Herbert Xu a écrit :
> On Tue, Nov 17, 2009 at 04:26:04AM -0800, David Miller wrote:
>> Really, the link watch stuff is just due for a redesign.  I don't
>> think a simple hack is going to cut it this time, sorry Eric :-)
>
> I have no objections against any redesigns, but since the only
> caller of linkwatch_forget_dev runs in process context with the
> RTNL, it could also legally emit those events.

Thanks guys, here an updated version then, before linkwatch surgery ?

In this version, I force the event to be sent synchronously.

[PATCH net-next-2.6] linkwatch: linkwatch_forget_dev() to speedup device dismantle

time ip link del eth3.103 ; time ip link del eth3.104 ; time ip link del eth3.105

real	0m0.266s
user	0m0.000s
sys	0m0.001s

real	0m0.770s
user	0m0.000s
sys	0m0.000s

real	0m1.022s
user	0m0.000s
sys	0m0.000s

One problem of current schem in vlan dismantle phase is the
holding of device done by following chain :

vlan_dev_stop() ->
	netif_carrier_off(dev) ->
		linkwatch_fire_event(dev) ->
			dev_hold() ...

And __linkwatch_run_queue() runs up to one second later...

A generic fix to this problem is to add a linkwatch_forget_dev() method
to unlink the device from the list of watched devices.

dev->link_watch_next becomes dev->link_watch_list (and use a bit more memory),
to be able to unlink device in O(1).

After patch :
time ip link del eth3.103 ; time ip link del eth3.104 ; time ip link del eth3.105

real    0m0.024s
user    0m0.000s
sys     0m0.000s

real    0m0.032s
user    0m0.000s
sys     0m0.001s

real    0m0.033s
user    0m0.000s
sys     0m0.000s
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some drivers ndo_get_stats() method need to perform txqueue stats folding.

Move folding from dev_get_stats() to a new dev_txq_stats_fold() function
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Recent changes in the TX error propagation require additional checking
and masking of values returned from hard_start_xmit(), mainly to
separate cases where skb was consumed. This aim can be simplified by
changing the order of NETDEV_TX and NET_XMIT codes, because the latter
are treated similarly to negative (ERRNO) values.

After this change much simpler dev_xmit_complete() is also used in
sch_direct_xmit(), so it is moved to netdevice.h.

Additionally NET_RX definitions in netdevice.h are moved up from
between TX codes to avoid confusion while reading the TX comment.
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

No longer need read_lock(&dev_base_lock), use RCU instead.
We also can avoid taking references on inet6_dev structs.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently the ->ndo_hard_start_xmit() callbacks are only permitted to return
one of the NETDEV_TX codes. This prevents any kind of error propagation for
virtual devices, like queue congestion of the underlying device in case of
layered devices, or unreachability in case of tunnels.

This patches changes the NET_XMIT codes to avoid clashes with the NETDEV_TX
codes and changes the two callers of dev_hard_start_xmit() to expect either
errno codes, NET_XMIT codes or NETDEV_TX codes as return value.

In case of qdisc_restart(), all non NETDEV_TX codes are mapped to NETDEV_TX_OK
since no error propagation is possible when using qdiscs. In case of
dev_queue_xmit(), the error is propagated upwards.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This adds an RCU macro for continuing search, useful for some
network devices like vlan.
Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
Acked-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This cleanup patch puts struct/union/enum opening braces,
in first line to ease grep games.

struct something
{

becomes :

struct something {
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adds RCU management to the list of netdevices.

Convert some for_each_netdev() users to RCU version, if
it can avoid read_lock-ing dev_base_lock

Ie:
	read_lock(&dev_base_loack);
	for_each_netdev(net, dev)
		some_action();
	read_unlock(&dev_base_lock);

becomes :

	rcu_read_lock();
	for_each_netdev_rcu(net, dev)
		some_action();
	rcu_read_unlock();
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some workloads hit dev_base_lock rwlock pretty hard.
We can use RCU lookups to avoid touching this rwlock
(and avoid touching netdevice refcount)

netdevices are already freed after a RCU grace period, so this patch
adds no penalty at device dismantle time.

However, it adds a synchronize_rcu() call in dev_change_name()
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This isn't beautifully abstracted, but it is simple,
simplifies uses and so far is only needed for the bonding driver.
Signed-off-by: NEric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This will allow drivers to adjust their receive path dynamically
based on whether GRO is being applied successfully.

Currently all in-tree callers ignore the return values of these
functions and do not need to be changed.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Acked-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This clarifies which return and parameter types are GRO result codes
and not RX result codes.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Acked-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some workloads hit dev_base_lock rwlock pretty hard.
We can use RCU lookups to avoid touching this rwlock.

netdevices are already freed after a RCU grace period, so this patch
adds no penalty at device dismantle time.

dev_ifname() converted to dev_get_by_index_rcu()
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add ndo_fcoe_get_wwn so Fiber Channel over Ethernet (FCoE) can make use of
the provided World Wide Port Name (WWPN) and World Wide Node Name (WWNN)
from the underlying network interface driver.
Signed-off-by: NYi Zou <yi.zou@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Introduce rollback_registered_many() and unregister_netdevice_many()

rollback_registered_many() is able to perform necessary steps at device dismantle
time, factorizing two expensive synchronize_net() calls.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patchs adds an unreg_list anchor to struct net_device, and
introduces an unregister_netdevice_queue() function, able to queue
a net_device to a list instead of immediately unregister it.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

nanodoc was missing an ndo_-prefix.
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The data center bridging ops structure can be const
Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
Acked-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Let attribute group vectors be declared "const".  We'd
like to let most attribute metadata live in read-only
sections... this is a start.
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This patch fixes commit e36b9d16. The approach
there is to call dev_close()/dev_open() whenever the device type is changed in
order to remap the device IP multicast addresses to HW multicast addresses.
This approach suffers from 2 drawbacks:

*. It assumes tha the device is UP when calling dev_close(), or otherwise
   dev_close() has no affect. It is worth to mention that initscripts (Redhat)
   and sysconfig (Suse) doesn't act the same in this matter. 
*. dev_close() has other side affects, like deleting entries from the routing
   table, which might be unnecessary.

The fix here is to directly remap the IP multicast addresses to HW multicast
addresses for a bonding device that changes its type, and nothing else.
Reported-by: NJason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: NMoni Shoua <monis@voltaire.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The Ethernet framing is used for a lot of devices these days. Most
prominent are WiFi and WiMAX based devices. However for userspace
application it is important to classify these devices correctly and
not only see them as Ethernet devices. The daemons like HAL, DeviceKit
or even NetworkManager with udev support tries to do the classification
in userspace with a lot trickery and extra system calls. This is not
good and actually reaches its limitations. Especially since the kernel
does know the type of the Ethernet device it is pretty stupid.

To solve this problem the underlying device type needs to be set and
then the value will be exported as DEVTYPE via uevents and available
within udev.

  # cat /sys/class/net/wlan0/uevent
  DEVTYPE=wlan
  INTERFACE=wlan0
  IFINDEX=5

This is similar to subsystems like USB and SCSI that distinguish
between hosts, devices, disks, partitions etc.

The new SET_NETDEV_DEVTYPE() is a convenience helper to set the actual
device type. All device types are free form, but for convenience the
same strings as used with RFKILL are choosen.
Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently the multiqueue integration with the qdisc API suffers from
a few problems:

- with multiple queues, all root qdiscs use the same handle. This means
  they can't be exposed to userspace in a backwards compatible fashion.

- all API operations always refer to queue number 0. Newly created
  qdiscs are automatically shared between all queues, its not possible
  to address individual queues or restore multiqueue behaviour once a
  shared qdisc has been attached.

- Dumps only contain the root qdisc of queue 0, in case of non-shared
  qdiscs this means the statistics are incomplete.

This patch reintroduces dev->qdisc, which points to the (single) root qdisc
from userspace's point of view. Currently it either points to the first
(non-shared) default qdisc, or a qdisc shared between all queues. The
following patches will introduce a classful dummy qdisc, which will be used
as root qdisc and contain the per-queue qdiscs as children.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add ndo_fcoe_enable/_disable to net_device_ops so the corresponding
HW can initialize itself for FCoE traffic or clean up after FCoE traffic is
done. This is expected to be called by the kernel FCoE stack upon receiving
a request for creating an FCoE instance on the corresponding netdev interface.
When implemented by the actual HW, the HW driver check the op code to perform
corresponding initialization or clean up for FCoE. The initialization normally
includes allocating extra queues for FCoE, setting corresponding HW registers
for FCoE, indicating FCoE offload features via netdev, etc. The clean-up would
include releasing the resources allocated for FCoE.
Signed-off-by: NYi Zou <yi.zou@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The transmit function should only return one of three possible values,
some drivers got confused and returned errno's or other values.
This changes the definition so that this can be caught at compile time.
Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Patch compiled and 32 simultaneous netperf testing ran fine.
Signed-off-by: NKrishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add NETIF_F_FCOE_MTU to indicate that the NIC can support a secondary MTU for
converged traffic of LAN and Fiber Channel over Ethernet (FCoE). The MTU for
FCoE is 2158 = 14 (FCoE header) + 24 (FC header) + 2112 (FC max payload) +
4 (FC CRC) + 4 (FCoE trailer).
Signed-off-by: NYi Zou <yi.zou@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

almost no users in the tree; and the few that use them treat them
like NET_RX_DROP.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch is inspired by patch recently posted by Johannes Berg. Basically what
my patch does is to group list and a count of addresses into newly introduced
structure netdev_hw_addr_list. This brings us two benefits:
1) struct net_device becames a bit nicer.
2) in the future there will be a possibility to operate with lists independently
   on netdevices (with exporting right functions).
I wanted to introduce this patch before I'll post a multicast lists conversion.
Signed-off-by: NJiri Pirko <jpirko@redhat.com>

 drivers/net/bnx2.c              |    4 +-
 drivers/net/e1000/e1000_main.c  |    4 +-
 drivers/net/ixgbe/ixgbe_main.c  |    6 +-
 drivers/net/mv643xx_eth.c       |    2 +-
 drivers/net/niu.c               |    4 +-
 drivers/net/virtio_net.c        |   10 ++--
 drivers/s390/net/qeth_l2_main.c |    2 +-
 include/linux/netdevice.h       |   17 +++--
 net/core/dev.c                  |  130 ++++++++++++++++++--------------------
 9 files changed, 89 insertions(+), 90 deletions(-)
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As it stands skb fraglists can get past the check in dev_queue_xmit
if the skb is marked as GSO.  In particular, if the packet doesn't
have the proper checksums for GSO, but can otherwise be handled by
the underlying device, we will not perform the fraglist check on it
at all.

If the underlying device cannot handle fraglists, then this will
break.

The fix is as simple as moving the fraglist check from the device
check into skb_gso_ok.

This has caused crashes with Xen when used together with GRO which
can generate GSO packets with fraglists.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch converts unicast address list to standard list_head using
previously introduced struct netdev_hw_addr. It also relaxes the
locking. Original spinlock (still used for multicast addresses) is not
needed and is no longer used for a protection of this list. All
reading and writing takes place under rtnl (with no changes).

I also removed a possibility to specify the length of the address
while adding or deleting unicast address. It's always dev->addr_len.

The convertion touched especially e1000 and ixgbe codes when the
change is not so trivial.
Signed-off-by: NJiri Pirko <jpirko@redhat.com>

 drivers/net/bnx2.c               |   13 +--
 drivers/net/e1000/e1000_main.c   |   24 +++--
 drivers/net/ixgbe/ixgbe_common.c |   14 ++--
 drivers/net/ixgbe/ixgbe_common.h |    4 +-
 drivers/net/ixgbe/ixgbe_main.c   |    6 +-
 drivers/net/ixgbe/ixgbe_type.h   |    4 +-
 drivers/net/macvlan.c            |   11 +-
 drivers/net/mv643xx_eth.c        |   11 +-
 drivers/net/niu.c                |    7 +-
 drivers/net/virtio_net.c         |    7 +-
 drivers/s390/net/qeth_l2_main.c  |    6 +-
 drivers/scsi/fcoe/fcoe.c         |   16 ++--
 include/linux/netdevice.h        |   18 ++--
 net/8021q/vlan.c                 |    4 +-
 net/8021q/vlan_dev.c             |   10 +-
 net/core/dev.c                   |  195 +++++++++++++++++++++++++++-----------
 net/dsa/slave.c                  |   10 +-
 net/packet/af_packet.c           |    4 +-
 18 files changed, 227 insertions(+), 137 deletions(-)
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

[PATCH net-next] bonding: allow bond in mode balance-alb to work properly in bridge -try4.3

(updated)
changes v4.2 -> v4.3
- memcpy the address always, not just in case it differs from master->dev_addr
- compare_ether_addr_64bits() is not used so there is no direct need to make new
  header file (I think it would be good to have bond stuff in separate file
  anyway).

changes v4.1 -> v4.2
- use skb->pkt_type == PACKET_HOST compare rather then comparing skb dest addr
  against skb->dev->dev_addr

The problem is described in following bugzilla:
https://bugzilla.redhat.com/show_bug.cgi?id=487763

Basically here's what's going on. In every mode, bonding interface uses the same
mac address for all enslaved devices (except fail_over_mac). Only balance-alb
will simultaneously use multiple MAC addresses across different slaves. When you
put this kind of bond device into a bridge it will only add one of mac adresses
into a hash list of mac addresses, say X. This mac address is marked as local.
But this bonding interface also has mac address Y. Now then packet arrives with
destination address Y, this address is not marked as local and the packed looks
like it needs to be forwarded. This packet is then lost which is wrong.

Notice that interfaces can be added and removed from bond while it is in bridge.

***
When the multiple addresses for bridge port approach failed to solve this issue
due to STP I started to think other way to solve this. I returned to previous
solution but tweaked one.

This patch solves the situation in the bonding without touching bridge code.
For every incoming frame to bonding the destination address is compared to
current address of the slave device from which tha packet came. If these two
match destination address is replaced by mac address of the master. This address
is known by bridge so it is delivered properly. Note that the comparsion is not
made directly, it's used skb->pkt_type == PACKET_HOST instead. This is "set"
previously in eth_type_trans().

I experimentally tried that this works as good as searching through the slave
list (v4 of this patch).

Jirka
Signed-off-by: NJiri Pirko <jpirko@redhat.com>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NAndy Gospodarek <andy@greyhouse.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NJiri Pirko <jpirko@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use ALIGN() and PTR_ALIGN() macros instead of handcoding them.

Get rid of NETDEV_ALIGN_CONST ugly define
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

For the overwhelming majority of cases, skb_gro_header's return
value cannot be NULL.  Yet we must check it because of its current
form.  This patch splits it up into multiple functions in order
to avoid this.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

By caching frag0_len, we can avoid checking both frag0 and the
length separately in skb_gro_header.  This helps as skb_gro_header
is called four times per packet which amounts to a few million
times at 10Gb/s.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently skb_gro_header is used for packets which put the hardware
header in skb->data with the rest in frags.  Since the drivers that
need this optimisation all provide completely non-linear packets,
we can gain extra optimisations by only performing the frag0
optimisation for completely non-linear packets.

In particular, we can simply test frag0 (instead of skb_headlen)
to see whether the optimisation is in force.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The function skb_gro_header is called four times per packet which
quickly adds up at 10Gb/s.  This patch inlines it to allow better
optimisations.

Some architectures perform multiplication for page_address, which
is done by each skb_gro_header invocation.  This patch caches that
value in skb->cb to avoid the unnecessary multiplications.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We would like to get rid of netdev->trans_start = jiffies; that about all net
drivers have to use in their start_xmit() function, and use txq->trans_start
instead.

This can be done generically in core network, as suggested by David.

Some devices, (particularly loopback) dont need trans_start update, because
they dont have transmit watchdog. We could add a new device flag, or rely
on fact that txq->tran_start can be updated is txq->xmit_lock_owner is
different than -1. Use a helper function to hide our choice.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

All drivers are already converted to new net_device_ops API
and nobody uses old API anymore.
Signed-off-by: NAlexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>