platform/kernel/linux-starfive.git
6 years agonet: amd: add missing of_node_put()
Yangtao Li [Thu, 22 Nov 2018 12:34:41 +0000 (07:34 -0500)]
net: amd: add missing of_node_put()

of_find_node_by_path() acquires a reference to the node
returned by it and that reference needs to be dropped by its caller.
This place doesn't do that, so fix it.

Signed-off-by: Yangtao Li <tiny.windzz@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoteam: no need to do team_notify_peers or team_mcast_rejoin when disabling port
Hangbin Liu [Thu, 22 Nov 2018 08:15:28 +0000 (16:15 +0800)]
team: no need to do team_notify_peers or team_mcast_rejoin when disabling port

team_notify_peers() will send ARP and NA to notify peers. team_mcast_rejoin()
will send multicast join group message to notify peers. We should do this when
enabling/changed to a new port. But it doesn't make sense to do it when a port
is disabled.

On the other hand, when we set mcast_rejoin_count to 2, and do a failover,
team_port_disable() will increase mcast_rejoin.count_pending to 2 and then
team_port_enable() will increase mcast_rejoin.count_pending to 4. We will send
4 mcast rejoin messages at latest, which will make user confused. The same
with notify_peers.count.

Fix it by deleting team_notify_peers() and team_mcast_rejoin() in
team_port_disable().

Reported-by: Liang Li <liali@redhat.com>
Fixes: fc423ff00df3a ("team: add peer notification")
Fixes: 492b200efdd20 ("team: add support for sending multicast rejoins")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agovirtio-net: fail XDP set if guest csum is negotiated
Jason Wang [Thu, 22 Nov 2018 06:36:31 +0000 (14:36 +0800)]
virtio-net: fail XDP set if guest csum is negotiated

We don't support partial csumed packet since its metadata will be lost
or incorrect during XDP processing. So fail the XDP set if guest_csum
feature is negotiated.

Fixes: f600b6905015 ("virtio_net: Add XDP support")
Reported-by: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Pavel Popa <pashinho1990@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agovirtio-net: disable guest csum during XDP set
Jason Wang [Thu, 22 Nov 2018 06:36:30 +0000 (14:36 +0800)]
virtio-net: disable guest csum during XDP set

We don't disable VIRTIO_NET_F_GUEST_CSUM if XDP was set. This means we
can receive partial csumed packets with metadata kept in the
vnet_hdr. This may have several side effects:

- It could be overridden by header adjustment, thus is might be not
  correct after XDP processing.
- There's no way to pass such metadata information through
  XDP_REDIRECT to another driver.
- XDP does not support checksum offload right now.

So simply disable guest csum if possible in this the case of XDP.

Fixes: 3f93522ffab2d ("virtio-net: switch off offloads on demand if possible on XDP set")
Reported-by: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Pavel Popa <pashinho1990@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/sched: act_police: add missing spinlock initialization
Davide Caratti [Wed, 21 Nov 2018 17:23:53 +0000 (18:23 +0100)]
net/sched: act_police: add missing spinlock initialization

commit f2cbd4852820 ("net/sched: act_police: fix race condition on state
variables") introduces a new spinlock, but forgets its initialization.
Ensure that tcf_police_init() initializes 'tcfp_lock' every time a 'police'
action is newly created, to avoid the following lockdep splat:

 INFO: trying to register non-static key.
 the code is fine but needs lockdep annotation.
 turning off the locking correctness validator.
 <...>
 Call Trace:
  dump_stack+0x85/0xcb
  register_lock_class+0x581/0x590
  __lock_acquire+0xd4/0x1330
  ? tcf_police_init+0x2fa/0x650 [act_police]
  ? lock_acquire+0x9e/0x1a0
  lock_acquire+0x9e/0x1a0
  ? tcf_police_init+0x2fa/0x650 [act_police]
  ? tcf_police_init+0x55a/0x650 [act_police]
  _raw_spin_lock_bh+0x34/0x40
  ? tcf_police_init+0x2fa/0x650 [act_police]
  tcf_police_init+0x2fa/0x650 [act_police]
  tcf_action_init_1+0x384/0x4c0
  tcf_action_init+0xf6/0x160
  tcf_action_add+0x73/0x170
  tc_ctl_action+0x122/0x160
  rtnetlink_rcv_msg+0x2a4/0x490
  ? netlink_deliver_tap+0x99/0x400
  ? validate_linkmsg+0x370/0x370
  netlink_rcv_skb+0x4d/0x130
  netlink_unicast+0x196/0x230
  netlink_sendmsg+0x2e5/0x3e0
  sock_sendmsg+0x36/0x40
  ___sys_sendmsg+0x280/0x2f0
  ? _raw_spin_unlock+0x24/0x30
  ? handle_pte_fault+0xafe/0xf30
  ? find_held_lock+0x2d/0x90
  ? syscall_trace_enter+0x1df/0x360
  ? __sys_sendmsg+0x5e/0xa0
  __sys_sendmsg+0x5e/0xa0
  do_syscall_64+0x60/0x210
  entry_SYSCALL_64_after_hwframe+0x49/0xbe
 RIP: 0033:0x7f1841c7cf10
 Code: c3 48 8b 05 82 6f 2c 00 f7 db 64 89 18 48 83 cb ff eb dd 0f 1f 80 00 00 00 00 83 3d 8d d0 2c 00 00 75 10 b8 2e 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 ae cc 00 00 48 89 04 24
 RSP: 002b:00007ffcf9df4d68 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
 RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007f1841c7cf10
 RDX: 0000000000000000 RSI: 00007ffcf9df4dc0 RDI: 0000000000000003
 RBP: 000000005bf56105 R08: 0000000000000002 R09: 00007ffcf9df8edc
 R10: 00007ffcf9df47e0 R11: 0000000000000246 R12: 0000000000671be0
 R13: 00007ffcf9df4e84 R14: 0000000000000008 R15: 0000000000000000

Fixes: f2cbd4852820 ("net/sched: act_police: fix race condition on state variables")
Reported-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: don't keep lonely packets forever in the gro hash
Paolo Abeni [Wed, 21 Nov 2018 17:21:35 +0000 (18:21 +0100)]
net: don't keep lonely packets forever in the gro hash

Eric noted that with UDP GRO and NAPI timeout, we could keep a single
UDP packet inside the GRO hash forever, if the related NAPI instance
calls napi_gro_complete() at an higher frequency than the NAPI timeout.
Willem noted that even TCP packets could be trapped there, till the
next retransmission.
This patch tries to address the issue, flushing the old packets -
those with a NAPI_GRO_CB age before the current jiffy - before scheduling
the NAPI timeout. The rationale is that such a timeout should be
well below a jiffy and we are not flushing packets eligible for sane GRO.

v1  -> v2:
 - clarified the commit message and comment

RFC -> v1:
 - added 'Fixes tags', cleaned-up the wording.

Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Fixes: 3b47d30396ba ("net: gro: add a per device gro flush timer")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/ipv6: re-do dad when interface has IFF_NOARP flag change
Hangbin Liu [Wed, 21 Nov 2018 13:52:33 +0000 (21:52 +0800)]
net/ipv6: re-do dad when interface has IFF_NOARP flag change

When we add a new IPv6 address, we should also join corresponding solicited-node
multicast address, unless the interface has IFF_NOARP flag, as function
addrconf_join_solict() did. But if we remove IFF_NOARP flag later, we do
not do dad and add the mcast address. So we will drop corresponding neighbour
discovery message that came from other nodes.

A typical example is after creating a ipvlan with mode l3, setting up an ipv6
address and changing the mode to l2. Then we will not be able to ping this
address as the interface doesn't join related solicited-node mcast address.

Fix it by re-doing dad when interface changed IFF_NOARP flag. Then we will add
corresponding mcast group and check if there is a duplicate address on the
network.

Reported-by: Jianlin Shi <jishi@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agopacket: copy user buffers before orphan or clone
Willem de Bruijn [Tue, 20 Nov 2018 18:00:18 +0000 (13:00 -0500)]
packet: copy user buffers before orphan or clone

tpacket_snd sends packets with user pages linked into skb frags. It
notifies that pages can be reused when the skb is released by setting
skb->destructor to tpacket_destruct_skb.

This can cause data corruption if the skb is orphaned (e.g., on
transmit through veth) or cloned (e.g., on mirror to another psock).

Create a kernel-private copy of data in these cases, same as tun/tap
zerocopy transmission. Reuse that infrastructure: mark the skb as
SKBTX_ZEROCOPY_FRAG, which will trigger copy in skb_orphan_frags(_rx).

Unlike other zerocopy packets, do not set shinfo destructor_arg to
struct ubuf_info. tpacket_destruct_skb already uses that ptr to notify
when the original skb is released and a timestamp is recorded. Do not
change this timestamp behavior. The ubuf_info->callback is not needed
anyway, as no zerocopy notification is expected.

Mark destructor_arg as not-a-uarg by setting the lower bit to 1. The
resulting value is not a valid ubuf_info pointer, nor a valid
tpacket_snd frame address. Add skb_zcopy_.._nouarg helpers for this.

The fix relies on features introduced in commit 52267790ef52 ("sock:
add MSG_ZEROCOPY"), so can be backported as is only to 4.14.

Tested with from `./in_netns.sh ./txring_overwrite` from
http://github.com/wdebruij/kerneltools/tests

Fixes: 69e3c75f4d54 ("net: TX_RING and packet mmap")
Reported-by: Anand H. Krishnan <anandhkrishnan@gmail.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'ibmvnic-Fix-queue-and-buffer-accounting-errors'
David S. Miller [Thu, 22 Nov 2018 19:53:26 +0000 (11:53 -0800)]
Merge branch 'ibmvnic-Fix-queue-and-buffer-accounting-errors'

Thomas Falcon says:

====================
ibmvnic: Fix queue and buffer accounting errors

This series includes two small fixes. The first resolves a typo bug
in the code to clean up unused RX buffers during device queue removal.
The second ensures that device queue memory is updated to reflect new
supported queue ring sizes after migration to other backing hardware.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoibmvnic: Update driver queues after change in ring size support
Thomas Falcon [Wed, 21 Nov 2018 17:17:59 +0000 (11:17 -0600)]
ibmvnic: Update driver queues after change in ring size support

During device reset, queue memory is not being updated to accommodate
changes in ring buffer sizes supported by backing hardware. Track
any differences in ring buffer sizes following the reset and update
queue memory when possible.

Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoibmvnic: Fix RX queue buffer cleanup
Thomas Falcon [Wed, 21 Nov 2018 17:17:58 +0000 (11:17 -0600)]
ibmvnic: Fix RX queue buffer cleanup

The wrong index is used when cleaning up RX buffer objects during release
of RX queues. Update to use the correct index counter.

Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: thunderx: set xdp_prog to NULL if bpf_prog_add fails
Lorenzo Bianconi [Wed, 21 Nov 2018 15:32:10 +0000 (16:32 +0100)]
net: thunderx: set xdp_prog to NULL if bpf_prog_add fails

Set xdp_prog pointer to NULL if bpf_prog_add fails since that routine
reports the error code instead of NULL in case of failure and xdp_prog
pointer value is used in the driver to verify if XDP is currently
enabled.
Moreover report the error code to userspace if nicvf_xdp_setup fails

Fixes: 05c773f52b96 ("net: thunderx: Add basic XDP support")
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/dim: Update DIM start sample after each DIM iteration
Tal Gilboa [Wed, 21 Nov 2018 14:28:23 +0000 (16:28 +0200)]
net/dim: Update DIM start sample after each DIM iteration

On every iteration of net_dim, the algorithm may choose to
check for the system state by comparing current data sample
with previous data sample. After each of these comparison,
regardless of the action taken, the sample used as baseline
is needed to be updated.

This patch fixes a bug that causes DIM to take wrong decisions,
due to never updating the baseline sample for comparison between
iterations. This way, DIM always compares current sample with
zeros.

Although this is a functional fix, it also improves and stabilizes
performance as the algorithm works properly now.

Performance:
Tested single UDP TX stream with pktgen:
samples/pktgen/pktgen_sample03_burst_single_flow.sh -i p4p2 -d 1.1.1.1
-m 24:8a:07:88:26:8b -f 3 -b 128

ConnectX-5 100GbE packet rate improved from 15-19Mpps to 19-20Mpps.
Also, toggling between profiles is less frequent with the fix.

Fixes: 8115b750dbcb ("net/dim: use struct net_dim_sample as arg to net_dim")
Signed-off-by: Tal Gilboa <talgi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: faraday: ftmac100: remove netif_running(netdev) check before disabling interrupts
Vincent Chen [Wed, 21 Nov 2018 01:38:11 +0000 (09:38 +0800)]
net: faraday: ftmac100: remove netif_running(netdev) check before disabling interrupts

In the original ftmac100_interrupt(), the interrupts are only disabled when
the condition "netif_running(netdev)" is true. However, this condition
causes kerenl hang in the following case. When the user requests to
disable the network device, kernel will clear the bit __LINK_STATE_START
from the dev->state and then call the driver's ndo_stop function. Network
device interrupts are not blocked during this process. If an interrupt
occurs between clearing __LINK_STATE_START and stopping network device,
kernel cannot disable the interrupts due to the condition
"netif_running(netdev)" in the ISR. Hence, kernel will hang due to the
continuous interruption of the network device.

In order to solve the above problem, the interrupts of the network device
should always be disabled in the ISR without being restricted by the
condition "netif_running(netdev)".

[V2]
Remove unnecessary curly braces.

Signed-off-by: Vincent Chen <vincentc@andestech.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'smc-fixes'
David S. Miller [Thu, 22 Nov 2018 00:14:56 +0000 (16:14 -0800)]
Merge branch 'smc-fixes'

Ursula Braun says:

====================
net/smc: fixes 2018-11-12

here is V4 of some net/smc fixes in different areas for the net tree.

v1->v2:
   do not define 8-byte alignment for union smcd_cdc_cursor in
   patch 4/5 "net/smc: atomic SMCD cursor handling"
v2->v3:
   stay with 8-byte alignment for union smcd_cdc_cursor in
   patch 4/5 "net/smc: atomic SMCD cursor handling", but get rid of
   __packed for struct smcd_cdc_msg
v3->v4:
   get rid of another __packed for struct smc_cdc_msg in
   patch 4/5 "net/smc: atomic SMCD cursor handling"
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/smc: use after free fix in smc_wr_tx_put_slot()
Ursula Braun [Tue, 20 Nov 2018 15:46:43 +0000 (16:46 +0100)]
net/smc: use after free fix in smc_wr_tx_put_slot()

In smc_wr_tx_put_slot() field pend->idx is used after being
cleared. That means always idx 0 is cleared in the wr_tx_mask.
This results in a broken administration of available WR send
payload buffers.

Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/smc: atomic SMCD cursor handling
Ursula Braun [Tue, 20 Nov 2018 15:46:42 +0000 (16:46 +0100)]
net/smc: atomic SMCD cursor handling

Running uperf tests with SMCD on LPARs results in corrupted cursors.
SMCD cursors should be treated atomically to fix cursor corruption.

Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/smc: add SMC-D shutdown signal
Hans Wippel [Tue, 20 Nov 2018 15:46:41 +0000 (16:46 +0100)]
net/smc: add SMC-D shutdown signal

When a SMC-D link group is freed, a shutdown signal should be sent to
the peer to indicate that the link group is invalid. This patch adds the
shutdown signal to the SMC code.

Signed-off-by: Hans Wippel <hwippel@linux.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/smc: use queue pair number when matching link group
Karsten Graul [Tue, 20 Nov 2018 15:46:40 +0000 (16:46 +0100)]
net/smc: use queue pair number when matching link group

When searching for an existing link group the queue pair number is also
to be taken into consideration. When the SMC server sends a new number
in a CLC packet (keeping all other values equal) then a new link group
is to be created on the SMC client side.

Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/smc: abort CLC connection in smc_release
Hans Wippel [Tue, 20 Nov 2018 15:46:39 +0000 (16:46 +0100)]
net/smc: abort CLC connection in smc_release

In case of a non-blocking SMC socket, the initial CLC handshake is
performed over a blocking TCP connection in a worker. If the SMC socket
is released, smc_release has to wait for the blocking CLC socket
operations (e.g., kernel_connect) inside the worker.

This patch aborts a CLC connection when the respective non-blocking SMC
socket is released to avoid waiting on socket operations or timeouts.

Signed-off-by: Hans Wippel <hwippel@linux.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge tag 'wireless-drivers-for-davem-2018-11-20' of git://git.kernel.org/pub/scm...
David S. Miller [Wed, 21 Nov 2018 23:51:16 +0000 (15:51 -0800)]
Merge tag 'wireless-drivers-for-davem-2018-11-20' of git://git./linux/kernel/git/kvalo/wireless-drivers

Kalle Valo says:

====================
wireless-drivers fixes for 4.20

First set of fixes for 4.20, this time we have quite a few them but
all very small.

ath9k

* fix a locking regression found by a static checker

wlcore

* fix a crash which was a regression with wakeirq handling

brcm80211

* yet another fix for 160 MHz channel handling

mt76

* fix a longstaning build problem when CONFIG_LEDS_CLASS is disabled

* don't use uninitialised mutex

iwlwifi

* do note that the iwlwifi merge tag (commit 4ec321c14693) seems to
  contain wrong list of changes so ignore that

* fix ACPI data handling, a memory leak and other smaller fixes

ath10k

* fix a crash during suspend which was a recent regression
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotcp: defer SACK compression after DupThresh
Eric Dumazet [Tue, 20 Nov 2018 13:53:59 +0000 (05:53 -0800)]
tcp: defer SACK compression after DupThresh

Jean-Louis reported a TCP regression and bisected to recent SACK
compression.

After a loss episode (receiver not able to keep up and dropping
packets because its backlog is full), linux TCP stack is sending
a single SACK (DUPACK).

Sender waits a full RTO timer before recovering losses.

While RFC 6675 says in section 5, "Algorithm Details",

   (2) If DupAcks < DupThresh but IsLost (HighACK + 1) returns true --
       indicating at least three segments have arrived above the current
       cumulative acknowledgment point, which is taken to indicate loss
       -- go to step (4).
...
   (4) Invoke fast retransmit and enter loss recovery as follows:

there are old TCP stacks not implementing this strategy, and
still counting the dupacks before starting fast retransmit.

While these stacks probably perform poorly when receivers implement
LRO/GRO, we should be a little more gentle to them.

This patch makes sure we do not enable SACK compression unless
3 dupacks have been sent since last rcv_nxt update.

Ideally we should even rearm the timer to send one or two
more DUPACK if no more packets are coming, but that will
be work aiming for linux-4.21.

Many thanks to Jean-Louis for bisecting the issue, providing
packet captures and testing this patch.

Fixes: 5d9f4262b7ea ("tcp: add SACK compression")
Reported-by: Jean-Louis Dupond <jean-louis@dupond.be>
Tested-by: Jean-Louis Dupond <jean-louis@dupond.be>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: skb_scrub_packet(): Scrub offload_fwd_mark
Petr Machata [Tue, 20 Nov 2018 11:39:56 +0000 (11:39 +0000)]
net: skb_scrub_packet(): Scrub offload_fwd_mark

When a packet is trapped and the corresponding SKB marked as
already-forwarded, it retains this marking even after it is forwarded
across veth links into another bridge. There, since it ingresses the
bridge over veth, which doesn't have offload_fwd_mark, it triggers a
warning in nbp_switchdev_frame_mark().

Then nbp_switchdev_allowed_egress() decides not to allow egress from
this bridge through another veth, because the SKB is already marked, and
the mark (of 0) of course matches. Thus the packet is incorrectly
blocked.

Solve by resetting offload_fwd_mark() in skb_scrub_packet(). That
function is called from tunnels and also from veth, and thus catches the
cases where traffic is forwarded between bridges and transformed in a
way that invalidates the marking.

Fixes: 6bc506b4fb06 ("bridge: switchdev: Add forward mark support for stacked devices")
Fixes: abf4bb6b63d0 ("skbuff: Add the offload_mr_fwd_mark field")
Signed-off-by: Petr Machata <petrm@mellanox.com>
Suggested-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/sched: act_police: fix race condition on state variables
Davide Caratti [Tue, 20 Nov 2018 21:18:44 +0000 (22:18 +0100)]
net/sched: act_police: fix race condition on state variables

after 'police' configuration parameters were converted to use RCU instead
of spinlock, the state variables used to compute the traffic rate (namely
'tcfp_toks', 'tcfp_ptoks' and 'tcfp_t_c') are erroneously read/updated in
the traffic path without any protection.

Use a dedicated spinlock to avoid race conditions on these variables, and
ensure proper cache-line alignment. In this way, 'police' is still faster
than what we observed when 'tcf_lock' was used in the traffic path _ i.e.
reverting commit 2d550dbad83c ("net/sched: act_police: don't use spinlock
in the data path"). Moreover, we preserve the throughput improvement that
was obtained after 'police' started using per-cpu counters, when 'avrate'
is used instead of 'rate'.

Changes since v1 (thanks to Eric Dumazet):
- call ktime_get_ns() before acquiring the lock in the traffic path
- use a dedicated spinlock instead of tcf_lock
- improve cache-line usage

Fixes: 2d550dbad83c ("net/sched: act_police: don't use spinlock in the data path")
Reported-and-suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
6 years agoMAINTAINERS: add myself as co-maintainer for r8169
Heiner Kallweit [Tue, 20 Nov 2018 20:22:50 +0000 (21:22 +0100)]
MAINTAINERS: add myself as co-maintainer for r8169

Meanwhile I know the driver quite well and I refactored bigger parts
of it. As a result people contact me already with r8169 questions.
Therefore I'd volunteer to become co-maintainer of the driver also
officially.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotcp: Fix SOF_TIMESTAMPING_RX_HARDWARE to use the latest timestamp during TCP coalescing
Stephen Mallon [Tue, 20 Nov 2018 08:15:02 +0000 (19:15 +1100)]
tcp: Fix SOF_TIMESTAMPING_RX_HARDWARE to use the latest timestamp during TCP coalescing

During tcp coalescing ensure that the skb hardware timestamp refers to the
highest sequence number data.
Previously only the software timestamp was updated during coalescing.

Signed-off-by: Stephen Mallon <stephen.mallon@sydney.edu.au>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotg3: Add PHY reset for 5717/5719/5720 in change ring and flow control paths
Siva Reddy Kallam [Tue, 20 Nov 2018 04:34:04 +0000 (10:04 +0530)]
tg3: Add PHY reset for 5717/5719/5720 in change ring and flow control paths

This patch has the fix to avoid PHY lockup with 5717/5719/5720 in change
ring and flow control paths. This patch solves the RX hang while doing
continuous ring or flow control parameters with heavy traffic from peer.

Signed-off-by: Siva Reddy Kallam <siva.kallam@broadcom.com>
Acked-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge tag 'mlx5-fixes-2018-11-19' of git://git.kernel.org/pub/scm/linux/kernel/git...
David S. Miller [Tue, 20 Nov 2018 03:04:33 +0000 (19:04 -0800)]
Merge tag 'mlx5-fixes-2018-11-19' of git://git./linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
Mellanox, mlx5 fixes 2018-11-19

The following fixes are for mlx5 core and netdev driver.

For -stable v4.16
bc7fda7d4637 ('net/mlx5e: IPoIB, Reset QP after channels are closed')

For -stable v4.17
36917a270395 ('net/mlx5: IPSec, Fix the SA context hash key')

For -stable v4.18
6492a432be3a ('net/mlx5e: Always use the match level enum when parsing TC rule match')
c3f81be236b1 ('net/mlx5e: Removed unnecessary warnings in FEC caps query')
c5ce2e736b64 ('net/mlx5e: Fix selftest for small MTUs')

For -stable v4.19
effcd896b25e ('net/mlx5e: Adjust to max number of channles when re-attaching')
394cbc5acd68 ('net/mlx5e: RX, verify received packet size in Linear Striding RQ')
447cbb3613c8 ('net/mlx5e: Don't match on vlan non-existence if ethertype is wildcarded')
c223c1574612 ('net/mlx5e: Claim TC hw offloads support only under a proper build config')

Please pull and let me know if there's any problem.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/ibmnvic: Fix deadlock problem in reset
Juliet Kim [Mon, 19 Nov 2018 21:59:22 +0000 (15:59 -0600)]
net/ibmnvic: Fix deadlock problem in reset

This patch changes to use rtnl_lock only during a reset to avoid
deadlock that could occur when a thread operating close is holding
rtnl_lock and waiting for reset_lock acquired by another thread,
which is waiting for rtnl_lock in order to set the number of tx/rx
queues during a reset.

Also, we now setting the number of tx/rx queues during a soft reset
for failover or LPM events.

Signed-off-by: Juliet Kim <julietk@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'qed-Fix-Queue-Manager-getters'
David S. Miller [Tue, 20 Nov 2018 02:38:16 +0000 (18:38 -0800)]
Merge branch 'qed-Fix-Queue-Manager-getters'

Denis Bolotin says:

====================
qed: Fix Queue Manager getters

This patch series fixes various queue manager getter functions. It is
important to make sure the getter's caller will receive a valid queue even
in error case to prevent more serious bugs.
Please consider applying to net.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoqed: Fix QM getters to always return a valid pq
Denis Bolotin [Mon, 19 Nov 2018 14:28:31 +0000 (16:28 +0200)]
qed: Fix QM getters to always return a valid pq

The getter callers doesn't know the valid Physical Queues (PQ) values.
This patch makes sure that a valid PQ will always be returned.

The patch consists of 3 fixes:

 - When qed_init_qm_get_idx_from_flags() receives a disabled flag, it
   returned PQ 0, which can potentially be another function's pq. Verify
   that flag is enabled, otherwise return default start_pq.

 - When qed_init_qm_get_idx_from_flags() receives an unknown flag, it
   returned NULL and could lead to a segmentation fault. Return default
   start_pq instead.

 - A modulo operation was added to MCOS/VFS PQ getters to make sure the
   PQ returned is in range of the required flag.

Fixes: b5a9ee7cf3be ("qed: Revise QM cofiguration")
Signed-off-by: Denis Bolotin <denis.bolotin@cavium.com>
Signed-off-by: Michal Kalderon <michal.kalderon@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoqed: Fix bitmap_weight() check
Denis Bolotin [Mon, 19 Nov 2018 14:28:30 +0000 (16:28 +0200)]
qed: Fix bitmap_weight() check

Fix the condition which verifies that only one flag is set. The API
bitmap_weight() should receive size in bits instead of bytes.

Fixes: b5a9ee7cf3be ("qed: Revise QM cofiguration")
Signed-off-by: Denis Bolotin <denis.bolotin@cavium.com>
Signed-off-by: Michal Kalderon <michal.kalderon@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/mlx5e: Fix failing ethtool query on FEC query error
Shay Agroskin [Thu, 8 Nov 2018 12:23:36 +0000 (14:23 +0200)]
net/mlx5e: Fix failing ethtool query on FEC query error

If FEC caps query fails when executing 'ethtool <interface>'
the whole callback fails unnecessarily, fixed that by replacing the
error return code with debug logging only.

Fixes: 6cfa94605091 ("net/mlx5e: Ethtool driver callback for query/set FEC policy")
Signed-off-by: Shay Agroskin <shayag@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Removed unnecessary warnings in FEC caps query
Shay Agroskin [Sun, 28 Oct 2018 07:06:11 +0000 (09:06 +0200)]
net/mlx5e: Removed unnecessary warnings in FEC caps query

Querying interface FEC caps with 'ethtool [int]' after link reset
throws warning regading link speed.
This warning is not needed as there is already an indication in
user space that the link is not up.

Fixes: 0696d60853d5 ("net/mlx5e: Receive buffer configuration")
Signed-off-by: Shay Agroskin <shayag@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Fix wrong field name in FEC related functions
Shay Agroskin [Sun, 28 Oct 2018 07:17:29 +0000 (09:17 +0200)]
net/mlx5e: Fix wrong field name in FEC related functions

This bug would result in reading wrong FEC capabilities for 10G/40G.

Fixes: 2095b2641477 ("net/mlx5e: Add port FEC get/set functions")
Signed-off-by: Shay Agroskin <shayag@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Fix a bug in turning off FEC policy in unsupported speeds
Shay Agroskin [Sun, 28 Oct 2018 14:13:46 +0000 (16:13 +0200)]
net/mlx5e: Fix a bug in turning off FEC policy in unsupported speeds

Some speeds don't support turning FEC policy off. In case a requested
FEC policy is not supported for a speed (including current speed), its new
FEC policy would be:
no FEC - if disabling FEC is supported for that speed
unchanged - else

Fixes: 2095b2641477 ("net/mlx5e: Add port FEC get/set functions")
Signed-off-by: Shay Agroskin <shayag@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agoMerge branch 'ena-hibernation-and-rmmod-bug-fixes'
David S. Miller [Mon, 19 Nov 2018 23:13:00 +0000 (15:13 -0800)]
Merge branch 'ena-hibernation-and-rmmod-bug-fixes'

Arthur Kiyanovski says:

====================
net: ena: hibernation and rmmod bug fixes

This patchset includes 2 bug fixes:
1. A fix to a crash during resume from hibernation.
2. A fix to an illegal memory access during driver removal (e.g. during rmmod)
   which might cause a crash in certain systems.

The subminor number in the driver version is also promoted to indicate driver
was changed.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: ena: update driver version from 2.0.1 to 2.0.2
Arthur Kiyanovski [Mon, 19 Nov 2018 10:05:22 +0000 (12:05 +0200)]
net: ena: update driver version from 2.0.1 to 2.0.2

Update driver version due to critical bug fixes.

Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: ena: fix crash during ena_remove()
Arthur Kiyanovski [Mon, 19 Nov 2018 10:05:21 +0000 (12:05 +0200)]
net: ena: fix crash during ena_remove()

In ena_remove() we have the following stack call:
ena_remove()
  unregister_netdev()
  ena_destroy_device()
    netif_carrier_off()

Calling netif_carrier_off() causes linkwatch to try to handle the
link change event on the already unregistered netdev, which leads
to a read from an unreadable memory address.

This patch switches the order of the two functions, so that
netif_carrier_off() is called on a regiestered netdev.

To accomplish this fix we also had to:
1. Remove the set bit ENA_FLAG_TRIGGER_RESET
2. Add a sanitiy check in ena_close()
both to prevent double device reset (when calling unregister_netdev()
ena_close is called, but the device was already deleted in
ena_destroy_device()).
3. Set the admin_queue running state to false to avoid using it after
device was reset (for example when calling ena_destroy_all_io_queues()
right after ena_com_dev_reset() in ena_down)

Fixes: 944b28aa2982 ("net: ena: fix missing lock during device destruction")
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: ena: fix crash during failed resume from hibernation
Arthur Kiyanovski [Mon, 19 Nov 2018 10:05:20 +0000 (12:05 +0200)]
net: ena: fix crash during failed resume from hibernation

During resume from hibernation if ena_restore_device fails,
ena_com_dev_reset() is called, and uses the readless read mechanism,
which was already destroyed by the call to
ena_com_mmio_reg_read_request_destroy(). This causes a NULL pointer
reference.

In this commit we switch the call order of the above two functions
to avoid this crash.

Fixes: d7703ddbd7c9 ("net: ena: fix rare bug when failed restart/resume is followed by driver removal")
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agosctp: not increase stream's incnt before sending addstrm_in request
Xin Long [Sun, 18 Nov 2018 13:59:49 +0000 (21:59 +0800)]
sctp: not increase stream's incnt before sending addstrm_in request

Different from processing the addstrm_out request, The receiver handles
an addstrm_in request by sending back an addstrm_out request to the
sender who will increase its stream's in and incnt later.

Now stream->incnt has been increased since it sent out the addstrm_in
request in sctp_send_add_streams(), with the wrong stream->incnt will
even cause crash when copying stream info from the old stream's in to
the new one's in sctp_process_strreset_addstrm_out().

This patch is to fix it by simply removing the stream->incnt change
from sctp_send_add_streams().

Fixes: 242bd2d519d7 ("sctp: implement sender-side procedures for Add Incoming/Outgoing Streams Request Parameter")
Reported-by: Jianwen Ji <jiji@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/mlx5e: Fix selftest for small MTUs
Valentine Fatiev [Wed, 17 Oct 2018 08:45:07 +0000 (11:45 +0300)]
net/mlx5e: Fix selftest for small MTUs

Loopback test had fixed packet size, which can be bigger than configured
MTU. Shorten the loopback packet size to be bigger than minimal MTU
allowed by the device. Text field removed from struct 'mlx5ehdr'
as redundant to allow send small packets as minimal allowed MTU.

Fixes: d605d66 ("net/mlx5e: Add support for ethtool self diagnostics test")
Signed-off-by: Valentine Fatiev <valentinef@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: RX, verify received packet size in Linear Striding RQ
Moshe Shemesh [Thu, 11 Oct 2018 04:31:10 +0000 (07:31 +0300)]
net/mlx5e: RX, verify received packet size in Linear Striding RQ

In case of striding RQ, we use  MPWRQ (Multi Packet WQE RQ), which means
that WQE (RX descriptor) can be used for many packets and so the WQE is
much bigger than MTU.  In virtualization setups where the port mtu can
be larger than the vf mtu, if received packet is bigger than MTU, it
won't be dropped by HW on too small receive WQE. If we use linear SKB in
striding RQ, since each stride has room for mtu size payload and skb
info, an oversized packet can lead to crash for crossing allocated page
boundary upon the call to build_skb. So driver needs to check packet
size and drop it.

Introduce new SW rx counter, rx_oversize_pkts_sw_drop, which counts the
number of packets dropped by the driver for being too large.

As a new field is added to the RQ struct, re-open the channels whenever
this field is being used in datapath (i.e., in the case of linear
Striding RQ).

Fixes: 619a8f2a42f1 ("net/mlx5e: Use linear SKB in Striding RQ")
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Apply the correct check for supporting TC esw rules split
Roi Dayan [Tue, 23 Oct 2018 14:30:04 +0000 (17:30 +0300)]
net/mlx5e: Apply the correct check for supporting TC esw rules split

The mirror and not the output count is the one denoting a split.
Fix to condition the offload attempt on the mirror count being > 0
along the firmware to have the related capability.

Fixes: 592d36515969 ("net/mlx5e: Parse mirroring action for offloaded TC eswitch flows")
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Yossi Kuperman <yossiku@mellanox.com>
Reviewed-by: Chris Mi <chrism@mellanox.com>
Acked-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Adjust to max number of channles when re-attaching
Yuval Avnery [Tue, 16 Oct 2018 20:20:20 +0000 (15:20 -0500)]
net/mlx5e: Adjust to max number of channles when re-attaching

When core driver enters deattach/attach flow after pci reset,
Number of logical CPUs may have changed.
As a result we need to update the cpu affiliated resource tables.
1. indirect rqt list
2. eq table

Reproduction (PowerPC):
echo 1000 > /sys/kernel/debug/powerpc/eeh_max_freezes
ppc64_cpu --smt=on
# Restart driver
modprobe -r ... ; modprobe ...
# Link up
ifconfig ...
# Only physical CPUs
ppc64_cpu --smt=off
# Inject PCI errors so PCI will reset - calling the pci error handler
echo 0x8000000000000000 > /sys/kernel/debug/powerpc/<PCI BUS>/err_injct_inboundA

Call trace when trying to add non-existing rqs to an indirect rqt:
mlx5e_redirect_rqt+0x84/0x260 [mlx5_core] (unreliable)
mlx5e_redirect_rqts+0x188/0x190 [mlx5_core]
mlx5e_activate_priv_channels+0x488/0x570 [mlx5_core]
mlx5e_open_locked+0xbc/0x140 [mlx5_core]
mlx5e_open+0x50/0x130 [mlx5_core]
mlx5e_nic_enable+0x174/0x1b0 [mlx5_core]
mlx5e_attach_netdev+0x154/0x290 [mlx5_core]
mlx5e_attach+0x88/0xd0 [mlx5_core]
mlx5_attach_device+0x168/0x1e0 [mlx5_core]
mlx5_load_one+0x1140/0x1210 [mlx5_core]
mlx5_pci_resume+0x6c/0xf0 [mlx5_core]

Create cq will fail when trying to use non-existing EQ.

Fixes: 89d44f0a6c73 ("net/mlx5_core: Add pci error handlers to mlx5_core driver")
Signed-off-by: Yuval Avnery <yuvalav@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Always use the match level enum when parsing TC rule match
Or Gerlitz [Sun, 28 Oct 2018 10:27:29 +0000 (12:27 +0200)]
net/mlx5e: Always use the match level enum when parsing TC rule match

We get the match level (none, l2, l3, l4) while going over the match
dissectors of an offloaded tc rule. When doing this, the match level
enum and the not min inline enum values should be used, fix that.

This worked accidentally b/c both enums have the same numerical values.

Fixes: d708f902989b ('net/mlx5e: Get the required HW match level while parsing TC flow matches')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Claim TC hw offloads support only under a proper build config
Or Gerlitz [Thu, 18 Oct 2018 10:31:27 +0000 (12:31 +0200)]
net/mlx5e: Claim TC hw offloads support only under a proper build config

Currently, we are only supporting tc hw offloads when the eswitch
support is compiled in, but we are not gating the adevertizment
of the NETIF_F_HW_TC feature on this config being set.

Fix it, and while doing that, also avoid dealing with the feature
on ethtool when the config is not set.

Fixes: e8f887ac6a45 ('net/mlx5e: Introduce tc offload support')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: Don't match on vlan non-existence if ethertype is wildcarded
Or Gerlitz [Thu, 25 Oct 2018 15:41:58 +0000 (15:41 +0000)]
net/mlx5e: Don't match on vlan non-existence if ethertype is wildcarded

For the "all" ethertype we should not care whether the packet has
vlans. Besides being wrong, the way we did it caused FW error
for rules such as:

tc filter add dev eth0 protocol all parent ffff: \
prio 1 flower skip_sw action drop

b/c the matching meta-data (outer headers bit in struct mlx5_flow_spec)
wasn't set. Fix that by matching on vlan non-existence only if we were
also told to match on the ethertype.

Fixes: cee26487620b ('net/mlx5e: Set vlan masks for all offloaded TC rules')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reported-by: Slava Ovsiienko <viacheslavo@mellanox.com>
Reviewed-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5e: IPoIB, Reset QP after channels are closed
Denis Drozdov [Thu, 27 Sep 2018 11:17:54 +0000 (14:17 +0300)]
net/mlx5e: IPoIB, Reset QP after channels are closed

The mlx5e channels should be closed before mlx5i_uninit_underlay_qp
puts the QP into RST (reset) state during mlx5i_close. Currently QP
state incorrectly set to RST before channels got deactivated and closed,
since mlx5_post_send request expects QP in RTS (Ready To Send) state.

The fix is to keep QP in RTS state until mlx5e channels get closed
and to reset QP afterwards.

Also this fix is simply correct in order to keep the open/close flow
symmetric, i.e mlx5i_init_underlay_qp() is called first thing at open,
the correct thing to do is to call mlx5i_uninit_underlay_qp() last thing
at close, which is exactly what this patch is doing.

Fixes: dae37456c8ac ("net/mlx5: Support for attaching multiple underlay QPs to root flow table")
Signed-off-by: Denis Drozdov <denisd@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agonet/mlx5: IPSec, Fix the SA context hash key
Raed Salem [Thu, 18 Oct 2018 05:55:21 +0000 (08:55 +0300)]
net/mlx5: IPSec, Fix the SA context hash key

The commit "net/mlx5: Refactor accel IPSec code" introduced a
bug where asynchronous short time change in hash key value
by create/release SA context might happen during an asynchronous
hash resize operation this could cause a subsequent remove SA
context operation to fail as the key value used during resize is
not the same key value used when remove SA context operation is
invoked.

This commit fixes the bug by defining the SA context hash key
such that it includes only fields that never change during the
lifetime of the SA context object.

Fixes: d6c4f0298cec ("net/mlx5: Refactor accel IPSec code")
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Aviad Yehezkel <aviadye@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
6 years agoRevert "sctp: remove sctp_transport_pmtu_check"
Xin Long [Sun, 18 Nov 2018 08:14:47 +0000 (16:14 +0800)]
Revert "sctp: remove sctp_transport_pmtu_check"

This reverts commit 22d7be267eaa8114dcc28d66c1c347f667d7878a.

The dst's mtu in transport can be updated by a non sctp place like
in xfrm where the MTU information didn't get synced between asoc,
transport and dst, so it is still needed to do the pmtu check
in sctp_packet_config.

Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agosctp: not allow to set asoc prsctp_enable by sockopt
Xin Long [Sun, 18 Nov 2018 07:21:53 +0000 (15:21 +0800)]
sctp: not allow to set asoc prsctp_enable by sockopt

As rfc7496#section4.5 says about SCTP_PR_SUPPORTED:

   This socket option allows the enabling or disabling of the
   negotiation of PR-SCTP support for future associations.  For existing
   associations, it allows one to query whether or not PR-SCTP support
   was negotiated on a particular association.

It means only sctp sock's prsctp_enable can be set.

Note that for the limitation of SCTP_{CURRENT|ALL}_ASSOC, we will
add it when introducing SCTP_{FUTURE|CURRENT|ALL}_ASSOC for linux
sctp in another patchset.

v1->v2:
  - drop the params.assoc_id check as Neil suggested.

Fixes: 28aa4c26fce2 ("sctp: add SCTP_PR_SUPPORTED on sctp sockopt")
Reported-by: Ying Xu <yinxu@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agosctp: count sk_wmem_alloc by skb truesize in sctp_packet_transmit
Xin Long [Sun, 18 Nov 2018 07:07:38 +0000 (15:07 +0800)]
sctp: count sk_wmem_alloc by skb truesize in sctp_packet_transmit

Now sctp increases sk_wmem_alloc by 1 when doing set_owner_w for the
skb allocked in sctp_packet_transmit and decreases by 1 when freeing
this skb.

But when this skb goes through networking stack, some subcomponents
might change skb->truesize and add the same amount on sk_wmem_alloc.
However sctp doesn't know the amount to decrease by, it would cause
a leak on sk->sk_wmem_alloc and the sock can never be freed.

Xiumei found this issue when it hit esp_output_head() by using sctp
over ipsec, where skb->truesize is added and so is sk->sk_wmem_alloc.

Since sctp has used sk_wmem_queued to count for writable space since
Commit cd305c74b0f8 ("sctp: use sk_wmem_queued to check for writable
space"), it's ok to fix it by counting sk_wmem_alloc by skb truesize
in sctp_packet_transmit.

Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible")
Reported-by: Xiumei Mu <xmu@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMAINTAINERS: Add myself as third phylib maintainer
Heiner Kallweit [Mon, 19 Nov 2018 07:01:03 +0000 (08:01 +0100)]
MAINTAINERS: Add myself as third phylib maintainer

Add myself as third phylib maintainer.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Acked-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Mon, 19 Nov 2018 17:24:04 +0000 (09:24 -0800)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) Fix some potentially uninitialized variables and use-after-free in
    kvaser_usb can drier, from Jimmy Assarsson.

 2) Fix leaks in qed driver, from Denis Bolotin.

 3) Socket leak in l2tp, from Xin Long.

 4) RSS context allocation fix in bnxt_en from Michael Chan.

 5) Fix cxgb4 build errors, from Ganesh Goudar.

 6) Route leaks in ipv6 when removing exceptions, from Xin Long.

 7) Memory leak in IDR allocation handling of act_pedit, from Davide
    Caratti.

 8) Use-after-free of bridge vlan stats, from Nikolay Aleksandrov.

 9) When MTU is locked, do not force DF bit on ipv4 tunnels. From
    Sabrina Dubroca.

10) When NAPI cached skb is reused, we must set it to the proper initial
    state which includes skb->pkt_type. From Eric Dumazet.

11) Lockdep and non-linear SKB handling fix in tipc from Jon Maloy.

12) Set RX queue properly in various tuntap receive paths, from Matthew
    Cover.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (61 commits)
  tuntap: fix multiqueue rx
  ipv6: Fix PMTU updates for UDP/raw sockets in presence of VRF
  tipc: don't assume linear buffer when reading ancillary data
  tipc: fix lockdep warning when reinitilaizing sockets
  net-gro: reset skb->pkt_type in napi_reuse_skb()
  tc-testing: tdc.py: Guard against lack of returncode in executed command
  tc-testing: tdc.py: ignore errors when decoding stdout/stderr
  ip_tunnel: don't force DF when MTU is locked
  MAINTAINERS: Add entry for CAKE qdisc
  net: bridge: fix vlan stats use-after-free on destruction
  socket: do a generic_file_splice_read when proto_ops has no splice_read
  net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs
  Revert "net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs"
  net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs
  net/sched: act_pedit: fix memory leak when IDR allocation fails
  net: lantiq: Fix returned value in case of error in 'xrx200_probe()'
  ipv6: fix a dst leak when removing its exception
  net: mvneta: Don't advertise 2.5G modes
  drivers/net/ethernet/qlogic/qed/qed_rdma.h: fix typo
  net/mlx4: Fix UBSAN warning of signed integer overflow
  ...

6 years agotuntap: fix multiqueue rx
Matthew Cover [Sun, 18 Nov 2018 07:46:00 +0000 (00:46 -0700)]
tuntap: fix multiqueue rx

When writing packets to a descriptor associated with a combined queue, the
packets should end up on that queue.

Before this change all packets written to any descriptor associated with a
tap interface end up on rx-0, even when the descriptor is associated with a
different queue.

The rx traffic can be generated by either of the following.
  1. a simple tap program which spins up multiple queues and writes packets
     to each of the file descriptors
  2. tx from a qemu vm with a tap multiqueue netdev

The queue for rx traffic can be observed by either of the following (done
on the hypervisor in the qemu case).
  1. a simple netmap program which opens and reads from per-queue
     descriptors
  2. configuring RPS and doing per-cpu captures with rxtxcpu

Alternatively, if you printk() the return value of skb_get_rx_queue() just
before each instance of netif_receive_skb() in tun.c, you will get 65535
for every skb.

Calling skb_record_rx_queue() to set the rx queue to the queue_index fixes
the association between descriptor and rx queue.

Signed-off-by: Matthew Cover <matthew.cover@stackpath.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoipv6: Fix PMTU updates for UDP/raw sockets in presence of VRF
David Ahern [Sun, 18 Nov 2018 18:45:30 +0000 (10:45 -0800)]
ipv6: Fix PMTU updates for UDP/raw sockets in presence of VRF

Preethi reported that PMTU discovery for UDP/raw applications is not
working in the presence of VRF when the socket is not bound to a device.
The problem is that ip6_sk_update_pmtu does not consider the L3 domain
of the skb device if the socket is not bound. Update the function to
set oif to the L3 master device if relevant.

Fixes: ca254490c8df ("net: Add VRF support to IPv6 stack")
Reported-by: Preethi Ramachandra <preethir@juniper.net>
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoLinux 4.20-rc3
Linus Torvalds [Sun, 18 Nov 2018 21:33:44 +0000 (13:33 -0800)]
Linux 4.20-rc3

6 years agoMerge tag 'libnvdimm-fixes-4.20-rc3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 18 Nov 2018 20:21:09 +0000 (12:21 -0800)]
Merge tag 'libnvdimm-fixes-4.20-rc3' of git://git./linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm fixes from Dan Williams:
 "A small batch of fixes for v4.20-rc3.

  The overflow continuation fix addresses something that has been broken
  for several releases. Arguably it could wait even longer, but it's a
  one line fix and this finishes the last of the known address range
  scrub bug reports. The revert addresses a lockdep regression. The unit
  tests are not critical to fix, but no reason to hold this fix back.

  Summary:

   - Address Range Scrub overflow continuation handling has been broken
     since it was initially merged. It was only recently that error
     injection and platform-BIOS support enabled this corner case to be
     exercised.

   - The recent attempt to provide more isolation for the kernel Address
     Range Scrub state machine from userapace initiated sessions
     triggers a lockdep report. Revert and try again at the next merge
     window.

   - Fix a kasan reported buffer overflow in libnvdimm unit test
     infrastrucutre (nfit_test)"

* tag 'libnvdimm-fixes-4.20-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  Revert "acpi, nfit: Further restrict userspace ARS start requests"
  acpi, nfit: Fix ARS overflow continuation
  tools/testing/nvdimm: Fix the array size for dimm devices.

6 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Sun, 18 Nov 2018 19:31:26 +0000 (11:31 -0800)]
Merge branch 'akpm' (patches from Andrew)

Merge misc fixes from Andrew Morton:
 "16 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  mm/memblock.c: fix a typo in __next_mem_pfn_range() comments
  mm, page_alloc: check for max order in hot path
  scripts/spdxcheck.py: make python3 compliant
  tmpfs: make lseek(SEEK_DATA/SEK_HOLE) return ENXIO with a negative offset
  lib/ubsan.c: don't mark __ubsan_handle_builtin_unreachable as noreturn
  mm/vmstat.c: fix NUMA statistics updates
  mm/gup.c: fix follow_page_mask() kerneldoc comment
  ocfs2: free up write context when direct IO failed
  scripts/faddr2line: fix location of start_kernel in comment
  mm: don't reclaim inodes with many attached pages
  mm, memory_hotplug: check zone_movable in has_unmovable_pages
  mm/swapfile.c: use kvzalloc for swap_info_struct allocation
  MAINTAINERS: update OMAP MMC entry
  hugetlbfs: fix kernel BUG at fs/hugetlbfs/inode.c:444!
  kernel/sched/psi.c: simplify cgroup_move_task()
  z3fold: fix possible reclaim races

6 years agoMerge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 18 Nov 2018 18:58:20 +0000 (10:58 -0800)]
Merge branch 'sched-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull scheduler fix from Ingo Molnar:
 "Fix an exec() related scalability/performance regression, which was
  caused by incorrectly calculating load and migrating tasks on exec()
  when they shouldn't be"

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/fair: Fix cpu_util_wake() for 'execl' type workloads

6 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 18 Nov 2018 18:54:59 +0000 (10:54 -0800)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf fixes from Ingo Molnar:
 "Fix uncore PMU enumeration for CofeeLake CPUs"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86/intel/uncore: Support CoffeeLake 8th CBOX
  perf/x86/intel/uncore: Add more IMC PCI IDs for KabyLake and CoffeeLake CPUs

6 years agoMerge branch 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 18 Nov 2018 18:52:26 +0000 (10:52 -0800)]
Merge branch 'efi-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull EFI fixes from Ingo Molnar:
 "Misc fixes: two warning splat fixes, a leak fix and persistent memory
  allocation fixes for ARM"

* 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi: Permit calling efi_mem_reserve_persistent() from atomic context
  efi/arm: Defer persistent reservations until after paging_init()
  efi/arm/libstub: Pack FDT after populating it
  efi/arm: Revert deferred unmap of early memmap mapping
  efi: Fix debugobjects warning on 'efi_rts_work'

6 years agoMerge branch 'spectre' of git://git.armlinux.org.uk/~rmk/linux-arm
Linus Torvalds [Sun, 18 Nov 2018 18:45:09 +0000 (10:45 -0800)]
Merge branch 'spectre' of git://git.armlinux.org.uk/~rmk/linux-arm

Pull ARM spectre updates from Russell King:
 "These are the currently known final bits that resolve the Spectre
  issues. big.Little systems used to be sufficiently identical in that
  there were no differences between individual CPUs in the system that
  mattered to the kernel. With the advent of the Spectre problem, the
  CPUs now have differences in how the workaround is applied.

  As a result of previous Spectre patches, these systems ended up
  reporting quite a lot of:

     "CPUx: Spectre v2: incorrect context switching function, system vulnerable"

  messages due to the action of the big.Little switcher causing the CPUs
  to be re-initialised regularly. This series resolves that issue by
  making the CPU vtable unique to each CPU.

  However, since this is used very early, before per-cpu is setup,
  per-cpu can't be used. We also have a problem that two of the methods
  are not called from preempt-safe paths, but thankfully these remain
  identical between all CPUs in the system. To make sure, we validate
  that these are identical during boot"

* 'spectre' of git://git.armlinux.org.uk/~rmk/linux-arm:
  ARM: spectre-v2: per-CPU vtables to work around big.Little systems
  ARM: add PROC_VTABLE and PROC_TABLE macros
  ARM: clean up per-processor check_bugs method call
  ARM: split out processor lookup
  ARM: make lookup_processor_type() non-__init

6 years agomm/memblock.c: fix a typo in __next_mem_pfn_range() comments
Chen Chang [Fri, 16 Nov 2018 23:08:57 +0000 (15:08 -0800)]
mm/memblock.c: fix a typo in __next_mem_pfn_range() comments

Link: http://lkml.kernel.org/r/20181107100247.13359-1-rainccrun@gmail.com
Signed-off-by: Chen Chang <rainccrun@gmail.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agomm, page_alloc: check for max order in hot path
Michal Hocko [Fri, 16 Nov 2018 23:08:53 +0000 (15:08 -0800)]
mm, page_alloc: check for max order in hot path

Konstantin has noticed that kvmalloc might trigger the following
warning:

  WARNING: CPU: 0 PID: 6676 at mm/vmstat.c:986 __fragmentation_index+0x54/0x60
  [...]
  Call Trace:
   fragmentation_index+0x76/0x90
   compaction_suitable+0x4f/0xf0
   shrink_node+0x295/0x310
   node_reclaim+0x205/0x250
   get_page_from_freelist+0x649/0xad0
   __alloc_pages_nodemask+0x12a/0x2a0
   kmalloc_large_node+0x47/0x90
   __kmalloc_node+0x22b/0x2e0
   kvmalloc_node+0x3e/0x70
   xt_alloc_table_info+0x3a/0x80 [x_tables]
   do_ip6t_set_ctl+0xcd/0x1c0 [ip6_tables]
   nf_setsockopt+0x44/0x60
   SyS_setsockopt+0x6f/0xc0
   do_syscall_64+0x67/0x120
   entry_SYSCALL_64_after_hwframe+0x3d/0xa2

the problem is that we only check for an out of bound order in the slow
path and the node reclaim might happen from the fast path already.  This
is fixable by making sure that kvmalloc doesn't ever use kmalloc for
requests that are larger than KMALLOC_MAX_SIZE but this also shows that
the code is rather fragile.  A recent UBSAN report just underlines that
by the following report

  UBSAN: Undefined behaviour in mm/page_alloc.c:3117:19
  shift exponent 51 is too large for 32-bit type 'int'
  CPU: 0 PID: 6520 Comm: syz-executor1 Not tainted 4.19.0-rc2 #1
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
  Call Trace:
   __dump_stack lib/dump_stack.c:77 [inline]
   dump_stack+0xd2/0x148 lib/dump_stack.c:113
   ubsan_epilogue+0x12/0x94 lib/ubsan.c:159
   __ubsan_handle_shift_out_of_bounds+0x2b6/0x30b lib/ubsan.c:425
   __zone_watermark_ok+0x2c7/0x400 mm/page_alloc.c:3117
   zone_watermark_fast mm/page_alloc.c:3216 [inline]
   get_page_from_freelist+0xc49/0x44c0 mm/page_alloc.c:3300
   __alloc_pages_nodemask+0x21e/0x640 mm/page_alloc.c:4370
   alloc_pages_current+0xcc/0x210 mm/mempolicy.c:2093
   alloc_pages include/linux/gfp.h:509 [inline]
   __get_free_pages+0x12/0x60 mm/page_alloc.c:4414
   dma_mem_alloc+0x36/0x50 arch/x86/include/asm/floppy.h:156
   raw_cmd_copyin drivers/block/floppy.c:3159 [inline]
   raw_cmd_ioctl drivers/block/floppy.c:3206 [inline]
   fd_locked_ioctl+0xa00/0x2c10 drivers/block/floppy.c:3544
   fd_ioctl+0x40/0x60 drivers/block/floppy.c:3571
   __blkdev_driver_ioctl block/ioctl.c:303 [inline]
   blkdev_ioctl+0xb3c/0x1a30 block/ioctl.c:601
   block_ioctl+0x105/0x150 fs/block_dev.c:1883
   vfs_ioctl fs/ioctl.c:46 [inline]
   do_vfs_ioctl+0x1c0/0x1150 fs/ioctl.c:687
   ksys_ioctl+0x9e/0xb0 fs/ioctl.c:702
   __do_sys_ioctl fs/ioctl.c:709 [inline]
   __se_sys_ioctl fs/ioctl.c:707 [inline]
   __x64_sys_ioctl+0x7e/0xc0 fs/ioctl.c:707
   do_syscall_64+0xc4/0x510 arch/x86/entry/common.c:290
   entry_SYSCALL_64_after_hwframe+0x49/0xbe

Note that this is not a kvmalloc path.  It is just that the fast path
really depends on having sanitzed order as well.  Therefore move the
order check to the fast path.

Link: http://lkml.kernel.org/r/20181113094305.GM15120@dhcp22.suse.cz
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reported-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Reported-by: Kyungtae Kim <kt0755@gmail.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Pavel Tatashin <pavel.tatashin@microsoft.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: Aaron Lu <aaron.lu@intel.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Byoungyoung Lee <lifeasageek@gmail.com>
Cc: "Dae R. Jeong" <threeearcat@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agoscripts/spdxcheck.py: make python3 compliant
Uwe Kleine-König [Fri, 16 Nov 2018 23:08:43 +0000 (15:08 -0800)]
scripts/spdxcheck.py: make python3 compliant

Without this change the following happens when using Python3 (3.6.6):

$ echo "GPL-2.0" | python3 scripts/spdxcheck.py -
FAIL: 'str' object has no attribute 'decode'
Traceback (most recent call last):
  File "scripts/spdxcheck.py", line 253, in <module>
    parser.parse_lines(sys.stdin, args.maxlines, '-')
  File "scripts/spdxcheck.py", line 171, in parse_lines
    line = line.decode(locale.getpreferredencoding(False), errors='ignore')
AttributeError: 'str' object has no attribute 'decode'

So as the line is already a string, there is no need to decode it and
the line can be dropped.

/usr/bin/python on Arch is Python 3.  So this would indeed be worth
going into 4.19.

Link: http://lkml.kernel.org/r/20181023070802.22558-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Joe Perches <joe@perches.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agotmpfs: make lseek(SEEK_DATA/SEK_HOLE) return ENXIO with a negative offset
Yufen Yu [Fri, 16 Nov 2018 23:08:39 +0000 (15:08 -0800)]
tmpfs: make lseek(SEEK_DATA/SEK_HOLE) return ENXIO with a negative offset

Other filesystems such as ext4, f2fs and ubifs all return ENXIO when
lseek (SEEK_DATA or SEEK_HOLE) requests a negative offset.

man 2 lseek says

:      EINVAL whence  is  not  valid.   Or: the resulting file offset would be
:             negative, or beyond the end of a seekable device.
:
:      ENXIO  whence is SEEK_DATA or SEEK_HOLE, and the file offset is  beyond
:             the end of the file.

Make tmpfs return ENXIO under these circumstances as well.  After this,
tmpfs also passes xfstests's generic/448.

[akpm@linux-foundation.org: rewrite changelog]
Link: http://lkml.kernel.org/r/1540434176-14349-1-git-send-email-yuyufen@huawei.com
Signed-off-by: Yufen Yu <yuyufen@huawei.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Hugh Dickins <hughd@google.com>
Cc: William Kucharski <william.kucharski@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agolib/ubsan.c: don't mark __ubsan_handle_builtin_unreachable as noreturn
Arnd Bergmann [Fri, 16 Nov 2018 23:08:35 +0000 (15:08 -0800)]
lib/ubsan.c: don't mark __ubsan_handle_builtin_unreachable as noreturn

gcc-8 complains about the prototype for this function:

  lib/ubsan.c:432:1: error: ignoring attribute 'noreturn' in declaration of a built-in function '__ubsan_handle_builtin_unreachable' because it conflicts with attribute 'const' [-Werror=attributes]

This is actually a GCC's bug. In GCC internals
__ubsan_handle_builtin_unreachable() declared with both 'noreturn' and
'const' attributes instead of only 'noreturn':

   https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84210

Workaround this by removing the noreturn attribute.

[aryabinin: add information about GCC bug in changelog]
Link: http://lkml.kernel.org/r/20181107144516.4587-1-aryabinin@virtuozzo.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Acked-by: Olof Johansson <olof@lixom.net>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agomm/vmstat.c: fix NUMA statistics updates
Janne Huttunen [Fri, 16 Nov 2018 23:08:32 +0000 (15:08 -0800)]
mm/vmstat.c: fix NUMA statistics updates

Scan through the whole array to see if an update is needed.  While we're
at it, use sizeof() to be safe against any possible type changes in the
future.

The bug here is that we wouldn't sync per-cpu counters into global ones
if there was an update of numa_stats for higher cpus.  Highly
theoretical one though because it is much more probable that zone_stats
are updated so we would refresh anyway.  So I wouldn't bother to mark
this for stable, yet something nice to fix.

[mhocko@suse.com: changelog enhancement]
Link: http://lkml.kernel.org/r/1541601517-17282-1-git-send-email-janne.huttunen@nokia.com
Fixes: 1d90ca897cb0 ("mm: update NUMA counter threshold size")
Signed-off-by: Janne Huttunen <janne.huttunen@nokia.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agomm/gup.c: fix follow_page_mask() kerneldoc comment
Mike Rapoport [Fri, 16 Nov 2018 23:08:29 +0000 (15:08 -0800)]
mm/gup.c: fix follow_page_mask() kerneldoc comment

Commit df06b37ffe5a ("mm/gup: cache dev_pagemap while pinning pages")
modified the signature of follow_page_mask() but left the parameter
description behind.

Update the description to make the code and comments agree again.

While at it, update formatting of the return value description to match
Documentation/doc-guide/kernel-doc.rst guidelines.

Link: http://lkml.kernel.org/r/1541603316-27832-1-git-send-email-rppt@linux.ibm.com
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agoocfs2: free up write context when direct IO failed
Wengang Wang [Fri, 16 Nov 2018 23:08:25 +0000 (15:08 -0800)]
ocfs2: free up write context when direct IO failed

The write context should also be freed even when direct IO failed.
Otherwise a memory leak is introduced and entries remain in
oi->ip_unwritten_list causing the following BUG later in unlink path:

  ERROR: bug expression: !list_empty(&oi->ip_unwritten_list)
  ERROR: Clear inode of 215043, inode has unwritten extents
  ...
  Call Trace:
  ? __set_current_blocked+0x42/0x68
  ocfs2_evict_inode+0x91/0x6a0 [ocfs2]
  ? bit_waitqueue+0x40/0x33
  evict+0xdb/0x1af
  iput+0x1a2/0x1f7
  do_unlinkat+0x194/0x28f
  SyS_unlinkat+0x1b/0x2f
  do_syscall_64+0x79/0x1ae
  entry_SYSCALL_64_after_hwframe+0x151/0x0

This patch also logs, with frequency limit, direct IO failures.

Link: http://lkml.kernel.org/r/20181102170632.25921-1-wen.gang.wang@oracle.com
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com>
Reviewed-by: Junxiao Bi <junxiao.bi@oracle.com>
Reviewed-by: Changwei Ge <ge.changwei@h3c.com>
Reviewed-by: Joseph Qi <jiangqi903@gmail.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agoscripts/faddr2line: fix location of start_kernel in comment
Randy Dunlap [Fri, 16 Nov 2018 23:08:22 +0000 (15:08 -0800)]
scripts/faddr2line: fix location of start_kernel in comment

Fix a source file reference location to the correct path name.

Link: http://lkml.kernel.org/r/1d50bd3d-178e-dcd8-779f-9711887440eb@infradead.org
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agomm: don't reclaim inodes with many attached pages
Roman Gushchin [Fri, 16 Nov 2018 23:08:18 +0000 (15:08 -0800)]
mm: don't reclaim inodes with many attached pages

Spock reported that commit 172b06c32b94 ("mm: slowly shrink slabs with a
relatively small number of objects") leads to a regression on his setup:
periodically the majority of the pagecache is evicted without an obvious
reason, while before the change the amount of free memory was balancing
around the watermark.

The reason behind is that the mentioned above change created some
minimal background pressure on the inode cache.  The problem is that if
an inode is considered to be reclaimed, all belonging pagecache page are
stripped, no matter how many of them are there.  So, if a huge
multi-gigabyte file is cached in the memory, and the goal is to reclaim
only few slab objects (unused inodes), we still can eventually evict all
gigabytes of the pagecache at once.

The workload described by Spock has few large non-mapped files in the
pagecache, so it's especially noticeable.

To solve the problem let's postpone the reclaim of inodes, which have
more than 1 attached page.  Let's wait until the pagecache pages will be
evicted naturally by scanning the corresponding LRU lists, and only then
reclaim the inode structure.

Link: http://lkml.kernel.org/r/20181023164302.20436-1-guro@fb.com
Signed-off-by: Roman Gushchin <guro@fb.com>
Reported-by: Spock <dairinin@gmail.com>
Tested-by: Spock <dairinin@gmail.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: <stable@vger.kernel.org> [4.19.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agomm, memory_hotplug: check zone_movable in has_unmovable_pages
Michal Hocko [Fri, 16 Nov 2018 23:08:15 +0000 (15:08 -0800)]
mm, memory_hotplug: check zone_movable in has_unmovable_pages

Page state checks are racy.  Under a heavy memory workload (e.g.  stress
-m 200 -t 2h) it is quite easy to hit a race window when the page is
allocated but its state is not fully populated yet.  A debugging patch to
dump the struct page state shows

  has_unmovable_pages: pfn:0x10dfec00, found:0x1, count:0x0
  page:ffffea0437fb0000 count:1 mapcount:1 mapping:ffff880e05239841 index:0x7f26e5000 compound_mapcount: 1
  flags: 0x5fffffc0090034(uptodate|lru|active|head|swapbacked)

Note that the state has been checked for both PageLRU and PageSwapBacked
already.  Closing this race completely would require some sort of retry
logic.  This can be tricky and error prone (think of potential endless
or long taking loops).

Workaround this problem for movable zones at least.  Such a zone should
only contain movable pages.  Commit 15c30bc09085 ("mm, memory_hotplug:
make has_unmovable_pages more robust") has told us that this is not
strictly true though.  Bootmem pages should be marked reserved though so
we can move the original check after the PageReserved check.  Pages from
other zones are still prone to races but we even do not pretend that
memory hotremove works for those so pre-mature failure doesn't hurt that
much.

Link: http://lkml.kernel.org/r/20181106095524.14629-1-mhocko@kernel.org
Fixes: 15c30bc09085 ("mm, memory_hotplug: make has_unmovable_pages more robust")
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reported-by: Baoquan He <bhe@redhat.com>
Tested-by: Baoquan He <bhe@redhat.com>
Acked-by: Baoquan He <bhe@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Acked-by: Balbir Singh <bsingharora@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agomm/swapfile.c: use kvzalloc for swap_info_struct allocation
Vasily Averin [Fri, 16 Nov 2018 23:08:11 +0000 (15:08 -0800)]
mm/swapfile.c: use kvzalloc for swap_info_struct allocation

Commit a2468cc9bfdf ("swap: choose swap device according to numa node")
changed 'avail_lists' field of 'struct swap_info_struct' to an array.
In popular linux distros it increased size of swap_info_struct up to 40
Kbytes and now swap_info_struct allocation requires order-4 page.
Switch to kvzmalloc allows to avoid unexpected allocation failures.

Link: http://lkml.kernel.org/r/fc23172d-3c75-21e2-d551-8b1808cbe593@virtuozzo.com
Fixes: a2468cc9bfdf ("swap: choose swap device according to numa node")
Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Acked-by: Aaron Lu <aaron.lu@intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Huang Ying <ying.huang@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agoMAINTAINERS: update OMAP MMC entry
Aaro Koskinen [Fri, 16 Nov 2018 23:08:08 +0000 (15:08 -0800)]
MAINTAINERS: update OMAP MMC entry

Jarkko's e-mail address hasn't worked for a long time.  We still want to
keep this driver working as it is critical for some of the OMAP boards.
I use and test this driver frequently, so change myself as a maintainer
with "Odd Fixes" status.

Link: http://lkml.kernel.org/r/20181106222750.12939-1-aaro.koskinen@iki.fi
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agohugetlbfs: fix kernel BUG at fs/hugetlbfs/inode.c:444!
Mike Kravetz [Fri, 16 Nov 2018 23:08:04 +0000 (15:08 -0800)]
hugetlbfs: fix kernel BUG at fs/hugetlbfs/inode.c:444!

This bug has been experienced several times by the Oracle DB team.  The
BUG is in remove_inode_hugepages() as follows:

/*
 * If page is mapped, it was faulted in after being
 * unmapped in caller.  Unmap (again) now after taking
 * the fault mutex.  The mutex will prevent faults
 * until we finish removing the page.
 *
 * This race can only happen in the hole punch case.
 * Getting here in a truncate operation is a bug.
 */
if (unlikely(page_mapped(page))) {
BUG_ON(truncate_op);

In this case, the elevated map count is not the result of a race.
Rather it was incorrectly incremented as the result of a bug in the huge
pmd sharing code.  Consider the following:

 - Process A maps a hugetlbfs file of sufficient size and alignment
   (PUD_SIZE) that a pmd page could be shared.

 - Process B maps the same hugetlbfs file with the same size and
   alignment such that a pmd page is shared.

 - Process B then calls mprotect() to change protections for the mapping
   with the shared pmd. As a result, the pmd is 'unshared'.

 - Process B then calls mprotect() again to chage protections for the
   mapping back to their original value. pmd remains unshared.

 - Process B then forks and process C is created. During the fork
   process, we do dup_mm -> dup_mmap -> copy_page_range to copy page
   tables. Copying page tables for hugetlb mappings is done in the
   routine copy_hugetlb_page_range.

In copy_hugetlb_page_range(), the destination pte is obtained by:

dst_pte = huge_pte_alloc(dst, addr, sz);

If pmd sharing is possible, the returned pointer will be to a pte in an
existing page table.  In the situation above, process C could share with
either process A or process B.  Since process A is first in the list,
the returned pte is a pointer to a pte in process A's page table.

However, the check for pmd sharing in copy_hugetlb_page_range is:

/* If the pagetables are shared don't copy or take references */
if (dst_pte == src_pte)
continue;

Since process C is sharing with process A instead of process B, the
above test fails.  The code in copy_hugetlb_page_range which follows
assumes dst_pte points to a huge_pte_none pte.  It copies the pte entry
from src_pte to dst_pte and increments this map count of the associated
page.  This is how we end up with an elevated map count.

To solve, check the dst_pte entry for huge_pte_none.  If !none, this
implies PMD sharing so do not copy.

Link: http://lkml.kernel.org/r/20181105212315.14125-1-mike.kravetz@oracle.com
Fixes: c5c99429fa57 ("fix hugepages leak due to pagetable page sharing")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Prakash Sangappa <prakash.sangappa@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agokernel/sched/psi.c: simplify cgroup_move_task()
Olof Johansson [Fri, 16 Nov 2018 23:08:00 +0000 (15:08 -0800)]
kernel/sched/psi.c: simplify cgroup_move_task()

The existing code triggered an invalid warning about 'rq' possibly being
used uninitialized.  Instead of doing the silly warning suppression by
initializa it to NULL, refactor the code to bail out early instead.

Warning was:

  kernel/sched/psi.c: In function `cgroup_move_task':
  kernel/sched/psi.c:639:13: warning: `rq' may be used uninitialized in this function [-Wmaybe-uninitialized]

Link: http://lkml.kernel.org/r/20181103183339.8669-1-olof@lixom.net
Fixes: 2ce7135adc9ad ("psi: cgroup support")
Signed-off-by: Olof Johansson <olof@lixom.net>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agoz3fold: fix possible reclaim races
Vitaly Wool [Fri, 16 Nov 2018 23:07:56 +0000 (15:07 -0800)]
z3fold: fix possible reclaim races

Reclaim and free can race on an object which is basically fine but in
order for reclaim to be able to map "freed" object we need to encode
object length in the handle.  handle_to_chunks() is then introduced to
extract object length from a handle and use it during mapping.

Moreover, to avoid racing on a z3fold "headless" page release, we should
not try to free that page in z3fold_free() if the reclaim bit is set.
Also, in the unlikely case of trying to reclaim a page being freed, we
should not proceed with that page.

While at it, fix the page accounting in reclaim function.

This patch supersedes "[PATCH] z3fold: fix reclaim lock-ups".

Link: http://lkml.kernel.org/r/20181105162225.74e8837d03583a9b707cf559@gmail.com
Signed-off-by: Vitaly Wool <vitaly.vul@sony.com>
Signed-off-by: Jongseok Kim <ks77sj@gmail.com>
Reported-by-by: Jongseok Kim <ks77sj@gmail.com>
Reviewed-by: Snild Dolkow <snild@sony.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
6 years agotipc: don't assume linear buffer when reading ancillary data
Jon Maloy [Sat, 17 Nov 2018 17:17:06 +0000 (12:17 -0500)]
tipc: don't assume linear buffer when reading ancillary data

The code for reading ancillary data from a received buffer is assuming
the buffer is linear. To make this assumption true we have to linearize
the buffer before message data is read.

Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotipc: fix lockdep warning when reinitilaizing sockets
Jon Maloy [Fri, 16 Nov 2018 21:55:04 +0000 (16:55 -0500)]
tipc: fix lockdep warning when reinitilaizing sockets

We get the following warning:

[   47.926140] 32-bit node address hash set to 2010a0a
[   47.927202]
[   47.927433] ================================
[   47.928050] WARNING: inconsistent lock state
[   47.928661] 4.19.0+ #37 Tainted: G            E
[   47.929346] --------------------------------
[   47.929954] inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
[   47.930116] swapper/3/0 [HC0[0]:SC1[3]:HE1:SE0] takes:
[   47.930116] 00000000af8bc31e (&(&ht->lock)->rlock){+.?.}, at: rhashtable_walk_enter+0x36/0xb0
[   47.930116] {SOFTIRQ-ON-W} state was registered at:
[   47.930116]   _raw_spin_lock+0x29/0x60
[   47.930116]   rht_deferred_worker+0x556/0x810
[   47.930116]   process_one_work+0x1f5/0x540
[   47.930116]   worker_thread+0x64/0x3e0
[   47.930116]   kthread+0x112/0x150
[   47.930116]   ret_from_fork+0x3a/0x50
[   47.930116] irq event stamp: 14044
[   47.930116] hardirqs last  enabled at (14044): [<ffffffff9a07fbba>] __local_bh_enable_ip+0x7a/0xf0
[   47.938117] hardirqs last disabled at (14043): [<ffffffff9a07fb81>] __local_bh_enable_ip+0x41/0xf0
[   47.938117] softirqs last  enabled at (14028): [<ffffffff9a0803ee>] irq_enter+0x5e/0x60
[   47.938117] softirqs last disabled at (14029): [<ffffffff9a0804a5>] irq_exit+0xb5/0xc0
[   47.938117]
[   47.938117] other info that might help us debug this:
[   47.938117]  Possible unsafe locking scenario:
[   47.938117]
[   47.938117]        CPU0
[   47.938117]        ----
[   47.938117]   lock(&(&ht->lock)->rlock);
[   47.938117]   <Interrupt>
[   47.938117]     lock(&(&ht->lock)->rlock);
[   47.938117]
[   47.938117]  *** DEADLOCK ***
[   47.938117]
[   47.938117] 2 locks held by swapper/3/0:
[   47.938117]  #0: 0000000062c64f90 ((&d->timer)){+.-.}, at: call_timer_fn+0x5/0x280
[   47.938117]  #1: 00000000ee39619c (&(&d->lock)->rlock){+.-.}, at: tipc_disc_timeout+0xc8/0x540 [tipc]
[   47.938117]
[   47.938117] stack backtrace:
[   47.938117] CPU: 3 PID: 0 Comm: swapper/3 Tainted: G            E     4.19.0+ #37
[   47.938117] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   47.938117] Call Trace:
[   47.938117]  <IRQ>
[   47.938117]  dump_stack+0x5e/0x8b
[   47.938117]  print_usage_bug+0x1ed/0x1ff
[   47.938117]  mark_lock+0x5b5/0x630
[   47.938117]  __lock_acquire+0x4c0/0x18f0
[   47.938117]  ? lock_acquire+0xa6/0x180
[   47.938117]  lock_acquire+0xa6/0x180
[   47.938117]  ? rhashtable_walk_enter+0x36/0xb0
[   47.938117]  _raw_spin_lock+0x29/0x60
[   47.938117]  ? rhashtable_walk_enter+0x36/0xb0
[   47.938117]  rhashtable_walk_enter+0x36/0xb0
[   47.938117]  tipc_sk_reinit+0xb0/0x410 [tipc]
[   47.938117]  ? mark_held_locks+0x6f/0x90
[   47.938117]  ? __local_bh_enable_ip+0x7a/0xf0
[   47.938117]  ? lockdep_hardirqs_on+0x20/0x1a0
[   47.938117]  tipc_net_finalize+0xbf/0x180 [tipc]
[   47.938117]  tipc_disc_timeout+0x509/0x540 [tipc]
[   47.938117]  ? call_timer_fn+0x5/0x280
[   47.938117]  ? tipc_disc_msg_xmit.isra.19+0xa0/0xa0 [tipc]
[   47.938117]  ? tipc_disc_msg_xmit.isra.19+0xa0/0xa0 [tipc]
[   47.938117]  call_timer_fn+0xa1/0x280
[   47.938117]  ? tipc_disc_msg_xmit.isra.19+0xa0/0xa0 [tipc]
[   47.938117]  run_timer_softirq+0x1f2/0x4d0
[   47.938117]  __do_softirq+0xfc/0x413
[   47.938117]  irq_exit+0xb5/0xc0
[   47.938117]  smp_apic_timer_interrupt+0xac/0x210
[   47.938117]  apic_timer_interrupt+0xf/0x20
[   47.938117]  </IRQ>
[   47.938117] RIP: 0010:default_idle+0x1c/0x140
[   47.938117] Code: 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 44 00 00 41 54 55 53 65 8b 2d d8 2b 74 65 0f 1f 44 00 00 e8 c6 2c 8b ff fb f4 <65> 8b 2d c5 2b 74 65 0f 1f 44 00 00 5b 5d 41 5c c3 65 8b 05 b4 2b
[   47.938117] RSP: 0018:ffffaf6ac0207ec8 EFLAGS: 00000206 ORIG_RAX: ffffffffffffff13
[   47.938117] RAX: ffff8f5b3735e200 RBX: 0000000000000003 RCX: 0000000000000001
[   47.938117] RDX: 0000000000000001 RSI: 0000000000000001 RDI: ffff8f5b3735e200
[   47.938117] RBP: 0000000000000003 R08: 0000000000000001 R09: 0000000000000000
[   47.938117] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
[   47.938117] R13: 0000000000000000 R14: ffff8f5b3735e200 R15: ffff8f5b3735e200
[   47.938117]  ? default_idle+0x1a/0x140
[   47.938117]  do_idle+0x1bc/0x280
[   47.938117]  cpu_startup_entry+0x19/0x20
[   47.938117]  start_secondary+0x187/0x1c0
[   47.938117]  secondary_startup_64+0xa4/0xb0

The reason seems to be that tipc_net_finalize()->tipc_sk_reinit() is
calling the function rhashtable_walk_enter() within a timer interrupt.
We fix this by executing tipc_net_finalize() in work queue context.

Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet-gro: reset skb->pkt_type in napi_reuse_skb()
Eric Dumazet [Sun, 18 Nov 2018 05:57:02 +0000 (21:57 -0800)]
net-gro: reset skb->pkt_type in napi_reuse_skb()

eth_type_trans() assumes initial value for skb->pkt_type
is PACKET_HOST.

This is indeed the value right after a fresh skb allocation.

However, it is possible that GRO merged a packet with a different
value (like PACKET_OTHERHOST in case macvlan is used), so
we need to make sure napi->skb will have pkt_type set back to
PACKET_HOST.

Otherwise, valid packets might be dropped by the stack because
their pkt_type is not PACKET_HOST.

napi_reuse_skb() was added in commit 96e93eab2033 ("gro: Add
internal interfaces for VLAN"), but this bug always has
been there.

Fixes: 96e93eab2033 ("gro: Add internal interfaces for VLAN")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge branch 'tdc-fixes'
David S. Miller [Sun, 18 Nov 2018 05:54:53 +0000 (21:54 -0800)]
Merge branch 'tdc-fixes'

Lucas Bates says:

====================
Prevent uncaught exceptions in tdc

This patch series addresses two potential bugs in tdc that can
cause exceptions to be raised in certain circumstances.  These
exceptions are generally not handled, so instead we will prevent
them from being raised.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotc-testing: tdc.py: Guard against lack of returncode in executed command
Brenda J. Butler [Fri, 16 Nov 2018 22:37:56 +0000 (17:37 -0500)]
tc-testing: tdc.py: Guard against lack of returncode in executed command

Add some defensive coding in case one of the subprocesses created by tdc
returns nothing. If no object is returned from exec_cmd, then tdc will
halt with an unhandled exception.

Signed-off-by: Brenda J. Butler <bjb@mojatatu.com>
Signed-off-by: Lucas Bates <lucasb@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agotc-testing: tdc.py: ignore errors when decoding stdout/stderr
Lucas Bates [Fri, 16 Nov 2018 22:37:55 +0000 (17:37 -0500)]
tc-testing: tdc.py: ignore errors when decoding stdout/stderr

Prevent exceptions from being raised while decoding output
from an executed command. There is no impact on tdc's
execution and the verify command phase would fail the pattern
match.

Signed-off-by: Lucas Bates <lucasb@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoip_tunnel: don't force DF when MTU is locked
Sabrina Dubroca [Fri, 16 Nov 2018 15:58:19 +0000 (16:58 +0100)]
ip_tunnel: don't force DF when MTU is locked

The various types of tunnels running over IPv4 can ask to set the DF
bit to do PMTU discovery. However, PMTU discovery is subject to the
threshold set by the net.ipv4.route.min_pmtu sysctl, and is also
disabled on routes with "mtu lock". In those cases, we shouldn't set
the DF bit.

This patch makes setting the DF bit conditional on the route's MTU
locking state.

This issue seems to be older than git history.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMAINTAINERS: Add entry for CAKE qdisc
Toke Høiland-Jørgensen [Fri, 16 Nov 2018 20:13:59 +0000 (12:13 -0800)]
MAINTAINERS: Add entry for CAKE qdisc

We would like the existing community to be kept in the loop for any new
developments on CAKE; and I certainly plan to keep maintaining it. Reflect
this in MAINTAINERS.

Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: bridge: fix vlan stats use-after-free on destruction
Nikolay Aleksandrov [Fri, 16 Nov 2018 16:50:01 +0000 (18:50 +0200)]
net: bridge: fix vlan stats use-after-free on destruction

Syzbot reported a use-after-free of the global vlan context on port vlan
destruction. When I added per-port vlan stats I missed the fact that the
global vlan context can be freed before the per-port vlan rcu callback.
There're a few different ways to deal with this, I've chosen to add a
new private flag that is set only when per-port stats are allocated so
we can directly check it on destruction without dereferencing the global
context at all. The new field in net_bridge_vlan uses a hole.

v2: cosmetic change, move the check to br_process_vlan_info where the
    other checks are done
v3: add change log in the patch, add private (in-kernel only) flags in a
    hole in net_bridge_vlan struct and use that instead of mixing
    user-space flags with private flags

Fixes: 9163a0fc1f0c ("net: bridge: add support for per-port vlan stats")
Reported-by: syzbot+04681da557a0e49a52e5@syzkaller.appspotmail.com
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agosocket: do a generic_file_splice_read when proto_ops has no splice_read
Slavomir Kaslev [Fri, 16 Nov 2018 09:27:53 +0000 (11:27 +0200)]
socket: do a generic_file_splice_read when proto_ops has no splice_read

splice(2) fails with -EINVAL when called reading on a socket with no splice_read
set in its proto_ops (such as vsock sockets). Switch this to fallbacks to a
generic_file_splice_read instead.

Signed-off-by: Slavomir Kaslev <kaslevs@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: mdio-gpio: Fix working over slow can_sleep GPIOs
Martin Schiller [Fri, 16 Nov 2018 07:38:36 +0000 (08:38 +0100)]
net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs

Up until commit 7e5fbd1e0700 ("net: mdio-gpio: Convert to use gpiod
functions where possible"), the _cansleep variants of the gpio_ API was
used. After that commit and the change to gpiod_ API, the _cansleep()
was dropped. This then results in WARN_ON() when used with GPIO
devices which do sleep. Add back the _cansleep() to avoid this.

Fixes: 7e5fbd1e0700 ("net: mdio-gpio: Convert to use gpiod functions where possible")
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoRevert "net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs"
David S. Miller [Sat, 17 Nov 2018 07:04:37 +0000 (23:04 -0800)]
Revert "net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs"

This reverts commit dfa0d55ff6be64e7b6881212a291cb95f8da3b08.

Discussion still ongoing, I shouldn't have applied this.

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge tag 'batadv-net-for-davem-20181114' of git://git.open-mesh.org/linux-merge
David S. Miller [Sat, 17 Nov 2018 04:26:30 +0000 (20:26 -0800)]
Merge tag 'batadv-net-for-davem-20181114' of git://git.open-mesh.org/linux-merge

Simon Wunderlich says:

====================
Here are two batman-adv bugfixes:

 - Explicitly pad short ELP packets with zeros, by Sven Eckelmann

 - Fix packet size calculation when merging fragments,
   by Sven Eckelmann
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: phy: mdio-gpio: Fix working over slow can_sleep GPIOs
Martin Schiller [Wed, 14 Nov 2018 11:54:49 +0000 (12:54 +0100)]
net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs

This commit re-enables support for slow GPIO pins. It was initially
introduced by commit 2d6c9091ab76 ("net: mdio-gpio: support access that
may sleep") and got lost by commit 7e5fbd1e0700 ("net: mdio-gpio:
Convert to use gpiod functions where possible").

Also add a warning about slow GPIO pins like it is done in i2c-gpio.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet/sched: act_pedit: fix memory leak when IDR allocation fails
Davide Caratti [Wed, 14 Nov 2018 11:17:25 +0000 (12:17 +0100)]
net/sched: act_pedit: fix memory leak when IDR allocation fails

tcf_idr_check_alloc() can return a negative value, on allocation failures
(-ENOMEM) or IDR exhaustion (-ENOSPC): don't leak keys_ex in these cases.

Fixes: 0190c1d452a9 ("net: sched: atomically check-allocate action")
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: lantiq: Fix returned value in case of error in 'xrx200_probe()'
Christophe JAILLET [Tue, 13 Nov 2018 17:42:24 +0000 (18:42 +0100)]
net: lantiq: Fix returned value in case of error in 'xrx200_probe()'

Return 'err' in the error handling path instead of 0.
Return explicitly 0 in the normal path, instead of 'err', which is known
to be 0 at this point.

Fixes: fe1a56420cf2 ("net: lantiq: Add Lantiq / Intel VRX200 Ethernet driver")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoipv6: fix a dst leak when removing its exception
Xin Long [Tue, 13 Nov 2018 16:48:28 +0000 (00:48 +0800)]
ipv6: fix a dst leak when removing its exception

These is no need to hold dst before calling rt6_remove_exception_rt().
The call to dst_hold_safe() in ip6_link_failure() was for ip6_del_rt(),
which has been removed in Commit 93531c674315 ("net/ipv6: separate
handling of FIB entries from dst based routes"). Otherwise, it will
cause a dst leak.

This patch is to simply remove the dst_hold_safe() call before calling
rt6_remove_exception_rt() and also do the same in ip6_del_cached_rt().
It's safe, because the removal of the exception that holds its dst's
refcnt is protected by rt6_exception_lock.

Fixes: 93531c674315 ("net/ipv6: separate handling of FIB entries from dst based routes")
Fixes: 23fb93a4d3f1 ("net/ipv6: Cleanup exception and cache route handling")
Reported-by: Li Shuang <shuali@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agonet: mvneta: Don't advertise 2.5G modes
Maxime Chevallier [Fri, 9 Nov 2018 08:17:33 +0000 (09:17 +0100)]
net: mvneta: Don't advertise 2.5G modes

Using 2.5G speed relies on the SerDes lanes being configured
accordingly. The lanes have to be reconfigured to switch between
1G and 2.5G, and for now only the bootloader does this configuration.

In the case we add a Comphy driver to handle switching the lanes
dynamically, it's better for now to stick with supporting only 1G and
add advertisement for 2.5G once we really are capable of handling both
speeds without problem.

Since the interface mode is initialy taken from the DT, we want to make
sure that adding comphy support won't break boards that don't update
their dtb.

Fixes: da58a931f248 ("net: mvneta: Add support for 2500Mbps SGMII")
Reported-by: Andrew Lunn <andrew@lunn.ch>
Reported-by: Russell King <linux@armlinux.org.uk>
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
6 years agoMerge tag 'fsnotify_for_v4.20-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 16 Nov 2018 19:18:36 +0000 (13:18 -0600)]
Merge tag 'fsnotify_for_v4.20-rc3' of git://git./linux/kernel/git/jack/linux-fs

Pull fsnotify fix from Jan Kara:
 "One small fsnotify fix for duplicate events"

* tag 'fsnotify_for_v4.20-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  fanotify: fix handling of events on child sub-directory

6 years agoMerge tag 'gfs2-4.20.fixes3' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2...
Linus Torvalds [Fri, 16 Nov 2018 17:38:14 +0000 (11:38 -0600)]
Merge tag 'gfs2-4.20.fixes3' of git://git./linux/kernel/git/gfs2/linux-gfs2

Pull bfs2 fixes from Andreas Gruenbacher:
 "Fix two bugs leading to leaked buffer head references:

   - gfs2: Put bitmap buffers in put_super
   - gfs2: Fix iomap buffer head reference counting bug

  And one bug leading to significant slow-downs when deleting large
  files:

   - gfs2: Fix metadata read-ahead during truncate (2)"

* tag 'gfs2-4.20.fixes3' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
  gfs2: Fix iomap buffer head reference counting bug
  gfs2: Fix metadata read-ahead during truncate (2)
  gfs2: Put bitmap buffers in put_super