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>
The earlier patch to fix the deadlock between a network device going
away and writing to sysfs attributes was incomplete.
- It did not set signal_pending so we would leak ERSTARTSYS to user space.
- It used ERESTARTSYS which only restarts if sigaction configures it to.
- It did not cover store and show for ifalias.
So fix all of these up and use the new helper restart_syscall so we get
the details correct on what it takes.
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gen_estimator can overflow bps (bytes per second) with Gb links, while
it was designed with a u32 API, with a theorical limit of 34360Mbit
(2^32 bytes)
Using 64 bit intermediate avbps/brate counters can allow us to reach
this theorical limit.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
offsetof(struct net_device, features)=0x44
offsetof(struct net_device, stats.tx_packets)=0x54
offsetof(struct net_device, stats.tx_bytes)=0x5c
offsetof(struct net_device, stats.tx_dropped)=0x6c
Network drivers that touch dev->stats.tx_packets/stats.tx_bytes in their
tx path can slow down SMP operations, since they dirty a cache line
that should stay shared (dev->features is needed in rx and tx paths)
We could move away stats field in net_device but it wont help that much.
(Two cache lines dirtied in tx path, we can do one only)
Better solution is to add tx_packets/tx_bytes/tx_dropped in struct
netdev_queue because this structure is already touched in tx path and
counters updates will then be free (no increase in size)
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When TCP frees up write buffer space, avoid waking up tasks that have
done a poll() or select() on the same socket specifying read-side
events.
This is an extension of a read-side patch by Eric Dumazet.
Signed-off-by: John Dykstra <john.dykstra1@gmail.com>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It looks like the dev in netpoll_poll can be NULL - at lease it's
checked at the function beginning. Thus the dev->netde_ops dereference
looks dangerous.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add missed checking of dev_addr_init return value in alloc_netdev_mq.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
net/core/dev.c | 15 ++++++++++++---
1 files changed, 12 insertions(+), 3 deletions(-)
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit ead2ceb0ec ("Network Drop Monitor:
Adding kfree_skb_clean for non-drops and modifying end-of-line points
for skbs") established new conventions for identifying dropped packets.
Align skb_kill_datagram() with these conventions so that packets that
get dropped just before the copy to userspace are properly tracked.
Signed-off-by: John Dykstra <john.dykstra1@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit ac45f602ee ("net: infrastructure
for hardware time stamping") added two skb initialization actions to
__alloc_skb(), which need to be added to skb_recycle_check() as well.
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
v5 -> v6 (current):
-removed so far unused static functions
-corrected dev_addr_del_multiple to call del instead of add
v4 -> v5:
-added device address type (suggested by davem)
-removed refcounting (better to have simplier code then safe potentially few
bytes)
v3 -> v4:
-changed kzalloc to kmalloc in __hw_addr_add_ii()
-ASSERT_RTNL() avoided in dev_addr_flush() and dev_addr_init()
v2 -> v3:
-removed unnecessary rcu read locking
-moved dev_addr_flush() calling to ensure no null dereference of dev_addr
v1 -> v2:
-added forgotten ASSERT_RTNL to dev_addr_init and dev_addr_flush
-removed unnecessary rcu_read locking in dev_addr_init
-use compare_ether_addr_64bits instead of compare_ether_addr
-use L1_CACHE_BYTES as size for allocating struct netdev_hw_addr
-use call_rcu instead of rcu_synchronize
-moved is_etherdev_addr into __KERNEL__ ifdef
This patch introduces a new list in struct net_device and brings a set of
functions to handle the work with device address list. The list is a replacement
for the original dev_addr field and because in some situations there is need to
carry several device addresses with the net device. To be backward compatible,
dev_addr is made to point to the first member of the list so original drivers
sees no difference.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net_create() will be used by C/R to create fresh netns on restart.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
copy_net_ns() doesn't copy anything, it creates fresh netns, so
get/put of old netns isn't needed.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Based almost entirely upon a patch by Eric Dumazet.
The common case is to have num-tx-queues <= num_rx_queues
and even if num_tx_queues is larger it will not be significantly
larger.
Therefore, a subtraction loop is always going to be faster than
modulus.
Signed-off-by: David S. Miller <davem@davemloft.net>
When skb_rx_queue_recorded() is true, we dont want to use jash distribution
as the device driver exactly told us which queue was selected at RX time.
jhash makes a statistical shuffle, but this wont work with 8 static inputs.
Later improvements would be to compute reciprocal value of real_num_tx_queues
to avoid a divide here. But this computation should be done once,
when real_num_tx_queues is set. This needs a separate patch, and a new
field in struct net_device.
Reported-by: Andrew Dickinson <andrew@whydna.net>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Lennert Buytenhek wrote:
> Since 4fb6699481 ("net: Optimize memory
> usage when splicing from sockets.") I'm seeing this oops (e.g. in
> 2.6.30-rc3) when splicing from a TCP socket to /dev/null on a driver
> (mv643xx_eth) that uses LRO in the skb mode (lro_receive_skb) rather
> than the frag mode:
My patch incorrectly assumed skb->sk was always valid, but for
"frag_listed" skbs we can only use skb->sk of their parent.
Reported-by: Lennert Buytenhek <buytenh@wantstofly.org>
Debugged-by: Lennert Buytenhek <buytenh@wantstofly.org>
Tested-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In 2.6.25 we added UDP mem accounting.
This unfortunatly added a penalty when a frame is transmitted, since
we have at TX completion time to call sock_wfree() to perform necessary
memory accounting. This calls sock_def_write_space() and utimately
scheduler if any thread is waiting on the socket.
Thread(s) waiting for an incoming frame was scheduled, then had to sleep
again as event was meaningless.
(All threads waiting on a socket are using same sk_sleep anchor)
This adds lot of extra wakeups and increases latencies, as noted
by Christoph Lameter, and slows down softirq handler.
Reference : http://marc.info/?l=linux-netdev&m=124060437012283&w=2
Fortunatly, Davide Libenzi recently added concept of keyed wakeups
into kernel, and particularly for sockets (see commit
37e5540b3c
epoll keyed wakeups: make sockets use keyed wakeups)
Davide goal was to optimize epoll, but this new wakeup infrastructure
can help non epoll users as well, if they care to setup an appropriate
handler.
This patch introduces new DEFINE_WAIT_FUNC() helper and uses it
in wait_for_packet(), so that only relevant event can wakeup a thread
blocked in this function.
Trace of function calls from bnx2 TX completion bnx2_poll_work() is :
__kfree_skb()
skb_release_head_state()
sock_wfree()
sock_def_write_space()
__wake_up_sync_key()
__wake_up_common()
receiver_wake_function() : Stops here since thread is waiting for an INPUT
Reported-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The skb_gro_* code fails to handle the case where a header starts
in the linear area but ends in the frags area. Since the goal
of skb_gro_* is to optimise the case of completely non-linear
packets, we can simply bail out if we have anything in the linear
area.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
When I initially implemented this protocol, I disregarded the use of netlink
attribute headers, thinking for my purposes they weren't needed. I've come to
find out that, as I'm starting to work with sending down messages with
associated data (like config messages), the kernel code spits out warnings about
trailing data in a netlink skb that doesn't have an associated header on it. As
such, I'm going to start including attribute headers in my netlink transaction,
and so for completeness, I should likely include them on messages bound from the
kernel to user space. This patch adds that header to the kernel, and bumps the
protocol version accordingly
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
aio_write gets const struct iovec * but tun_chr_aio_write casts this to struct
iovec * and modifies the iovec. As a result, attempts to use io_submit
to send packets to a tun device fail with weird errors such as EINVAL.
Since tun is the only user of skb_copy_datagram_from_iovec, we can
fix this simply by changing the later so that it does not
touch the iovec passed to it.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There's an skb_copy_datagram_iovec() to copy out of a paged skb,
but it modifies the iovec, and does not support starting
at an offset in the destination. We want both in tun.c, so let's
add the function.
It's a carbon copy of skb_copy_datagram_iovec() with enough changes to
be annoying.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This loop over fragments in napi_fraginfo_skb() was "interesting".
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Sidorenko reported:
"while experimenting with 'netem' we have found some strange behaviour. It
seemed that ingress delay as measured by 'ping' command shows up on some
hosts but not on others.
After some investigation I have found that the problem is that skbuff->tstamp
field value depends on whether there are any packet sniffers enabled. That
is:
- if any ptype_all handler is registered, the tstamp field is as expected
- if there are no ptype_all handlers, the tstamp field does not show the delay"
This patch prevents unnecessary update of tstamp in dev_queue_xmit_nit()
on ingress path (with act_mirred) adding a check, so minimal overhead on
the fast path, but only when sniffers etc. are active.
Since netem at ingress seems to logically emulate a network before a host,
tstamp is zeroed to trigger the update and pretend delays are from the
outside.
Reported-by: Alex Sidorenko <alexandre.sidorenko@hp.com>
Tested-by: Alex Sidorenko <alexandre.sidorenko@hp.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It turns out that copying a 16-byte area at ~800k times a second
can be really expensive :) This patch redesigns the frags GRO
interface to avoid copying that area twice.
The two disciples of the frags interface have been converted.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since everybody has been focusing on baremetal GRO performance
no one noticed when I added a bug that zapped gso_size for all
GRO packets. This only gets picked up when you forward the skb
out of an interface.
Thanks to Mark Wagner for noticing this bug when testing kvm.
Reported-by: Mark Wagner <mwagner@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently netif_device_attach/detach are only stopping one queue. They
should be starting and stopping all the queues on a given device.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (54 commits)
glge: remove unused #include <version.h>
dnet: remove unused #include <version.h>
tcp: miscounts due to tcp_fragment pcount reset
tcp: add helper for counter tweaking due mid-wq change
hso: fix for the 'invalid frame length' messages
hso: fix for crash when unplugging the device
fsl_pq_mdio: Fix compile failure
fsl_pq_mdio: Revive UCC MDIO support
ucc_geth: Pass proper device to DMA routines, otherwise oops happens
i.MX31: Fixing cs89x0 network building to i.MX31ADS
tc35815: Fix build error if NAPI enabled
hso: add Vendor/Product ID's for new devices
ucc_geth: Remove unused header
gianfar: Remove unused header
kaweth: Fix locking to be SMP-safe
net: allow multiple dev per napi with GRO
r8169: reset IntrStatus after chip reset
ixgbe: Fix potential memory leak/driver panic issue while setting up Tx & Rx ring parameters
ixgbe: fix ethtool -A|a behavior
ixgbe: Patch to fix driver panic while freeing up tx & rx resources
...
GRO assumes that there is a one-to-one relationship between NAPI
structure and network device. Some devices like sky2 share multiple
devices on a single interrupt so only have one NAPI handler. Rather than
split GRO from NAPI, just have GRO assume if device changes that
it is a different flow.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for event-aware wakeups to the sockets code. Events are
delivered to the wakeup target, so that epoll can avoid spurious wakeups
for non-interesting events.
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Acked-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: William Lee Irwin III <wli@movementarian.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
wireless: remove duplicated .ndo_set_mac_address
netfilter: xtables: fix IPv6 dependency in the cluster match
tg3: Add GRO support.
niu: Add GRO support.
ucc_geth: Fix use-after-of_node_put() in ucc_geth_probe().
gianfar: Fix use-after-of_node_put() in gfar_of_init().
kernel: remove HIPQUAD()
netpoll: store local and remote ip in net-endian
netfilter: fix endian bug in conntrack printks
dmascc: fix incomplete conversion to network_device_ops
gso: Fix support for linear packets
skbuff.h: fix missing kernel-doc
ni5010: convert to net_device_ops
Setting ->owner as done currently (pde->owner = THIS_MODULE) is racy
as correctly noted at bug #12454. Someone can lookup entry with NULL
->owner, thus not pinning enything, and release it later resulting
in module refcount underflow.
We can keep ->owner and supply it at registration time like ->proc_fops
and ->data.
But this leaves ->owner as easy-manipulative field (just one C assignment)
and somebody will forget to unpin previous/pin current module when
switching ->owner. ->proc_fops is declared as "const" which should give
some thoughts.
->read_proc/->write_proc were just fixed to not require ->owner for
protection.
rmmod'ed directories will be empty and return "." and ".." -- no harm.
And directories with tricky enough readdir and lookup shouldn't be modular.
We definitely don't want such modular code.
Removing ->owner will also make PDE smaller.
So, let's nuke it.
Kudos to Jeff Layton for reminding about this, let's say, oversight.
http://bugzilla.kernel.org/show_bug.cgi?id=12454
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Allows for the removal of byteswapping in some places and
the removal of HIPQUAD (replaced by %pI4).
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When GRO/frag_list support was added to GSO, I made an error
which broke the support for segmenting linear GSO packets (GSO
packets are normally non-linear in the payload).
These days most of these packets are constructed by the tun
driver, which prefers to allocate linear memory if possible.
This is fixed in the latest kernel, but for 2.6.29 and earlier
it is still the norm.
Therefore this bug causes failures with GSO when used with tun
in 2.6.29.
Reported-by: James Huang <jamesclhuang@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
When I fixed the GRO crash in the legacy receive path I used
napi_complete to replace __napi_complete. Unfortunately they're
not the same when NETPOLL is enabled, which may result in us
not calling __napi_complete at all.
What's more, we really do need to keep the __napi_complete call
within the IRQ-off section since in theory an IRQ can occur in
between and fill up the backlog to the maximum, causing us to
lock up.
Since we can't seem to find a fix that works properly right now,
this patch reverts all the GRO support from the netif_rx path.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no gain using prefetch() in dev_hard_start_xmit(), since
we already had to read ops->ndo_select_queue pointer in dev_pick_tx(),
and both pointers are probably located in the same cache line.
This prefetch call slows down fast path because of a stall in address
computation.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some minor changes to queue hashing:
1. Use const on accessor functions
2. Export skb_tx_hash for use in drivers (see ixgbe)
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct sk_buff pointers should be freed with kfree_skb.
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet