linux-stable.git/drivers/net/tun.c, branch linux-4.8.y

tun: fix transmit timestamp support

2016-08-24T06:09:27+00:00

Instead of using sock_tx_timestamp, use skb_tx_timestamp to record
software transmit timestamp of a packet.

sock_tx_timestamp resets and overrides the tx_flags of the skb.
The function is intended to be called from within the protocol
layer when creating the skb, not from a device driver. This is
inconsistent with other drivers and will cause issues for TCP.

In TCP, we intend to sample the timestamps for the last byte
for each sendmsg/sendpage. For that reason, tcp_sendmsg calls
tcp_tx_timestamp only with the last skb that it generates.
For example, if a 128KB message is split into two 64KB packets
we want to sample the SND timestamp of the last packet. The current
code in the tun driver, however, will result in sampling the SND
timestamp for both packets.

Also, when the last packet is split into smaller packets for
retranmission (see tcp_fragment), the tun driver will record
timestamps for all of the retransmitted packets and not only the
last packet.

Fixes: eda297729171 (tun: Support software transmit time stamping.)
Signed-off-by: Soheil Hassas Yeganeh 
Signed-off-by: Francis Yan 
Acked-by: Eric Dumazet 
Signed-off-by: David S. Miller

tun: Don't assume type tun in tun_device_event

2016-07-09T03:58:57+00:00

The referenced change added a netlink notifier for processing
device queue size events.  These events are fired for all devices
but the registered callback assumed they only occurred for tun
devices.  This fix adds a check (borrowed from macvtap.c) to discard
non-tun device events.

For reference, this fixes the following splat:
[   71.505935] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010
[   71.513870] IP: [] tun_device_event+0x110/0x340
[   71.519906] PGD 3f41f56067 PUD 3f264b7067 PMD 0
[   71.524497] Oops: 0002 [#1] SMP DEBUG_PAGEALLOC
[   71.529374] gsmi: Log Shutdown Reason 0x03
[   71.533417] Modules linked in:[   71.533826] mlx4_en: eth1: Link Up

[   71.539616]  bonding w1_therm wire cdc_acm ehci_pci ehci_hcd mlx4_en ib_uverbs mlx4_ib ib_core mlx4_core
[   71.549282] CPU: 12 PID: 7915 Comm: set.ixion-haswe Not tainted 4.7.0-dbx-DEV #8
[   71.556586] Hardware name: Intel Grantley,Wellsburg/Ixion_IT_15, BIOS 2.58.0 05/03/2016
[   71.564495] task: ffff887f00bb20c0 ti: ffff887f00798000 task.ti: ffff887f00798000
[   71.571894] RIP: 0010:[]  [] tun_device_event+0x110/0x340
[   71.580327] RSP: 0018:ffff887f0079bbd8  EFLAGS: 00010202
[   71.585576] RAX: fffffffffffffae8 RBX: ffff887ef6d03378 RCX: 0000000000000000
[   71.592624] RDX: 0000000000000000 RSI: 0000000000000028 RDI: 0000000000000000
[   71.599675] RBP: ffff887f0079bc48 R08: 0000000000000000 R09: 0000000000000001
[   71.606730] R10: 0000000000000004 R11: 0000000000000000 R12: 0000000000000010
[   71.613780] R13: 0000000000000000 R14: 0000000000000001 R15: ffff887f0079bd00
[   71.620832] FS:  00007f5cdc581700(0000) GS:ffff883f7f700000(0000) knlGS:0000000000000000
[   71.628826] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   71.634500] CR2: 0000000000000010 CR3: 0000003f3eb62000 CR4: 00000000001406e0
[   71.641549] Stack:
[   71.643533]  ffff887f0079bc08 0000000000000246 000000000000001e ffff887ef6d00000
[   71.650871]  ffff887f0079bd00 0000000000000000 0000000000000000 ffffffff00000000
[   71.658210]  ffff887f0079bc48 ffffffff81d24070 00000000fffffff9 ffffffff81cec7a0
[   71.665549] Call Trace:
[   71.667975]  [] notifier_call_chain+0x5d/0x80
[   71.673823]  [] ? show_tx_maxrate+0x30/0x30
[   71.679502]  [] __raw_notifier_call_chain+0xe/0x10
[   71.685778]  [] raw_notifier_call_chain+0x16/0x20
[   71.691976]  [] call_netdevice_notifiers_info+0x40/0x70
[   71.698681]  [] call_netdevice_notifiers+0x16/0x20
[   71.704956]  [] change_tx_queue_len+0x66/0x90
[   71.710807]  [] netdev_store.isra.5+0xbf/0xd0
[   71.716658]  [] tx_queue_len_store+0x50/0x60
[   71.722431]  [] dev_attr_store+0x18/0x30
[   71.727857]  [] sysfs_kf_write+0x4f/0x70
[   71.733274]  [] kernfs_fop_write+0x147/0x1d0
[   71.739045]  [] ? rcu_read_lock_sched_held+0x8f/0xa0
[   71.745499]  [] __vfs_write+0x28/0x120
[   71.750748]  [] ? percpu_down_read+0x57/0x90
[   71.756516]  [] ? __sb_start_write+0xc8/0xe0
[   71.762278]  [] ? __sb_start_write+0xc8/0xe0
[   71.768038]  [] vfs_write+0xbe/0x1b0
[   71.773113]  [] SyS_write+0x52/0xa0
[   71.778110]  [] entry_SYSCALL_64_fastpath+0x18/0xa8
[   71.784472] Code: 45 31 f6 48 8b 93 78 33 00 00 48 81 c3 78 33 00 00 48 39 d3 48 8d 82 e8 fa ff ff 74 25 48 8d b0 40 05 00 00 49 63 d6 41 83 c6 01 <49> 89 34 d4 48 8b 90 18 05 00 00 48 39 d3 48 8d 82 e8 fa ff ff
[   71.803655] RIP  [] tun_device_event+0x110/0x340
[   71.809769]  RSP 
[   71.813213] CR2: 0000000000000010
[   71.816512] ---[ end trace 4db6449606319f73 ]---

Fixes: 1576d9860599 ("tun: switch to use skb array for tx")
Signed-off-by: Craig Gallek 
Acked-by: Jason Wang 
Signed-off-by: David S. Miller

tun: fix build warnings

2016-07-05T00:19:27+00:00

Stephen Rothwell reports a build warnings(powerpc ppc64_defconfig)

drivers/net/tun.c: In function 'tun_do_read.part.5':
/home/sfr/next/next/drivers/net/tun.c:1491:6: warning: 'err' may be
used uninitialized in this function [-Wmaybe-uninitialized]
   int err;

This is because tun_ring_recv() may return an uninitialized err, fix this.

Reported-by: Stephen Rothwell 
Signed-off-by: Jason Wang 
Signed-off-by: David S. Miller

tun: switch to use skb array for tx

2016-07-01T09:32:17+00:00

We used to queue tx packets in sk_receive_queue, this is less
efficient since it requires spinlocks to synchronize between producer
and consumer.

This patch tries to address this by:

- switch from sk_receive_queue to a skb_array, and resize it when
  tx_queue_len was changed.
- introduce a new proto_ops peek_len which was used for peeking the
  skb length.
- implement a tun version of peek_len for vhost_net to use and convert
  vhost_net to use peek_len if possible.

Pktgen test shows about 15.3% improvement on guest receiving pps for small
buffers:

Before: ~1300000pps
After : ~1500000pps

Signed-off-by: Jason Wang 
Signed-off-by: David S. Miller

tun: fix csum generation for tap devices

2016-06-15T21:00:33+00:00

The commit 34166093639b ("tuntap: use common code for virtio_net_hdr
and skb GSO conversion") replaced the tun code for header manipulation
with the generic helpers. While doing so, it implictly moved the
skb_partial_csum_set() invocation after eth_type_trans(), which
invalidate the current gso start/offset values.
Fix it by moving the helper invocation before the mac pulling.

Fixes: 34166093639 ("tuntap: use common code for virtio_net_hdr and skb GSO conversion")
Signed-off-by: Paolo Abeni 
Acked-by: Mike Rapoport 
Signed-off-by: David S. Miller

tuntap: use common code for virtio_net_hdr and skb GSO conversion

2016-06-11T06:03:55+00:00

Replace open coded conversion between virtio_net_hdr to skb GSO info with
virtio_net_hdr_{from,to}_skb

Signed-off-by: Mike Rapoport 
Signed-off-by: David S. Miller

tuntap: correctly wake up process during uninit

2016-05-20T23:28:37+00:00

We used to check dev->reg_state against NETREG_REGISTERED after each
time we are woke up. But after commit 9e641bdcfa4e ("net-tun:
restructure tun_do_read for better sleep/wakeup efficiency"), it uses
skb_recv_datagram() which does not check dev->reg_state. This will
result if we delete a tun/tap device after a process is blocked in the
reading. The device will wait for the reference count which was held
by that process for ever.

Fixes this by using RCV_SHUTDOWN which will be checked during
sk_recv_datagram() before trying to wake up the process during uninit.

Fixes: 9e641bdcfa4e ("net-tun: restructure tun_do_read for better
sleep/wakeup efficiency")
Cc: Eric Dumazet 
Cc: Xi Wang 
Cc: Michael S. Tsirkin 
Signed-off-by: Jason Wang 
Acked-by: Eric Dumazet 
Acked-by: Michael S. Tsirkin 
Signed-off-by: David S. Miller

tuntap: calculate rps hash only when needed

2016-04-28T20:38:54+00:00

There's no need to calculate rps hash if it was not enabled. So this
patch export rps_needed and check it before trying to get rps
hash. Tests (using pktgen to inject packets to guest) shows this can
improve pps about 13% (when rps is disabled).

Before:
~1150000 pps
After:
~1300000 pps

Cc: Michael S. Tsirkin 
Signed-off-by: Jason Wang 
----
Changes from V1:
- Fix build when CONFIG_RPS is not set
Signed-off-by: David S. Miller

tun: don't require serialization lock on tx

2016-04-18T18:36:26+00:00

The current tun_net_xmit() implementation don't need any external
lock since it relies on rcu protection for the tun data structure
and on socket queue lock for skb queuing.

This patch set the NETIF_F_LLTX feature bit in the tun device, so
that on xmit, in absence of qdisc, no serialization lock is acquired
by the caller.

The user space can remove the default tun qdisc with:

tc qdisc replace dev  root noqueue

Signed-off-by: Paolo Abeni 
Acked-by: Hannes Frederic Sowa 
Acked-by: Eric Dumazet 
Acked-by: Michael S. Tsirkin 
Signed-off-by: David S. Miller

tun: use per cpu variables for stats accounting

2016-04-15T02:55:25+00:00

Currently the tun device accounting uses dev->stats without applying any
kind of protection, regardless that accounting happens in preemptible
process context.
This patch move the tun stats to a per cpu data structure, and protect
the updates with  u64_stats_update_begin()/u64_stats_update_end() or
this_cpu_inc according to the stat type. The per cpu stats are
aggregated by the newly added ndo_get_stats64 ops.

Signed-off-by: Paolo Abeni 
Signed-off-by: David S. Miller