Guillaume Nault says:
====================
ppp: fix deadlock upon recursive xmit
This series fixes the issue reported by Feng where packets looping
through a ppp device makes the module deadlock:
https://marc.info/?l=linux-netdev&m=147134567319038&w=2
The problem can occur on virtual interfaces (e.g. PPP over L2TP, or
PPPoE on vxlan devices), when a PPP packet is routed back to the PPP
interface.
PPP's xmit path isn't reentrant, so patch #1 uses a per-cpu variable
to detect and break recursion. Patch #2 sets the NETIF_F_LLTX flag to
avoid lock inversion issues between ppp and txqueue locks.
There are multiple entry points to the PPP xmit path. This series has
been tested with lockdep and should address recursion issues no matter
how the packet entered the path.
A similar issue in L2TP is not covered by this series:
l2tp_xmit_skb() also isn't reentrant, and it can be called as part of
PPP's xmit path (pppol2tp_xmit()), or directly from the L2TP socket
(l2tp_ppp_sendmsg()). If a packet is sent by l2tp_ppp_sendmsg() and
routed to the parent PPP interface, then it's going to hit
l2tp_xmit_skb() again.
Breaking recursion as done in ppp_generic is not enough, because we'd
still have a lock inversion issue (locking in l2tp_xmit_skb() can
happen before or after locking in ppp_generic). The best approach would
be to use the ip_tunnel functions and remove the socket locking in
l2tp_xmit_skb(). But that'd be something for net-next.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
In case of misconfiguration, a virtual PPP channel might send packets
back to their parent PPP interface. This typically happens in
misconfigured L2TP setups, where PPP's peer IP address is set with the
IP of the L2TP peer.
When that happens the system hangs due to PPP trying to recursively
lock its xmit path.
[ 243.332155] BUG: spinlock recursion on CPU#1, accel-pppd/926
[ 243.333272] lock: 0xffff880033d90f18, .magic: dead4ead, .owner: accel-pppd/926, .owner_cpu: 1
[ 243.334859] CPU: 1 PID: 926 Comm: accel-pppd Not tainted 4.8.0-rc2 #1
[ 243.336010] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Debian-1.8.2-1 04/01/2014
[ 243.336018] ffff7fffffffffff ffff8800319a77a0 ffffffff8128de85 ffff880033d90f18
[ 243.336018] ffff880033ad8000 ffff8800319a77d8 ffffffff810ad7c0 ffffffff0000039e
[ 243.336018] ffff880033d90f18 ffff880033d90f60 ffff880033d90f18 ffff880033d90f28
[ 243.336018] Call Trace:
[ 243.336018] [<ffffffff8128de85>] dump_stack+0x4f/0x65
[ 243.336018] [<ffffffff810ad7c0>] spin_dump+0xe1/0xeb
[ 243.336018] [<ffffffff810ad7f0>] spin_bug+0x26/0x28
[ 243.336018] [<ffffffff810ad8b9>] do_raw_spin_lock+0x5c/0x160
[ 243.336018] [<ffffffff815522aa>] _raw_spin_lock_bh+0x35/0x3c
[ 243.336018] [<ffffffffa01a88e2>] ? ppp_push+0xa7/0x82d [ppp_generic]
[ 243.336018] [<ffffffffa01a88e2>] ppp_push+0xa7/0x82d [ppp_generic]
[ 243.336018] [<ffffffff810adada>] ? do_raw_spin_unlock+0xc2/0xcc
[ 243.336018] [<ffffffff81084962>] ? preempt_count_sub+0x13/0xc7
[ 243.336018] [<ffffffff81552438>] ? _raw_spin_unlock_irqrestore+0x34/0x49
[ 243.336018] [<ffffffffa01ac657>] ppp_xmit_process+0x48/0x877 [ppp_generic]
[ 243.336018] [<ffffffff81084962>] ? preempt_count_sub+0x13/0xc7
[ 243.336018] [<ffffffff81408cd3>] ? skb_queue_tail+0x71/0x7c
[ 243.336018] [<ffffffffa01ad1c5>] ppp_start_xmit+0x21b/0x22a [ppp_generic]
[ 243.336018] [<ffffffff81426af1>] dev_hard_start_xmit+0x15e/0x32c
[ 243.336018] [<ffffffff81454ed7>] sch_direct_xmit+0xd6/0x221
[ 243.336018] [<ffffffff814273a8>] __dev_queue_xmit+0x52a/0x820
[ 243.336018] [<ffffffff814276a9>] dev_queue_xmit+0xb/0xd
[ 243.336018] [<ffffffff81430a3c>] neigh_direct_output+0xc/0xe
[ 243.336018] [<ffffffff8146b5d7>] ip_finish_output2+0x4d2/0x548
[ 243.336018] [<ffffffff8146a8e6>] ? dst_mtu+0x29/0x2e
[ 243.336018] [<ffffffff8146d49c>] ip_finish_output+0x152/0x15e
[ 243.336018] [<ffffffff8146df84>] ? ip_output+0x74/0x96
[ 243.336018] [<ffffffff8146df9c>] ip_output+0x8c/0x96
[ 243.336018] [<ffffffff8146d55e>] ip_local_out+0x41/0x4a
[ 243.336018] [<ffffffff8146dd15>] ip_queue_xmit+0x531/0x5c5
[ 243.336018] [<ffffffff814a82cd>] ? udp_set_csum+0x207/0x21e
[ 243.336018] [<ffffffffa01f2f04>] l2tp_xmit_skb+0x582/0x5d7 [l2tp_core]
[ 243.336018] [<ffffffffa01ea458>] pppol2tp_xmit+0x1eb/0x257 [l2tp_ppp]
[ 243.336018] [<ffffffffa01acf17>] ppp_channel_push+0x91/0x102 [ppp_generic]
[ 243.336018] [<ffffffffa01ad2d8>] ppp_write+0x104/0x11c [ppp_generic]
[ 243.336018] [<ffffffff811a3c1e>] __vfs_write+0x56/0x120
[ 243.336018] [<ffffffff81239801>] ? fsnotify_perm+0x27/0x95
[ 243.336018] [<ffffffff8123ab01>] ? security_file_permission+0x4d/0x54
[ 243.336018] [<ffffffff811a4ca4>] vfs_write+0xbd/0x11b
[ 243.336018] [<ffffffff811a5a0a>] SyS_write+0x5e/0x96
[ 243.336018] [<ffffffff81552a1b>] entry_SYSCALL_64_fastpath+0x13/0x94
The main entry points for sending packets over a PPP unit are the
.write() and .ndo_start_xmit() callbacks (simplified view):
.write(unit fd) or .ndo_start_xmit()
\
CALL ppp_xmit_process()
\
LOCK unit's xmit path (ppp->wlock)
|
CALL ppp_push()
\
LOCK channel's xmit path (chan->downl)
|
CALL lower layer's .start_xmit() callback
\
... might recursively call .ndo_start_xmit() ...
/
RETURN from .start_xmit()
|
UNLOCK channel's xmit path
/
RETURN from ppp_push()
|
UNLOCK unit's xmit path
/
RETURN from ppp_xmit_process()
Packets can also be directly sent on channels (e.g. LCP packets):
.write(channel fd) or ppp_output_wakeup()
\
CALL ppp_channel_push()
\
LOCK channel's xmit path (chan->downl)
|
CALL lower layer's .start_xmit() callback
\
... might call .ndo_start_xmit() ...
/
RETURN from .start_xmit()
|
UNLOCK channel's xmit path
/
RETURN from ppp_channel_push()
Key points about the lower layer's .start_xmit() callback:
* It can be called directly by a channel fd .write() or by
ppp_output_wakeup() or indirectly by a unit fd .write() or by
.ndo_start_xmit().
* In any case, it's always called with chan->downl held.
* It might route the packet back to its parent unit using
.ndo_start_xmit() as entry point.
This patch detects and breaks recursion in ppp_xmit_process(). This
function is a good candidate for the task because it's called early
enough after .ndo_start_xmit(), it's always part of the recursion
loop and it's on the path of whatever entry point is used to send
a packet on a PPP unit.
Recursion detection is done using the per-cpu ppp_xmit_recursion
variable.
Since ppp_channel_push() too locks the channel's xmit path and calls
the lower layer's .start_xmit() callback, we need to also increment
ppp_xmit_recursion there. However there's no need to check for
recursion, as it's out of the recursion loop.
Reported-by: Feng Gao <gfree.wind@gmail.com>
Signed-off-by: Guillaume Nault <g.nault@alphalink.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
Casting away const is bad practice. Since this is ARM specific driver
don't have hardware actually test this.
Having getter functions for ops is really unnecessary code bloat, but
not going to touch that.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Vivien Didelot says:
====================
net: dsa: add MDB support
This patchset adds the switchdev MDB object support to the DSA layer.
The MDB support for the mv88e6xxx driver is very similar to the FDB
support. The FDB operations care about unicast addresses while the MDB
operations care about multicast addresses.
Both operation set load/purge/dump the Address Translation Table (ATU),
thus common code is used.
Changes in v2 based on Andrew's comments:
- drop "group" in multicast database related doc and comment
- change _one for more relevant _fid in mv88e6xxx_port_db_dump_one
- return -EOPNOTSUPP if switchdev obj ID is neither _FDB nor _MDB
====================
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for the MDB operations. This consists of
loading/purging/dumping multicast addresses for a given port in the ATU.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The MDB support for the mv88e6xxx driver will be very similar to the FDB
support, since it consists of loading/purging/dumping address to/from
the Address Translation Unit (ATU).
Prepare the support for MDB by making the FDB code accessing the ATU
generic. The FDB operations now provide access to the unicast addresses
while the MDB operations will provide access to the multicast addresses.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add SWITCHDEV_OBJ_ID_PORT_MDB support to the DSA layer.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Raghu Vatsavayi says:
====================
liquidio CN23XX support
Following patchset adds support for new device "CN23XX" in
liquidio family of adapters. As adviced by you I have split
the previous V3 patch of 18 patches into two halves. This
first patchset has first 10 patches, which are tested against
net-next. I will post the second half after this one.
This V4 patch also addressed all the comments from previous
submission:
1) Avoid busy loop while reading registers.
2) Other minor comments about debug messages and constants.
Please apply patches in following order as some of the
patches depend on earlier patches.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Add firmware download support for cn23xx device.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds support msix interrupt for cn23xx device.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds support for cn23xx queue manipulation.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adds support for initializing cn23xx device registers
related to mac, input/output and pf global config.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for cn23xx device init and sriov queue config.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for cn23xx specific queue definitions and
features.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds register definitions and structures for new
device cn23xx.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support of common irq enable functionality for both
iq(instruction queue) and oq(output queue).
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch contains changes for firmware version management.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Consolidate common functionality of various devices
from different files into lio_core.c/octeon_console.c.
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Check for ethtool_ops structures that are only stored in the ethtool_ops
field of a net_device structure or passed as the second argument to
netdev_set_default_ethtool_ops. These contexts are declared const, so
ethtool_ops structures that have these properties can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r disable optional_qualifier@
identifier i;
position p;
@@
static struct ethtool_ops i@p = { ... };
@ok1@
identifier r.i;
struct net_device e;
position p;
@@
e.ethtool_ops = &i@p;
@ok2@
identifier r.i;
expression e;
position p;
@@
netdev_set_default_ethtool_ops(e, &i@p)
@bad@
position p != {r.p,ok1.p,ok2.p};
identifier r.i;
@@
i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct ethtool_ops i = { ... };
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
Check for ethtool_ops structures that are only stored in the ethtool_ops
field of a net_device structure or passed as the second argument to
netdev_set_default_ethtool_ops. These contexts are declared const, so
ethtool_ops structures that have these properties can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r disable optional_qualifier@
identifier i;
position p;
@@
static struct ethtool_ops i@p = { ... };
@ok1@
identifier r.i;
struct net_device e;
position p;
@@
e.ethtool_ops = &i@p;
@ok2@
identifier r.i;
expression e;
position p;
@@
netdev_set_default_ethtool_ops(e, &i@p)
@bad@
position p != {r.p,ok1.p,ok2.p};
identifier r.i;
@@
i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct ethtool_ops i = { ... };
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
Check for ethtool_ops structures that are only stored in the ethtool_ops
field of a net_device structure or passed as the second argument to
netdev_set_default_ethtool_ops. These contexts are declared const, so
ethtool_ops structures that have these properties can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r disable optional_qualifier@
identifier i;
position p;
@@
static struct ethtool_ops i@p = { ... };
@ok1@
identifier r.i;
struct net_device e;
position p;
@@
e.ethtool_ops = &i@p;
@ok2@
identifier r.i;
expression e;
position p;
@@
netdev_set_default_ethtool_ops(e, &i@p)
@bad@
position p != {r.p,ok1.p,ok2.p};
identifier r.i;
@@
i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct ethtool_ops i = { ... };
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
Check for ethtool_ops structures that are only stored in the ethtool_ops
field of a net_device structure or passed as the second argument to
netdev_set_default_ethtool_ops. These contexts are declared const, so
ethtool_ops structures that have these properties can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r disable optional_qualifier@
identifier i;
position p;
@@
static struct ethtool_ops i@p = { ... };
@ok1@
identifier r.i;
struct net_device e;
position p;
@@
e.ethtool_ops = &i@p;
@ok2@
identifier r.i;
expression e;
position p;
@@
netdev_set_default_ethtool_ops(e, &i@p)
@bad@
position p != {r.p,ok1.p,ok2.p};
identifier r.i;
@@
i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct ethtool_ops i = { ... };
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
Check for ethtool_ops structures that are only stored in the ethtool_ops
field of a net_device structure or passed as the second argument to
netdev_set_default_ethtool_ops. These contexts are declared const, so
ethtool_ops structures that have these properties can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r disable optional_qualifier@
identifier i;
position p;
@@
static struct ethtool_ops i@p = { ... };
@ok1@
identifier r.i;
struct net_device e;
position p;
@@
e.ethtool_ops = &i@p;
@ok2@
identifier r.i;
expression e;
position p;
@@
netdev_set_default_ethtool_ops(e, &i@p)
@bad@
position p != {r.p,ok1.p,ok2.p};
identifier r.i;
@@
i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct ethtool_ops i = { ... };
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Acked-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The addition of the per-queue statistics introduced a harmless warning
on all 32-bit architectures:
drivers/net/ethernet/qlogic/qede/qede_ethtool.c: In function 'qede_get_ethtool_stats':
drivers/net/ethernet/qlogic/qede/qede_ethtool.c:244:31: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
buf[cnt++] = QEDE_TQSTATS_DATA(edev,
^
drivers/net/ethernet/qlogic/qede/qede_ethtool.c:244:22: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
buf[cnt++] = QEDE_TQSTATS_DATA(edev,
^
This changes the cast to 'void *' to shut up the warning, which
avoids the assumptions on the size of the pointer type.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 68db9ec2df ("qede: Add support for per-queue stats.")
Signed-off-by: David S. Miller <davem@davemloft.net>
When CONFIG_PM_SLEEP is disabled, we get a couple of harmless warnings:
drivers/net/ethernet/renesas/ravb_main.c:2117:12: error: 'ravb_resume' defined but not used [-Werror=unused-function]
drivers/net/ethernet/renesas/ravb_main.c:2104:12: error: 'ravb_suspend' defined but not used [-Werror=unused-function]
The simplest solution here is to replace the #ifdef with __maybe_unused
annotations, which lets the compiler do the right thing by itself.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 0184165b2f ("ravb: add sleep PM suspend/resume support")
Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern says:
====================
net: mpls: fragmentation and gso fixes for locally originated traffic
This series fixes mtu and fragmentation for tunnels using lwtunnel
output redirect, and fixes GSO for MPLS for locally originated traffic
reported by Lennert Buytenhek.
A follow on series will address fragmentation and GSO for forwarded
MPLS traffic. Hardware offload of GSO with MPLS also needs to be
addressed.
Simon: Can you verify this works with OVS for single and multiple
labels?
v4
- more updates to mpls_gso_segment per Alex's comments (thanks, Alex)
- updates to teaching OVS about marking MPLS labels as the network header
v3
- updates to mpls_gso_segment per Alex's comments
- dropped skb->encapsulation = 1 from mpls_xmit per Alex's comment
v2
- consistent use of network_header in skb to fix GSO for MPLS
- update MPLS code in OVS to network_header and inner_network_header
====================
Tested-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
veth does not really transmit packets only moves the skb from one
netdev to another so gso and checksum is not really needed. Add
the features to mpls_features to get the same benefit and performance
with MPLS as without it.
Reported-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As reported by Lennert the MPLS GSO code is failing to properly segment
large packets. There are a couple of problems:
1. the inner protocol is not set so the gso segment functions for inner
protocol layers are not getting run, and
2 MPLS labels for packets that use the "native" (non-OVS) MPLS code
are not properly accounted for in mpls_gso_segment.
The MPLS GSO code was added for OVS. It is re-using skb_mac_gso_segment
to call the gso segment functions for the higher layer protocols. That
means skb_mac_gso_segment is called twice -- once with the network
protocol set to MPLS and again with the network protocol set to the
inner protocol.
This patch sets the inner skb protocol addressing item 1 above and sets
the network_header and inner_network_header to mark where the MPLS labels
start and end. The MPLS code in OVS is also updated to set the two
network markers.
>From there the MPLS GSO code uses the difference between the network
header and the inner network header to know the size of the MPLS header
that was pushed. It then pulls the MPLS header, resets the mac_len and
protocol for the inner protocol and then calls skb_mac_gso_segment
to segment the skb.
Afterward the inner protocol segmentation is done the skb protocol
is set to mpls for each segment and the network and mac headers
restored.
Reported-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Today mpls iptunnel lwtunnel_output redirect expects the tunnel
output function to handle fragmentation. This is ok but can be
avoided if we did not do the mpls output redirect too early.
ie we could wait until ip fragmentation is done and then call
mpls output for each ip fragment.
To make this work we will need,
1) the lwtunnel state to carry encap headroom
2) and do the redirect to the encap output handler on the ip fragment
(essentially do the output redirect after fragmentation)
This patch adds tunnel headroom in lwtstate to make sure we
account for tunnel data in mtu calculations during fragmentation
and adds new xmit redirect handler to redirect to lwtunnel xmit func
after ip fragmentation.
This includes IPV6 and some mtu fixes and testing from David Ahern.
Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After commit 145dd5f9c8 ("net: flush the softnet backlog in process
context"), we can easily batch calls to flush_all_backlogs() for all
devices processed in rollback_registered_many()
Tested:
Before patch, on an idle host.
modprobe dummy numdummies=10000
perf stat -e context-switches -a rmmod dummy
Performance counter stats for 'system wide':
1,211,798 context-switches
1.302137465 seconds time elapsed
After patch:
perf stat -e context-switches -a rmmod dummy
Performance counter stats for 'system wide':
225,523 context-switches
0.721623566 seconds time elapsed
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove a useless log message and improve the logic for setting
a PHY address from the contents of the MNG_IF_SEL register.
Signed-off-by: Mark Rustad <mark.d.rustad@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=H5Xf
-----END PGP SIGNATURE-----
Merge tag 'rxrpc-rewrite-20160830-1' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs
David Howells says:
====================
rxrpc: Preparation for removal of use of skbs from AFS
Here's a set of patches that prepare the way for the removal of the use of
sk_buffs from fs/afs (they'll be entirely retained within net/rxrpc):
(1) Fix a potential NULL-pointer deref in rxrpc_abort_calls().
(2) Condense all the terminal call state machine states to a single one
plus supplementary info.
(3) Add a trace point for rxrpc call usage debugging.
(4) Cleanups and missing headers.
(5) Provide a way for AFS to ask about a call's peer address without
having an sk_buff to query.
(6) Use call->peer directly rather than going via call->conn (which might
be NULL).
(7) Pass struct socket * to various rxrpc kernel interface functions so
they can use that directly rather than getting it from the rxrpc_call
struct.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Trivial fix to spelling mistake in dev_warn message.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pass struct socket * to more rxrpc kernel interface functions. They should
be starting from this rather than the socket pointer in the rxrpc_call
struct if they need to access the socket.
I have left:
rxrpc_kernel_is_data_last()
rxrpc_kernel_get_abort_code()
rxrpc_kernel_get_error_number()
rxrpc_kernel_free_skb()
rxrpc_kernel_data_consumed()
unmodified as they're all about to be removed (and, in any case, don't
touch the socket).
Signed-off-by: David Howells <dhowells@redhat.com>
Provide a function so that kernel users, such as AFS, can ask for the peer
address of a call:
void rxrpc_kernel_get_peer(struct rxrpc_call *call,
struct sockaddr_rxrpc *_srx);
In the future the kernel service won't get sk_buffs to look inside.
Further, this allows us to hide any canonicalisation inside AF_RXRPC for
when IPv6 support is added.
Also propagate this through to afs_find_server() and issue a warning if we
can't handle the address family yet.
Signed-off-by: David Howells <dhowells@redhat.com>
Condense the terminal states of a call state machine to a single state,
plus a separate completion type value. The value is then set, along with
error and abort code values, only when the call is transitioned to the
completion state.
Helpers are provided to simplify this.
Signed-off-by: David Howells <dhowells@redhat.com>
The call pointer in a channel on a connection will be NULL if there's no
active call on that channel. rxrpc_abort_calls() needs to check for this
before trying to take the call's state_lock.
Signed-off-by: David Howells <dhowells@redhat.com>
The nf_log_set is an interface function, so it should do the strict sanity
check of parameters. Convert the return value of nf_log_set as int instead
of void. When the pf is invalid, return -EOPNOTSUPP.
Signed-off-by: Gao Feng <fgao@ikuai8.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
There is one macro ARPHRD_ETHER which defines the ethernet proto for ARP,
so we could use it instead of the literal number '1'.
Signed-off-by: Gao Feng <fgao@ikuai8.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
After timer removal this just calls nf_ct_delete so remove the __ prefix
version and make nf_ct_kill a shorthand for nf_ct_delete.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
If we evicted a large fraction of the scanned conntrack entries re-schedule
the next gc cycle for immediate execution.
This triggers during tests where load is high, then drops to zero and
many connections will be in TW/CLOSE state with < 30 second timeouts.
Without this change it will take several minutes until conntrack count
comes back to normal.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Conntrack gc worker to evict stale entries.
GC happens once every 5 seconds, but we only scan at most 1/64th of the
table (and not more than 8k) buckets to avoid hogging cpu.
This means that a complete scan of the table will take several minutes
of wall-clock time.
Considering that the gc run will never have to evict any entries
during normal operation because those will happen from packet path
this should be fine.
We only need gc to make sure userspace (conntrack event listeners)
eventually learn of the timeout, and for resource reclaim in case the
system becomes idle.
We do not disable BH and cond_resched for every bucket so this should
not introduce noticeable latencies either.
A followup patch will add a small change to speed up GC for the extreme
case where most entries are timed out on an otherwise idle system.
v2: Use cond_resched_rcu_qs & add comment wrt. missing restart on
nulls value change in gc worker, suggested by Eric Dumazet.
v3: don't call cancel_delayed_work_sync twice (again, Eric).
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
When dumping we already have to look at the entire table, so we might
as well toss those entries whose timeout value is in the past.
We also look at every entry during resize operations.
However, eviction there is not as simple because we hold the
global resize lock so we can't evict without adding a 'expired' list
to drop from later. Considering that resizes are very rare it doesn't
seem worth doing it.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
With stats enabled this eats 80 bytes on x86_64 per nf_conn entry, as
Eric Dumazet pointed out during netfilter workshop 2016.
Eric also says: "Another reason was the fact that Thomas was about to
change max timer range [..]" (500462a9de, 'timers: Switch to
a non-cascading wheel').
Remove the timer and use a 32bit jiffies value containing timestamp until
entry is valid.
During conntrack lookup, even before doing tuple comparision, check
the timeout value and evict the entry in case it is too old.
The dying bit is used as a synchronization point to avoid races where
multiple cpus try to evict the same entry.
Because lookup is always lockless, we need to bump the refcnt once
when we evict, else we could try to evict already-dead entry that
is being recycled.
This is the standard/expected way when conntrack entries are destroyed.
Followup patches will introduce garbage colliction via work queue
and further places where we can reap obsoleted entries (e.g. during
netlink dumps), this is needed to avoid expired conntracks from hanging
around for too long when lookup rate is low after a busy period.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
David S. Miller