One point of contention in high network loads is the dst_release() performed
when a transmited skb is freed. This is because NIC tx completion calls
dev_kree_skb() long after original call to dev_queue_xmit(skb).
CPU cache is cold and the atomic op in dst_release() stalls. On SMP, this is
quite visible if one CPU is 100% handling softirqs for a network device,
since dst_clone() is done by other cpus, involving cache line ping pongs.
It seems right place to release dst is in dev_hard_start_xmit(), for most
devices but ones that are virtual, and some exceptions.
David Miller suggested to define a new device flag, set in alloc_netdev_mq()
(so that most devices set it at init time), and carefuly unset in devices
which dont want a NULL skb->dst in their ndo_start_xmit().
List of devices that must clear this flag is :
- loopback device, because it calls netif_rx() and quoting Patrick :
"ip_route_input() doesn't accept loopback addresses, so loopback packets
already need to have a dst_entry attached."
- appletalk/ipddp.c : needs skb->dst in its xmit function
- And all devices that call again dev_queue_xmit() from their xmit function
(as some classifiers need skb->dst) : bonding, vlan, macvlan, eql, ifb, hdlc_fr
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can in some situations drop packets in netif_rx()
loopback driver does not report these (unlikely) drops to its stats,
and incorrectly change packets/bytes counts.
After this patch applied, "ifconfig lo" can reports these drops as in :
# ifconfig lo
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:692562900 errors:3228 dropped:3228 overruns:0 frame:0
TX packets:692562900 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:2865674174 (2.6 GiB) TX bytes:2865674174 (2.6 GiB)
I initialy chose to reflect those errors only in tx_dropped/tx_errors, but David
convinced me that it was really RX errors, as loopback_xmit() really starts
a RX process. (calling eth_type_trans() for example, that itself pulls the ethernet header)
These errors are accounted in rx_dropped/rx_errors.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch moves neigh_setup and hard_start_xmit into the network device ops
structure. For bisection, fix all the previously converted drivers as well.
Bonding driver took the biggest hit on this.
Added a prefetch of the hard_start_xmit in the fast path to try and reduce
any impact this would have.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
First device to convert over is the loopback device.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I was recently hunting a bug that occurred in network namespace
cleanup. In looking at the code it became apparrent that we have
and will continue to have cases where if we have anything going
on in a network namespace there will be assumptions that the
loopback device is present. Things like sending igmp unsubscribe
messages when we bring down network devices invokes the routing
code which assumes that at least the loopback driver is present.
Therefore to avoid magic initcall ordering hackery that is hard
to follow and hard to get right insert a call to register the
loopback device directly from net_dev_init(). This guarantes
that the loopback device is the first device registered and
the last network device to go away.
But do it carefully so we register the loopback device after
we clear dev_boot_phase.
Signed-off-by: Eric W. Biederman <ebiederm@maxwell.aristanetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I was recently hunting a bug that occurred in network namespace
cleanup. In looking at the code it became apparrent that we have
and will continue to have cases where if we have anything going
on in a network namespace there will be assumptions that the
loopback device is present. Things like sending igmp unsubscribe
messages when we bring down network devices invokes the routing
code which assumes that at least the loopback driver is present.
Therefore to avoid magic initcall ordering hackery that is hard
to follow and hard to get right insert a call to register the
loopback device directly from net_dev_init(). This guarantes
that the loopback device is the first device registered and
the last network device to go away.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The generic packet receive code takes care of setting
netdev->last_rx when necessary, for the sake of the
bonding ARP monitor.
Drivers need not do it any more.
Some cases had to be skipped over because the drivers
were making use of the ->last_rx value themselves.
Signed-off-by: David S. Miller <davem@davemloft.net>
As noticed by Saikiran Madugula, commit 7447ef63cf
("loopback: Remove rest of LOOPBACK_TSO code.") got rid of
emulate_large_send_offload() but didn't get rid of the call
site as well.
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that the network stack can handle inbound packets with partial
checksums, we should no longer clobber the ip_summed field in the
loopback driver. This is because CHECKSUM_UNNECESSARY implies that
the checksum field is actually valid which is not true for loopback
packets since it's only partial (and thus complemented).
This allows packets from lo to then be SNATed to an external source
while still preserving the checksum's validity.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
It hasn't been enabled for a long time and the generic GSO
engine is better documentation of what is expected of a
device implementing TSO.
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch enables TSO since the loopback device is naturally
capable of handling packets of any size. This also means that
we won't enable GSO on lo which is good until GSO is fixed to
preserve netfilter state as netfilter treats loopback packets
in a special way.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The private area of a netdev is now at a fixed offset once more.
Unfortunately, some assumptions that netdev_priv() == netdev->priv
crept back into the tree. In particular this happened in the
loopback driver. Make it use netdev->ml_priv.
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce per-net_device inlines: dev_net(), dev_net_set().
Without CONFIG_NET_NS, no namespace other than &init_net exists.
Let's explicitly define them to help compiler optimizations.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
The LOOPBACK_OVERHEAD is not used in this file at all.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is probably a result of the changes from commit
854d836 - [NET]: Dynamically allocate the loopback device, part 2
Signed-off-by: Emil Medve <Emilian.Medve@Freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
This patch reverts Eric's commit 2b008b0a8e
It diets .text & .data section of the kernel if CONFIG_NET_NS is not set.
This is safe after list operations cleanup.
Signed-of-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is not safe to to place struct pernet_operations in a special section.
We need struct pernet_operations to last until we call unregister_pernet_subsys.
Which doesn't happen until module unload.
So marking struct pernet_operations is a disaster for modules in two ways.
- We discard it before we call the exit method it points to.
- Because I keep struct pernet_operations on a linked list discarding
it for compiled in code removes elements in the middle of a linked
list and does horrible things for linked insert.
So this looks safe assuming __exit_refok is not discarded
for modules.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the loopback device is failed to initialize inside the new
namespaces, panic() is called. Do not do it when the namespace
in question is not the init_net.
Plus cleanup the error path a bit.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
With the net namespaces many code leaved the __init section,
thus making the kernel occupy more memory than it did before.
Since we have a config option that prohibits the namespace
creation, the functions that initialize/finalize some netns
stuff are simply not needed and can be freed after the boot.
Currently, this is almost not noticeable, since few calls
are no longer in __init, but when the namespaces will be
merged it will be possible to free more code. I propose to
use the __net_init, __net_exit and __net_initdata "attributes"
for functions/variables that are not used if the CONFIG_NET_NS
is not set to save more space in memory.
The exiting functions cannot just reside in the __exit section,
as noticed by David, since the init section will have
references on it and the compilation will fail due to modpost
checks. These references can exist, since the init namespace
never dies and the exit callbacks are never called. So I
introduce the __exit_refok attribute just like it is already
done with the __init_refok.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
A hint as to why it is safe to use per cpu variables,
and note that we actually can have multiple instances
of the loopback device now.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Loopback device is special. It should be initialized at the very
beginning. Initialization order has been changed by
Eric W. Biederman <ebiederm@xmission.com> and this change is non-obvious
and important enough to add proper comment.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since hardware header operations are part of the protocol class
not the device instance, make them into a separate object and
save memory.
Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch makes loopback_dev per network namespace. Adding
code to create a different loopback device for each network
namespace and adding the code to free a loopback device
when a network namespace exits.
This patch modifies all users the loopback_dev so they
access it as init_net.loopback_dev, keeping all of the
code compiling and working. A later pass will be needed to
update the users to use something other than the initial network
namespace.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch add support for dynamically allocating the statistics counters
for the loopback device and adds appropriate device methods for allocating
and freeing the loopback device.
This completes support for creating multiple instances of the loopback
device, in preparation for creating per network namespace instances.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Doing this makes loopback.c a better example of how to do a
simple network device, and it removes the special case
single static allocation of a struct net_device, hopefully
making maintenance easier.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-By: Kirill Korotaev <dev@sw.ru>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
This patch replaces all occurences to the static variable
loopback_dev to a pointer loopback_dev. That provides the
mindless, trivial, uninteressting change part for the dynamic
allocation for the loopback.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Acked-By: Kirill Korotaev <dev@sw.ru>
Acked-by: Benjamin Thery <benjamin.thery@bull.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
For the operations
get-tx-csum
get-sg
get-tso
get-ufo
the default ethtool_op_xxx behavior is fine for all drivers, so we
permit op==NULL to imply the default behavior.
This provides a more uniform behavior across all drivers, eliminating
ethtool(8) "ioctl not supported" errors on older drivers that had
not been updated for the latest sub-ioctls.
The ethtool_op_xxx() functions are left exported, in case anyone
wishes to call them directly from a driver-private implementation --
a not-uncommon case. Should an ethtool_op_xxx() helper remain unused
for a while, except by net/core/ethtool.c, we can un-export it at a
later date.
[ Resolved conflicts with set/get value ethtool patch... -DaveM ]
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The core patchset of the network namespace sent by
Eric Biederman does not do dynamic loopback creation.
So there is no call to alloc_netdev_mq which fills the
network namespace field of the netdevice.
This patch assign the loopback to the init network namespace.
Signed-off-by: Daniel Lezcano <dlezcano@fr.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces NETIF_F_NETNS_LOCAL a flag to indicate
a network device is local to a single network namespace and
should never be moved. Useful for pseudo devices that we
need an instance in each network namespace (like the loopback
device) and for any device we find that cannot handle multiple
network namespaces so we may trap them in the initial network
namespace.
This patch introduces the function dev_change_net_namespace
a function used to move a network device from one network
namespace to another. To the network device nothing
special appears to happen, to the components of the network
stack it appears as if the network device was unregistered
in the network namespace it is in, and a new device
was registered in the network namespace the device
was moved to.
This patch sets up a namespace device destructor that
upon the exit of a network namespace moves all of the
movable network devices to the initial network namespace
so they are not lost.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Because IPv4 and IPv6 both depend on the presence of the loopback
device to function, failure in registration the loopback device should
be fatal.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rusty added a new 'stats' field to struct net_device.
loopback driver can use it instead of declaring another struct
net_device_stats This saves some memory.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To clearly state the intent of copying to linear sk_buffs, _offset being a
overly long variant but interesting for the sake of saving some bytes.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
For the places where we need a pointer to the network header, it is still legal
to touch skb->nh.raw directly if just adding to, subtracting from or setting it
to another layer header.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For the common, open coded 'skb->nh.raw = skb->data' operation, so that we can
later turn skb->nh.raw into a offset, reducing the size of struct sk_buff in
64bit land while possibly keeping it as a pointer on 32bit.
This one touches just the most simple case, next will handle the slightly more
"complex" cases.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For the cases where we want to set skb->mac.raw to an offset from skb->data.
Simple cases first, the memmove ones and specially pktgen will be left for later.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
One less thing for drivers writers to worry about.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch converts drivers/net/loopback.c to using module_init().
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
As BHs are off in loopback_xmit(), preemption cannot occurs, so we can
use __get_cpu_var() instead of per_cpu() (and avoid a
preempt_enable()/preempt_disable() pair)
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We dont need a full struct net_device_stats (currently 23 long : 184 bytes on
x86_64) per possible CPU, but only two counters : bytes and packets
We save few CPU cycles too in loopback_xmit() not updating 4 fields, but 2.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The loopback device status structure is a singleton and doesn't
need to be allocated. Add ethtool_ops hooks to show checksum always on,
and make ethtool_ops const.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the wrapper function skb_is_gso which can be used instead
of directly testing skb_shinfo(skb)->gso_size. This makes things a little
nicer and allows us to change the primary key for indicating whether an skb
is GSO (if we ever want to do that).
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Having separate fields in sk_buff for TSO/UFO (tso_size/ufo_size) is not
going to scale if we add any more segmentation methods (e.g., DCCP). So
let's merge them.
They were used to tell the protocol of a packet. This function has been
subsumed by the new gso_type field. This is essentially a set of netdev
feature bits (shifted by 16 bits) that are required to process a specific
skb. As such it's easy to tell whether a given device can process a GSO
skb: you just have to and the gso_type field and the netdev's features
field.
I've made gso_type a conjunction. The idea is that you have a base type
(e.g., SKB_GSO_TCPV4) that can be modified further to support new features.
For example, if we add a hardware TSO type that supports ECN, they would
declare NETIF_F_TSO | NETIF_F_TSO_ECN. All TSO packets with CWR set would
have a gso_type of SKB_GSO_TCPV4 | SKB_GSO_TCPV4_ECN while all other TSO
packets would be SKB_GSO_TCPV4. This means that only the CWR packets need
to be emulated in software.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we stop allocating percpu memory for not-possible CPUs we must not touch
the percpu data for not-possible CPUs at all. The correct way of doing this
is to test cpu_possible() or to use for_each_cpu().
This patch is a kernel-wide sweep of all instances of NR_CPUS. I found very
few instances of this bug, if any. But the patch converts lots of open-coded
test to use the preferred helper macros.
Cc: Mikael Starvik <starvik@axis.com>
Cc: David Howells <dhowells@redhat.com>
Acked-by: Kyle McMartin <kyle@parisc-linux.org>
Cc: Anton Blanchard <anton@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Christian Zankel <chris@zankel.net>
Cc: Philippe Elie <phil.el@wanadoo.fr>
Cc: Nathan Scott <nathans@sgi.com>
Cc: Jens Axboe <axboe@suse.de>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Eric Dumazet