linux/net/ipv4/ip_options.c, branch v3.2.41

ipv4: Fix wrong order of ip_rt_get_source() and update iph->daddr.

2012-03-01T00:31:02Z

[ Upstream commit 5dc7883f2a7c25f8df40d7479687153558cd531b ] This patch fix a bug which introduced by commit ac8a4810 (ipv4: Save nexthop address of LSRR/SSRR option to IPCB.).In that patch, we saved the nexthop of SRR in ip_option->nexthop and update iph->daddr until we get to ip_forward_options(), but we need to update it before ip_rt_get_source(), otherwise we may get a wrong src. Signed-off-by: Li Wei Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

ipv4: Save nexthop address of LSRR/SSRR option to IPCB.

2011-11-24T00:19:32Z

We can not update iph->daddr in ip_options_rcv_srr(), It is too early. When some exception ocurred later (eg. in ip_forward() when goto sr_failed) we need the ip header be identical to the original one as ICMP need it. Add a field 'nexthop' in struct ip_options to save nexthop of LSRR or SSRR option. Signed-off-by: Li Wei Signed-off-by: David S. Miller

ipv4: fix for ip_options_rcv_srr() daddr update.

2011-11-09T20:59:00Z

When opt->srr_is_hit is set skb_rtable(skb) has been updated for 'nexthop' and iph->daddr should always equals to skb_rtable->rt_dst holds, We need update iph->daddr either. Signed-off-by: Li Wei Signed-off-by: David S. Miller

ip_options_compile: properly handle unaligned pointer

2011-05-31T22:11:02Z

The current code takes an unaligned pointer and does htonl() on it to make it big-endian, then does a memcpy(). The problem is that the compiler decides that since the pointer is to a __be32, it is legal to optimize the copy into a processor word store. However, on an architecture that does not handled unaligned writes in kernel space, this produces an unaligned exception fault. The solution is to track the pointer as a "char *" (which removes a bunch of unpleasant casts in any case), and then just use put_unaligned_be32() to write the value to memory. Signed-off-by: Chris Metcalf Signed-off-by: David S. Miller

ipv4: Remove rt->rt_dst reference from ip_forward_options().

2011-05-13T21:31:02Z

At this point iph->daddr equals what rt->rt_dst would hold. Signed-off-by: David S. Miller

ipv4: Remove route key identity dependencies in ip_rt_get_source().

2011-05-13T21:29:41Z

Pass in the sk_buff so that we can fetch the necessary keys from the packet header when working with input routes. Signed-off-by: David S. Miller

ipv4: Kill spurious write to iph->daddr in ip_forward_options().

2011-05-13T21:15:50Z

This code block executes when opt->srr_is_hit is set. It will be set only by ip_options_rcv_srr(). ip_options_rcv_srr() walks until it hits a matching nexthop in the SRR option addresses, and when it matches one 1) looks up the route for that nexthop and 2) on route lookup success it writes that nexthop value into iph->daddr. ip_forward_options() runs later, and again walks the SRR option addresses looking for the option matching the destination of the route stored in skb_rtable(). This route will be the same exact one looked up for the nexthop by ip_options_rcv_srr(). Therefore "rt->rt_dst == iph->daddr" must be true. All it really needs to do is record the route's source address in the matching SRR option adddress. It need not write iph->daddr again, since that has already been done by ip_options_rcv_srr() as detailed above. Signed-off-by: David S. Miller

ipv4: Simplify iph->daddr overwrite in ip_options_rcv_srr().

2011-05-12T23:30:58Z

We already copy the 4-byte nexthop from the options block into local variable "nexthop" for the route lookup. Re-use that variable instead of memcpy()'ing again when assigning to iph->daddr after the route lookup succeeds. Signed-off-by: David S. Miller

ipv4: Kill spurious opt->srr check in ip_options_rcv_srr().

2011-05-12T23:26:57Z

All call sites conditionalize the call to ip_options_rcv_srr() with a check of opt->srr, so no need to check it again there. Signed-off-by: David S. Miller

inet: add RCU protection to inet->opt

2011-04-28T20:16:35Z

We lack proper synchronization to manipulate inet->opt ip_options Problem is ip_make_skb() calls ip_setup_cork() and ip_setup_cork() possibly makes a copy of ipc->opt (struct ip_options), without any protection against another thread manipulating inet->opt. Another thread can change inet->opt pointer and free old one under us. Use RCU to protect inet->opt (changed to inet->inet_opt). Instead of handling atomic refcounts, just copy ip_options when necessary, to avoid cache line dirtying. We cant insert an rcu_head in struct ip_options since its included in skb->cb[], so this patch is large because I had to introduce a new ip_options_rcu structure. Signed-off-by: Eric Dumazet Cc: Herbert Xu Signed-off-by: David S. Miller