All drivers implement ethtool get_perm_addr the same way -- by calling
the generic function. So we can inline the generic function into the
caller and avoid going through the drivers.
Signed-off-by: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: David S. Miller <davem@davemloft.net>
During the transition to the ethtool_ops way of doing things, we supported
calling the device's ->do_ioctl method to allow unconverted drivers to
continue working. Those days are long behind us, all in-tree drivers
use the ethtool_ops way, and so we no longer need to support this.
The bonding driver is the biggest beneficiary of this; it no longer
needs to call ioctl() as a fallback if ethtool_ops aren't supported.
Also put a proper copyright statement on ethtool.c.
Signed-off-by: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: David S. Miller <davem@davemloft.net>
Non-static inline code usually doesn't makes sense.
In this case making is static and non-inline is the correct solution.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds code to allow errors to be passed up from event
handlers of NETDEV_REGISTER and NETDEV_CHANGENAME. It also adds
the notifier_from_errno/notifier_to_errnor helpers to pass the
errno value up to the notifier caller.
If an error is detected when a device is registered, it causes
that operation to fail. A NETDEV_UNREGISTER will be sent to
all event handlers.
Similarly if NETDEV_CHANGENAME fails the original name is restored
and a new NETDEV_CHANGENAME event is sent.
As such all event handlers must be idempotent with respect to
these events.
When an event handler is registered NETDEV_REGISTER events are
sent for all devices currently registered. Should any of them
fail, we will send NETDEV_GOING_DOWN/NETDEV_DOWN/NETDEV_UNREGISTER
events to that handler for the devices which have already been
registered with it. The handler registration itself will fail.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we added name-based hashing the dev_base_lock was designated as the
lock to take when changing the name hash list. Unfortunately, because
it was a preexisting lock that just happened to be taken in the right
spots we neglected to take it in dev_change_name.
The race can affect calles of __dev_get_by_name that do so without taking
the RTNL. They may end up walking down the wrong hash chain and end up
missing the device that they're looking for.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch makes register_netdevice call dev->uninit if the regsitration
fails after dev->init has completed successfully. Very few drivers use
the init/uninit calls but at least one (drivers/net/wan/sealevel.c) may
leak without this change.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix kernel-doc omissions in net/:
Warning(linux-2.6.23-rc1//net/core/dev.c:2728): No description found for parameter 'addr'
Warning(linux-2.6.23-rc1//net/core/dev.c:2752): No description found for parameter 'addr'
Warning(linux-2.6.23-rc1//net/core/dev.c:3839): No description found for parameter 'net_dma'
Warning(linux-2.6.23-rc1//net/core/dev.c:3877): No description found for parameter 'state'
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add missing entries to af_family_clock_key_strings[].
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Slab destructors were no longer supported after Christoph's
c59def9f22 change. They've been
BUGs for both slab and slub, and slob never supported them
either.
This rips out support for the dtor pointer from kmem_cache_create()
completely and fixes up every single callsite in the kernel (there were
about 224, not including the slab allocator definitions themselves,
or the documentation references).
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
the two init sites resulted in inconsistend names for the lock class.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
this two functions could share the dev->_xmit_lock acquired context.
Signed-off-by: Denis Cheng <crquan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Because this function is only called by unregister_netdevice,
this moving could make this non-global function static,
and also remove its declaration in netdevice.h;
Any further, function __dev_addr_discard is also just called by
dev_mc_discard and dev_unicast_discard, keeping this two functions
both in one c file could make __dev_addr_discard also static
and remove its declaration in netdevice.h;
Futhermore, the sequential call to dev_unicast_discard and then
dev_mc_discard in unregister_netdevice have a similar mechanism that:
(netif_tx_lock_bh / __dev_addr_discard / netif_tx_unlock_bh),
they should merged into one to eliminate duplicates in acquiring and
releasing the dev->_xmit_lock, this would be done in my following patch.
Signed-off-by: Denis Cheng <crquan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
-Fixes ABBA deadlock noted by Patrick McHardy <kaber@trash.net>:
> There is at least one ABBA deadlock, est_timer() does:
> read_lock(&est_lock)
> spin_lock(e->stats_lock) (which is dev->queue_lock)
>
> and qdisc_destroy calls htb_destroy under dev->queue_lock, which
> calls htb_destroy_class, then gen_kill_estimator and this
> write_locks est_lock.
To fix the ABBA deadlock the rate estimators are now kept on an rcu list.
-The est_lock changes the use from protecting the list to protecting
the update to the 'bstat' pointer in order to avoid NULL dereferencing.
-The 'interval' member of the gen_estimator structure removed as it is
not needed.
Signed-off-by: Ranko Zivojnovic <ranko@spidernet.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the freezer treats all tasks as freezable, except for the kernel
threads that explicitly set the PF_NOFREEZE flag for themselves. This
approach is problematic, since it requires every kernel thread to either
set PF_NOFREEZE explicitly, or call try_to_freeze(), even if it doesn't
care for the freezing of tasks at all.
It seems better to only require the kernel threads that want to or need to
be frozen to use some freezer-related code and to remove any
freezer-related code from the other (nonfreezable) kernel threads, which is
done in this patch.
The patch causes all kernel threads to be nonfreezable by default (ie. to
have PF_NOFREEZE set by default) and introduces the set_freezable()
function that should be called by the freezable kernel threads in order to
unset PF_NOFREEZE. It also makes all of the currently freezable kernel
threads call set_freezable(), so it shouldn't cause any (intentional)
change of behaviour to appear. Additionally, it updates documentation to
describe the freezing of tasks more accurately.
[akpm@linux-foundation.org: build fixes]
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Nigel Cunningham <nigel@nigel.suspend2.net>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This reverts commit 29578624e3.
Ingo Molnar reports complete breakage with his e1000 card (no
networking, card reports transmit timeouts), and bisected it down to
this commit. Let's figure out what went wrong, but not keep breaking
machines until we do.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Olaf Kirch <olaf.kirch@oracle.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Part two in the O_CLOEXEC saga: adding support for file descriptors received
through Unix domain sockets.
The patch is once again pretty minimal, it introduces a new flag for recvmsg
and passes it just like the existing MSG_CMSG_COMPAT flag. I think this bit
is not used otherwise but the networking people will know better.
This new flag is not recognized by recvfrom and recv. These functions cannot
be used for that purpose and the asymmetry this introduces is not worse than
the already existing MSG_CMSG_COMPAT situations.
The patch must be applied on the patch which introduced O_CLOEXEC. It has to
remove static from the new get_unused_fd_flags function but since scm.c cannot
live in a module the function still hasn't to be exported.
Here's a test program to make sure the code works. It's so much longer than
the actual patch...
#include <errno.h>
#include <error.h>
#include <fcntl.h>
#include <stdio.h>
#include <string.h>
#include <unistd.h>
#include <sys/socket.h>
#include <sys/un.h>
#ifndef O_CLOEXEC
# define O_CLOEXEC 02000000
#endif
#ifndef MSG_CMSG_CLOEXEC
# define MSG_CMSG_CLOEXEC 0x40000000
#endif
int
main (int argc, char *argv[])
{
if (argc > 1)
{
int fd = atol (argv[1]);
printf ("child: fd = %d\n", fd);
if (fcntl (fd, F_GETFD) == 0 || errno != EBADF)
{
puts ("file descriptor valid in child");
return 1;
}
return 0;
}
struct sockaddr_un sun;
strcpy (sun.sun_path, "./testsocket");
sun.sun_family = AF_UNIX;
char databuf[] = "hello";
struct iovec iov[1];
iov[0].iov_base = databuf;
iov[0].iov_len = sizeof (databuf);
union
{
struct cmsghdr hdr;
char bytes[CMSG_SPACE (sizeof (int))];
} buf;
struct msghdr msg = { .msg_iov = iov, .msg_iovlen = 1,
.msg_control = buf.bytes,
.msg_controllen = sizeof (buf) };
struct cmsghdr *cmsg = CMSG_FIRSTHDR (&msg);
cmsg->cmsg_level = SOL_SOCKET;
cmsg->cmsg_type = SCM_RIGHTS;
cmsg->cmsg_len = CMSG_LEN (sizeof (int));
msg.msg_controllen = cmsg->cmsg_len;
pid_t child = fork ();
if (child == -1)
error (1, errno, "fork");
if (child == 0)
{
int sock = socket (PF_UNIX, SOCK_STREAM, 0);
if (sock < 0)
error (1, errno, "socket");
if (bind (sock, (struct sockaddr *) &sun, sizeof (sun)) < 0)
error (1, errno, "bind");
if (listen (sock, SOMAXCONN) < 0)
error (1, errno, "listen");
int conn = accept (sock, NULL, NULL);
if (conn == -1)
error (1, errno, "accept");
*(int *) CMSG_DATA (cmsg) = sock;
if (sendmsg (conn, &msg, MSG_NOSIGNAL) < 0)
error (1, errno, "sendmsg");
return 0;
}
/* For a test suite this should be more robust like a
barrier in shared memory. */
sleep (1);
int sock = socket (PF_UNIX, SOCK_STREAM, 0);
if (sock < 0)
error (1, errno, "socket");
if (connect (sock, (struct sockaddr *) &sun, sizeof (sun)) < 0)
error (1, errno, "connect");
unlink (sun.sun_path);
*(int *) CMSG_DATA (cmsg) = -1;
if (recvmsg (sock, &msg, MSG_CMSG_CLOEXEC) < 0)
error (1, errno, "recvmsg");
int fd = *(int *) CMSG_DATA (cmsg);
if (fd == -1)
error (1, 0, "no descriptor received");
char fdname[20];
snprintf (fdname, sizeof (fdname), "%d", fd);
execl ("/proc/self/exe", argv[0], fdname, NULL);
puts ("execl failed");
return 1;
}
[akpm@linux-foundation.org: Fix fastcall inconsistency noted by Michael Buesch]
[akpm@linux-foundation.org: build fix]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Michael Buesch <mb@bu3sch.de>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add ethtool utility function to set or clear IPV6_CSUM feature flag.
Modify tg3.c and bnx2.c to use this function when doing ethtool -K
to change tx checksum.
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add macvlan driver, which allows to create virtual ethernet devices
based on MAC address.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The method drivers currently use to synchronize multicast lists is not
very pretty:
- walk the multicast list
- search each entry on a copy of the previous list
- if new add to lower device
- walk the copy of the previous list
- search each entry on the current list
- if removed delete from lower device
- copy entire list
This patch adds a new field to struct dev_addr_list to store the
synchronization state and adds two helper functions for synchronization
and cleanup.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently the set_multicast_list (and set_rx_mode) callbacks are
responsible for configuring the device according to the IFF_PROMISC,
IFF_MULTICAST and IFF_ALLMULTI flags and the mc_list (and uc_list in
case of set_rx_mode).
These callbacks can be invoked from BH context without the rtnl_mutex
by dev_mc_add/dev_mc_delete, which makes reading the device flags and
promiscous/allmulti count racy. For real hardware drivers that just
commit all changes to the hardware this is not a real problem since
the stack guarantees to call them for every change, so at least the
final call will not race and commit the correct configuration to the
hardware.
For software devices that want to synchronize promiscous and multicast
state to an underlying device however this can cause corruption of the
underlying device's flags or promisc/allmulti counts.
When the software device is concurrently put in promiscous or allmulti
mode while set_multicast_list is invoked from bottem half context, the
device might synchronize the change to the underlying device without
holding the rtnl_mutex, which races with concurrent changes to the
underlying device.
Add a dev->change_rx_flags hook that is invoked when any of the flags
that affect rx filtering change (under the rtnl_mutex), which allows
drivers to perform synchronization immediately and only synchronize
the address lists in set_multicast_list/set_rx_mode.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
* 'ioat-md-accel-for-linus' of git://lost.foo-projects.org/~dwillia2/git/iop: (28 commits)
ioatdma: add the unisys "i/oat" pci vendor/device id
ARM: Add drivers/dma to arch/arm/Kconfig
iop3xx: surface the iop3xx DMA and AAU units to the iop-adma driver
iop13xx: surface the iop13xx adma units to the iop-adma driver
dmaengine: driver for the iop32x, iop33x, and iop13xx raid engines
md: remove raid5 compute_block and compute_parity5
md: handle_stripe5 - request io processing in raid5_run_ops
md: handle_stripe5 - add request/completion logic for async expand ops
md: handle_stripe5 - add request/completion logic for async read ops
md: handle_stripe5 - add request/completion logic for async check ops
md: handle_stripe5 - add request/completion logic for async compute ops
md: handle_stripe5 - add request/completion logic for async write ops
md: common infrastructure for running operations with raid5_run_ops
md: raid5_run_ops - run stripe operations outside sh->lock
raid5: replace custom debug PRINTKs with standard pr_debug
raid5: refactor handle_stripe5 and handle_stripe6 (v3)
async_tx: add the async_tx api
xor: make 'xor_blocks' a library routine for use with async_tx
dmaengine: make clients responsible for managing channels
dmaengine: refactor dmaengine around dma_async_tx_descriptor
...
The current implementation assumes that a channel will only be used by one
client at a time. In order to enable channel sharing the dmaengine core is
changed to a model where clients subscribe to channel-available-events.
Instead of tracking how many channels a client wants and how many it has
received the core just broadcasts the available channels and lets the
clients optionally take a reference. The core learns about the clients'
needs at dma_event_callback time.
In support of multiple operation types, clients can specify a capability
mask to only be notified of channels that satisfy a certain set of
capabilities.
Changelog:
* removed DMA_TX_ARRAY_INIT, no longer needed
* dma_client_chan_free -> dma_chan_release: switch to global reference
counting only at device unregistration time, before it was also happening
at client unregistration time
* clients now return dma_state_client to dmaengine (ack, dup, nak)
* checkpatch.pl fixes
* fixup merge with git-ioat
Cc: Chris Leech <christopher.leech@intel.com>
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: David S. Miller <davem@davemloft.net>
Drivers need to validate the initial addresses in their netlink attribute
validation function or manually reject them if they can't support this.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
All drivers need to unregister their devices in the module unload function.
While doing so they must hold the rtnl and atomically unregister the
rtnl_link ops as well. This makes the rtnl_link_unregister function that
takes the rtnl itself completely useless.
Provide default newlink/dellink functions, make __rtnl_link_unregister and
rtnl_link_unregister unregister all devices with matching rtnl_link_ops and
change the existing users to take advantage of that.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Keep netpoll/poll_napi from messing with the poll_list.
Only net_rx_action is allowed to manipulate the list.
Signed-off-by: Olaf Kirch <olaf.kirch@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As noticed by Jarek Poplawski <jarkao2@o2.pl>, the timer removal in
gen_kill_estimator races with the timer function rearming the timer.
Check whether the timer list is empty before rearming the timer
in the timer function to fix this.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Jarek Poplawski <jarkao2@o2.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
93ec2c723e applied excessive duct tape to
the netpoll beast's netpoll_cleanup(), thus substituting one leak with
another, and opening up a little buglet :-)
net_device->npinfo (netpoll_info) is a shared and refcounted object and
cannot simply be set NULL the first time netpoll_cleanup() is called.
Otherwise, further netpoll_cleanup()'s see np->dev->npinfo == NULL and
become no-ops, thus leaking. And it's a bug too: the first call to
netpoll_cleanup() would thus (annoyingly) "disable" other (still alive)
netpolls too. Maybe nobody noticed this because netconsole (only user
of netpoll) never supported multiple netpoll objects earlier.
This is a trivial and obvious one-line fixlet.
Signed-off-by: Satyam Sharma <ssatyam@cse.iitk.ac.in>
Signed-off-by: David S. Miller <davem@davemloft.net>
- save 4 bytes
- it's read-mostly.
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Vasily Averin <vvs@sw.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
This includes /proc/net/protocols, /proc/net/rxrpc_calls and
/proc/net/rxrpc_connections files.
All three need seq_list_start_head to show some header.
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The TRACE target can be used to follow IP and IPv6 packets through
the ruleset.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Signed-off-by: Patrick NcHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Added transport mode ESP support for starters. I will send more of
these modes and types once i have resolved the tunnel mode isses.
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
By default all flows in pktgen are randomly selected.
This patch introduces ability to have all defined flows to
be sent sequentially. Robert defined randomness to be the
default behavior.
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
Track the extra packet overhead for VLAN tags, MPLS, IPSEC etc
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a reference to an existing address is increased or decreased without
hitting zero, the address count is incorrectly adjusted.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the multiqueue hardware device support API to the core network
stack. Allow drivers to allocate multiple queues and manage them at
the netdev level if they choose to do so.
Added a new field to sk_buff, namely queue_mapping, for drivers to
know which tx_ring to select based on OS classification of the flow.
Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes a boolean error in the new TX checksum check
that causes bogus TSO packets to be generated.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for configuring secondary unicast addresses on network
devices. To support this devices capable of filtering multiple
unicast addresses need to change their set_multicast_list function
to configure unicast filters as well and assign it to dev->set_rx_mode
instead of dev->set_multicast_list. Other devices are put into promiscous
mode when secondary unicast addresses are present.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use generic net_device address lists for multicast list handling.
Some defines are used to keep drivers working.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce struct dev_addr_list and list maintenance functions
based on dev_mc_list and the related functions. This will be
used by follow-up patches for both multicast and secondary
unicast addresses.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
dev_mc_add/dev_mc_delete take care of uploading the list when
necessary and thats the only interface other code should use.
Also remove two incorrect calls in DECnet.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The existing model for checksum offload does not correctly handle
devices that can offload IPV4 and IPV6 only. The NETIF_F_HW_CSUM flag
implies device can do any arbitrary protocol.
This patch:
* adds NETIF_F_IPV6_CSUM for those devices
* fixes bnx2 and tg3 devices that need it
* add NETIF_F_IPV6_CSUM to ipv6 output (incl GSO)
* fixes assumptions about NETIF_F_ALL_CSUM in nat
* adjusts bridge union of checksumming computation
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a nested compat attribute type that can be used to convert
attributes that contain a structure to nested attributes in a
backwards compatible way.
The attribute looks like this:
struct {
[ compat contents ]
struct rtattr {
.rta_len = total size,
.rta_type = type,
} rta;
struct old_structure struct;
[ nested top-level attribute ]
struct rtattr {
.rta_len = nest size,
.rta_type = type,
} nest_attr;
[ optional 0 .. n nested attributes ]
struct rtattr {
.rta_len = private attribute len,
.rta_type = private attribute typ,
} nested_attr;
struct nested_data data;
};
Since both userspace and kernel deal correctly with attributes that are
larger than expected old versions will just parse the compat part and
ignore the rest.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently NAT (and others) that want to modify cloned skbs copy them,
even if in the vast majority of cases its not necessary because the
skb is a clone made by TCP and the portion NAT wants to modify is
actually writable because TCP release the header reference before
cloning.
The problem is that there is no clean way for NAT to find out how
long the writable header area is, so this patch introduces skb->hdr_len
to hold this length. When a headerless skb is cloned skb->hdr_len
is set to the current headroom, for regular clones it is copied from
the original. A new function skb_clone_writable(skb, len) returns
whether the skb is writable up to len bytes from skb->data. To avoid
enlarging the skb the mac_len field is reduced to 16 bit and the
new hdr_len field is put in the remaining 16 bit.
I've done a few rough benchmarks of NAT (not with this exact patch,
but a very similar one). As expected it saves huge amounts of system
time in case of sendfile, bringing it down to basically the same
amount as without NAT, with sendmsg it only helps on loopback,
probably because of the large MTU.
Transmit a 1GB file using sendfile/sendmsg over eth0/lo with and
without NAT:
- sendfile eth0, no NAT: sys 0m0.388s
- sendfile eth0, NAT: sys 0m1.835s
- sendfile eth0: NAT + path: sys 0m0.370s (~ -80%)
- sendfile lo, no NAT: sys 0m0.258s
- sendfile lo, NAT: sys 0m2.609s
- sendfile lo, NAT + patch: sys 0m0.260s (~ -90%)
- sendmsg eth0, no NAT: sys 0m2.508s
- sendmsg eth0, NAT: sys 0m2.539s
- sendmsg eth0, NAT + patch: sys 0m2.445s (no change)
- sendmsg lo, no NAT: sys 0m2.151s
- sendmsg lo, NAT: sys 0m3.557s
- sendmsg lo, NAT + patch: sys 0m2.159s (~ -40%)
I expect other users can see a similar performance improvement,
packet mangling iptables targets, ipip and ip_gre come to mind ..
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
