linux/include/net/netns, branch v2.6.26

[NETNS]: The generic per-net pointers.

2008-04-15T07:36:08Z

Add the elastic array of void * pointer to the struct net. The access rules are simple: 1. register the ops with register_pernet_gen_device to get the id of your private pointer 2. call net_assign_generic() to put the private data on the struct net (most preferably this should be done in the ->init callback of the ops registered) 3. do not store any private reference on the net_generic array; 4. do not change this pointer while the net is alive; 5. use the net_generic() to get the pointer. When adding a new pointer, I copy the old array, replace it with a new one and schedule the old for kfree after an RCU grace period. Since the net_generic explores the net->gen array inside rcu read section and once set the net->gen->ptr[x] pointer never changes, this grants us a safe access to generic pointers. Quoting Paul: "... RCU is protecting -only- the net_generic structure that net_generic() is traversing, and the [pointer] returned by net_generic() is protected by a reference counter in the upper-level struct net." Signed-off-by: Pavel Emelyanov Acked-by: Paul E. McKenney Signed-off-by: David S. Miller

[NETNS][DCCPV6]: Move the dccp_v6_ctl_sk on the struct net.

2008-04-14T05:32:25Z

And replace all its usage with init_net's socket. Signed-off-by: Pavel Emelyanov Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller

[NETNS][DCCPV4]: Move the dccp_v4_ctl_sk on the struct net.

2008-04-14T05:29:37Z

And replace all its usage with init_net's socket. Signed-off-by: Pavel Emelyanov Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller

[NETNS]: Add an empty netns_dccp structure on struct net.

2008-04-14T05:28:42Z

According to the overall struct net design, it will be filled with DCCP-related members. Signed-off-by: Pavel Emelyanov Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller

[NETNS]: Create tcp control socket in the each namespace.

2008-04-03T21:31:33Z

Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller

[SOCK][NETNS]: Add the percpu prot_inuse counter in the struct net.

2008-04-01T02:42:16Z

Such an accounting would cost us two more dereferences to get the percpu variable from the struct net, so I make sock_prot_inuse_get and _add calls work differently depending on CONFIG_NET_NS - without it old optimized routines are used. The per-cpu counter for init_net is prepared in core_initcall, so that even af_inet, that starts as fs_initcall, will already have the init_net prepared. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller

[NETNS]: Introduce a netns_core structure.

2008-04-01T02:41:14Z

There's already some stuff on the struct net, that should better be folded into netns_core structure. I'm making the per-proto inuse counter be per-net also, which is also a candidate for this, so introduce this structure and populate it a bit. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller

[NETNS][ICMP]: Make ctl tables for ICMP sysctls per-net.

2008-03-26T08:56:24Z

Add some flesh to ipv4_sysctl_init_net and ipv4_sysctl_exit_net, i.e. copy the table, alter .data pointers and register it per-net. Other ipv4_table's sysctls are now global, but this is going to change once sysctl permissions patches migrate from -mm tree to mainline in 2.6.26 merge window :) Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller

[NETNS][ICMP]: Move ICMP sysctls on struct net.

2008-03-26T08:55:37Z

Initialization is moved to icmp_sk_init, all the places, that refer to them use init_net for now. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller

[NETNS][IPV6] mcast - handle several network namespace

2008-03-07T19:16:55Z

This patch make use of the network namespace information at the right places to handle the multicast for several network namespaces. It makes the socket control to be per namespace too. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller