- 16 Apr, 2008 40 commits
-
-
John Heffner authored
This change is necessary to allow cwnd to grow during persistent reordering. Cwnd moderation is applied when in the disorder state and an ack that fills the hole comes in. If the hole was greater than 3 packets, but less than tp->reordering, cwnd will shrink when it should not have. Signed-off-by:
John Heffner <jheffner@napa.(none)> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Pavel Emelyanov authored
The alloc_netdev_mq() tries to produce 32-bytes alignment for both the net_device itself and its private data. The second alignment is achieved by adding the NETDEV_ALIGN_CONST to the whole size of the memory to be allocated. However, for those devices that do not need the private area, this addition just makes the net_device weight 1024 + 32 = 1068 bytes, i.e. consume twice as much memory. Since loopback device is such (sizeof_priv == 0 for it), and each net namespace creates one, this can save a noticeable amount of memory for kernel with net namespaces turned on. After this set the lo device is actually allocated from a size-1024 kmem cache on i386 box even with NETPOLL and WIRELESS_EXT turned on. Signed-off-by:
Pavel Emelyanov <xemul@openvz.org> Signed-off-by:
-
Denis V. Lunev authored
dev_set_net is called for - just allocated devices - devices moving from one namespace to another release_net has proper check inside to distinguish these cases. Signed-off-by:
Denis V. Lunev <den@openvz.org> Signed-off-by:
-
Denis V. Lunev authored
Signed-off-by:
-
Denis V. Lunev authored
Signed-off-by:
-
Denis V. Lunev authored
Signed-off-by:
-
Denis V. Lunev authored
Signed-off-by:
-
Denis V. Lunev authored
Signed-off-by:
-
Denis V. Lunev authored
Protocol control sockets and netlink kernel sockets should not prevent the namespace stop request. They are initialized and disposed in a special way by sk_change_net/sk_release_kernel. Signed-off-by:
-
Denis V. Lunev authored
Make release_net/hold_net noop for performance-hungry people. This is a debug staff and should be used in the debug mode only. Add check for net != NULL in hold/release calls. This will be required later on. [ Added minor simplifications suggested by Brian Haley. -DaveM ] Signed-off-by:
-
Pavel Emelyanov authored
And no need in some IPPROTO_XXX enabling, since ipv6 code doesn't have any filtering. So, just set proper net and mark device with NETNS_LOCAL. Signed-off-by:
-
Pavel Emelyanov authored
All the ip_route_output_key(), dev_get_by_...() and ipv6_chk_addr() calls are now stubbed with init_net. Fortunately, all the places already have where to get the proper net from. Signed-off-by:
-
Pavel Emelyanov authored
Move hashes in the struct ip6_tnl_net, replace tnls_xxx[] with ip6n->tnlx_xxx[] and handle init and exit appropriately. Signed-off-by:
-
Pavel Emelyanov authored
All the code, that reference it already has the ip6_tnl_net pointer, so s/ip6_fb_tnl_dev/ip6n->fb_tnl_dev/ and move creation/releasing code into net init/exit ops. Signed-off-by:
-
Pavel Emelyanov authored
Calls to ip6_tnl_lookup were stubbed with init_net - give them a proper one. Signed-off-by:
-
Pavel Emelyanov authored
Hashes and fallback device used in them will be per-net. Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
Pavel Emelyanov authored
This makes sit-generated traffic enter the namespace. Signed-off-by:
-
Pavel Emelyanov authored
Set proper net and mark a new device as NETNS_LOCAL before registering. Signed-off-by:
-
Pavel Emelyanov authored
I.e. replace init_net stubs in ip_route_output_key() calls. Signed-off-by:
-
Pavel Emelyanov authored
Just move all the hashes on the sit_net structure and patch the rest of the code appropriately. Signed-off-by:
-
Pavel Emelyanov authored
Allocate and register one in sit_init_net, use sitn->fb_tunnel_dev over the code and unregister one in sit_exit_net. Signed-off-by:
-
Pavel Emelyanov authored
Replace introduced in the previous patch init_net stubs with the proper net pointer. Signed-off-by:
-
Pavel Emelyanov authored
... to make them prepared for future hashes and fallback device move on the struct sit_net. Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
Pavel Emelyanov authored
This one was also disabled by default for sanity. Signed-off-by:
-
Pavel Emelyanov authored
I.e. set the proper net and mark as NETNS_LOCAL. Signed-off-by:
-
Pavel Emelyanov authored
As for the IPIP tunnel, there are some ip_route_output_key() calls in there that require a proper net so give one to them. And a proper net for the __get_dev_by_index hanging around. Signed-off-by:
-
Pavel Emelyanov authored
Very similar to what was done for the IPIP code. Signed-off-by:
-
Pavel Emelyanov authored
Everything is prepared for this change now. Create on in init callback, use it over the code and destroy on net exit. Signed-off-by:
-
Pavel Emelyanov authored
This is the part#2 of the patch #2 - get the proper net for these functions. This change in a separate patch in order not to get lost in a large previous patch. Signed-off-by:
-
Pavel Emelyanov authored
The fallback device and hashes are to become per-net, but many code doesn't have anything to get the struct net pointer from. So pass the proper net there with an extra argument. Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
Pavel Emelyanov authored
This one was disabled by default for sanity. Signed-off-by:
-
Pavel Emelyanov authored
Set the proper net before calling register_netdev and disable the tunnel device netns changing. Signed-off-by:
-
Pavel Emelyanov authored
There are some ip_route_output_key() calls in there that require a proper net so give one to them. Besides - give a proper net to a single __get_dev_by_index call in ipip_tunnel_bind_dev(). Signed-off-by:
-
Pavel Emelyanov authored
Either net or ipip_net already exists in all the required places, so just use one. Besides, tune net_init and net_exit calls to respectively initialize the hashes and destroy devices. Signed-off-by:
-
Pavel Emelyanov authored
This is the part#2 of the previous patch - get the proper net for these functions. I make it in a separate patch, so that this change does not get lost in a large previous patch. Signed-off-by:
-
Pavel Emelyanov authored
The hashes of tunnels will be per-net too, so prepare all the functions that uses them for this change by adding an argument. Use init_net temporarily in places, where the net does not exist explicitly yet. Signed-off-by:
-
Pavel Emelyanov authored
Create on in ipip_init_net(), use it all over the code (the proper place to get the net from already exists) and destroy in ipip_net_exit(). Signed-off-by:
-
