
+#include <net/cipso_ipv4.h>
+
+#include "netlabel_domainhash.h"
+#include "netlabel_user.h"
+#include "netlabel_mgmt.h"
+
+/* NetLabel Generic NETLINK CIPSOv4 family */
+static struct genl_family netlbl_mgmt_gnl_family = {
+ .id = GENL_ID_GENERATE,
+ .hdrsize = 0,
+ .name = NETLBL_NLTYPE_MGMT_NAME,
+ .version = NETLBL_PROTO_VERSION,
+ .maxattr = 0,
+};
+
+
+/*
+ * Local Prototypes
+ */
+
+static void netlbl_mgmt_send_ack(const struct genl_info *info,
+ const u32 ret_code);
+
+
+/*
+ * NetLabel Command Handlers
+ */
+
+/**
+ * netlbl_mgmt_add - Handle an ADD message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated ADD message and add the domains from the message
+ * to the hash table. See netlabel.h for a description of the message format.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val;
+ unsigned char *msg_ptr = netlbl_netlink_payload_data(skb);
+ u32 msg_len = netlbl_netlink_payload_len(skb);
+ u32 count;
+ struct netlbl_dom_map *entry = NULL;
+ u32 iter;
+ u32 tmp_val;
+ u32 rcu_locked = 0;
+
+ ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
+ if (ret_val != 0)
+ goto add_failure;
+
+ if (msg_len < 4)
+ goto add_failure;
+ count = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+
+ for (iter = 0; iter < count && msg_len > 0; iter++, entry = NULL) {
+ if (msg_len < 4)
+ goto add_failure;
+ tmp_val = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (tmp_val == 0)
+ goto add_failure;
+ entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+ if (entry == NULL) {
+ ret_val = -ENOMEM;
+ goto add_failure;
+ }
+
+ if (msg_len < tmp_val)
+ goto add_failure;
+ entry->domain = kmalloc(tmp_val, GFP_KERNEL);
+ if (entry->domain == NULL) {
+ ret_val = -ENOMEM;
+ goto add_failure;
+ }
+ strncpy(entry->domain, msg_ptr, tmp_val);
+ entry->domain[tmp_val - 1] = '\0';
+ msg_ptr += tmp_val;
+ msg_len -= tmp_val;
+
+ if (msg_len < 4)
+ goto add_failure;
+ tmp_val = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ entry->type = tmp_val;
+ switch (tmp_val) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ if (msg_len < 4)
+ goto add_failure;
+ tmp_val = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ /* XXX - we should be holding a rcu_read_lock() here
+ while we hold the result but since the entry
+ will always be deleted when the CIPSO DOI
+ is deleted we aren't going to keep the lock */
+ rcu_read_lock();
+ entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val);
+ if (entry->type_def.cipsov4 == NULL) {
+ rcu_read_unlock();
+ goto add_failure;
+ }
+ rcu_locked = 1;
+ break;
+ default:
+ goto add_failure;
+ }
+
+ ret_val = netlbl_domhsh_add(entry);
+ if (rcu_locked) {
+ rcu_read_unlock();
+ rcu_locked = 0;
+ }
+ if (ret_val != 0)
+ goto add_failure;
+ }
+
+ netlbl_mgmt_send_ack(info, NETLBL_E_OK);
+ return 0;
+
+add_failure:
+ if (entry) {
+ if (entry->domain)
+ kfree(entry->domain);
+ kfree(entry);
+ }
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_remove - Handle a REMOVE message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated REMOVE message and remove the specified domain
+ * mappings. Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_remove(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val;
+ unsigned char *msg_ptr = netlbl_netlink_payload_data(skb);
+ u32 msg_len = netlbl_netlink_payload_len(skb);
+ u32 count;
+ u32 iter;
+ u32 tmp_val;
+
+ ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
+ if (ret_val != 0)
+ goto remove_return;
+
+ if (msg_len < 4)
+ goto remove_return;
+ count = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+
+ for (iter = 0; iter < count && msg_len > 0; iter++) {
+ if (msg_len < 4)
+ goto remove_return;
+ tmp_val = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ if (tmp_val == 0 ||
+ msg_len < tmp_val || msg_ptr[tmp_val - 1] != '\0')
+ goto remove_return;
+ ret_val = netlbl_domhsh_remove(msg_ptr);
+ msg_ptr += tmp_val;
+ msg_len -= tmp_val;
+ if (ret_val != 0)
+ goto remove_return;
+ }
+
+ ret_val = 0;
+
+remove_return:
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_list - Handle a LIST message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated LIST message and dumps the domain hash table in a
+ * form suitable for use in a kernel generated LIST message. Returns zero on
+ * success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_list(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val = -ENOMEM;
+ struct sk_buff *ans_skb;
+ unsigned char *buf_ptr;
+
+ ans_skb = netlbl_domhsh_dump(NLMSG_SPACE(GENL_HDRLEN));
+ if (ans_skb == NULL)
+ goto list_failure;
+ buf_ptr = skb_push(ans_skb, NLMSG_SPACE(GENL_HDRLEN));
+ __netlbl_netlink_hdr(buf_ptr,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_mgmt_gnl_family.id,
+ NLBL_MGMT_C_LIST,
+ ans_skb->len);
+
+ ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+ if (ret_val != 0)
+ goto list_failure;
+
+ return 0;
+
+list_failure:
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_adddef - Handle an ADDDEF message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated ADDDEF message and respond accordingly. Returns
+ * zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val;
+ unsigned char *msg_ptr = netlbl_netlink_payload_data(skb);
+ u32 msg_len = netlbl_netlink_payload_len(skb);
+ struct netlbl_dom_map *entry = NULL;
+ u32 tmp_val;
+ u32 rcu_locked = 0;
+
+ ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
+ if (ret_val != 0)
+ goto adddef_failure;
+
+ if (msg_len < 4)
+ goto adddef_failure;
+ tmp_val = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+
+ entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+ if (entry == NULL) {
+ ret_val = -ENOMEM;
+ goto adddef_failure;
+ }
+ entry->type = tmp_val;
+
+ switch (entry->type) {
+ case NETLBL_NLTYPE_UNLABELED:
+ break;
+ case NETLBL_NLTYPE_CIPSOV4:
+ if (msg_len < 4)
+ goto adddef_failure;
+ tmp_val = netlbl_getinc_u32(&msg_ptr);
+ msg_len -= 4;
+ /* XXX - we should be holding a rcu_read_lock here while we
+ hold the result but since the entry will always be
+ deleted when the CIPSO DOI is deleted we are going
+ to skip the lock */
+ rcu_read_lock();
+ entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val);
+ if (entry->type_def.cipsov4 == NULL) {
+ rcu_read_unlock();
+ goto adddef_failure;
+ }
+ rcu_locked = 1;
+ break;
+ default:
+ ret_val = -EINVAL;
+ goto adddef_failure;
+ }
+
+ ret_val = netlbl_domhsh_add_default(entry);
+ if (rcu_locked) {
+ rcu_read_unlock();
+ rcu_locked = 0;
+ }
+ if (ret_val != 0)
+ goto adddef_failure;
+
+ netlbl_mgmt_send_ack(info, NETLBL_E_OK);
+ return 0;
+
+adddef_failure:
+ if (entry)
+ kfree(entry);
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_removedef - Handle a REMOVEDEF message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated REMOVEDEF message and remove the default domain
+ * mapping. Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_removedef(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val;
+
+ ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
+ if (ret_val != 0)
+ goto removedef_return;
+
+ ret_val = netlbl_domhsh_remove_default();
+
+removedef_return:
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_listdef - Handle a LISTDEF message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated LISTDEF message and dumps the default domain
+ * mapping in a form suitable for use in a kernel generated LISTDEF message.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val = -ENOMEM;
+ struct sk_buff *ans_skb;
+ unsigned char *buf_ptr;
+
+ ans_skb = netlbl_domhsh_dump_default(NLMSG_SPACE(GENL_HDRLEN));
+ if (ans_skb == NULL)
+ goto listdef_failure;
+ buf_ptr = skb_push(ans_skb, NLMSG_SPACE(GENL_HDRLEN));
+ __netlbl_netlink_hdr(buf_ptr,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_mgmt_gnl_family.id,
+ NLBL_MGMT_C_LISTDEF,
+ ans_skb->len);
+
+ ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+ if (ret_val != 0)
+ goto listdef_failure;
+
+ return 0;
+
+listdef_failure:
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_modules - Handle a MODULES message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated MODULES message and respond accordingly.
+ *
+ */
+static int netlbl_mgmt_modules(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val = -ENOMEM;
+ u32 mod_count;
+ size_t msg_size;
+ size_t data_size;
+ struct sk_buff *ans_skb = NULL;
+ unsigned char *data;
+
+ /* unlabeled + cipsov4 */
+ mod_count = 2;
+
+ data_size = GENL_HDRLEN + 4 + mod_count * 4;
+ msg_size = NLMSG_SPACE(data_size);
+
+ ans_skb = alloc_skb(msg_size, GFP_KERNEL);
+ if (ans_skb == NULL)
+ goto modules_failure;
+
+ data = netlbl_netlink_hdr_put(ans_skb,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_mgmt_gnl_family.id,
+ NLBL_MGMT_C_MODULES,
+ data_size);
+ if (data == NULL)
+ goto modules_failure;
+
+ netlbl_putinc_u32(&data, mod_count);
+ netlbl_putinc_u32(&data, NETLBL_NLTYPE_UNLABELED);
+ netlbl_putinc_u32(&data, NETLBL_NLTYPE_CIPSOV4);
+
+ ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+ if (ret_val != 0)
+ goto modules_failure;
+
+ return 0;
+
+modules_failure:
+ if (ans_skb)
+ kfree_skb(ans_skb);
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_version - Handle a VERSION message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated VERSION message and respond accordingly. Returns
+ * zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_version(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val = -ENOMEM;
+ size_t data_size;
+ size_t msg_size;
+ struct sk_buff *ans_skb = NULL;
+ unsigned char *data;
+
+ data_size = GENL_HDRLEN + 4;
+ msg_size = NLMSG_SPACE(data_size);
+
+ ans_skb = alloc_skb(msg_size, GFP_KERNEL);
+ if (ans_skb == NULL)
+ goto version_failure;
+ data = netlbl_netlink_hdr_put(ans_skb,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_mgmt_gnl_family.id,
+ NLBL_MGMT_C_VERSION,
+ data_size);
+ if (data == NULL)
+ goto version_failure;
+
+ netlbl_putinc_u32(&data, NETLBL_PROTO_VERSION);
+
+ ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+ if (ret_val != 0)
+ goto version_failure;
+
+ return 0;
+
+version_failure:
+ if (ans_skb)
+ kfree_skb(ans_skb);
+ netlbl_mgmt_send_ack(info, -ret_val);
+ return ret_val;
+}
+
+
+/*
+ * NetLabel Generic NETLINK Command Definitions
+ */
+
+static struct genl_ops netlbl_mgmt_genl_c_add = {
+ .cmd = NLBL_MGMT_C_ADD,
+ .flags = 0,
+ .doit = netlbl_mgmt_add,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_mgmt_genl_c_remove = {
+ .cmd = NLBL_MGMT_C_REMOVE,
+ .flags = 0,
+ .doit = netlbl_mgmt_remove,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_mgmt_genl_c_list = {
+ .cmd = NLBL_MGMT_C_LIST,
+ .flags = 0,
+ .doit = netlbl_mgmt_list,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_mgmt_genl_c_adddef = {
+ .cmd = NLBL_MGMT_C_ADDDEF,
+ .flags = 0,
+ .doit = netlbl_mgmt_adddef,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_mgmt_genl_c_removedef = {
+ .cmd = NLBL_MGMT_C_REMOVEDEF,
+ .flags = 0,
+ .doit = netlbl_mgmt_removedef,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_mgmt_genl_c_listdef = {
+ .cmd = NLBL_MGMT_C_LISTDEF,
+ .flags = 0,
+ .doit = netlbl_mgmt_listdef,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_mgmt_genl_c_modules = {
+ .cmd = NLBL_MGMT_C_MODULES,
+ .flags = 0,
+ .doit = netlbl_mgmt_modules,
+ .dumpit = NULL,
+};
+
+static struct genl_ops netlbl_mgmt_genl_c_version = {
+ .cmd = NLBL_MGMT_C_VERSION,
+ .flags = 0,
+ .doit = netlbl_mgmt_version,
+ .dumpit = NULL,
+};
+
+/*
+ * NetLabel Generic NETLINK Protocol Functions
+ */
+
+/**
+ * netlbl_mgmt_send_ack - Send an ACK message
+ * @info: the generic NETLINK information
+ * @ret_code: return code to use
+ *
+ * Description:
+ * This function sends an ACK message to the sender of the NETLINK message
+ * specified by @info.
+ *
+ */
+static void netlbl_mgmt_send_ack(const struct genl_info *info,
+ const u32 ret_code)
+{
+ size_t msg_size;
+ size_t data_size;
+ struct sk_buff *skb;
+ unsigned char *data;
+
+ data_size = GENL_HDRLEN + 8;
+ msg_size = NLMSG_SPACE(data_size);
+
+ skb = alloc_skb(msg_size, GFP_KERNEL);
+ if (skb == NULL)
+ return;
+
+ data = netlbl_netlink_hdr_put(skb,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_mgmt_gnl_family.id,
+ NLBL_MGMT_C_ACK,
+ data_size);
+ if (data == NULL)
+ goto send_ack_failure;
+
+ netlbl_putinc_u32(&data, info->snd_seq);
+ netlbl_putinc_u32(&data, ret_code);
+
+ netlbl_netlink_snd(skb, info->snd_pid);
+ return;
+
+send_ack_failure:
+ kfree_skb(skb);
+}
+
+/**
+ * netlbl_mgmt_genl_init - Register the NetLabel management component
+ *
+ * Description:
+ * Register the NetLabel management component with the Generic NETLINK
+ * mechanism. Returns zero on success, negative values on failure.
+ *
+ */
+int netlbl_mgmt_genl_init(void)
+{
+ int ret_val;
+
+ ret_val = genl_register_family(&netlbl_mgmt_gnl_family);
+ if (ret_val != 0)
+ return ret_val;
+
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_add);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_remove);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_list);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_adddef);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_removedef);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_listdef);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_modules);
+ if (ret_val != 0)
+ goto genl_init_failure;
+ ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_version);
+ if (ret_val != 0)
+ goto genl_init_failure;
+
+ return 0;
+
+genl_init_failure:
+ netlbl_mgmt_genl_exit();
+ return ret_val;
+}
+
+/**
+ * netlbl_mgmt_genl_exit - Unregister the NetLabel management component
+ *
+ * Description:
+ * Unregister the NetLabel management component with the Generic NETLINK
+ * mechanism. Returns zero on success, negative values on failure.
+ *
+ */
+void netlbl_mgmt_genl_exit(void)
+{
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_version);
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_modules);
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_listdef);
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_removedef);
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_adddef);
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_list);
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_remove);
+ genl_unregister_ops(&netlbl_mgmt_gnl_family,
+ &netlbl_mgmt_genl_c_add);
+
+ genl_unregister_family(&netlbl_mgmt_gnl_family);
+ netlbl_mgmt_gnl_family.id = GENL_ID_GENERATE;
+}
+
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_mgmt.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_mgmt.h
@@ -0,0 +1,250 @@
+/*
+ * NetLabel Management Support
+ *
+ * This file defines the management functions for the NetLabel system. The
+ * NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_MGMT_H
+#define _NETLABEL_MGMT_H
+
+#include <net/netlabel.h>
+
+/*
+ * The following NetLabel payloads are supported by the management interface,
+ * all of which are preceeded by the nlmsghdr struct.
+ *
+ * o ACK:
+ * Sent by the kernel in response to an applications message, applications
+ * should never send this message.
+ *
+ * +----------------------+-----------------------+
+ * | seq number (32 bits) | return code (32 bits) |
+ * +----------------------+-----------------------+
+ *
+ * seq number: the sequence number of the original message, taken from the
+ * nlmsghdr structure
+ * return code: return value, based on errno values
+ *
+ * o ADD:
+ * Sent by an application to add a domain mapping to the NetLabel system.
+ * The kernel should respond with an ACK.
+ *
+ * +-------------------+
+ * | domains (32 bits) | ...
+ * +-------------------+
+ *
+ * domains: the number of domains in the message
+ *
+ * +-------------------------+--------------------------+
+ * | domain length (32 bits) | domain string (variable) | ...
+ * +-------------------------+--------------------------+
+ *
+ * +-------------------------+-------------- ---- --- -- -
+ * | protocol type (32 bits) | mapping data ... repeated
+ * +-------------------------+-------------- ---- --- -- -
+ *
+ * domain length: the length of the "domain string" in bytes
+ * domain string: the domain string, NULL terminated
+ * protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
+ * mapping data: specific to the map type (see below)
+ *
+ * NETLBL_NLTYPE_UNLABELED
+ *
+ * No mapping data for this protocol type.
+ *
+ * NETLBL_NLTYPE_CIPSOV4
+ *
+ * +---------------+
+ * | doi (32 bits) |
+ * +---------------+
+ *
+ * doi: the CIPSO DOI value
+ *
+ * o REMOVE:
+ * Sent by an application to remove a domain mapping from the NetLabel
+ * system. The kernel should ACK this message.
+ *
+ * +-------------------+
+ * | domains (32 bits) | ...
+ * +-------------------+
+ *
+ * domains: the number of domains in the message
+ *
+ * +-------------------------+--------------------------+
+ * | domain length (32 bits) | domain string (variable) | ...
+ * +-------------------------+--------------------------+
+ *
+ * domain length: the length of the "domain string" in bytes
+ * domain string: the domain string, NULL terminated
+ *
+ * o LIST:
+ * This message can be sent either from an application or by the kernel in
+ * response to an application generated LIST message. When sent by an
+ * application there is no payload. The kernel should respond to a LIST
+ * message either with a LIST message on success or an ACK message on
+ * failure.
+ *
+ * +-------------------+
+ * | domains (32 bits) | ...
+ * +-------------------+
+ *
+ * domains: the number of domains in the message
+ *
+ * +-------------------------+--------------------------+
+ * | domain length (32 bits) | domain string (variable) | ...
+ * +-------------------------+--------------------------+
+ *
+ * +-------------------------+-------------- ---- --- -- -
+ * | protocol type (32 bits) | mapping data ... repeated
+ * +-------------------------+-------------- ---- --- -- -
+ *
+ * domain length: the length of the "domain string" in bytes
+ * domain string: the domain string, NULL terminated
+ * protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
+ * mapping data: specific to the map type (see below)
+ *
+ * NETLBL_NLTYPE_UNLABELED
+ *
+ * No mapping data for this protocol type.
+ *
+ * NETLBL_NLTYPE_CIPSOV4
+ *
+ * +----------------+---------------+
+ * | type (32 bits) | doi (32 bits) |
+ * +----------------+---------------+
+ *
+ * type: the CIPSO mapping table type (defined in the cipso_ipv4.h header
+ * as CIPSO_V4_MAP_*)
+ * doi: the CIPSO DOI value
+ *
+ * o ADDDEF:
+ * Sent by an application to set the default domain mapping for the NetLabel
+ * system. The kernel should respond with an ACK.
+ *
+ * +-------------------------+-------------- ---- --- -- -
+ * | protocol type (32 bits) | mapping data ... repeated
+ * +-------------------------+-------------- ---- --- -- -
+ *
+ * protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
+ * mapping data: specific to the map type (see below)
+ *
+ * NETLBL_NLTYPE_UNLABELED
+ *
+ * No mapping data for this protocol type.
+ *
+ * NETLBL_NLTYPE_CIPSOV4
+ *
+ * +---------------+
+ * | doi (32 bits) |
+ * +---------------+
+ *
+ * doi: the CIPSO DOI value
+ *
+ * o REMOVEDEF:
+ * Sent by an application to remove the default domain mapping from the
+ * NetLabel system, there is no payload. The kernel should ACK this message.
+ *
+ * o LISTDEF:
+ * This message can be sent either from an application or by the kernel in
+ * response to an application generated LISTDEF message. When sent by an
+ * application there is no payload. The kernel should respond to a
+ * LISTDEF message either with a LISTDEF message on success or an ACK message
+ * on failure.
+ *
+ * +-------------------------+-------------- ---- --- -- -
+ * | protocol type (32 bits) | mapping data ... repeated
+ * +-------------------------+-------------- ---- --- -- -
+ *
+ * protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
+ * mapping data: specific to the map type (see below)
+ *
+ * NETLBL_NLTYPE_UNLABELED
+ *
+ * No mapping data for this protocol type.
+ *
+ * NETLBL_NLTYPE_CIPSOV4
+ *
+ * +----------------+---------------+
+ * | type (32 bits) | doi (32 bits) |
+ * +----------------+---------------+
+ *
+ * type: the CIPSO mapping table type (defined in the cipso_ipv4.h header
+ * as CIPSO_V4_MAP_*)
+ * doi: the CIPSO DOI value
+ *
+ * o MODULES:
+ * Sent by an application to request a list of configured NetLabel modules
+ * in the kernel. When sent by an application there is no payload.
+ *
+ * +-------------------+
+ * | modules (32 bits) | ...
+ * +-------------------+
+ *
+ * modules: the number of modules in the message, if this is an application
+ * generated message and the value is zero then return a list of
+ * the configured modules
+ *
+ * +------------------+
+ * | module (32 bits) | ... repeated
+ * +------------------+
+ *
+ * module: the module number as defined by NETLBL_NLTYPE_*
+ *
+ * o VERSION:
+ * Sent by an application to request the NetLabel version string. When sent
+ * by an application there is no payload. This message type is also used by
+ * the kernel to respond to an VERSION request.
+ *
+ * +-------------------+
+ * | version (32 bits) |
+ * +-------------------+
+ *
+ * version: the protocol version number
+ *
+ */
+
+/* NetLabel Management commands */
+enum {
+ NLBL_MGMT_C_UNSPEC,
+ NLBL_MGMT_C_ACK,
+ NLBL_MGMT_C_ADD,
+ NLBL_MGMT_C_REMOVE,
+ NLBL_MGMT_C_LIST,
+ NLBL_MGMT_C_ADDDEF,
+ NLBL_MGMT_C_REMOVEDEF,
+ NLBL_MGMT_C_LISTDEF,
+ NLBL_MGMT_C_MODULES,
+ NLBL_MGMT_C_VERSION,
+ __NLBL_MGMT_C_MAX,
+};
+#define NLBL_MGMT_C_MAX (__NLBL_MGMT_C_MAX - 1)
+
+/* NetLabel protocol functions */
+int netlbl_mgmt_genl_init(void);
+void netlbl_mgmt_genl_exit(void);
+
+#endif
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_unlabeled.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_unlabeled.h
@@ -0,0 +1,84 @@
+/*
+ * NetLabel Unlabeled Support
+ *
+ * This file defines functions for dealing with unlabeled packets for the
+ * NetLabel system. The NetLabel system manages static and dynamic label
+ * mappings for network protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_UNLABELED_H
+#define _NETLABEL_UNLABELED_H
+
+#include <net/netlabel.h>
+
+/*
+ * The following NetLabel payloads are supported by the Unlabeled subsystem.
+ *
+ * o ACK:
+ * Sent by the kernel in response to an applications message, applications
+ * should never send this message.
+ *
+ * +----------------------+-----------------------+
+ * | seq number (32 bits) | return code (32 bits) |
+ * +----------------------+-----------------------+
+ *
+ * seq number: the sequence number of the original message, taken from the
+ * nlmsghdr structure
+ * return code: return value, based on errno values
+ *
+ * o ACCEPT
+ * This message is sent from an application to specify if the kernel should
+ * allow unlabled packets to pass if they do not match any of the static
+ * mappings defined in the unlabeled module.
+ *
+ * +-----------------+
+ * | allow (32 bits) |
+ * +-----------------+
+ *
+ * allow: if true (1) then allow the packets to pass, if false (0) then
+ * reject the packets
+ *
+ */
+
+/* NetLabel Unlabeled commands */
+enum {
+ NLBL_UNLABEL_C_UNSPEC,
+ NLBL_UNLABEL_C_ACK,
+ NLBL_UNLABEL_C_ACCEPT,
+ __NLBL_UNLABEL_C_MAX,
+};
+#define NLBL_UNLABEL_C_MAX (__NLBL_UNLABEL_C_MAX - 1)
+
+/* NetLabel protocol functions */
+int netlbl_unlabel_genl_init(void);
+void netlbl_unlabel_genl_exit(void);
+
+/* Process Unlabeled incoming network packets */
+int netlbl_unlabel_getattr(struct netlbl_lsm_secattr *secattr);
+
+/* Set the default configuration to allow Unlabeled packets */
+int netlbl_unlabel_defconf(void);
+
+#endif
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_user.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_user.c
@@ -0,0 +1,131 @@
+/*
+ * NetLabel NETLINK Interface
+ *
+ * This file defines the NETLINK interface for the NetLabel system. The
+ * NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/socket.h>
+#include <net/sock.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+#include <net/netlabel.h>
+#include <asm/bug.h>
+
+#include "netlabel_mgmt.h"
+#include "netlabel_unlabeled.h"
+#include "netlabel_cipso_v4.h"
+
+
+/*
+ * NetLabel NETLINK Setup Functions
+ */
+
+/**
+ * netlbl_netlink_init - Initialize the NETLINK communication channel
+ *
+ * Description:
+ * Call out to the NetLabel components so they can register their families and
+ * commands with the Generic NETLINK mechanism. Returns zero on success and
+ * non-zero on failure.
+ *
+ */
+int netlbl_netlink_init(void)
+{
+ int ret_val;
+
+ ret_val = netlbl_mgmt_genl_init();
+ if (ret_val != 0)
+ goto init_failure;
+
+ ret_val = netlbl_cipsov4_genl_init();
+ if (ret_val != 0)
+ goto init_failure;
+
+ ret_val = netlbl_unlabel_genl_init();
+ if (ret_val != 0)
+ goto init_failure;
+
+ return 0;
+
+init_failure:
+ netlbl_unlabel_genl_exit();
+ return ret_val;
+}
+
+/**
+ * netlbl_exit - Shutdown the NETLINK communication channel
+ *
+ * Description:
+ * Call out to the NetLabel components so they can deregister themselves from
+ * the Generic NETLINK mechanism.
+ *
+ */
+void netlbl_netlink_exit(void)
+{
+ netlbl_unlabel_genl_exit();
+ netlbl_cipsov4_genl_exit();
+ netlbl_mgmt_genl_exit();
+}
+
+/*
+ * NETLINK I/O Functions
+ */
+
+/**
+ * netlbl_netlink_snd - Send a NetLabel message
+ * @skb: NetLabel message
+ * @pid: destination PID
+ *
+ * Description:
+ * Sends a unicast NetLabel message over the NETLINK socket.
+ *
+ */
+int netlbl_netlink_snd(struct sk_buff *skb, const u32 pid)
+{
+ return genlmsg_unicast(skb, pid);
+}
+
+/**
+ * netlbl_netlink_snd - Send a NetLabel message
+ * @skb: NetLabel message
+ * @pid: sending PID
+ * @group: multicast group id
+ *
+ * Description:
+ * Sends a multicast NetLabel message over the NETLINK socket to all members
+ * of @group except @pid.
+ *
+ */
+int netlbl_netlink_snd_multicast(struct sk_buff *skb,
+ const u32 pid,
+ const u32 group)
+{
+ return genlmsg_multicast(skb, pid, group);
+}
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_user.h
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_user.h
@@ -0,0 +1,175 @@
+/*
+ * NetLabel NETLINK Interface
+ *
+ * This file defines the NETLINK interface for the NetLabel system. The
+ * NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_USER_H
+#define _NETLABEL_USER_H
+
+#include <linux/skbuff.h>
+#include <linux/capability.h>
+#include <linux/genetlink.h>
+#include <net/netlabel.h>
+
+/* NetLabel NETLINK helper functions */
+
+/**
+ * netlbl_netlink_cap_check - Check the NETLINK msg capabilities
+ * @skb: the NETLINK buffer
+ * @req_cap: the required capability
+ *
+ * Description:
+ * Check the NETLINK buffer's capabilities against the required capabilities.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static inline int netlbl_netlink_cap_check(const struct sk_buff *skb,
+ const kernel_cap_t req_cap)
+{
+ if (cap_raised(NETLINK_CB(skb).eff_cap, req_cap))
+ return 0;
+ return -EPERM;
+}
+
+/**
+ * netlbl_netlink_hdr_put - Write the NETLINK buffers into a buffer
+ * @skb: the buffer
+ * @pid: the PID of the receipient
+ * @seq: the sequence number
+ * @flags: NETLINK message flags
+ * @type: the generic NETLINK message family type
+ * @cmd: command
+ * @len: length
+ *
+ * Description:
+ * Write both a NETLINK nlmsghdr structure and a Generic NETLINK genlmsghdr
+ * struct to a buffer. Returns a pointer to the start of the payload buffer
+ * on success or NULL on failure.
+ *
+ */
+static inline void __netlbl_netlink_hdr(unsigned char *buffer,
+ const u32 pid,
+ const u32 seq,
+ const int flags,
+ const int type,
+ const u8 cmd,
+ const int len)
+{
+ struct nlmsghdr *nlh;
+ struct genlmsghdr *hdr;
+
+ nlh = (struct nlmsghdr *)buffer;
+ nlh->nlmsg_type = type;
+ nlh->nlmsg_len = len;
+ nlh->nlmsg_flags = flags;
+ nlh->nlmsg_pid = pid;
+ nlh->nlmsg_seq = seq;
+
+ hdr = nlmsg_data(nlh);
+ hdr->cmd = cmd;
+ hdr->version = NETLBL_PROTO_VERSION;
+ hdr->reserved = 0;
+}
+
+/**
+ * netlbl_netlink_hdr_put - Write the NETLINK buffers into a sk_buff
+ * @skb: the packet
+ * @pid: the PID of the receipient
+ * @seq: the sequence number
+ * @flags: NETLINK message flags
+ * @type: the generic NETLINK message family type
+ * @cmd: command
+ * @len: length
+ *
+ * Description:
+ * Write both a NETLINK nlmsghdr structure and a Generic NETLINK genlmsghdr
+ * struct to the packet. Returns a pointer to the start of the payload buffer
+ * on success or NULL on failure.
+ *
+ */
+static inline unsigned char *netlbl_netlink_hdr_put(struct sk_buff *skb,
+ const u32 pid,
+ const u32 seq,
+ const int flags,
+ const int type,
+ const u8 cmd,
+ const int len)
+{
+ struct nlmsghdr *nlh;
+ struct genlmsghdr *hdr;
+
+ nlh = nlmsg_put(skb, pid, seq, type, len, flags);
+ if (nlh == NULL)
+ return NULL;
+
+ hdr = nlmsg_data(nlh);
+ hdr->cmd = cmd;
+ hdr->version = NETLBL_PROTO_VERSION;
+ hdr->reserved = 0;
+
+ return (unsigned char *)hdr + GENL_HDRLEN;
+}
+
+/**
+ * netlbl_netlink_payload_len - Return the length of the payload
+ * @skb: the NETLINK buffer
+ *
+ * Description:
+ * This function returns the length of the NetLabel payload.
+ *
+ */
+static inline u32 netlbl_netlink_payload_len(const struct sk_buff *skb)
+{
+ return nlmsg_len((struct nlmsghdr *)skb->data) - GENL_HDRLEN;
+}
+
+/**
+ * netlbl_netlink_payload_data - Returns a pointer to the start of the payload
+ * @skb: the NETLINK buffer
+ *
+ * Description:
+ * This function returns a pointer to the start of the NetLabel payload.
+ *
+ */
+static inline unsigned char *netlbl_netlink_payload_data(
+ const struct sk_buff *skb)
+{
+ return nlmsg_data((struct nlmsghdr *)skb->data) + GENL_HDRLEN;
+}
+
+
+/* NetLabel NETLINK I/O functions */
+
+int netlbl_netlink_init(void);
+void netlbl_netlink_exit(void);
+int netlbl_netlink_snd(struct sk_buff *skb, const u32 pid);
+int netlbl_netlink_snd_multicast(struct sk_buff *skb,
+ const u32 pid,
+ const u32 group);
+
+#endif
--
paul moore
linux security @ hp
^ permalink raw reply
* [RFC 7/8] NetLabel: unlabeled packet handling
From: paul.moore @ 2006-06-27 22:56 UTC (permalink / raw)
To: redhat-lspp, netdev; +Cc: sgrubb, pratt
In-Reply-To: <20060627225631.538814000@flek.zko.hp.com>
[-- Attachment #1: netlabel-unlabeled --]
[-- Type: text/plain, Size: 7060 bytes --]
Add unlabeled packet support to the NetLabel subsystem. NetLabel does not do
any processing on unlabled packets, but it must support passing unlabled
packets on both the inbound and outbound sides.
---
net/netlabel/netlabel_unlabeled.c | 258 ++++++++++++++++++++++++++++++++++++++
1 files changed, 258 insertions(+)
Index: linux-2.6.17.i686-quilt/net/netlabel/netlabel_unlabeled.c
===================================================================
--- /dev/null
+++ linux-2.6.17.i686-quilt/net/netlabel/netlabel_unlabeled.c
@@ -0,0 +1,258 @@
+/*
+ * NetLabel Unlabeled Support
+ *
+ * This file defines functions for dealing with unlabeled packets for the
+ * NetLabel system. The NetLabel system manages static and dynamic label
+ * mappings for network protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/socket.h>
+#include <linux/string.h>
+#include <linux/skbuff.h>
+#include <net/sock.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+
+#include <net/netlabel.h>
+#include <asm/bug.h>
+
+#include "netlabel_user.h"
+#include "netlabel_domainhash.h"
+#include "netlabel_unlabeled.h"
+
+/* Accept unlabeled packets flag */
+static atomic_t netlabel_unlabel_accept_flg = ATOMIC_INIT(0);
+
+/* NetLabel Generic NETLINK CIPSOv4 family */
+static struct genl_family netlbl_unlabel_gnl_family = {
+ .id = GENL_ID_GENERATE,
+ .hdrsize = 0,
+ .name = NETLBL_NLTYPE_UNLABELED_NAME,
+ .version = NETLBL_PROTO_VERSION,
+ .maxattr = 0,
+};
+
+
+/*
+ * Local Prototypes
+ */
+
+static void netlbl_unlabel_send_ack(const struct genl_info *info,
+ const u32 ret_code);
+
+
+/*
+ * NetLabel Command Handlers
+ */
+
+/**
+ * netlbl_unlabel_accept - Handle an ACCEPT message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated ACCEPT message and set the accept flag accordingly.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_unlabel_accept(struct sk_buff *skb, struct genl_info *info)
+{
+ int ret_val;
+ unsigned char *msg = netlbl_netlink_payload_data(skb);
+ u32 value;
+
+ ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
+ if (ret_val != 0)
+ return ret_val;
+
+ if (netlbl_netlink_payload_len(skb) == 4) {
+ value = netlbl_get_u32(msg);
+ if (value == 1 || value == 0) {
+ atomic_set(&netlabel_unlabel_accept_flg, value);
+ netlbl_unlabel_send_ack(info, NETLBL_E_OK);
+ return 0;
+ }
+ }
+
+ netlbl_unlabel_send_ack(info, EINVAL);
+ return -EINVAL;
+}
+
+
+/*
+ * NetLabel Generic NETLINK Command Definitions
+ */
+
+static struct genl_ops netlbl_unlabel_genl_c_accept = {
+ .cmd = NLBL_UNLABEL_C_ACCEPT,
+ .flags = 0,
+ .doit = netlbl_unlabel_accept,
+ .dumpit = NULL,
+};
+
+/*
+ * NetLabel Generic NETLINK Protocol Functions
+ */
+
+/**
+ * netlbl_unlabel_send_ack - Send an ACK message
+ * @info: the generic NETLINK information
+ * @ret_code: return code to use
+ *
+ * Description:
+ * This function sends an ACK message to the sender of the NETLINK message
+ * specified by @info.
+ *
+ */
+static void netlbl_unlabel_send_ack(const struct genl_info *info,
+ const u32 ret_code)
+{
+ size_t msg_size;
+ size_t data_size;
+ struct sk_buff *skb;
+ unsigned char *data;
+
+ data_size = GENL_HDRLEN + 8;
+ msg_size = NLMSG_SPACE(data_size);
+
+ skb = alloc_skb(msg_size, GFP_KERNEL);
+ if (skb == NULL)
+ return;
+
+ data = netlbl_netlink_hdr_put(skb,
+ info->snd_pid,
+ 0,
+ 0,
+ netlbl_unlabel_gnl_family.id,
+ NLBL_UNLABEL_C_ACK,
+ data_size);
+ if (data == NULL)
+ goto send_ack_failure;
+
+ netlbl_putinc_u32(&data, info->snd_seq);
+ netlbl_putinc_u32(&data, ret_code);
+
+ netlbl_netlink_snd(skb, info->snd_pid);
+ return;
+
+send_ack_failure:
+ kfree_skb(skb);
+}
+
+/**
+ * netlbl_unlabel_genl_init - Register the Unlabeled NetLabel component
+ *
+ * Description:
+ * Register the unlabeled packet NetLabel component with the Generic NETLINK
+ * mechanism. Returns zero on success, negative values on failure.
+ *
+ */
+int netlbl_unlabel_genl_init(void)
+{
+ int ret_val;
+
+ ret_val = genl_register_family(&netlbl_unlabel_gnl_family);
+ if (ret_val != 0)
+ return ret_val;
+
+ ret_val = genl_register_ops(&netlbl_unlabel_gnl_family,
+ &netlbl_unlabel_genl_c_accept);
+ if (ret_val != 0)
+ return ret_val;
+
+ return 0;
+}
+
+/**
+ * netlbl_unlabel_genl_exit - Unregister the Unlabeled NetLabel component
+ *
+ * Description:
+ * Unregister the unlabeled packet NetLabel component with the Generic NETLINK
+ * mechanism. Returns zero on success, negative values on failure.
+ *
+ */
+void netlbl_unlabel_genl_exit(void)
+{
+ genl_unregister_ops(&netlbl_unlabel_gnl_family,
+ &netlbl_unlabel_genl_c_accept);
+
+ genl_unregister_family(&netlbl_unlabel_gnl_family);
+ netlbl_unlabel_gnl_family.id = GENL_ID_GENERATE;
+}
+
+/*
+ * NetLabel KAPI Hooks
+ */
+
+/**
+ * netlbl_unlabel_getattr - Get the security attributes for an unlabled packet
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Determine the security attributes, if any, for an unlabled packet and return
+ * them in @secattr. Returns zero on success and negative values on failure.
+ *
+ */
+int netlbl_unlabel_getattr(struct netlbl_lsm_secattr *secattr)
+{
+ BUG_ON(secattr == NULL);
+
+ if (atomic_read(&netlabel_unlabel_accept_flg) == 1) {
+ memset(secattr, 0, sizeof(*secattr));
+ return 0;
+ }
+
+ return -ENOMSG;
+}
+
+/**
+ * netlbl_unlabel_defconf - Set the default config to allow unlabeled packets
+ *
+ * Description:
+ * Set the default NetLabel configuration to allow incoming unlabeled packets
+ * and to send unlabeled network traffic by default.
+ *
+ */
+int netlbl_unlabel_defconf(void)
+{
+ int ret_val;
+ struct netlbl_dom_map *entry;
+
+ entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+ if (entry == NULL)
+ return -ENOMEM;
+ entry->type = NETLBL_NLTYPE_UNLABELED;
+ ret_val = netlbl_domhsh_add_default(entry);
+ if (ret_val != 0)
+ return ret_val;
+
+ atomic_set(&netlabel_unlabel_accept_flg, 1);
+
+ return 0;
+}
--
paul moore
linux security @ hp
^ permalink raw reply
* Re: [patch 2/6] [Network namespace] Network device sharing by view
From: Herbert Poetzl @ 2006-06-27 23:07 UTC (permalink / raw)
To: Eric W. Biederman
Cc: Kirill Korotaev, Daniel Lezcano, Andrey Savochkin, linux-kernel,
netdev, serue, haveblue, clg, Andrew Morton, devel, sam, viro,
Alexey Kuznetsov
In-Reply-To: <m1y7vilfyk.fsf@ebiederm.dsl.xmission.com>
On Tue, Jun 27, 2006 at 10:29:39AM -0600, Eric W. Biederman wrote:
> Herbert Poetzl <herbert@13thfloor.at> writes:
>
> > On Tue, Jun 27, 2006 at 01:54:51PM +0400, Kirill Korotaev wrote:
> >> >>My point is that if you make namespace tagging at routing time, and
> >> >>your packets are being routed only once, you lose the ability
> >> >>to have separate routing tables in each namespace.
> >> >
> >> >
> >> >Right. What is the advantage of having separate the routing tables ?
> >
> >> it is impossible to have bridged networking, tun/tap and many other
> >> features without it. I even doubt that it is possible to introduce
> >> private netfilter rules w/o virtualization of routing.
> >
> > why? iptables work quite fine on a typical linux
> > system when you 'delegate' certain functionality
> > to certain chains (i.e. doesn't require access to
> > _all_ of them)
> >
> >> The question is do we want to have fully featured namespaces which
> >> allow to create isolated virtual environments with semantics and
> >> behaviour of standalone linux box or do we want to introduce some
> >> hacks with new rules/restrictions to meet ones goals only?
> >
> > well, soemtimes 'hacks' are not only simpler but also
> > a much better solution for a given problem than the
> > straight forward approach ...
>
> Well I would like to see a hack that qualifies.
> I watched the linux-vserver irc channel for a while and almost
> every network problem was caused by the change in semantics
> vserver provides.
the problem here is not the change in semantics compared
to a real linux system (as there basically is none) but
compared to _other_ technologies like UML or QEMU, which
add the need for bridging and additional interfaces, while
Linux-VServer only focuses on the IP layer ...
> In this case when you allow a guest more than one IP your hack
> while easy to maintain becomes much more complex.
why? a set of IPs is quite similar to a single IP (which
is actually a subset), so no real change there, only
IP_ANY means something different for a guest ...
> Especially as you address each case people care about one at a time.
hmm?
> In one shot this goes the entire way. Given how many people miss that
> you do the work at layer 2 than at layer 3 I would not call this the
> straight forward approach. The straight forward implementation yes,
> but not the straight forward approach.
seems I lost you here ...
> > for example, you won't have multiple routing tables
> > in a kernel where this feature is disabled, no?
> > so why should it affect a guest, or require modified
> > apps inside a guest when we would decide to provide
> > only a single routing table?
> >
> >> From my POV, fully virtualized namespaces are the future.
> >
> > the future is already there, it's called Xen or UML, or QEMU :)
>
> Yep. And now we need it to run fast.
hmm, maybe you should try to optimize linux for Xen then,
as I'm sure it will provide the optimal virtualization
and has all the features folks are looking for (regarding
virtualization)
I thought we are trying to figure a light-weight subset
of isolation and virtualization technologies and methods
which make sense to have in mainline ...
> >> It is what makes virtualization solution usable (w/o apps
> >> modifications), provides all the features and doesn't require much
> >> efforts from people to be used.
> >
> > and what if they want to use virtualization inside
> > their guests? where do you draw the line?
>
> The implementation doesn't have any problems with guests inside
> of guests.
>
> The only reason to restrict guests inside of guests is because
> the we aren't certain which permissions make sense.
well, we have not even touched the permission issues yet
best,
Herbert
> Eric
^ permalink raw reply
* Re: [patch 2/6] [Network namespace] Network device sharing by view
From: Dave Hansen @ 2006-06-27 23:12 UTC (permalink / raw)
To: Herbert Poetzl
Cc: Ben Greear, Eric W. Biederman, Daniel Lezcano, Andrey Savochkin,
linux-kernel, netdev, serue, clg, Andrew Morton, dev, devel, sam,
viro, Alexey Kuznetsov
In-Reply-To: <20060627225213.GB2612@MAIL.13thfloor.at>
On Wed, 2006-06-28 at 00:52 +0200, Herbert Poetzl wrote:
> seriously, what I think Eric meant was that it
> might be nice (especially for migration purposes)
> to keep the device namespace completely virtualized
> and not just isolated ...
It might be nice, but it is probably unneeded for an initial
implementation. In practice, a cluster doing
checkpoint/restart/migration will already have a system in place for
assigning unique IPs or other identifiers to each container. It could
just as easily make sure to assign unique network device names to
containers.
The issues really only come into play when you have an unstructured set
of machines and you want to migrate between them without having prepared
them with any kind of unique net device names beforehand.
It may look weird, but do application really *need* to see eth0 rather
than eth858354?
-- Dave
^ permalink raw reply
* Re: [Repost PATCH 6/6] PMC MSP85x0 gigabit ethernet driver
From: Francois Romieu @ 2006-06-27 23:20 UTC (permalink / raw)
To: Kiran Thota; +Cc: 'Yoichi Yuasa', linux-mips, netdev, Raj Palani, ralf
In-Reply-To: <C28979E4F697C249ABDA83AC0C33CDF80B6BC8@sjc1exm07.pmc_nt.nt.pmc-sierra.bc.ca>
Kiran Thota <Kiran_Thota@pmc-sierra.com> :
[...]
> +/*
> + * Allocate the SKBs for the Rx ring. Also used
> + * for refilling the queue
> + */
> +
> +static int msp85x0_ge_rx_task(struct net_device *netdev,
> + msp85x0_ge_port_info *msp85x0_ge_eth)
> +{
> + struct device *device = &msp85x0_ge_device[msp85x0_ge_eth->port_num]->dev;
> + volatile msp85x0_ge_rx_desc *rx_desc;
> + struct sk_buff *skb;
> + int rx_used_desc;
> + int count = 0;
> + oom_flag=0;
Global variable.
[...]
> + if((rx_used_desc + 1) == MSP85x0_GE_RX_QUEUE)
> + msp85x0_ge_eth->rx_used_desc_q =0;
> + else
> + msp85x0_ge_eth->rx_used_desc_q = (rx_used_desc + 1);
Consider greping drivers/net for NEXT_TX or RING_NEXT.
[...]
> +static void msp85x0_port_init(struct net_device *netdev,
> + msp85x0_ge_port_info * msp85x0_ge_eth)
> +{
> + unsigned long reg_data;
> + unsigned int port_num;
> +
> + port_num = msp85x0_ge_eth->port_num;
> + for (port_num = 0; port_num < NO_PORTS; port_num++)
There is something strange with port_num here.
[...]
> +static int start_tx_and_rx_activity(struct net_device *netdev)
> +{
The returned value is not used.
[...]
> +static int trtg_block_enable(struct net_device *netdev)
> +{
The returned value is not used.
[...]
> +static int enable_tx_and_rx_interrupts(struct net_device *netdev)
> +{
The returned value is not used.
[...]
> +static int xdma_config(struct net_device *netdev)
> +{
The indentation of this function is mostly broken.
[...]
> +static int msp85x0_ge_port_start(struct net_device *netdev)
> +{
The returned value is not used.
[...]
> +static int msp85x0_eth_setup_tx_rx_fifo(struct net_device *dev)
> +{
The returned value is not used.
[...]
> +static int msp85x0_ge_eth_open(struct net_device *netdev)
> +{
[...]
> + /* Fill the Rx ring with the SKBs */
> + msp85x0_ge_port_start(netdev);
[...]
> + if (!(phy_reg & 0x0400)) {
> + netif_carrier_off(netdev);
> + netif_stop_queue(netdev);
> + return MSP85x0_ERROR;
skb leak
[...]
> +int msp85x0_ge_start_xmit(struct sk_buff *skb, struct net_device *netdev)
> +{
static
This function ought to use NETDEV_TX_OK/NETDEV_TX_BUSY (should not happen).
[...]
> +static int msp85x0_ge_free_tx_queue(struct net_device *netdev)
> +{
> + msp85x0_ge_port_info *msp85x0_ge_eth = netdev_priv(netdev);
> + int pkts,port_num = msp85x0_ge_eth->port_num;
> + int tx_desc_used;
> + struct sk_buff *skb;
> +
> + /* Take the lock */
> + pkts=get_tx_pkt_count(port_num);
> + while(pkts)
> + {
> + pkts--;
> + tx_desc_used = msp85x0_ge_eth->tx_used_desc_q;
> +
> + /* return right away */
> + if (tx_desc_used == msp85x0_ge_eth->tx_curr_desc_q)
> + break;
> +
> + skb = msp85x0_ge_eth->tx_skb[tx_desc_used];
> + dev_kfree_skb_irq(skb);
msp85x0_ge_free_tx_queue() is issued in msp85x0_ge_start_xmit(), thus
not in irq context.
[...]
> +static int msp85x0_ge_receive_queue(struct net_device *netdev)
> +{
Indentation needs to fixed in this function.
[...]
> + if (packet.cmd_sts & (MSP85x0_GE_RX_PERR | MSP85x0_GE_RX_OVERFLOW_ERROR | MSP85x0_GE_RX_TRUNC | MSP85x0_GE_RX_CRC_ERROR))
> + {
> + if(packet.cmd_sts & MSP85x0_GE_RX_OVERFLOW_ERROR)
> + stats->rx_over_errors++;
> + else if(packet.cmd_sts & MSP85x0_GE_RX_TRUNC)
> + stats->rx_frame_errors++;
> + else
> + stats->rx_errors++;
> + dev_kfree_skb_any(skb);
It's called in ->poll(), outside of in_irq().
dev->last_rx should be updated after netif_receive_skb().
[...]
> +static int msp85x0_ge_poll(struct net_device *netdev, int *budget)
> +{
[...]
> + spin_lock_irqsave(&msp85x0_ge_eth->lock,flags);
Afaik poll takes place with irq enabled: no need to save/restore.
[...]
> +/* Don't Re-Initialize the port, Just start from where it stops */
> +static int msp85x0_ge_eth_reopen(struct net_device *netdev)
> +{
> + msp85x0_ge_port_info *msp85x0_ge_eth = netdev_priv(netdev);
> + unsigned int reg_data,irq;
> + int retval;
> +
> + irq = MSP85x0_ETH_PORT_IRQ;
> +
> + retval = request_irq(irq, INTERRUPT_HANDLER,
> + SA_INTERRUPT | SA_SAMPLE_RANDOM | SA_SHIRQ, netdev->name, netdev);
/me scratches head...
msp85x0_ge_change_mtu() does _not_ free_irqv and it issues
msp85x0_ge_eth_reopen().
I noticed this comment in msp85x0_ge_eth_stop():
/* This to work around to solve the msp85x0 shutdown and bringup sequence */
Can you elaborate ?
Random remarks:
- drivers/net/msp85x0_ge.h includes a lot of
#define MSP85x0_GE_MSTATX_SOMETHING
Your customers would surely appreciate extended stats through ethtool.
grep for get_ethtool_stats in drivers/net
- You should be able to sprinkle a few NET_IP_ALIGN here and there.
- I won't complain if you feel an urge to remove the _ge_ part in
msp85x0_ge_whatever
--
Ueimor
^ permalink raw reply
* Re: [patch 2/6] [Network namespace] Network device sharing by view
From: Alexey Kuznetsov @ 2006-06-27 23:42 UTC (permalink / raw)
To: Dave Hansen
Cc: Herbert Poetzl, Ben Greear, Eric W. Biederman, Daniel Lezcano,
Andrey Savochkin, linux-kernel, netdev, serue, clg, Andrew Morton,
dev, devel, sam, viro, Alexey Kuznetsov
In-Reply-To: <1151449973.24103.51.camel@localhost.localdomain>
Hello!
> It may look weird, but do application really *need* to see eth0 rather
> than eth858354?
Applications do not care, humans do. :-)
What's about applications they just need to see exactly the same device
after migration. Not only name, but f.e. also its ifindex. If you do not
create a separate namespace for netdevices, you will inevitably end up
with some strange hack sort of VPIDs to translate (or to partition) ifindices
or to tell that "ping -I eth858354 xxx" is too coimplicated application
to survive migration.
Alexey
^ permalink raw reply
* Please pull 'upstream' branch of wireless-2.6 (revised)
From: John W. Linville @ 2006-06-28 0:48 UTC (permalink / raw)
To: jeff; +Cc: netdev
In-Reply-To: <20060626212547.GE30706@tuxdriver.com>
On Mon, Jun 26, 2006 at 05:25:52PM -0400, John W. Linville wrote:
> Michael Buesch:
> bcm43xx: suspend MAC while executing long pwork
The above patch ruffled some feathers on netdev. In the interest of
moving things along, I have pulled that patch out of wireless-2.6.
I expect it will be back soon, probably with some additional changes
to satisfy concerns raised on the mailing list.
NOTE: While I was mucking around, I pulled a bunch of patches from
the master branch out into driver-specific branches for adm8211,
prism54usb, tiacx, and zd1211rw. Then I rebuilt the master branch
by pulling from the driver branches. This is intended to ease the
merging of individual drivers upstream (e.g. zd1211rw and maybe tiacx
in the near future).
Those working off my upstream branch or off Linus' tree should be
unaffected. Anyone who works off my master branch may need to rebase,
especially if they want me to be able to pull from them (due to dirty
history). I apologize for the hassle and appreciate your cooperation!
Thanks,
John
---
The following changes since commit fcc18e83e1f6fd9fa6b333735bf0fcd530655511:
Malcolm Parsons:
uclinux: use PER_LINUX_32BIT in binfmt_flat
are found in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6.git upstream
Daniel Drake:
bcm43xx: use softmac-suggested TX rate
bcm43xx: enable shared key authentication
Eric Sesterhenn:
skb used after passing to netif_rx in net/ieee80211/ieee80211_rx.c
Faidon Liambotis:
Add two PLX device IDs
Hong Liu:
ieee80211: fix not allocating IV+ICV space when usingencryption in ieee80211_tx_frame
Horms:
CONFIG_WIRELESS_EXT is neccessary after all
John W. Linville:
softmac: fix build-break from 881ee6999d66c8fc903b429b73bbe6045b38c549
Joseph Jezak:
SoftMAC: Prevent multiple authentication attempts on the same network
SoftMAC: Add network to ieee80211softmac_call_events when associate times out
Larry Finger:
Convert bcm43xx-softmac to use the ieee80211_is_valid_channel routine
2.6.17 missing a call to ieee80211softmac_capabilities from ieee80211softmac_assoc_req
Michael Buesch:
bcm43xx: workaround init_board vs. IRQ race
drivers/net/wireless/bcm43xx/bcm43xx_main.c | 31 ++++++++++++---------
drivers/net/wireless/bcm43xx/bcm43xx_main.h | 24 ----------------
drivers/net/wireless/bcm43xx/bcm43xx_radio.c | 7 +----
drivers/net/wireless/bcm43xx/bcm43xx_wx.c | 2 +
drivers/net/wireless/bcm43xx/bcm43xx_xmit.c | 5 +++
drivers/net/wireless/hostap/hostap_plx.c | 2 +
include/net/ieee80211softmac.h | 1 +
net/ieee80211/ieee80211_rx.c | 4 ++-
net/ieee80211/ieee80211_tx.c | 15 +++++++---
net/ieee80211/softmac/ieee80211softmac_assoc.c | 31 ++++++++++++++++-----
net/ieee80211/softmac/ieee80211softmac_auth.c | 4 +--
net/ieee80211/softmac/ieee80211softmac_io.c | 3 ++
net/ieee80211/softmac/ieee80211softmac_wx.c | 36 +++++++++++++++++++++++-
13 files changed, 105 insertions(+), 60 deletions(-)
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c
index 085d785..1cd47c5 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c
@@ -1885,6 +1885,15 @@ static irqreturn_t bcm43xx_interrupt_han
spin_lock(&bcm->irq_lock);
+ /* Only accept IRQs, if we are initialized properly.
+ * This avoids an RX race while initializing.
+ * We should probably not enable IRQs before we are initialized
+ * completely, but some careful work is needed to fix this. I think it
+ * is best to stay with this cheap workaround for now... .
+ */
+ if (unlikely(bcm43xx_status(bcm) != BCM43xx_STAT_INITIALIZED))
+ goto out;
+
reason = bcm43xx_read32(bcm, BCM43xx_MMIO_GEN_IRQ_REASON);
if (reason == 0xffffffff) {
/* irq not for us (shared irq) */
@@ -1906,19 +1915,11 @@ static irqreturn_t bcm43xx_interrupt_han
bcm43xx_interrupt_ack(bcm, reason);
- /* Only accept IRQs, if we are initialized properly.
- * This avoids an RX race while initializing.
- * We should probably not enable IRQs before we are initialized
- * completely, but some careful work is needed to fix this. I think it
- * is best to stay with this cheap workaround for now... .
- */
- if (likely(bcm43xx_status(bcm) == BCM43xx_STAT_INITIALIZED)) {
- /* disable all IRQs. They are enabled again in the bottom half. */
- bcm->irq_savedstate = bcm43xx_interrupt_disable(bcm, BCM43xx_IRQ_ALL);
- /* save the reason code and call our bottom half. */
- bcm->irq_reason = reason;
- tasklet_schedule(&bcm->isr_tasklet);
- }
+ /* disable all IRQs. They are enabled again in the bottom half. */
+ bcm->irq_savedstate = bcm43xx_interrupt_disable(bcm, BCM43xx_IRQ_ALL);
+ /* save the reason code and call our bottom half. */
+ bcm->irq_reason = reason;
+ tasklet_schedule(&bcm->isr_tasklet);
out:
mmiowb();
@@ -3661,6 +3662,10 @@ static void bcm43xx_ieee80211_set_securi
secinfo->encrypt = sec->encrypt;
dprintk(", .encrypt = %d", sec->encrypt);
}
+ if (sec->flags & SEC_AUTH_MODE) {
+ secinfo->auth_mode = sec->auth_mode;
+ dprintk(", .auth_mode = %d\n", sec->auth_mode);
+ }
dprintk("\n");
if (bcm43xx_status(bcm) == BCM43xx_STAT_INITIALIZED &&
!bcm->ieee->host_encrypt) {
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.h b/drivers/net/wireless/bcm43xx/bcm43xx_main.h
index 30a202b..1164936 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_main.h
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.h
@@ -112,30 +112,6 @@ int bcm43xx_channel_to_freq(struct bcm43
return bcm43xx_channel_to_freq_bg(channel);
}
-/* Lightweight function to check if a channel number is valid.
- * Note that this does _NOT_ check for geographical restrictions!
- */
-static inline
-int bcm43xx_is_valid_channel_a(u8 channel)
-{
- return (channel >= IEEE80211_52GHZ_MIN_CHANNEL
- && channel <= IEEE80211_52GHZ_MAX_CHANNEL);
-}
-static inline
-int bcm43xx_is_valid_channel_bg(u8 channel)
-{
- return (channel >= IEEE80211_24GHZ_MIN_CHANNEL
- && channel <= IEEE80211_24GHZ_MAX_CHANNEL);
-}
-static inline
-int bcm43xx_is_valid_channel(struct bcm43xx_private *bcm,
- u8 channel)
-{
- if (bcm43xx_current_phy(bcm)->type == BCM43xx_PHYTYPE_A)
- return bcm43xx_is_valid_channel_a(channel);
- return bcm43xx_is_valid_channel_bg(channel);
-}
-
void bcm43xx_tsf_read(struct bcm43xx_private *bcm, u64 *tsf);
void bcm43xx_tsf_write(struct bcm43xx_private *bcm, u64 tsf);
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_radio.c b/drivers/net/wireless/bcm43xx/bcm43xx_radio.c
index af5c0bf..bb9c484 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_radio.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_radio.c
@@ -1594,11 +1594,11 @@ int bcm43xx_radio_selectchannel(struct b
u16 r8, tmp;
u16 freq;
+ if (!ieee80211_is_valid_channel(bcm->ieee, channel))
+ return -EINVAL;
if ((radio->manufact == 0x17F) &&
(radio->version == 0x2060) &&
(radio->revision == 1)) {
- if (channel > 200)
- return -EINVAL;
freq = channel2freq_a(channel);
r8 = bcm43xx_radio_read16(bcm, 0x0008);
@@ -1651,9 +1651,6 @@ int bcm43xx_radio_selectchannel(struct b
TODO(); //TODO: TSSI2dbm workaround
bcm43xx_phy_xmitpower(bcm);//FIXME correct?
} else {
- if ((channel < 1) || (channel > 14))
- return -EINVAL;
-
if (synthetic_pu_workaround)
bcm43xx_synth_pu_workaround(bcm, channel);
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c
index c35cb3a..5c36e29 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c
@@ -119,7 +119,7 @@ static int bcm43xx_wx_set_channelfreq(st
channel = bcm43xx_freq_to_channel(bcm, data->freq.m);
freq = data->freq.m;
}
- if (!bcm43xx_is_valid_channel(bcm, channel))
+ if (!ieee80211_is_valid_channel(bcm->ieee, channel))
goto out_unlock;
if (bcm43xx_status(bcm) == BCM43xx_STAT_INITIALIZED) {
//ieee80211softmac_disassoc(softmac, $REASON);
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c b/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c
index d8ece28..6dbd855 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c
@@ -296,11 +296,14 @@ void bcm43xx_generate_txhdr(struct bcm43
u16 control = 0;
u16 wsec_rate = 0;
u16 encrypt_frame;
+ const u16 ftype = WLAN_FC_GET_TYPE(le16_to_cpu(wireless_header->frame_ctl));
+ const int is_mgt = (ftype == IEEE80211_FTYPE_MGMT);
/* Now construct the TX header. */
memset(txhdr, 0, sizeof(*txhdr));
- bitrate = bcm->softmac->txrates.default_rate;
+ bitrate = ieee80211softmac_suggest_txrate(bcm->softmac,
+ is_multicast_ether_addr(wireless_header->addr1), is_mgt);
ofdm_modulation = !(ieee80211_is_cck_rate(bitrate));
fallback_bitrate = bcm43xx_calc_fallback_rate(bitrate);
fallback_ofdm_modulation = !(ieee80211_is_cck_rate(fallback_bitrate));
diff --git a/drivers/net/wireless/hostap/hostap_plx.c b/drivers/net/wireless/hostap/hostap_plx.c
index edaaa94..39c545e 100644
--- a/drivers/net/wireless/hostap/hostap_plx.c
+++ b/drivers/net/wireless/hostap/hostap_plx.c
@@ -67,10 +67,12 @@ static struct pci_device_id prism2_plx_i
PLXDEV(0x10b7, 0x7770, "3Com AirConnect PCI 777A"),
PLXDEV(0x111a, 0x1023, "Siemens SpeedStream SS1023"),
PLXDEV(0x126c, 0x8030, "Nortel emobility"),
+ PLXDEV(0x1562, 0x0001, "Symbol LA-4123"),
PLXDEV(0x1385, 0x4100, "Netgear MA301"),
PLXDEV(0x15e8, 0x0130, "National Datacomm NCP130 (PLX9052)"),
PLXDEV(0x15e8, 0x0131, "National Datacomm NCP130 (TMD7160)"),
PLXDEV(0x1638, 0x1100, "Eumitcom WL11000"),
+ PLXDEV(0x16ab, 0x1100, "Global Sun Tech GL24110P"),
PLXDEV(0x16ab, 0x1101, "Global Sun Tech GL24110P (?)"),
PLXDEV(0x16ab, 0x1102, "Linksys WPC11 with WDT11"),
PLXDEV(0x16ab, 0x1103, "Longshine 8031"),
diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h
index 7a483ab..00ad810 100644
--- a/include/net/ieee80211softmac.h
+++ b/include/net/ieee80211softmac.h
@@ -104,6 +104,7 @@ struct ieee80211softmac_assoc_info {
*/
u8 static_essid:1,
associating:1,
+ assoc_wait:1,
bssvalid:1,
bssfixed:1;
diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c
index 2bf567f..b2fabbb 100644
--- a/net/ieee80211/ieee80211_rx.c
+++ b/net/ieee80211/ieee80211_rx.c
@@ -369,6 +369,7 @@ #endif
/* Put this code here so that we avoid duplicating it in all
* Rx paths. - Jean II */
+#ifdef CONFIG_WIRELESS_EXT
#ifdef IW_WIRELESS_SPY /* defined in iw_handler.h */
/* If spy monitoring on */
if (ieee->spy_data.spy_number > 0) {
@@ -397,15 +398,16 @@ #ifdef IW_WIRELESS_SPY /* defined in iw
wireless_spy_update(ieee->dev, hdr->addr2, &wstats);
}
#endif /* IW_WIRELESS_SPY */
+#endif /* CONFIG_WIRELESS_EXT */
#ifdef NOT_YET
hostap_update_rx_stats(local->ap, hdr, rx_stats);
#endif
if (ieee->iw_mode == IW_MODE_MONITOR) {
- ieee80211_monitor_rx(ieee, skb, rx_stats);
stats->rx_packets++;
stats->rx_bytes += skb->len;
+ ieee80211_monitor_rx(ieee, skb, rx_stats);
return 1;
}
diff --git a/net/ieee80211/ieee80211_tx.c b/net/ieee80211/ieee80211_tx.c
index 6a5de1b..e63bf77 100644
--- a/net/ieee80211/ieee80211_tx.c
+++ b/net/ieee80211/ieee80211_tx.c
@@ -563,10 +563,13 @@ int ieee80211_tx_frame(struct ieee80211_
struct net_device_stats *stats = &ieee->stats;
struct sk_buff *skb_frag;
int priority = -1;
+ int fraglen = total_len;
+ int headroom = ieee->tx_headroom;
+ struct ieee80211_crypt_data *crypt = ieee->crypt[ieee->tx_keyidx];
spin_lock_irqsave(&ieee->lock, flags);
- if (encrypt_mpdu && !ieee->sec.encrypt)
+ if (encrypt_mpdu && (!ieee->sec.encrypt || !crypt))
encrypt_mpdu = 0;
/* If there is no driver handler to take the TXB, dont' bother
@@ -582,20 +585,24 @@ int ieee80211_tx_frame(struct ieee80211_
goto success;
}
- if (encrypt_mpdu)
+ if (encrypt_mpdu) {
frame->frame_ctl |= cpu_to_le16(IEEE80211_FCTL_PROTECTED);
+ fraglen += crypt->ops->extra_mpdu_prefix_len +
+ crypt->ops->extra_mpdu_postfix_len;
+ headroom += crypt->ops->extra_mpdu_prefix_len;
+ }
/* When we allocate the TXB we allocate enough space for the reserve
* and full fragment bytes (bytes_per_frag doesn't include prefix,
* postfix, header, FCS, etc.) */
- txb = ieee80211_alloc_txb(1, total_len, ieee->tx_headroom, GFP_ATOMIC);
+ txb = ieee80211_alloc_txb(1, fraglen, headroom, GFP_ATOMIC);
if (unlikely(!txb)) {
printk(KERN_WARNING "%s: Could not allocate TXB\n",
ieee->dev->name);
goto failed;
}
txb->encrypted = 0;
- txb->payload_size = total_len;
+ txb->payload_size = fraglen;
skb_frag = txb->fragments[0];
diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c
index 5e9a906..44215ce 100644
--- a/net/ieee80211/softmac/ieee80211softmac_assoc.c
+++ b/net/ieee80211/softmac/ieee80211softmac_assoc.c
@@ -47,9 +47,7 @@ ieee80211softmac_assoc(struct ieee80211s
dprintk(KERN_INFO PFX "sent association request!\n");
- /* Change the state to associating */
spin_lock_irqsave(&mac->lock, flags);
- mac->associnfo.associating = 1;
mac->associated = 0; /* just to make sure */
/* Set a timer for timeout */
@@ -63,6 +61,7 @@ void
ieee80211softmac_assoc_timeout(void *d)
{
struct ieee80211softmac_device *mac = (struct ieee80211softmac_device *)d;
+ struct ieee80211softmac_network *n;
unsigned long flags;
spin_lock_irqsave(&mac->lock, flags);
@@ -75,11 +74,12 @@ ieee80211softmac_assoc_timeout(void *d)
mac->associnfo.associating = 0;
mac->associnfo.bssvalid = 0;
mac->associated = 0;
+
+ n = ieee80211softmac_get_network_by_bssid_locked(mac, mac->associnfo.bssid);
spin_unlock_irqrestore(&mac->lock, flags);
dprintk(KERN_INFO PFX "assoc request timed out!\n");
- /* FIXME: we need to know the network here. that requires a bit of restructuring */
- ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT, NULL);
+ ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT, n);
}
void
@@ -203,6 +203,10 @@ ieee80211softmac_assoc_work(void *d)
if (mac->associated)
ieee80211softmac_send_disassoc_req(mac, WLAN_REASON_DISASSOC_STA_HAS_LEFT);
+ spin_lock_irqsave(&mac->lock, flags);
+ mac->associnfo.associating = 1;
+ spin_unlock_irqrestore(&mac->lock, flags);
+
/* try to find the requested network in our list, if we found one already */
if (bssvalid || mac->associnfo.bssfixed)
found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid);
@@ -295,19 +299,32 @@ ieee80211softmac_assoc_work(void *d)
memcpy(mac->associnfo.associate_essid.data, found->essid.data, IW_ESSID_MAX_SIZE + 1);
/* we found a network! authenticate (if necessary) and associate to it. */
- if (!found->authenticated) {
+ if (found->authenticating) {
+ dprintk(KERN_INFO PFX "Already requested authentication, waiting...\n");
+ if(!mac->associnfo.assoc_wait) {
+ mac->associnfo.assoc_wait = 1;
+ ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL);
+ }
+ return;
+ }
+ if (!found->authenticated && !found->authenticating) {
/* This relies on the fact that _auth_req only queues the work,
* otherwise adding the notification would be racy. */
if (!ieee80211softmac_auth_req(mac, found)) {
- dprintk(KERN_INFO PFX "cannot associate without being authenticated, requested authentication\n");
- ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL);
+ if(!mac->associnfo.assoc_wait) {
+ dprintk(KERN_INFO PFX "Cannot associate without being authenticated, requested authentication\n");
+ mac->associnfo.assoc_wait = 1;
+ ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL);
+ }
} else {
printkl(KERN_WARNING PFX "Not authenticated, but requesting authentication failed. Giving up to associate\n");
+ mac->associnfo.assoc_wait = 0;
ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED, found);
}
return;
}
/* finally! now we can start associating */
+ mac->associnfo.assoc_wait = 0;
ieee80211softmac_assoc(mac, found);
}
diff --git a/net/ieee80211/softmac/ieee80211softmac_auth.c b/net/ieee80211/softmac/ieee80211softmac_auth.c
index 90b8484..ebc33ca 100644
--- a/net/ieee80211/softmac/ieee80211softmac_auth.c
+++ b/net/ieee80211/softmac/ieee80211softmac_auth.c
@@ -36,8 +36,9 @@ ieee80211softmac_auth_req(struct ieee802
struct ieee80211softmac_auth_queue_item *auth;
unsigned long flags;
- if (net->authenticating)
+ if (net->authenticating || net->authenticated)
return 0;
+ net->authenticating = 1;
/* Add the network if it's not already added */
ieee80211softmac_add_network(mac, net);
@@ -92,7 +93,6 @@ ieee80211softmac_auth_queue(void *data)
return;
}
net->authenticated = 0;
- net->authenticating = 1;
/* add a timeout call so we eventually give up waiting for an auth reply */
schedule_delayed_work(&auth->work, IEEE80211SOFTMAC_AUTH_TIMEOUT);
auth->retry--;
diff --git a/net/ieee80211/softmac/ieee80211softmac_io.c b/net/ieee80211/softmac/ieee80211softmac_io.c
index 0954161..8cc8b20 100644
--- a/net/ieee80211/softmac/ieee80211softmac_io.c
+++ b/net/ieee80211/softmac/ieee80211softmac_io.c
@@ -229,6 +229,9 @@ ieee80211softmac_assoc_req(struct ieee80
return 0;
ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_ASSOC_REQ, net->bssid, net->bssid);
+ /* Fill in the capabilities */
+ (*pkt)->capability = ieee80211softmac_capabilities(mac, net);
+
/* Fill in Listen Interval (?) */
(*pkt)->listen_interval = cpu_to_le16(10);
diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c
index 0e65ff4..75320b6 100644
--- a/net/ieee80211/softmac/ieee80211softmac_wx.c
+++ b/net/ieee80211/softmac/ieee80211softmac_wx.c
@@ -70,12 +70,44 @@ ieee80211softmac_wx_set_essid(struct net
char *extra)
{
struct ieee80211softmac_device *sm = ieee80211_priv(net_dev);
+ struct ieee80211softmac_network *n;
+ struct ieee80211softmac_auth_queue_item *authptr;
int length = 0;
unsigned long flags;
-
+
+ /* Check if we're already associating to this or another network
+ * If it's another network, cancel and start over with our new network
+ * If it's our network, ignore the change, we're already doing it!
+ */
+ if((sm->associnfo.associating || sm->associated) &&
+ (data->essid.flags && data->essid.length && extra)) {
+ /* Get the associating network */
+ n = ieee80211softmac_get_network_by_bssid(sm, sm->associnfo.bssid);
+ if(n && n->essid.len == (data->essid.length - 1) &&
+ !memcmp(n->essid.data, extra, n->essid.len)) {
+ dprintk(KERN_INFO PFX "Already associating or associated to "MAC_FMT"\n",
+ MAC_ARG(sm->associnfo.bssid));
+ return 0;
+ } else {
+ dprintk(KERN_INFO PFX "Canceling existing associate request!\n");
+ spin_lock_irqsave(&sm->lock,flags);
+ /* Cancel assoc work */
+ cancel_delayed_work(&sm->associnfo.work);
+ /* We don't have to do this, but it's a little cleaner */
+ list_for_each_entry(authptr, &sm->auth_queue, list)
+ cancel_delayed_work(&authptr->work);
+ sm->associnfo.bssvalid = 0;
+ sm->associnfo.bssfixed = 0;
+ spin_unlock_irqrestore(&sm->lock,flags);
+ flush_scheduled_work();
+ }
+ }
+
+
spin_lock_irqsave(&sm->lock, flags);
-
+
sm->associnfo.static_essid = 0;
+ sm->associnfo.assoc_wait = 0;
if (data->essid.flags && data->essid.length && extra /*required?*/) {
length = min(data->essid.length - 1, IW_ESSID_MAX_SIZE);
--
John W. Linville
linville@tuxdriver.com
^ permalink raw reply related
* [Patch 1/1] AF_UNIX Datagram getpeersec (with latest updates)
From: Catherine Zhang @ 2006-06-28 1:34 UTC (permalink / raw)
To: netdev, davem, jmorris, chrisw, herbert, sds, tjaeger, akpm
Cc: latten, sergeh, gcwilson, czhang.us
Hi,
This patch combines all previous updates. Many thanks to James, Dave,
and Stephen for their modifications and comments!
cheers,
Catherine
--
From: cxzhang@watson.ibm.com
This patch implements an API whereby an application can determine the
label of its peer's Unix datagram sockets via the auxiliary data mechanism of
recvmsg.
Patch purpose:
This patch enables a security-aware application to retrieve the
security context of the peer of a Unix datagram socket. The application
can then use this security context to determine the security context for
processing on behalf of the peer who sent the packet.
Patch design and implementation:
The design and implementation is very similar to the UDP case for INET
sockets. Basically we build upon the existing Unix domain socket API for
retrieving user credentials. Linux offers the API for obtaining user
credentials via ancillary messages (i.e., out of band/control messages
that are bundled together with a normal message). To retrieve the security
context, the application first indicates to the kernel such desire by
setting the SO_PASSSEC option via getsockopt. Then the application
retrieves the security context using the auxiliary data mechanism.
An example server application for Unix datagram socket should look like this:
toggle = 1;
toggle_len = sizeof(toggle);
setsockopt(sockfd, SOL_SOCKET, SO_PASSSEC, &toggle, &toggle_len);
recvmsg(sockfd, &msg_hdr, 0);
if (msg_hdr.msg_controllen > sizeof(struct cmsghdr)) {
cmsg_hdr = CMSG_FIRSTHDR(&msg_hdr);
if (cmsg_hdr->cmsg_len <= CMSG_LEN(sizeof(scontext)) &&
cmsg_hdr->cmsg_level == SOL_SOCKET &&
cmsg_hdr->cmsg_type == SCM_SECURITY) {
memcpy(&scontext, CMSG_DATA(cmsg_hdr), sizeof(scontext));
}
}
sock_setsockopt is enhanced with a new socket option SOCK_PASSSEC to allow
a server socket to receive security context of the peer.
Testing:
We have tested the patch by setting up Unix datagram client and server
applications. We verified that the server can retrieve the security context
using the auxiliary data mechanism of recvmsg.
---
include/asm-alpha/socket.h | 1 +
include/asm-arm/socket.h | 1 +
include/asm-arm26/socket.h | 1 +
include/asm-cris/socket.h | 1 +
include/asm-frv/socket.h | 1 +
include/asm-h8300/socket.h | 1 +
include/asm-i386/socket.h | 1 +
include/asm-ia64/socket.h | 1 +
include/asm-m32r/socket.h | 1 +
include/asm-m68k/socket.h | 1 +
include/asm-mips/socket.h | 1 +
include/asm-parisc/socket.h | 1 +
include/asm-powerpc/socket.h | 1 +
include/asm-s390/socket.h | 1 +
include/asm-sh/socket.h | 1 +
include/asm-sparc/socket.h | 1 +
include/asm-sparc64/socket.h | 1 +
include/asm-v850/socket.h | 1 +
include/asm-x86_64/socket.h | 1 +
include/asm-xtensa/socket.h | 1 +
include/linux/net.h | 1 +
include/linux/selinux.h | 15 +++++++++++++++
include/net/af_unix.h | 6 ++++++
include/net/scm.h | 17 +++++++++++++++++
net/core/sock.c | 11 +++++++++++
net/unix/af_unix.c | 27 +++++++++++++++++++++++++++
security/selinux/exports.c | 11 +++++++++++
security/selinux/hooks.c | 8 +++++++-
28 files changed, 115 insertions(+), 1 deletion(-)
diff -puN include/asm-alpha/socket.h~lsm-secpeer-unix include/asm-alpha/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-alpha/socket.h~lsm-secpeer-unix 2006-06-27 18:14:52.586456256 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-alpha/socket.h 2006-06-27 18:16:31.488420872 -0400
@@ -51,6 +51,7 @@
#define SCM_TIMESTAMP SO_TIMESTAMP
#define SO_PEERSEC 30
+#define SO_PASSSEC 34
/* Security levels - as per NRL IPv6 - don't actually do anything */
#define SO_SECURITY_AUTHENTICATION 19
diff -puN include/asm-arm/socket.h~lsm-secpeer-unix include/asm-arm/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-arm/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.052800968 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-arm/socket.h 2006-06-27 18:16:31.489420720 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-arm26/socket.h~lsm-secpeer-unix include/asm-arm26/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-arm26/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.095794432 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-arm26/socket.h 2006-06-27 18:16:31.489420720 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-cris/socket.h~lsm-secpeer-unix include/asm-cris/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-cris/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.132788808 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-cris/socket.h 2006-06-27 18:16:31.490420568 -0400
@@ -50,6 +50,7 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-frv/socket.h~lsm-secpeer-unix include/asm-frv/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-frv/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.169783184 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-frv/socket.h 2006-06-27 18:16:31.490420568 -0400
@@ -48,6 +48,7 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-h8300/socket.h~lsm-secpeer-unix include/asm-h8300/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-h8300/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.210776952 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-h8300/socket.h 2006-06-27 18:16:31.490420568 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-i386/socket.h~lsm-secpeer-unix include/asm-i386/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-i386/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.248771176 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-i386/socket.h 2006-06-27 18:16:31.491420416 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-ia64/socket.h~lsm-secpeer-unix include/asm-ia64/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-ia64/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.286765400 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-ia64/socket.h 2006-06-27 18:16:31.491420416 -0400
@@ -57,5 +57,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_IA64_SOCKET_H */
diff -puN include/asm-m32r/socket.h~lsm-secpeer-unix include/asm-m32r/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-m32r/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.317760688 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-m32r/socket.h 2006-06-27 18:16:31.492420264 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_M32R_SOCKET_H */
diff -puN include/asm-m68k/socket.h~lsm-secpeer-unix include/asm-m68k/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-m68k/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.357754608 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-m68k/socket.h 2006-06-27 18:16:31.492420264 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-mips/socket.h~lsm-secpeer-unix include/asm-mips/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-mips/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.397748528 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-mips/socket.h 2006-06-27 18:16:31.493420112 -0400
@@ -69,6 +69,7 @@ To add: #define SO_REUSEPORT 0x0200 /* A
#define SO_PEERSEC 30
#define SO_SNDBUFFORCE 31
#define SO_RCVBUFFORCE 33
+#define SO_PASSSEC 34
#ifdef __KERNEL__
diff -puN include/asm-parisc/socket.h~lsm-secpeer-unix include/asm-parisc/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-parisc/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.433743056 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-parisc/socket.h 2006-06-27 18:16:31.493420112 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 0x401c
#define SO_PEERSEC 0x401d
+#define SO_PASSSEC 0x401e
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-powerpc/socket.h~lsm-secpeer-unix include/asm-powerpc/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-powerpc/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.469737584 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-powerpc/socket.h 2006-06-27 18:16:31.494419960 -0400
@@ -55,5 +55,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_POWERPC_SOCKET_H */
diff -puN include/asm-s390/socket.h~lsm-secpeer-unix include/asm-s390/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-s390/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.510731352 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-s390/socket.h 2006-06-27 18:16:31.494419960 -0400
@@ -56,5 +56,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-sh/socket.h~lsm-secpeer-unix include/asm-sh/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-sh/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.543726336 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-sh/socket.h 2006-06-27 18:16:31.494419960 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* __ASM_SH_SOCKET_H */
diff -puN include/asm-sparc/socket.h~lsm-secpeer-unix include/asm-sparc/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-sparc/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.572721928 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-sparc/socket.h 2006-06-27 18:16:31.495419808 -0400
@@ -48,6 +48,7 @@
#define SCM_TIMESTAMP SO_TIMESTAMP
#define SO_PEERSEC 0x001e
+#define SO_PASSSEC 0x001f
/* Security levels - as per NRL IPv6 - don't actually do anything */
#define SO_SECURITY_AUTHENTICATION 0x5001
diff -puN include/asm-sparc64/socket.h~lsm-secpeer-unix include/asm-sparc64/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-sparc64/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.603717216 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-sparc64/socket.h 2006-06-27 18:16:31.495419808 -0400
@@ -48,6 +48,7 @@
#define SCM_TIMESTAMP SO_TIMESTAMP
#define SO_PEERSEC 0x001e
+#define SO_PASSSEC 0x001f
/* Security levels - as per NRL IPv6 - don't actually do anything */
#define SO_SECURITY_AUTHENTICATION 0x5001
diff -puN include/asm-v850/socket.h~lsm-secpeer-unix include/asm-v850/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-v850/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.637712048 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-v850/socket.h 2006-06-27 18:16:31.496419656 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* __V850_SOCKET_H__ */
diff -puN include/asm-x86_64/socket.h~lsm-secpeer-unix include/asm-x86_64/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-x86_64/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.674706424 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-x86_64/socket.h 2006-06-27 18:16:31.496419656 -0400
@@ -48,5 +48,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _ASM_SOCKET_H */
diff -puN include/asm-xtensa/socket.h~lsm-secpeer-unix include/asm-xtensa/socket.h
--- linux-2.6.17-rc6-mm2-JM/include/asm-xtensa/socket.h~lsm-secpeer-unix 2006-06-27 18:15:10.711700800 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/asm-xtensa/socket.h 2006-06-27 18:16:31.497419504 -0400
@@ -59,5 +59,6 @@
#define SO_ACCEPTCONN 30
#define SO_PEERSEC 31
+#define SO_PASSSEC 34
#endif /* _XTENSA_SOCKET_H */
diff -puN include/linux/net.h~lsm-secpeer-unix include/linux/net.h
--- linux-2.6.17-rc6-mm2-JM/include/linux/net.h~lsm-secpeer-unix 2006-06-27 18:15:10.740696392 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/linux/net.h 2006-06-27 18:16:31.497419504 -0400
@@ -61,6 +61,7 @@ typedef enum {
#define SOCK_ASYNC_WAITDATA 1
#define SOCK_NOSPACE 2
#define SOCK_PASSCRED 3
+#define SOCK_PASSSEC 4
#ifndef ARCH_HAS_SOCKET_TYPES
/**
diff -puN include/linux/selinux.h~lsm-secpeer-unix include/linux/selinux.h
--- linux-2.6.17-rc6-mm2-JM/include/linux/selinux.h~lsm-secpeer-unix 2006-06-27 18:15:10.769691984 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/linux/selinux.h 2006-06-27 18:16:31.498419352 -0400
@@ -18,6 +18,7 @@ struct selinux_audit_rule;
struct audit_context;
struct inode;
struct kern_ipc_perm;
+struct socket;
#ifdef CONFIG_SECURITY_SELINUX
@@ -119,6 +120,15 @@ void selinux_get_ipc_sid(const struct ke
void selinux_get_task_sid(struct task_struct *tsk, u32 *sid);
/**
+ * selinux_get_sock_sid - return the SID of socket
+ * @sock: the socket whose SID will be returned
+ * @sid: pointer to security context ID to be filled in.
+ *
+ * Returns nothing
+ */
+void selinux_get_sock_sid(struct socket *sock, u32 *sid);
+
+/**
* selinux_string_to_sid - map a security context string to a security ID
* @str: the security context string to be mapped
* @sid: ID value returned via this.
@@ -193,6 +203,11 @@ static inline void selinux_get_task_sid(
*sid = 0;
}
+static inline void selinux_get_sock_sid(struct socket *sock, u32 *sid)
+{
+ *sid = 0;
+}
+
static inline int selinux_string_to_sid(const char *str, u32 *sid)
{
*sid = 0;
diff -puN include/net/af_unix.h~lsm-secpeer-unix include/net/af_unix.h
--- linux-2.6.17-rc6-mm2-JM/include/net/af_unix.h~lsm-secpeer-unix 2006-06-27 18:15:10.798687576 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/net/af_unix.h 2006-06-27 18:23:28.787981696 -0400
@@ -53,10 +53,16 @@ struct unix_address {
struct unix_skb_parms {
struct ucred creds; /* Skb credentials */
struct scm_fp_list *fp; /* Passed files */
+#ifdef CONFIG_SECURITY_NETWORK
+ char *secdata; /* Security context */
+ u32 seclen; /* Security length */
+#endif
};
#define UNIXCB(skb) (*(struct unix_skb_parms*)&((skb)->cb))
#define UNIXCREDS(skb) (&UNIXCB((skb)).creds)
+#define UNIXSECDATA(skb) (&UNIXCB((skb)).secdata)
+#define UNIXSECLEN(skb) (&UNIXCB((skb)).seclen)
#define unix_state_rlock(s) spin_lock(&unix_sk(s)->lock)
#define unix_state_runlock(s) spin_unlock(&unix_sk(s)->lock)
diff -puN include/net/scm.h~lsm-secpeer-unix include/net/scm.h
--- linux-2.6.17-rc6-mm2-JM/include/net/scm.h~lsm-secpeer-unix 2006-06-27 18:15:10.827683168 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/include/net/scm.h 2006-06-27 18:21:59.094617160 -0400
@@ -19,6 +19,10 @@ struct scm_cookie
{
struct ucred creds; /* Skb credentials */
struct scm_fp_list *fp; /* Passed files */
+#ifdef CONFIG_SECURITY_NETWORK
+ char *secdata; /* Security context */
+ u32 seclen; /* Security length */
+#endif
unsigned long seq; /* Connection seqno */
};
@@ -48,6 +52,17 @@ static __inline__ int scm_send(struct so
return __scm_send(sock, msg, scm);
}
+#ifdef CONFIG_SECURITY_NETWORK
+static inline void scm_passec(struct socket *sock, struct msghdr *msg, struct scm_cookie *scm)
+{
+ if (test_bit(SOCK_PASSSEC, &sock->flags) && scm->secdata != NULL)
+ put_cmsg(msg, SOL_SOCKET, SCM_SECURITY, scm->seclen, scm->secdata);
+}
+#else
+static inline void scm_passec(struct socket *sock, struct msghdr *msg, struct scm_cookie *scm)
+{ }
+#endif /* CONFIG_SECURITY_NETWORK */
+
static __inline__ void scm_recv(struct socket *sock, struct msghdr *msg,
struct scm_cookie *scm, int flags)
{
@@ -62,6 +77,8 @@ static __inline__ void scm_recv(struct s
if (test_bit(SOCK_PASSCRED, &sock->flags))
put_cmsg(msg, SOL_SOCKET, SCM_CREDENTIALS, sizeof(scm->creds), &scm->creds);
+ scm_passec(sock, msg, scm);
+
if (!scm->fp)
return;
diff -puN net/core/sock.c~lsm-secpeer-unix net/core/sock.c
--- linux-2.6.17-rc6-mm2-JM/net/core/sock.c~lsm-secpeer-unix 2006-06-27 18:15:10.856678760 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/net/core/sock.c 2006-06-27 18:16:31.500419048 -0400
@@ -565,6 +565,13 @@ set_rcvbuf:
ret = -ENONET;
break;
+ case SO_PASSSEC:
+ if (valbool)
+ set_bit(SOCK_PASSSEC, &sock->flags);
+ else
+ clear_bit(SOCK_PASSSEC, &sock->flags);
+ break;
+
/* We implement the SO_SNDLOWAT etc to
not be settable (1003.1g 5.3) */
default:
@@ -723,6 +730,10 @@ int sock_getsockopt(struct socket *sock,
v.val = sk->sk_state == TCP_LISTEN;
break;
+ case SO_PASSSEC:
+ v.val = test_bit(SOCK_PASSSEC, &sock->flags) ? 1 : 0;
+ break;
+
case SO_PEERSEC:
return security_socket_getpeersec_stream(sock, optval, optlen, len);
diff -puN net/unix/af_unix.c~lsm-secpeer-unix net/unix/af_unix.c
--- linux-2.6.17-rc6-mm2-JM/net/unix/af_unix.c~lsm-secpeer-unix 2006-06-27 18:15:10.885674352 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/net/unix/af_unix.c 2006-06-27 18:25:42.767613696 -0400
@@ -128,6 +128,30 @@ static atomic_t unix_nr_socks = ATOMIC_I
#define UNIX_ABSTRACT(sk) (unix_sk(sk)->addr->hash != UNIX_HASH_SIZE)
+#ifdef CONFIG_SECURITY_NETWORK
+static void unix_get_peersec_dgram(struct sk_buff *skb)
+{
+ int err;
+
+ err = security_socket_getpeersec_dgram(skb, UNIXSECDATA(skb),
+ UNIXSECLEN(skb));
+ if (err)
+ *(UNIXSECDATA(skb)) = NULL;
+}
+
+static inline void unix_set_secdata(struct scm_cookie *scm, struct sk_buff *skb)
+{
+ scm->secdata = *UNIXSECDATA(skb);
+ scm->seclen = *UNIXSECLEN(skb);
+}
+#else
+static void unix_get_peersec_dgram(struct sk_buff *skb)
+{ }
+
+static inline void unix_set_secdata(struct scm_cookie *scm, struct sk_buff *skb)
+{ }
+#endif /* CONFIG_SECURITY_NETWORK */
+
/*
* SMP locking strategy:
* hash table is protected with spinlock unix_table_lock
@@ -1291,6 +1315,8 @@ static int unix_dgram_sendmsg(struct kio
if (siocb->scm->fp)
unix_attach_fds(siocb->scm, skb);
+ unix_get_peersec_dgram(skb);
+
skb->h.raw = skb->data;
err = memcpy_fromiovec(skb_put(skb,len), msg->msg_iov, len);
if (err)
@@ -1570,6 +1596,7 @@ static int unix_dgram_recvmsg(struct kio
memset(&tmp_scm, 0, sizeof(tmp_scm));
}
siocb->scm->creds = *UNIXCREDS(skb);
+ unix_set_secdata(siocb->scm, skb);
if (!(flags & MSG_PEEK))
{
diff -puN security/selinux/exports.c~lsm-secpeer-unix security/selinux/exports.c
--- linux-2.6.17-rc6-mm2-JM/security/selinux/exports.c~lsm-secpeer-unix 2006-06-27 18:15:10.914669944 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/security/selinux/exports.c 2006-06-27 18:16:31.502418744 -0400
@@ -17,6 +17,7 @@
#include <linux/selinux.h>
#include <linux/fs.h>
#include <linux/ipc.h>
+#include <net/sock.h>
#include "security.h"
#include "objsec.h"
@@ -72,6 +73,16 @@ void selinux_get_task_sid(struct task_st
*sid = 0;
}
+void selinux_get_sock_sid(struct socket *sock, u32 *sid)
+{
+ if (selinux_enabled) {
+ const struct inode *inode = SOCK_INODE(sock);
+ selinux_get_inode_sid(inode, sid);
+ return;
+ }
+ *sid = 0;
+}
+
int selinux_string_to_sid(char *str, u32 *sid)
{
if (selinux_enabled)
diff -puN security/selinux/hooks.c~lsm-secpeer-unix security/selinux/hooks.c
--- linux-2.6.17-rc6-mm2-JM/security/selinux/hooks.c~lsm-secpeer-unix 2006-06-27 18:15:10.945665232 -0400
+++ linux-2.6.17-rc6-mm2-JM-cxzhang/security/selinux/hooks.c 2006-06-27 18:16:31.505418288 -0400
@@ -69,6 +69,7 @@
#include <linux/sysctl.h>
#include <linux/audit.h>
#include <linux/string.h>
+#include <linux/selinux.h>
#include "avc.h"
#include "objsec.h"
@@ -3396,7 +3397,12 @@ out:
static int selinux_socket_getpeersec_dgram(struct sk_buff *skb, char **secdata, u32 *seclen)
{
int err = 0;
- u32 peer_sid = selinux_socket_getpeer_dgram(skb);
+ u32 peer_sid;
+
+ if (skb->sk->sk_family == PF_UNIX)
+ selinux_get_sock_sid(skb->sk->sk_socket, &peer_sid);
+ else
+ peer_sid = selinux_socket_getpeer_dgram(skb);
if (peer_sid == SECSID_NULL)
return -EINVAL;
_
^ permalink raw reply
* Re: [Patch 1/1] AF_UNIX Datagram getpeersec (with latest updates)
From: James Morris @ 2006-06-28 1:33 UTC (permalink / raw)
To: Catherine Zhang
Cc: netdev, David S. Miller, chrisw, Herbert Xu, Stephen Smalley,
tjaeger, Andrew Morton, latten, sergeh, gcwilson, czhang.us
In-Reply-To: <20060628013428.GA31240@jiayuguan.watson.ibm.com>
On Tue, 27 Jun 2006, Catherine Zhang wrote:
> diff -puN security/selinux/exports.c~lsm-secpeer-unix security/selinux/exports.c
> --- linux-2.6.17-rc6-mm2-JM/security/selinux/exports.c~lsm-secpeer-unix 2006-06-27 18:15:10.914669944 -0400
> +++ linux-2.6.17-rc6-mm2-JM-cxzhang/security/selinux/exports.c 2006-06-27 18:16:31.502418744 -0400
> @@ -17,6 +17,7 @@
> #include <linux/selinux.h>
> #include <linux/fs.h>
> #include <linux/ipc.h>
> +#include <net/sock.h>
>
> #include "security.h"
> #include "objsec.h"
> @@ -72,6 +73,16 @@ void selinux_get_task_sid(struct task_st
> *sid = 0;
> }
>
> +void selinux_get_sock_sid(struct socket *sock, u32 *sid)
> +{
> + if (selinux_enabled) {
> + const struct inode *inode = SOCK_INODE(sock);
> + selinux_get_inode_sid(inode, sid);
> + return;
> + }
> + *sid = 0;
> +}
> +
Just one more thing, we don't need to export this function now.
- James
--
James Morris
<jmorris@namei.org>
^ permalink raw reply
* Re: [Patch 1/1] AF_UNIX Datagram getpeersec (with latest updates)
From: Xiaolan Zhang @ 2006-06-28 1:52 UTC (permalink / raw)
To: James Morris
Cc: Andrew Morton, chrisw, Catherine Zhang, czhang.us,
David S. Miller, George Wilson, Herbert Xu, latten, netdev,
Stephen Smalley, Serge E Hallyn, tjaeger
In-Reply-To: <Pine.LNX.4.64.0606272130080.6477@d.namei>
James Morris <jmorris@namei.org> wrote on 06/27/2006 09:33:17 PM:
> On Tue, 27 Jun 2006, Catherine Zhang wrote:
>
> > diff -puN security/selinux/exports.c~lsm-secpeer-unix
> security/selinux/exports.c
> > --- linux-2.6.17-rc6-mm2-JM/security/selinux/exports.c~lsm-
> secpeer-unix 2006-06-27 18:15:10.914669944 -0400
> > +++ linux-2.6.17-rc6-mm2-JM-cxzhang/security/selinux/exports.c
> 2006-06-27 18:16:31.502418744 -0400
> > @@ -17,6 +17,7 @@
> > #include <linux/selinux.h>
> > #include <linux/fs.h>
> > #include <linux/ipc.h>
> > +#include <net/sock.h>
> >
> > #include "security.h"
> > #include "objsec.h"
> > @@ -72,6 +73,16 @@ void selinux_get_task_sid(struct task_st
> > *sid = 0;
> > }
> >
> > +void selinux_get_sock_sid(struct socket *sock, u32 *sid)
> > +{
> > + if (selinux_enabled) {
> > + const struct inode *inode = SOCK_INODE(sock);
> > + selinux_get_inode_sid(inode, sid);
> > + return;
> > + }
> > + *sid = 0;
> > +}
> > +
>
>
> Just one more thing, we don't need to export this function now.
You mean moving it to security/selinux/hooks.c and making it static?
I think conceptually this is where it should reside -- auditing system
might need it in the future, for example.
thanks,
Catherine
^ permalink raw reply
* Re: [RFC] cirrus ep93xx ethernet driver
From: Lennert Buytenhek @ 2006-06-28 2:05 UTC (permalink / raw)
To: netdev
In-Reply-To: <20060626025924.GF6826@xi.wantstofly.org>
On Mon, Jun 26, 2006 at 04:59:24AM +0200, Lennert Buytenhek wrote:
> The cirrus ep93xx is an ARM SoC that includes an ethernet MAC --
> this patch adds a driver for that ethernet MAC.
Attached is a new version that optimises interrupt handling somewhat.
Since we clear RX status as the first thing we do in the poll handler,
we might as well read the read-to-clear version of the interrupt status
register in the interrupt handler and avoid the explicit clear in the
poll handler. This shaves close to a second off a 128M sendfile() test.
At ~40 seconds for a 128M sendfile (~3.2MB/sec), the network performance
of this CPU isn't impressive by any means, but given that the CPU only
runs at 200MHz and that the MAC doesn't do checksum offloading and insists
on 4-byte buffer alignment, we can't really do a whole lot better. The
performance with this driver is still a good deal better than with the
vendor driver, though -- for this particular test (128M sendfile), the
vendor driver needs 1m21s.
Apart from that it still uses numeric chip register addresses, I'm
quite happy with the driver as it is, it survives heavy beating and
is pretty stable.
Index: linux-2.6.17-git10/drivers/net/arm/Kconfig
===================================================================
--- linux-2.6.17-git10.orig/drivers/net/arm/Kconfig
+++ linux-2.6.17-git10/drivers/net/arm/Kconfig
@@ -39,3 +39,10 @@ config ARM_AT91_ETHER
help
If you wish to compile a kernel for the AT91RM9200 and enable
ethernet support, then you should always answer Y to this.
+
+config EP93XX_ETH
+ tristate "EP93xx Ethernet support"
+ depends on NET_ETHERNET && ARM && ARCH_EP93XX
+ help
+ This is a driver for the ethernet hardware included in EP93xx CPUs.
+ Say Y if you are building a kernel for EP93xx based devices.
Index: linux-2.6.17-git10/drivers/net/arm/Makefile
===================================================================
--- linux-2.6.17-git10.orig/drivers/net/arm/Makefile
+++ linux-2.6.17-git10/drivers/net/arm/Makefile
@@ -8,3 +8,4 @@ obj-$(CONFIG_ARM_ETHERH) += etherh.o
obj-$(CONFIG_ARM_ETHER3) += ether3.o
obj-$(CONFIG_ARM_ETHER1) += ether1.o
obj-$(CONFIG_ARM_AT91_ETHER) += at91_ether.o
+obj-$(CONFIG_EP93XX_ETH) += ep93xx_eth.o
Index: linux-2.6.17-git10/drivers/net/arm/ep93xx_eth.c
===================================================================
--- /dev/null
+++ linux-2.6.17-git10/drivers/net/arm/ep93xx_eth.c
@@ -0,0 +1,668 @@
+/*
+ * EP93xx ethernet network device driver
+ * Copyright (C) 2006 Lennert Buytenhek <buytenh@wantstofly.org>
+ * Dedicated to Marija Kulikova.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/config.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/init.h>
+#include <linux/moduleparam.h>
+#include <linux/platform_device.h>
+#include <asm/arch/ep93xx-regs.h>
+#include <asm/arch/platform.h>
+#include <asm/io.h>
+#include "ep93xx_eth.h"
+
+#define DRV_MODULE_VERSION "0.1"
+
+#define RX_QUEUE_ENTRIES 64
+#define TX_QUEUE_ENTRIES 8
+
+struct ep93xx_descs
+{
+ struct ep93xx_rdesc rdesc[RX_QUEUE_ENTRIES];
+ struct ep93xx_tdesc tdesc[TX_QUEUE_ENTRIES];
+ struct ep93xx_rstat rstat[RX_QUEUE_ENTRIES];
+ struct ep93xx_tstat tstat[TX_QUEUE_ENTRIES];
+};
+
+struct ep93xx_priv
+{
+ struct resource *res;
+ void *base_addr;
+ int irq;
+
+ struct ep93xx_descs *descs;
+ dma_addr_t descs_dma_addr;
+
+ void *rx_buf[RX_QUEUE_ENTRIES];
+ void *tx_buf[TX_QUEUE_ENTRIES];
+
+ int rx_pointer;
+ int tx_clean_pointer;
+ int tx_pointer;
+ int tx_pending;
+
+ struct net_device_stats stats;
+};
+
+#define rdb(ep, off) __raw_readb((ep)->base_addr + (off))
+#define rdw(ep, off) __raw_readw((ep)->base_addr + (off))
+#define rdl(ep, off) __raw_readl((ep)->base_addr + (off))
+#define wrb(ep, off, val) __raw_writeb((val), (ep)->base_addr + (off))
+#define wrw(ep, off, val) __raw_writew((val), (ep)->base_addr + (off))
+#define wrl(ep, off, val) __raw_writel((val), (ep)->base_addr + (off))
+
+static int ep93xx_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+ int entry;
+
+ if (unlikely(skb->len) > PAGE_SIZE) {
+ ep->stats.tx_dropped++;
+ dev_kfree_skb(skb);
+ return 0;
+ }
+
+ entry = ep->tx_pointer;
+ ep->tx_pointer = (ep->tx_pointer + 1) % TX_QUEUE_ENTRIES;
+
+ ep->descs->tdesc[entry].tdesc1 =
+ TDESC1_EOF | (entry << 16) | (skb->len & 0xfff);
+ skb_copy_and_csum_dev(skb, ep->tx_buf[entry]);
+ dma_sync_single(NULL, ep->descs->tdesc[entry].buf_addr,
+ skb->len, DMA_TO_DEVICE);
+ dev_kfree_skb(skb);
+
+ dev->trans_start = jiffies;
+
+ local_irq_disable();
+ ep->tx_pending++;
+ if (ep->tx_pending == TX_QUEUE_ENTRIES)
+ netif_stop_queue(dev);
+ local_irq_enable();
+
+ wrl(ep, 0x00bc, 1);
+
+ return 0;
+}
+
+static int ep93xx_rx(struct net_device *dev, int *budget)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+ int tail_offset;
+ int rx_done;
+ int processed;
+
+ tail_offset = rdl(ep, 0x00a8) - ep->descs_dma_addr;
+
+ rx_done = 0;
+ processed = 0;
+ while (*budget > 0) {
+ int entry;
+ struct ep93xx_rstat *rstat;
+ u32 rstat0;
+ u32 rstat1;
+ int length;
+ struct sk_buff *skb;
+
+ entry = ep->rx_pointer;
+ rstat = ep->descs->rstat + entry;
+ if ((void *)rstat - (void *)ep->descs == tail_offset) {
+ rx_done = 1;
+ break;
+ }
+
+ rstat0 = rstat->rstat0;
+ rstat1 = rstat->rstat1;
+ rstat->rstat0 = 0;
+ rstat->rstat1 = 0;
+
+ if (!(rstat0 & RSTAT0_RFP)) {
+ printk(KERN_CRIT "ep93xx_rx: buffer not done "
+ " %.8x %.8x\n", rstat0, rstat1);
+ BUG();
+ }
+ if (!(rstat0 & RSTAT0_EOF)) {
+ printk(KERN_CRIT "ep93xx_rx: not end-of-frame "
+ " %.8x %.8x\n", rstat0, rstat1);
+ BUG();
+ }
+ if (!(rstat0 & RSTAT0_EOB)) {
+ printk(KERN_CRIT "ep93xx_rx: not end-of-buffer "
+ " %.8x %.8x\n", rstat0, rstat1);
+ BUG();
+ }
+ if (!(rstat1 & RSTAT1_RFP)) {
+ printk(KERN_CRIT "ep93xx_rx: buffer1 not done "
+ " %.8x %.8x\n", rstat0, rstat1);
+ BUG();
+ }
+ if ((rstat1 & RSTAT1_BUFFER_INDEX) >> 16 != entry) {
+ printk(KERN_INFO "ep93xx_rx: entry mismatch "
+ " %.8x %.8x\n", rstat0, rstat1);
+ BUG();
+ }
+
+ if (!(rstat0 & RSTAT0_RWE)) {
+ printk(KERN_NOTICE "ep93xx_rx: receive error "
+ " %.8x %.8x\n", rstat0, rstat1);
+
+ ep->stats.rx_errors++;
+ if (rstat0 & RSTAT0_OE)
+ ep->stats.rx_fifo_errors++;
+ if (rstat0 & RSTAT0_FE)
+ ep->stats.rx_frame_errors++;
+ if (rstat0 & (RSTAT0_RUNT | RSTAT0_EDATA))
+ ep->stats.rx_length_errors++;
+ if (rstat0 & RSTAT0_CRCE)
+ ep->stats.rx_crc_errors++;
+ goto err;
+ }
+
+ length = rstat1 & RSTAT1_FRAME_LENGTH;
+ if (length < 4 || length > PAGE_SIZE) {
+ printk(KERN_NOTICE "ep93xx_rx: invalid length "
+ " %.8x %.8x\n", rstat0, rstat1);
+ goto err;
+ }
+
+ /* Strip FCS. */
+ if (rstat0 & RSTAT0_CRCI)
+ length -= 4;
+
+ skb = dev_alloc_skb(length + 2);
+ if (likely(skb != NULL)) {
+ skb->dev = dev;
+ skb_reserve(skb, 2);
+ dma_sync_single(NULL, ep->descs->rdesc[entry].buf_addr,
+ length, DMA_FROM_DEVICE);
+ eth_copy_and_sum(skb, ep->rx_buf[entry], length, 0);
+ skb_put(skb, length);
+ skb->protocol = eth_type_trans(skb, dev);
+
+ dev->last_rx = jiffies;
+
+ netif_receive_skb(skb);
+
+ ep->stats.rx_packets++;
+ ep->stats.rx_bytes += length;
+ } else {
+ ep->stats.rx_dropped++;
+ }
+
+err:
+ ep->rx_pointer = (entry + 1) % RX_QUEUE_ENTRIES;
+ processed++;
+ dev->quota--;
+ (*budget)--;
+ }
+
+ if (processed) {
+ wrw(ep, 0x009c, processed);
+ wrw(ep, 0x00ac, processed);
+ }
+
+ return !rx_done;
+}
+
+static int ep93xx_have_more_rx(struct ep93xx_priv *ep)
+{
+ struct ep93xx_rstat *rstat;
+ int tail_offset;
+
+ rstat = ep->descs->rstat + ep->rx_pointer;
+ tail_offset = rdl(ep, 0x00a8) - ep->descs_dma_addr;
+
+ return !((void *)rstat - (void *)ep->descs == tail_offset);
+}
+
+static int ep93xx_poll(struct net_device *dev, int *budget)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+
+ /* @@@ Have to stop polling when device is downed while we
+ * are polling. */
+
+poll_some_more:
+ if (ep93xx_rx(dev, budget))
+ return 1;
+
+ netif_rx_complete(dev);
+
+ local_irq_disable();
+ wrl(ep, 0x0024, 0x0000000f);
+ if (ep93xx_have_more_rx(ep)) {
+ wrl(ep, 0x0024, 0x00000008);
+ wrl(ep, 0x0028, 0x00000004);
+ local_irq_enable();
+
+ if (netif_rx_reschedule(dev, 0))
+ goto poll_some_more;
+
+ return 0;
+ }
+ local_irq_enable();
+
+ return 0;
+}
+
+static void ep93xx_tx_complete(struct net_device *dev)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+ int tail_offset;
+ int wake;
+
+ tail_offset = rdl(ep, 0x00c8) - ep->descs_dma_addr;
+
+ wake = 0;
+ while (1) {
+ int entry;
+ struct ep93xx_tstat *tstat;
+ u32 tstat0;
+
+ entry = ep->tx_clean_pointer;
+ tstat = ep->descs->tstat + entry;
+ if ((void *)tstat - (void *)ep->descs == tail_offset)
+ break;
+
+ tstat0 = tstat->tstat0;
+ tstat->tstat0 = 0;
+
+ if (!(tstat0 & TSTAT0_TXFP)) {
+ printk(KERN_CRIT "ep93xx_tx_complete: buffer not done "
+ " %.8x\n", tstat0);
+ BUG();
+ }
+ if (tstat0 & TSTAT0_FA) {
+ printk(KERN_CRIT "ep93xx_tx_complete: frame aborted "
+ " %.8x\n", tstat0);
+ BUG();
+ }
+ if ((tstat0 & TSTAT0_BUFFER_INDEX) != entry) {
+ printk(KERN_INFO "ep93xx_tx_complete: entry mismatch "
+ " %.8x\n", tstat0);
+ BUG();
+ }
+
+ if (tstat0 & TSTAT0_TXWE) {
+ int length = ep->descs->tdesc[entry].tdesc1 & 0xfff;
+
+ ep->stats.tx_packets++;
+ ep->stats.tx_bytes += length;
+ } else {
+ ep->stats.tx_errors++;
+ }
+ if (tstat0 & TSTAT0_LCRS)
+ ep->stats.tx_carrier_errors++;
+ if (tstat0 & TSTAT0_OW)
+ ep->stats.tx_window_errors++;
+ if (tstat0 & TSTAT0_TXU)
+ ep->stats.tx_fifo_errors++;
+ ep->stats.collisions += (tstat0 >> 16) & 0x1f;
+
+ ep->tx_clean_pointer = (entry + 1) % TX_QUEUE_ENTRIES;
+ if (ep->tx_pending == TX_QUEUE_ENTRIES)
+ wake = 1;
+ ep->tx_pending--;
+ }
+
+ if (wake)
+ netif_wake_queue(dev);
+}
+
+static irqreturn_t ep93xx_irq(int irq, void *dev_id, struct pt_regs *regs)
+{
+ struct net_device *dev = dev_id;
+ struct ep93xx_priv *ep = netdev_priv(dev);
+ u32 status;
+
+ status = rdl(ep, 0x002c);
+ if (status == 0)
+ return IRQ_NONE;
+
+ if (status & 0x00000004) {
+ if (likely(__netif_rx_schedule_prep(dev))) {
+ wrl(ep, 0x0024, 0x00000008);
+ __netif_rx_schedule(dev);
+ }
+ }
+
+ if (status & 0x00000008)
+ ep93xx_tx_complete(dev);
+
+ return IRQ_HANDLED;
+}
+
+static void ep93xx_free_buffers(struct ep93xx_priv *ep)
+{
+ int i;
+
+ for (i = 0; i < RX_QUEUE_ENTRIES; i++) {
+ dma_addr_t d;
+
+ d = ep->descs->rdesc[i].buf_addr;
+ if (d)
+ dma_unmap_single(NULL, d, PAGE_SIZE, DMA_FROM_DEVICE);
+
+ if (ep->rx_buf[i] != NULL)
+ free_page((unsigned long)ep->rx_buf[i]);
+ }
+
+ for (i = 0; i < TX_QUEUE_ENTRIES; i++) {
+ dma_addr_t d;
+
+ d = ep->descs->tdesc[i].buf_addr;
+ if (d)
+ dma_unmap_single(NULL, d, PAGE_SIZE, DMA_TO_DEVICE);
+
+ if (ep->tx_buf[i] != NULL)
+ free_page((unsigned long)ep->tx_buf[i]);
+ }
+
+ dma_free_coherent(NULL, sizeof(struct ep93xx_descs), ep->descs,
+ ep->descs_dma_addr);
+}
+
+static int ep93xx_alloc_buffers(struct ep93xx_priv *ep)
+{
+ int i;
+
+ ep->descs = dma_alloc_coherent(NULL, sizeof(struct ep93xx_descs),
+ &ep->descs_dma_addr, GFP_KERNEL | GFP_DMA);
+ if (ep->descs == NULL)
+ return 1;
+
+ for (i = 0; i < RX_QUEUE_ENTRIES; i++) {
+ void *page;
+ dma_addr_t d;
+
+ page = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+ if (page == NULL)
+ goto err;
+
+ d = dma_map_single(NULL, page, PAGE_SIZE, DMA_FROM_DEVICE);
+ if (dma_mapping_error(d)) {
+ free_page((unsigned long)page);
+ goto err;
+ }
+
+ ep->rx_buf[i] = page;
+ ep->descs->rdesc[i].buf_addr = d;
+ ep->descs->rdesc[i].rdesc1 = (i << 16) | PAGE_SIZE;
+ }
+
+ for (i = 0; i < TX_QUEUE_ENTRIES; i++) {
+ void *page;
+ dma_addr_t d;
+
+ page = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+ if (page == NULL)
+ goto err;
+
+ d = dma_map_single(NULL, page, PAGE_SIZE, DMA_TO_DEVICE);
+ if (dma_mapping_error(d)) {
+ free_page((unsigned long)page);
+ goto err;
+ }
+
+ ep->tx_buf[i] = page;
+ ep->descs->tdesc[i].buf_addr = d;
+ }
+
+ return 0;
+
+err:
+ ep93xx_free_buffers(ep);
+ return 1;
+}
+
+static void ep93xx_start_hw(struct net_device *dev)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+ unsigned long addr;
+
+ wrl(ep, 0x0020, 0x00000001);
+ while (rdl(ep, 0x0020) & 0x00000001)
+ ;
+
+ /* Receive descriptor ring. */
+ addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, rdesc);
+ wrl(ep, 0x0090, addr);
+ wrl(ep, 0x0098, addr);
+ wrw(ep, 0x0094, RX_QUEUE_ENTRIES * sizeof(struct ep93xx_rdesc));
+
+ /* Receive status ring. */
+ addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, rstat);
+ wrl(ep, 0x00a0, addr);
+ wrl(ep, 0x00a8, addr);
+ wrw(ep, 0x00a4, RX_QUEUE_ENTRIES * sizeof(struct ep93xx_rstat));
+
+ /* Transmit descriptor ring. */
+ addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, tdesc);
+ wrl(ep, 0x00b0, addr);
+ wrl(ep, 0x00b8, addr);
+ wrw(ep, 0x00b4, TX_QUEUE_ENTRIES * sizeof(struct ep93xx_tdesc));
+
+ /* Transmit status ring. */
+ addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, tstat);
+ wrl(ep, 0x00c0, addr);
+ wrl(ep, 0x00c8, addr);
+ wrw(ep, 0x00c4, TX_QUEUE_ENTRIES * sizeof(struct ep93xx_tstat));
+
+ wrl(ep, 0x0080, 0x00000101); // bmctl
+ wrl(ep, 0x0024, 0x0000000f); // inten
+ wrl(ep, 0x0064, 0x00000000); // gintmask
+ while (!(rdl(ep, 0x0084) & 0x00000008))
+ cpu_relax();
+ wrl(ep, 0x009c, RX_QUEUE_ENTRIES);
+ wrl(ep, 0x00ac, RX_QUEUE_ENTRIES);
+
+ wrb(ep, 0x0050, dev->dev_addr[0]);
+ wrb(ep, 0x0051, dev->dev_addr[1]);
+ wrb(ep, 0x0052, dev->dev_addr[2]);
+ wrb(ep, 0x0053, dev->dev_addr[3]);
+ wrb(ep, 0x0054, dev->dev_addr[4]);
+ wrb(ep, 0x0055, dev->dev_addr[5]);
+ wrl(ep, 0x004c, 0x00000000);
+
+ wrl(ep, 0x0000, 0x00073800);
+ wrl(ep, 0x0004, 0x00000001);
+}
+
+static void ep93xx_stop_hw(struct net_device *dev)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+
+ wrl(ep, 0x0020, 0x00000001);
+ while (rdl(ep, 0x0020) & 0x00000001)
+ ;
+}
+
+static int ep93xx_open(struct net_device *dev)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+ int err;
+
+ if (ep93xx_alloc_buffers(ep))
+ return -ENOMEM;
+
+ ep93xx_start_hw(dev);
+ ep->rx_pointer = 0;
+ ep->tx_clean_pointer = 0;
+ ep->tx_pointer = 0;
+ ep->tx_pending = 0;
+
+ err = request_irq(ep->irq, ep93xx_irq, SA_SHIRQ, "ep93xx_eth", dev);
+ if (err) {
+ ep93xx_stop_hw(dev);
+ ep93xx_free_buffers(ep);
+ return err;
+ }
+
+ wrl(ep, 0x0064, 0x00008000); // gintmask
+
+ netif_start_queue(dev);
+
+ return 0;
+}
+
+static int ep93xx_close(struct net_device *dev)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+
+ netif_stop_queue(dev);
+
+ wrl(ep, 0x0064, 0x00000000); // gintmask
+ free_irq(ep->irq, dev);
+ ep93xx_stop_hw(dev);
+ ep93xx_free_buffers(ep);
+
+ return 0;
+}
+
+static struct net_device_stats *ep93xx_get_stats(struct net_device *dev)
+{
+ struct ep93xx_priv *ep = netdev_priv(dev);
+ return &(ep->stats);
+}
+
+struct net_device *ep93xx_dev_alloc(struct ep93xx_eth_data *data)
+{
+ struct net_device *dev;
+ struct ep93xx_priv *ep;
+
+ dev = alloc_etherdev(sizeof(struct ep93xx_priv));
+ if (dev == NULL)
+ return NULL;
+ ep = netdev_priv(dev);
+
+ memcpy(dev->dev_addr, data->dev_addr, ETH_ALEN);
+
+ dev->hard_start_xmit = ep93xx_xmit;
+ dev->poll = ep93xx_poll;
+ dev->open = ep93xx_open;
+ dev->stop = ep93xx_close;
+ dev->get_stats = ep93xx_get_stats;
+
+ dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM;
+ dev->weight = 64;
+
+ return dev;
+}
+
+
+static int ep93xx_eth_remove(struct platform_device *pdev)
+{
+ struct net_device *dev;
+ struct ep93xx_priv *ep;
+
+ dev = platform_get_drvdata(pdev);
+ if (dev == NULL)
+ return 0;
+ platform_set_drvdata(pdev, NULL);
+
+ ep = netdev_priv(dev);
+
+ // @@@ force down
+ unregister_netdev(dev);
+ ep93xx_free_buffers(ep);
+
+ if (ep->base_addr != NULL)
+ iounmap(ep->base_addr);
+
+ if (ep->res != NULL) {
+ release_resource(ep->res);
+ kfree(ep->res);
+ }
+
+ free_netdev(dev);
+
+ return 0;
+}
+
+static int ep93xx_eth_probe(struct platform_device *pdev)
+{
+ struct ep93xx_eth_data *data;
+ struct net_device *dev;
+ struct ep93xx_priv *ep;
+ int err;
+
+ data = pdev->dev.platform_data;
+ if (pdev == NULL)
+ return -ENODEV;
+
+ printk(KERN_INFO "ep93xx-eth " DRV_MODULE_VERSION " loading\n");
+
+ dev = ep93xx_dev_alloc(data);
+ if (dev == NULL) {
+ err = -ENOMEM;
+ goto err_out;
+ }
+ ep = netdev_priv(dev);
+
+ platform_set_drvdata(pdev, dev);
+
+ ep->res = request_mem_region(pdev->resource[0].start,
+ pdev->resource[0].end - pdev->resource[0].start + 1,
+ pdev->dev.bus_id);
+ if (ep->res == NULL) {
+ dev_err(&pdev->dev, "Could not reserve memory region\n");
+ err = -ENOMEM;
+ goto err_out;
+ }
+
+ ep->base_addr = ioremap(pdev->resource[0].start,
+ pdev->resource[0].end - pdev->resource[0].start);
+ if (ep->base_addr == NULL) {
+ dev_err(&pdev->dev, "Failed to ioremap ethernet registers\n");
+ err = -EIO;
+ goto err_out;
+ }
+ ep->irq = pdev->resource[1].start;
+
+ err = register_netdev(dev);
+ if (err) {
+ dev_err(&pdev->dev, "Failed to register netdev\n");
+ goto err_out;
+ }
+
+ return 0;
+
+err_out:
+ ep93xx_eth_remove(pdev);
+ return err;
+}
+
+
+static struct platform_driver ep93xx_eth_driver = {
+ .probe = ep93xx_eth_probe,
+ .remove = ep93xx_eth_remove,
+ .driver = {
+ .name = "ep93xx-eth",
+ },
+};
+
+static int __init ep93xx_eth_init_module(void)
+{
+ return platform_driver_register(&ep93xx_eth_driver);
+}
+
+static void __exit ep93xx_eth_cleanup_module(void)
+{
+ platform_driver_unregister(&ep93xx_eth_driver);
+}
+
+module_init(ep93xx_eth_init_module);
+module_exit(ep93xx_eth_cleanup_module);
+MODULE_LICENSE("GPL");
Index: linux-2.6.17-git10/drivers/net/arm/ep93xx_eth.h
===================================================================
--- /dev/null
+++ linux-2.6.17-git10/drivers/net/arm/ep93xx_eth.h
@@ -0,0 +1,75 @@
+/*
+ * EP93xx ethernet network device driver
+ * Copyright (C) 2006 Lennert Buytenhek <buytenh@wantstofly.org>
+ * Dedicated to Marija Kulikova.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __EP93XX_ETH_H
+#define __EP93XX_ETH_H
+
+struct ep93xx_rdesc
+{
+ u32 buf_addr;
+ u32 rdesc1;
+};
+
+#define RDESC1_NSOF 0x80000000
+#define RDESC1_BUFFER_INDEX 0x7fff0000
+#define RDESC1_BUFFER_LENGTH 0x0000ffff
+
+struct ep93xx_rstat
+{
+ u32 rstat0;
+ u32 rstat1;
+};
+
+#define RSTAT0_RFP 0x80000000
+#define RSTAT0_RWE 0x40000000
+#define RSTAT0_EOF 0x20000000
+#define RSTAT0_EOB 0x10000000
+#define RSTAT0_AM 0x00c00000
+#define RSTAT0_RX_ERR 0x00200000
+#define RSTAT0_OE 0x00100000
+#define RSTAT0_FE 0x00080000
+#define RSTAT0_RUNT 0x00040000
+#define RSTAT0_EDATA 0x00020000
+#define RSTAT0_CRCE 0x00010000
+#define RSTAT0_CRCI 0x00008000
+#define RSTAT0_HTI 0x00003f00
+#define RSTAT1_RFP 0x80000000
+#define RSTAT1_BUFFER_INDEX 0x7fff0000
+#define RSTAT1_FRAME_LENGTH 0x0000ffff
+
+struct ep93xx_tdesc
+{
+ u32 buf_addr;
+ u32 tdesc1;
+};
+
+#define TDESC1_EOF 0x80000000
+#define TDESC1_BUFFER_INDEX 0x7fff0000
+#define TDESC1_BUFFER_ABORT 0x00008000
+#define TDESC1_BUFFER_LENGTH 0x00000fff
+
+struct ep93xx_tstat
+{
+ u32 tstat0;
+};
+
+#define TSTAT0_TXFP 0x80000000
+#define TSTAT0_TXWE 0x40000000
+#define TSTAT0_FA 0x20000000
+#define TSTAT0_LCRS 0x10000000
+#define TSTAT0_OW 0x04000000
+#define TSTAT0_TXU 0x02000000
+#define TSTAT0_ECOLL 0x01000000
+#define TSTAT0_NCOLL 0x001f0000
+#define TSTAT0_BUFFER_INDEX 0x00007fff
+
+
+#endif
Index: linux-2.6.17-git10/include/asm-arm/arch-ep93xx/ep93xx-regs.h
===================================================================
--- linux-2.6.17-git10.orig/include/asm-arm/arch-ep93xx/ep93xx-regs.h
+++ linux-2.6.17-git10/include/asm-arm/arch-ep93xx/ep93xx-regs.h
@@ -27,6 +27,7 @@
#define EP93XX_DMA_BASE (EP93XX_AHB_VIRT_BASE + 0x00000000)
#define EP93XX_ETHERNET_BASE (EP93XX_AHB_VIRT_BASE + 0x00010000)
+#define EP93XX_ETHERNET_PHYS_BASE (EP93XX_AHB_PHYS_BASE + 0x00010000)
#define EP93XX_USB_BASE (EP93XX_AHB_VIRT_BASE + 0x00020000)
#define EP93XX_USB_PHYS_BASE (EP93XX_AHB_PHYS_BASE + 0x00020000)
Index: linux-2.6.17-git10/include/asm-arm/arch-ep93xx/platform.h
===================================================================
--- linux-2.6.17-git10.orig/include/asm-arm/arch-ep93xx/platform.h
+++ linux-2.6.17-git10/include/asm-arm/arch-ep93xx/platform.h
@@ -11,5 +11,10 @@ void ep93xx_init_devices(void);
void ep93xx_clock_init(void);
extern struct sys_timer ep93xx_timer;
+struct ep93xx_eth_data
+{
+ unsigned char dev_addr[6];
+};
+
#endif
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox