openwrt/staging/blogic.git
5 years agomac80211: Use linked list instead of rhashtable walk for mesh tables
Herbert Xu [Thu, 14 Feb 2019 14:03:24 +0000 (22:03 +0800)]
mac80211: Use linked list instead of rhashtable walk for mesh tables

The mesh table code walks over hash tables for two purposes.  First of
all it's used as part of a netlink dump process, but it is also used
for looking up entries to delete using criteria other than the hash
key.

The second purpose is directly contrary to the design specification
of rhashtable walks.  It is only meant for use by netlink dumps.

This is because rhashtable is resizable and you cannot obtain a
stable walk over it during a resize process.

In fact mesh's use of rhashtable for dumping is bogus too.  Rather
than using rhashtable walk's iterator to keep track of the current
position, it always converts the current position to an integer
which defeats the purpose of the iterator.

Therefore this patch converts all uses of rhashtable walk into a
simple linked list.

This patch also adds a new spin lock to protect the hash table
insertion/removal as well as the walk list modifications.  In fact
the previous code was buggy as the removals can race with each
other, potentially resulting in a double-free.

Cc: stable@vger.kernel.org
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
5 years agonet: ethernet: freescale: set FEC ethtool regs version
Vivien Didelot [Thu, 14 Feb 2019 16:15:35 +0000 (11:15 -0500)]
net: ethernet: freescale: set FEC ethtool regs version

Currently the ethtool_regs version is set to 0 for FEC devices.

Use this field to store the register dump version exposed by the
kernel. The choosen version 2 corresponds to the kernel compile test:

        #if defined(CONFIG_M523x) || defined(CONFIG_M527x)
        || defined(CONFIG_M528x) || defined(CONFIG_M520x)
        || defined(CONFIG_M532x) || defined(CONFIG_ARM)
        || defined(CONFIG_ARM64) || defined(CONFIG_COMPILE_TEST)

and version 1 corresponds to the opposite. Binaries of ethtool unaware
of this version will dump the whole set as usual.

Signed-off-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns: Fix object reference leaks in hns_dsaf_roce_reset()
Huang Zijiang [Thu, 14 Feb 2019 06:41:45 +0000 (14:41 +0800)]
net: hns: Fix object reference leaks in hns_dsaf_roce_reset()

The of_find_device_by_node() takes a reference to the underlying device
structure, we should release that reference.

Signed-off-by: Huang Zijiang <huang.zijiang@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs
Jann Horn [Wed, 13 Feb 2019 21:45:59 +0000 (22:45 +0100)]
mm: page_alloc: fix ref bias in page_frag_alloc() for 1-byte allocs

The basic idea behind ->pagecnt_bias is: If we pre-allocate the maximum
number of references that we might need to create in the fastpath later,
the bump-allocation fastpath only has to modify the non-atomic bias value
that tracks the number of extra references we hold instead of the atomic
refcount. The maximum number of allocations we can serve (under the
assumption that no allocation is made with size 0) is nc->size, so that's
the bias used.

However, even when all memory in the allocation has been given away, a
reference to the page is still held; and in the `offset < 0` slowpath, the
page may be reused if everyone else has dropped their references.
This means that the necessary number of references is actually
`nc->size+1`.

Luckily, from a quick grep, it looks like the only path that can call
page_frag_alloc(fragsz=1) is TAP with the IFF_NAPI_FRAGS flag, which
requires CAP_NET_ADMIN in the init namespace and is only intended to be
used for kernel testing and fuzzing.

To test for this issue, put a `WARN_ON(page_ref_count(page) == 0)` in the
`offset < 0` path, below the virt_to_page() call, and then repeatedly call
writev() on a TAP device with IFF_TAP|IFF_NO_PI|IFF_NAPI_FRAGS|IFF_NAPI,
with a vector consisting of 15 elements containing 1 byte each.

Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'net-phy-fix-locking-issue'
David S. Miller [Thu, 14 Feb 2019 17:04:55 +0000 (12:04 -0500)]
Merge branch 'net-phy-fix-locking-issue'

Heiner Kallweit says:

====================
net: phy: fix locking issue

Russell pointed out that the locking used in phy_is_started() isn't
needed and misleading. This locking also contributes to a race fixed
with patch 2.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phy: fix potential race in the phylib state machine
Heiner Kallweit [Wed, 13 Feb 2019 19:12:54 +0000 (20:12 +0100)]
net: phy: fix potential race in the phylib state machine

Russell reported the following race in the phylib state machine
(quoting from his mail):

if (phy_polling_mode(phydev) && phy_is_started(phydev))
phy_queue_state_machine(phydev, PHY_STATE_TIME);

state = PHY_UP
thread 0 thread 1
phy_disconnect()
+-phy_is_started()
phy_is_started()                |
`-phy_stop()
  +-phydev->state = PHY_HALTED
  `-phy_stop_machine()
    `-cancel_delayed_work_sync()
phy_queue_state_machine()
`-mod_delayed_work()

At this point, the phydev->state_queue() has been added back onto the
system workqueue despite phy_stop_machine() having been called and
cancel_delayed_work_sync() called on it.

Fix this by protecting the complete operation in thread 0.

Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
Reported-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phy: don't use locking in phy_is_started
Heiner Kallweit [Wed, 13 Feb 2019 19:11:40 +0000 (20:11 +0100)]
net: phy: don't use locking in phy_is_started

Russell suggested to remove the locking from phy_is_started() because
the read is atomic anyway and actually the locking may be more
misleading.

Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
Suggested-by: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: fix timestamping Makefile
Deepa Dinamani [Wed, 13 Feb 2019 17:09:13 +0000 (09:09 -0800)]
selftests: fix timestamping Makefile

The clean target in the makefile conflicts with the generic
kselftests lib.mk, and fails to properly remove the compiled
test programs.

Remove the redundant rule, the TEST_GEN_FILES will be already
removed by the CLEAN macro in lib.mk.

Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Acked-by: Shuah Khan <shuah@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: dsa: bcm_sf2: potential array overflow in bcm_sf2_sw_suspend()
Dan Carpenter [Wed, 13 Feb 2019 08:23:04 +0000 (11:23 +0300)]
net: dsa: bcm_sf2: potential array overflow in bcm_sf2_sw_suspend()

The value of ->num_ports comes from bcm_sf2_sw_probe() and it is less
than or equal to DSA_MAX_PORTS.  The ds->ports[] array is used inside
the dsa_is_user_port() and dsa_is_cpu_port() functions.  The ds->ports[]
array is allocated in dsa_switch_alloc() and it has ds->num_ports
elements so this leads to a static checker warning about a potential out
of bounds read.

Fixes: 8cfa94984c9c ("net: dsa: bcm_sf2: add suspend/resume callbacks")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: fix possible overflow in __sk_mem_raise_allocated()
Eric Dumazet [Tue, 12 Feb 2019 20:26:27 +0000 (12:26 -0800)]
net: fix possible overflow in __sk_mem_raise_allocated()

With many active TCP sockets, fat TCP sockets could fool
__sk_mem_raise_allocated() thanks to an overflow.

They would increase their share of the memory, instead
of decreasing it.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agodsa: mv88e6xxx: Ensure all pending interrupts are handled prior to exit
John David Anglin [Mon, 11 Feb 2019 18:40:21 +0000 (13:40 -0500)]
dsa: mv88e6xxx: Ensure all pending interrupts are handled prior to exit

The GPIO interrupt controller on the espressobin board only supports edge interrupts.
If one enables the use of hardware interrupts in the device tree for the 88E6341, it is
possible to miss an edge.  When this happens, the INTn pin on the Marvell switch is
stuck low and no further interrupts occur.

I found after adding debug statements to mv88e6xxx_g1_irq_thread_work() that there is
a race in handling device interrupts (e.g. PHY link interrupts).  Some interrupts are
directly cleared by reading the Global 1 status register.  However, the device interrupt
flag, for example, is not cleared until all the unmasked SERDES and PHY ports are serviced.
This is done by reading the relevant SERDES and PHY status register.

The code only services interrupts whose status bit is set at the time of reading its status
register.  If an interrupt event occurs after its status is read and before all interrupts
are serviced, then this event will not be serviced and the INTn output pin will remain low.

This is not a problem with polling or level interrupts since the handler will be called
again to process the event.  However, it's a big problem when using level interrupts.

The fix presented here is to add a loop around the code servicing switch interrupts.  If
any pending interrupts remain after the current set has been handled, we loop and process
the new set.  If there are no pending interrupts after servicing, we are sure that INTn has
gone high and we will get an edge when a new event occurs.

Tested on espressobin board.

Fixes: dc30c35be720 ("net: dsa: mv88e6xxx: Implement interrupt support.")
Signed-off-by: John David Anglin <dave.anglin@bell.net>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phy: fix interrupt handling in non-started states
Heiner Kallweit [Tue, 12 Feb 2019 18:56:15 +0000 (19:56 +0100)]
net: phy: fix interrupt handling in non-started states

phylib enables interrupts before phy_start() has been called, and if
we receive an interrupt in a non-started state, the interrupt handler
returns IRQ_NONE. This causes problems with at least one Marvell chip
as reported by Andrew.
Fix this by handling interrupts the same as in phy_mac_interrupt(),
basically always running the phylib state machine. It knows when it
has to do something and when not.
This change allows to handle interrupts gracefully even if they
occur in a non-started state.

Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
Reported-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agosctp: set stream ext to NULL after freeing it in sctp_stream_outq_migrate
Xin Long [Tue, 12 Feb 2019 10:51:01 +0000 (18:51 +0800)]
sctp: set stream ext to NULL after freeing it in sctp_stream_outq_migrate

In sctp_stream_init(), after sctp_stream_outq_migrate() freed the
surplus streams' ext, but sctp_stream_alloc_out() returns -ENOMEM,
stream->outcnt will not be set to 'outcnt'.

With the bigger value on stream->outcnt, when closing the assoc and
freeing its streams, the ext of those surplus streams will be freed
again since those stream exts were not set to NULL after freeing in
sctp_stream_outq_migrate(). Then the invalid-free issue reported by
syzbot would be triggered.

We fix it by simply setting them to NULL after freeing.

Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations")
Reported-by: syzbot+58e480e7b28f2d890bfd@syzkaller.appspotmail.com
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agosctp: call gso_reset_checksum when computing checksum in sctp_gso_segment
Xin Long [Tue, 12 Feb 2019 10:47:30 +0000 (18:47 +0800)]
sctp: call gso_reset_checksum when computing checksum in sctp_gso_segment

Jianlin reported a panic when running sctp gso over gre over vlan device:

  [   84.772930] RIP: 0010:do_csum+0x6d/0x170
  [   84.790605] Call Trace:
  [   84.791054]  csum_partial+0xd/0x20
  [   84.791657]  gre_gso_segment+0x2c3/0x390
  [   84.792364]  inet_gso_segment+0x161/0x3e0
  [   84.793071]  skb_mac_gso_segment+0xb8/0x120
  [   84.793846]  __skb_gso_segment+0x7e/0x180
  [   84.794581]  validate_xmit_skb+0x141/0x2e0
  [   84.795297]  __dev_queue_xmit+0x258/0x8f0
  [   84.795949]  ? eth_header+0x26/0xc0
  [   84.796581]  ip_finish_output2+0x196/0x430
  [   84.797295]  ? skb_gso_validate_network_len+0x11/0x80
  [   84.798183]  ? ip_finish_output+0x169/0x270
  [   84.798875]  ip_output+0x6c/0xe0
  [   84.799413]  ? ip_append_data.part.50+0xc0/0xc0
  [   84.800145]  iptunnel_xmit+0x144/0x1c0
  [   84.800814]  ip_tunnel_xmit+0x62d/0x930 [ip_tunnel]
  [   84.801699]  gre_tap_xmit+0xac/0xf0 [ip_gre]
  [   84.802395]  dev_hard_start_xmit+0xa5/0x210
  [   84.803086]  sch_direct_xmit+0x14f/0x340
  [   84.803733]  __dev_queue_xmit+0x799/0x8f0
  [   84.804472]  ip_finish_output2+0x2e0/0x430
  [   84.805255]  ? skb_gso_validate_network_len+0x11/0x80
  [   84.806154]  ip_output+0x6c/0xe0
  [   84.806721]  ? ip_append_data.part.50+0xc0/0xc0
  [   84.807516]  sctp_packet_transmit+0x716/0xa10 [sctp]
  [   84.808337]  sctp_outq_flush+0xd7/0x880 [sctp]

It was caused by SKB_GSO_CB(skb)->csum_start not set in sctp_gso_segment.
sctp_gso_segment() calls skb_segment() with 'feature | NETIF_F_HW_CSUM',
which causes SKB_GSO_CB(skb)->csum_start not to be set in skb_segment().

For TCP/UDP, when feature supports HW_CSUM, CHECKSUM_PARTIAL will be set
and gso_reset_checksum will be called to set SKB_GSO_CB(skb)->csum_start.

So SCTP should do the same as TCP/UDP, to call gso_reset_checksum() when
computing checksum in sctp_gso_segment.

Reported-by: Jianlin Shi <jishi@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf
David S. Miller [Thu, 14 Feb 2019 00:13:47 +0000 (19:13 -0500)]
Merge git://git./pub/scm/linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
Netfilter/IPVS fixes for net

The following patchset contains Netfilter/IPVS fixes for net:

1) Missing structure initialization in ebtables causes splat with
   32-bit user level on a 64-bit kernel, from Francesco Ruggeri.

2) Missing dependency on nf_defrag in IPVS IPv6 codebase, from
   Andrea Claudi.

3) Fix possible use-after-free from release path of target extensions.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'mlx5-fixes-2019-02-13' of git://git.kernel.org/pub/scm/linux/kernel/git...
David S. Miller [Thu, 14 Feb 2019 00:07:47 +0000 (19:07 -0500)]
Merge tag 'mlx5-fixes-2019-02-13' of git://git./linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
Mellanox, mlx5 fixes 2019-02-13

This series introduces some fixes to mlx5 driver.
For more information please see tag log below.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx5e: XDP, fix redirect resources availability check
Saeed Mahameed [Tue, 12 Feb 2019 00:27:02 +0000 (16:27 -0800)]
net/mlx5e: XDP, fix redirect resources availability check

Currently mlx5 driver creates xdp redirect hw queues unconditionally on
netdevice open, This is great until someone starts redirecting XDP traffic
via ndo_xdp_xmit on mlx5 device and changes the device configuration at
the same time, this might cause crashes, since the other device's napi
is not aware of the mlx5 state change (resources un-availability).

To fix this we must synchronize with other devices napi's on the system.
Added a new flag under mlx5e_priv to determine XDP TX resources are
available, set/clear it up when necessary and use synchronize_rcu()
when the flag is turned off, so other napi's are in-sync with it, before
we actually cleanup the hw resources.

The flag is tested prior to committing to transmit on mlx5e_xdp_xmit, and
it is sufficient to determine if it safe to transmit or not. The other
two internal flags (MLX5E_STATE_OPENED and MLX5E_SQ_STATE_ENABLED) become
unnecessary. Thus, they are removed from data path.

Fixes: 58b99ee3e3eb ("net/mlx5e: Add support for XDP_REDIRECT in device-out side")
Reported-by: Toke Høiland-Jørgensen <toke@redhat.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Fix a compilation warning in events.c
Tariq Toukan [Tue, 22 Jan 2019 12:18:04 +0000 (14:18 +0200)]
net/mlx5: Fix a compilation warning in events.c

Eliminate the following compilation warning:

drivers/net/ethernet/mellanox/mlx5/core/events.c: warning: 'error_str'
may be used uninitialized in this function [-Wuninitialized]:  => 238:3

Fixes: c2fb3db22d35 ("net/mlx5: Rework handling of port module events")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Mikhael Goikhman <migo@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: No command allowed when command interface is not ready
Huy Nguyen [Thu, 7 Feb 2019 15:22:56 +0000 (09:22 -0600)]
net/mlx5: No command allowed when command interface is not ready

When EEH is injected and PCI bus stalls, mlx5's pci error detect
function is called to deactivate the command interface and tear down
the device. The issue is that there can be a thread that already
passed MLX5_DEVICE_STATE_INTERNAL_ERROR check, it will send the command
and stuck in the wait_func.

Solution:
Add function mlx5_cmd_flush to disable command interface and clear all
the pending commands. When device state is set to
MLX5_DEVICE_STATE_INTERNAL_ERROR, call mlx5_cmd_flush to ensure all
pending threads waiting for firmware commands completion are terminated.

Fixes: c1d4d2e92ad6 ("net/mlx5: Avoid calling sleeping function by the health poll thread")
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Fix NULL pointer derefernce in set channels error flow
Maria Pasechnik [Sun, 3 Feb 2019 15:55:09 +0000 (17:55 +0200)]
net/mlx5e: Fix NULL pointer derefernce in set channels error flow

New channels are applied to the priv channels only after they
are successfully opened. Then, the indirection table should be built
according to the new number of channels.
Currently, such build is preformed independently of whether the
channels opening is successful, and is not reverted on failure.

The bug is caused due to removal of rss params from channels struct
and moving it to priv struct. That change cause to independency between
channels and rss params.
This causes a crash on a later point, when accessing rqn of a non
existing channel.

This patch fixes it by moving the indirection table build right before
switching the priv channels to new channels struct, after the new set of
channels was successfully opened.

Fixes: bbeb53b8b2c9 ("net/mlx5e: Move RSS params to a dedicated struct")
Signed-off-by: Maria Pasechnik <mariap@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonetfilter: nft_compat: use-after-free when deleting targets
Pablo Neira Ayuso [Wed, 13 Feb 2019 12:03:53 +0000 (13:03 +0100)]
netfilter: nft_compat: use-after-free when deleting targets

Fetch pointer to module before target object is released.

Fixes: 29e3880109e3 ("netfilter: nf_tables: fix use-after-free when deleting compat expressions")
Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agoteam: avoid complex list operations in team_nl_cmd_options_set()
Cong Wang [Tue, 12 Feb 2019 05:59:51 +0000 (21:59 -0800)]
team: avoid complex list operations in team_nl_cmd_options_set()

The current opt_inst_list operations inside team_nl_cmd_options_set()
is too complex to track:

    LIST_HEAD(opt_inst_list);
    nla_for_each_nested(...) {
        list_for_each_entry(opt_inst, &team->option_inst_list, list) {
            if (__team_option_inst_tmp_find(&opt_inst_list, opt_inst))
                continue;
            list_add(&opt_inst->tmp_list, &opt_inst_list);
        }
    }
    team_nl_send_event_options_get(team, &opt_inst_list);

as while we retrieve 'opt_inst' from team->option_inst_list, it could
be added to the local 'opt_inst_list' for multiple times. The
__team_option_inst_tmp_find() doesn't work, as the setter
team_mode_option_set() still calls team->ops.exit() which uses
->tmp_list too in __team_options_change_check().

Simplify the list operations by moving the 'opt_inst_list' and
team_nl_send_event_options_get() into the nla_for_each_nested() loop so
that it can be guranteed that we won't insert a same list entry for
multiple times. Therefore, __team_option_inst_tmp_find() can be removed
too.

Fixes: 4fb0534fb7bb ("team: avoid adding twice the same option to the event list")
Fixes: 2fcdb2c9e659 ("team: allow to send multiple set events in one message")
Reported-by: syzbot+4d4af685432dc0e56c91@syzkaller.appspotmail.com
Reported-by: syzbot+68ee510075cf64260cc4@syzkaller.appspotmail.com
Cc: Jiri Pirko <jiri@resnulli.us>
Cc: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'net_sched-some-fixes-for-cls_tcindex'
David S. Miller [Tue, 12 Feb 2019 19:10:56 +0000 (14:10 -0500)]
Merge branch 'net_sched-some-fixes-for-cls_tcindex'

Cong Wang says:

====================
net_sched: some fixes for cls_tcindex

This patchset contains 3 bug fixes for tcindex filter. Please check
each patch for details.

v2: fix a compile error in patch 2
    drop netns refcnt in patch 1
====================

Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
5 years agonet_sched: fix two more memory leaks in cls_tcindex
Cong Wang [Mon, 11 Feb 2019 21:06:16 +0000 (13:06 -0800)]
net_sched: fix two more memory leaks in cls_tcindex

struct tcindex_filter_result contains two parts:
struct tcf_exts and struct tcf_result.

For the local variable 'cr', its exts part is never used but
initialized without being released properly on success path. So
just completely remove the exts part to fix this leak.

For the local variable 'new_filter_result', it is never properly
released if not used by 'r' on success path.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet_sched: fix a memory leak in cls_tcindex
Cong Wang [Mon, 11 Feb 2019 21:06:15 +0000 (13:06 -0800)]
net_sched: fix a memory leak in cls_tcindex

When tcindex_destroy() destroys all the filter results in
the perfect hash table, it invokes the walker to delete
each of them. However, results with class==0 are skipped
in either tcindex_walk() or tcindex_delete(), which causes
a memory leak reported by kmemleak.

This patch fixes it by skipping the walker and directly
deleting these filter results so we don't miss any filter
result.

As a result of this change, we have to initialize exts->net
properly in tcindex_alloc_perfect_hash(). For net-next, we
need to consider whether we should initialize ->net in
tcf_exts_init() instead, before that just directly test
CONFIG_NET_CLS_ACT=y.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet_sched: fix a race condition in tcindex_destroy()
Cong Wang [Mon, 11 Feb 2019 21:06:14 +0000 (13:06 -0800)]
net_sched: fix a race condition in tcindex_destroy()

tcindex_destroy() invokes tcindex_destroy_element() via
a walker to delete each filter result in its perfect hash
table, and tcindex_destroy_element() calls tcindex_delete()
which schedules tcf RCU works to do the final deletion work.
Unfortunately this races with the RCU callback
__tcindex_destroy(), which could lead to use-after-free as
reported by Adrian.

Fix this by migrating this RCU callback to tcf RCU work too,
as that workqueue is ordered, we will not have use-after-free.

Note, we don't need to hold netns refcnt because we don't call
tcf_exts_destroy() here.

Fixes: 27ce4f05e2ab ("net_sched: use tcf_queue_work() in tcindex filter")
Reported-by: Adrian <bugs@abtelecom.ro>
Cc: Ben Hutchings <ben@decadent.org.uk>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'ena-races'
David S. Miller [Tue, 12 Feb 2019 19:05:07 +0000 (14:05 -0500)]
Merge branch 'ena-races'

Arthur Kiyanovski says:

====================
net: ena: race condition bug fix and version update

This patchset includes a fix to a race condition that can cause
kernel panic, as well as a driver version update because of this
fix.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: ena: update driver version from 2.0.2 to 2.0.3
Arthur Kiyanovski [Mon, 11 Feb 2019 17:17:44 +0000 (19:17 +0200)]
net: ena: update driver version from 2.0.2 to 2.0.3

Update driver version due to bug fix.

Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: ena: fix race between link up and device initalization
Arthur Kiyanovski [Mon, 11 Feb 2019 17:17:43 +0000 (19:17 +0200)]
net: ena: fix race between link up and device initalization

Fix race condition between ena_update_on_link_change() and
ena_restore_device().

This race can occur if link notification arrives while the driver
is performing a reset sequence. In this case link can be set up,
enabling the device, before it is fully restored. If packets are
sent at this time, the driver might access uninitialized data
structures, causing kernel crash.

Move the clearing of ENA_FLAG_ONGOING_RESET and netif_carrier_on()
after ena_up() to ensure the device is ready when link is set up.

Fixes: d18e4f683445 ("net: ena: fix race condition between device reset and link up setup")
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/packet: fix 4gb buffer limit due to overflow check
Kal Conley [Sun, 10 Feb 2019 08:57:11 +0000 (09:57 +0100)]
net/packet: fix 4gb buffer limit due to overflow check

When calculating rb->frames_per_block * req->tp_block_nr the result
can overflow. Check it for overflow without limiting the total buffer
size to UINT_MAX.

This change fixes support for packet ring buffers >= UINT_MAX.

Fixes: 8f8d28e4d6d8 ("net/packet: fix overflow in check for tp_frame_nr")
Signed-off-by: Kal Conley <kal.conley@dectris.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoinet_diag: fix reporting cgroup classid and fallback to priority
Konstantin Khlebnikov [Sat, 9 Feb 2019 10:35:52 +0000 (13:35 +0300)]
inet_diag: fix reporting cgroup classid and fallback to priority

Field idiag_ext in struct inet_diag_req_v2 used as bitmap of requested
extensions has only 8 bits. Thus extensions starting from DCTCPINFO
cannot be requested directly. Some of them included into response
unconditionally or hook into some of lower 8 bits.

Extension INET_DIAG_CLASS_ID has not way to request from the beginning.

This patch bundle it with INET_DIAG_TCLASS (ipv6 tos), fixes space
reservation, and documents behavior for other extensions.

Also this patch adds fallback to reporting socket priority. This filed
is more widely used for traffic classification because ipv4 sockets
automatically maps TOS to priority and default qdisc pfifo_fast knows
about that. But priority could be changed via setsockopt SO_PRIORITY so
INET_DIAG_TOS isn't enough for predicting class.

Also cgroup2 obsoletes net_cls classid (it always zero), but we cannot
reuse this field for reporting cgroup2 id because it is 64-bit (ino+gen).

So, after this patch INET_DIAG_CLASS_ID will report socket priority
for most common setup when net_cls isn't set and/or cgroup2 in use.

Fixes: 0888e372c37f ("net: inet: diag: expose sockets cgroup classid")
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobatman-adv: fix uninit-value in batadv_interface_tx()
Eric Dumazet [Mon, 11 Feb 2019 22:41:22 +0000 (14:41 -0800)]
batman-adv: fix uninit-value in batadv_interface_tx()

KMSAN reported batadv_interface_tx() was possibly using a
garbage value [1]

batadv_get_vid() does have a pskb_may_pull() call
but batadv_interface_tx() does not actually make sure
this did not fail.

[1]
BUG: KMSAN: uninit-value in batadv_interface_tx+0x908/0x1e40 net/batman-adv/soft-interface.c:231
CPU: 0 PID: 10006 Comm: syz-executor469 Not tainted 4.20.0-rc7+ #5
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x173/0x1d0 lib/dump_stack.c:113
 kmsan_report+0x12e/0x2a0 mm/kmsan/kmsan.c:613
 __msan_warning+0x82/0xf0 mm/kmsan/kmsan_instr.c:313
 batadv_interface_tx+0x908/0x1e40 net/batman-adv/soft-interface.c:231
 __netdev_start_xmit include/linux/netdevice.h:4356 [inline]
 netdev_start_xmit include/linux/netdevice.h:4365 [inline]
 xmit_one net/core/dev.c:3257 [inline]
 dev_hard_start_xmit+0x607/0xc40 net/core/dev.c:3273
 __dev_queue_xmit+0x2e42/0x3bc0 net/core/dev.c:3843
 dev_queue_xmit+0x4b/0x60 net/core/dev.c:3876
 packet_snd net/packet/af_packet.c:2928 [inline]
 packet_sendmsg+0x8306/0x8f30 net/packet/af_packet.c:2953
 sock_sendmsg_nosec net/socket.c:621 [inline]
 sock_sendmsg net/socket.c:631 [inline]
 __sys_sendto+0x8c4/0xac0 net/socket.c:1788
 __do_sys_sendto net/socket.c:1800 [inline]
 __se_sys_sendto+0x107/0x130 net/socket.c:1796
 __x64_sys_sendto+0x6e/0x90 net/socket.c:1796
 do_syscall_64+0xbc/0xf0 arch/x86/entry/common.c:291
 entry_SYSCALL_64_after_hwframe+0x63/0xe7
RIP: 0033:0x441889
Code: 18 89 d0 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 bb 10 fc ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007ffdda6fd468 EFLAGS: 00000216 ORIG_RAX: 000000000000002c
RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 0000000000441889
RDX: 000000000000000e RSI: 00000000200000c0 RDI: 0000000000000003
RBP: 0000000000000003 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000216 R12: 00007ffdda6fd4c0
R13: 00007ffdda6fd4b0 R14: 0000000000000000 R15: 0000000000000000

Uninit was created at:
 kmsan_save_stack_with_flags mm/kmsan/kmsan.c:204 [inline]
 kmsan_internal_poison_shadow+0x92/0x150 mm/kmsan/kmsan.c:158
 kmsan_kmalloc+0xa6/0x130 mm/kmsan/kmsan_hooks.c:176
 kmsan_slab_alloc+0xe/0x10 mm/kmsan/kmsan_hooks.c:185
 slab_post_alloc_hook mm/slab.h:446 [inline]
 slab_alloc_node mm/slub.c:2759 [inline]
 __kmalloc_node_track_caller+0xe18/0x1030 mm/slub.c:4383
 __kmalloc_reserve net/core/skbuff.c:137 [inline]
 __alloc_skb+0x309/0xa20 net/core/skbuff.c:205
 alloc_skb include/linux/skbuff.h:998 [inline]
 alloc_skb_with_frags+0x1c7/0xac0 net/core/skbuff.c:5220
 sock_alloc_send_pskb+0xafd/0x10e0 net/core/sock.c:2083
 packet_alloc_skb net/packet/af_packet.c:2781 [inline]
 packet_snd net/packet/af_packet.c:2872 [inline]
 packet_sendmsg+0x661a/0x8f30 net/packet/af_packet.c:2953
 sock_sendmsg_nosec net/socket.c:621 [inline]
 sock_sendmsg net/socket.c:631 [inline]
 __sys_sendto+0x8c4/0xac0 net/socket.c:1788
 __do_sys_sendto net/socket.c:1800 [inline]
 __se_sys_sendto+0x107/0x130 net/socket.c:1796
 __x64_sys_sendto+0x6e/0x90 net/socket.c:1796
 do_syscall_64+0xbc/0xf0 arch/x86/entry/common.c:291
 entry_SYSCALL_64_after_hwframe+0x63/0xe7

Fixes: c6c8fea29769 ("net: Add batman-adv meshing protocol")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Marek Lindner <mareklindner@neomailbox.ch>
Cc: Simon Wunderlich <sw@simonwunderlich.de>
Cc: Antonio Quartulli <a@unstable.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv6: propagate genlmsg_reply return code
Li RongQing [Mon, 11 Feb 2019 11:32:20 +0000 (19:32 +0800)]
ipv6: propagate genlmsg_reply return code

genlmsg_reply can fail, so propagate its return code

Fixes: 915d7e5e593 ("ipv6: sr: add code base for control plane support of SR-IPv6")
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx4_en: Force CHECKSUM_NONE for short ethernet frames
Saeed Mahameed [Mon, 11 Feb 2019 16:04:17 +0000 (18:04 +0200)]
net/mlx4_en: Force CHECKSUM_NONE for short ethernet frames

When an ethernet frame is padded to meet the minimum ethernet frame
size, the padding octets are not covered by the hardware checksum.
Fortunately the padding octets are usually zero's, which don't affect
checksum. However, it is not guaranteed. For example, switches might
choose to make other use of these octets.
This repeatedly causes kernel hardware checksum fault.

Prior to the cited commit below, skb checksum was forced to be
CHECKSUM_NONE when padding is detected. After it, we need to keep
skb->csum updated. However, fixing up CHECKSUM_COMPLETE requires to
verify and parse IP headers, it does not worth the effort as the packets
are so small that CHECKSUM_COMPLETE has no significant advantage.

Future work: when reporting checksum complete is not an option for
IP non-TCP/UDP packets, we can actually fallback to report checksum
unnecessary, by looking at cqe IPOK bit.

Fixes: 88078d98d1bb ("net: pskb_trim_rcsum() and CHECKSUM_COMPLETE are friends")
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phylink: avoid resolving link state too early
Russell King [Mon, 11 Feb 2019 15:04:24 +0000 (15:04 +0000)]
net: phylink: avoid resolving link state too early

During testing on Armada 388 platforms, it was found with a certain
module configuration that it was possible to trigger a kernel oops
during the module load process, caused by the phylink resolver being
triggered for a currently disabled interface.

This problem was introduced by changing the way the SFP registration
works, which now can result in the sfp link down notification being
called during phylink_create().

Fixes: b5bfc21af5cb ("net: sfp: do not probe SFP module before we're attached")
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agogeneve: change NET_UDP_TUNNEL dependency to select
Matteo Croce [Mon, 11 Feb 2019 14:32:36 +0000 (15:32 +0100)]
geneve: change NET_UDP_TUNNEL dependency to select

Due to the depends on NET_UDP_TUNNEL, at the moment it is impossible to
compile GENEVE if no other protocol depending on NET_UDP_TUNNEL is
selected.

Fix this changing the depends to a select, and drop NET_IP_TUNNEL from the
select list, as it already depends on NET_UDP_TUNNEL.

Signed-off-by: Matteo Croce <mcroce@redhat.com>
Reviewed-and-tested-by: Andrea Claudi <aclaudi@redhat.com>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agosfc: initialise found bitmap in efx_ef10_mtd_probe
Bert Kenward [Tue, 12 Feb 2019 13:10:00 +0000 (13:10 +0000)]
sfc: initialise found bitmap in efx_ef10_mtd_probe

The bitmap of found partitions in efx_ef10_mtd_probe was not
initialised, causing partitions to be suppressed based off whatever
value was in the bitmap at the start.

Fixes: 3366463513f5 ("sfc: suppress duplicate nvmem partition types in efx_ef10_mtd_probe")
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'mac80211-for-davem-2019-02-12' of git://git.kernel.org/pub/scm/linux/kerne...
David S. Miller [Tue, 12 Feb 2019 16:43:19 +0000 (11:43 -0500)]
Merge tag 'mac80211-for-davem-2019-02-12' of git://git./linux/kernel/git/jberg/mac80211

Johannes Berg says:

====================
Just a few fixes:
 * aggregation session teardown with internal TXQs was
   continuing to send some frames marked as aggregation,
   fix from Ilan
 * IBSS join was missed during firmware restart, should
   such a thing happen
 * speculative execution based on the return value of
   cfg80211_classify8021d() - which is controlled by the
   sender of the packet - could be problematic in some
   code using it, prevent it
 * a few peer measurement fixes
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipvs: fix dependency on nf_defrag_ipv6
Andrea Claudi [Mon, 11 Feb 2019 15:14:39 +0000 (16:14 +0100)]
ipvs: fix dependency on nf_defrag_ipv6

ipvs relies on nf_defrag_ipv6 module to manage IPv6 fragmentation,
but lacks proper Kconfig dependencies and does not explicitly
request defrag features.

As a result, if netfilter hooks are not loaded, when IPv6 fragmented
packet are handled by ipvs only the first fragment makes through.

Fix it properly declaring the dependency on Kconfig and registering
netfilter hooks on ip_vs_add_service() and ip_vs_new_dest().

Reported-by: Li Shuang <shuali@redhat.com>
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agotipc: fix link session and re-establish issues
Tuong Lien [Mon, 11 Feb 2019 06:29:43 +0000 (13:29 +0700)]
tipc: fix link session and re-establish issues

When a link endpoint is re-created (e.g. after a node reboot or
interface reset), the link session number is varied by random, the peer
endpoint will be synced with this new session number before the link is
re-established.

However, there is a shortcoming in this mechanism that can lead to the
link never re-established or faced with a failure then. It happens when
the peer endpoint is ready in ESTABLISHING state, the 'peer_session' as
well as the 'in_session' flag have been set, but suddenly this link
endpoint leaves. When it comes back with a random session number, there
are two situations possible:

1/ If the random session number is larger than (or equal to) the
previous one, the peer endpoint will be updated with this new session
upon receipt of a RESET_MSG from this endpoint, and the link can be re-
established as normal. Otherwise, all the RESET_MSGs from this endpoint
will be rejected by the peer. In turn, when this link endpoint receives
one ACTIVATE_MSG from the peer, it will move to ESTABLISHED and start
to send STATE_MSGs, but again these messages will be dropped by the
peer due to wrong session.
The peer link endpoint can still become ESTABLISHED after receiving a
traffic message from this endpoint (e.g. a BCAST_PROTOCOL or
NAME_DISTRIBUTOR), but since all the STATE_MSGs are invalid, the link
will be forced down sooner or later!

Even in case the random session number is larger than the previous one,
it can be that the ACTIVATE_MSG from the peer arrives first, and this
link endpoint moves quickly to ESTABLISHED without sending out any
RESET_MSG yet. Consequently, the peer link will not be updated with the
new session number, and the same link failure scenario as above will
happen.

2/ Another situation can be that, the peer link endpoint was reset due
to any reasons in the meantime, its link state was set to RESET from
ESTABLISHING but still in session, i.e. the 'in_session' flag is not
reset...
Now, if the random session number from this endpoint is less than the
previous one, all the RESET_MSGs from this endpoint will be rejected by
the peer. In the other direction, when this link endpoint receives a
RESET_MSG from the peer, it moves to ESTABLISHING and starts to send
ACTIVATE_MSGs, but all these messages will be rejected by the peer too.
As a result, the link cannot be re-established but gets stuck with this
link endpoint in state ESTABLISHING and the peer in RESET!

Solution:

===========

This link endpoint should not go directly to ESTABLISHED when getting
ACTIVATE_MSG from the peer which may belong to the old session if the
link was re-created. To ensure the session to be correct before the
link is re-established, the peer endpoint in ESTABLISHING state will
send back the last session number in ACTIVATE_MSG for a verification at
this endpoint. Then, if needed, a new and more appropriate session
number will be regenerated to force a re-synch first.

In addition, when a link in ESTABLISHING state is reset, its state will
move to RESET according to the link FSM, along with resetting the
'in_session' flag (and the other data) as a normal link reset, it will
also be deleted if requested.

The solution is backward compatible.

Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: fix IPv6 prefix route residue
Zhiqiang Liu [Mon, 11 Feb 2019 02:57:46 +0000 (10:57 +0800)]
net: fix IPv6 prefix route residue

Follow those steps:
 # ip addr add 2001:123::1/32 dev eth0
 # ip addr add 2001:123:456::2/64 dev eth0
 # ip addr del 2001:123::1/32 dev eth0
 # ip addr del 2001:123:456::2/64 dev eth0
and then prefix route of 2001:123::1/32 will still exist.

This is because ipv6_prefix_equal in check_cleanup_prefix_route
func does not check whether two IPv6 addresses have the same
prefix length. If the prefix of one address starts with another
shorter address prefix, even though their prefix lengths are
different, the return value of ipv6_prefix_equal is true.

Here I add a check of whether two addresses have the same prefix
to decide whether their prefixes are equal.

Fixes: 5b84efecb7d9 ("ipv6 addrconf: don't cleanup prefix route for IFA_F_NOPREFIXROUTE")
Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
Reported-by: Wenhao Zhang <zhangwenhao8@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agotipc: fix skb may be leaky in tipc_link_input
Hoang Le [Mon, 11 Feb 2019 02:18:28 +0000 (09:18 +0700)]
tipc: fix skb may be leaky in tipc_link_input

When we free skb at tipc_data_input, we return a 'false' boolean.
Then, skb passed to subcalling tipc_link_input in tipc_link_rcv,

<snip>
1303 int tipc_link_rcv:
...
1354    if (!tipc_data_input(l, skb, l->inputq))
1355        rc |= tipc_link_input(l, skb, l->inputq);
</snip>

Fix it by simple changing to a 'true' boolean when skb is being free-ed.
Then, tipc_link_rcv will bypassed to subcalling tipc_link_input as above
condition.

Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <maloy@donjonn.com>
Signed-off-by: Hoang Le <hoang.h.le@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonetfilter: compat: initialize all fields in xt_init
Francesco Ruggeri [Sun, 10 Feb 2019 19:58:29 +0000 (11:58 -0800)]
netfilter: compat: initialize all fields in xt_init

If a non zero value happens to be in xt[NFPROTO_BRIDGE].cur at init
time, the following panic can be caused by running

% ebtables -t broute -F BROUTING

from a 32-bit user level on a 64-bit kernel. This patch replaces
kmalloc_array with kcalloc when allocating xt.

[  474.680846] BUG: unable to handle kernel paging request at 0000000009600920
[  474.687869] PGD 2037006067 P4D 2037006067 PUD 2038938067 PMD 0
[  474.693838] Oops: 0000 [#1] SMP
[  474.697055] CPU: 9 PID: 4662 Comm: ebtables Kdump: loaded Not tainted 4.19.17-11302235.AroraKernelnext.fc18.x86_64 #1
[  474.707721] Hardware name: Supermicro X9DRT/X9DRT, BIOS 3.0 06/28/2013
[  474.714313] RIP: 0010:xt_compat_calc_jump+0x2f/0x63 [x_tables]
[  474.720201] Code: 40 0f b6 ff 55 31 c0 48 6b ff 70 48 03 3d dc 45 00 00 48 89 e5 8b 4f 6c 4c 8b 47 60 ff c9 39 c8 7f 2f 8d 14 08 d1 fa 48 63 fa <41> 39 34 f8 4c 8d 0c fd 00 00 00 00 73 05 8d 42 01 eb e1 76 05 8d
[  474.739023] RSP: 0018:ffffc9000943fc58 EFLAGS: 00010207
[  474.744296] RAX: 0000000000000000 RBX: ffffc90006465000 RCX: 0000000002580249
[  474.751485] RDX: 00000000012c0124 RSI: fffffffff7be17e9 RDI: 00000000012c0124
[  474.758670] RBP: ffffc9000943fc58 R08: 0000000000000000 R09: ffffffff8117cf8f
[  474.765855] R10: ffffc90006477000 R11: 0000000000000000 R12: 0000000000000001
[  474.773048] R13: 0000000000000000 R14: ffffc9000943fcb8 R15: ffffc9000943fcb8
[  474.780234] FS:  0000000000000000(0000) GS:ffff88a03f840000(0063) knlGS:00000000f7ac7700
[  474.788612] CS:  0010 DS: 002b ES: 002b CR0: 0000000080050033
[  474.794632] CR2: 0000000009600920 CR3: 0000002037422006 CR4: 00000000000606e0
[  474.802052] Call Trace:
[  474.804789]  compat_do_replace+0x1fb/0x2a3 [ebtables]
[  474.810105]  compat_do_ebt_set_ctl+0x69/0xe6 [ebtables]
[  474.815605]  ? try_module_get+0x37/0x42
[  474.819716]  compat_nf_setsockopt+0x4f/0x6d
[  474.824172]  compat_ip_setsockopt+0x7e/0x8c
[  474.828641]  compat_raw_setsockopt+0x16/0x3a
[  474.833220]  compat_sock_common_setsockopt+0x1d/0x24
[  474.838458]  __compat_sys_setsockopt+0x17e/0x1b1
[  474.843343]  ? __check_object_size+0x76/0x19a
[  474.847960]  __ia32_compat_sys_socketcall+0x1cb/0x25b
[  474.853276]  do_fast_syscall_32+0xaf/0xf6
[  474.857548]  entry_SYSENTER_compat+0x6b/0x7a

Signed-off-by: Francesco Ruggeri <fruggeri@arista.com>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonet/x25: do not hold the cpu too long in x25_new_lci()
Eric Dumazet [Fri, 8 Feb 2019 20:41:05 +0000 (12:41 -0800)]
net/x25: do not hold the cpu too long in x25_new_lci()

Due to quadratic behavior of x25_new_lci(), syzbot was able
to trigger an rcu stall.

Fix this by not blocking BH for the whole duration of
the function, and inserting a reschedule point when possible.

If we care enough, using a bitmap could get rid of the quadratic
behavior.

syzbot report :

rcu: INFO: rcu_preempt self-detected stall on CPU
rcu:    0-...!: (10500 ticks this GP) idle=4fa/1/0x4000000000000002 softirq=283376/283376 fqs=0
rcu:     (t=10501 jiffies g=383105 q=136)
rcu: rcu_preempt kthread starved for 10502 jiffies! g383105 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402 ->cpu=0
rcu: RCU grace-period kthread stack dump:
rcu_preempt     I28928    10      2 0x80000000
Call Trace:
 context_switch kernel/sched/core.c:2844 [inline]
 __schedule+0x817/0x1cc0 kernel/sched/core.c:3485
 schedule+0x92/0x180 kernel/sched/core.c:3529
 schedule_timeout+0x4db/0xfd0 kernel/time/timer.c:1803
 rcu_gp_fqs_loop kernel/rcu/tree.c:1948 [inline]
 rcu_gp_kthread+0x956/0x17a0 kernel/rcu/tree.c:2105
 kthread+0x357/0x430 kernel/kthread.c:246
 ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:352
NMI backtrace for cpu 0
CPU: 0 PID: 8759 Comm: syz-executor2 Not tainted 5.0.0-rc4+ #51
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 <IRQ>
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x172/0x1f0 lib/dump_stack.c:113
 nmi_cpu_backtrace.cold+0x63/0xa4 lib/nmi_backtrace.c:101
 nmi_trigger_cpumask_backtrace+0x1be/0x236 lib/nmi_backtrace.c:62
 arch_trigger_cpumask_backtrace+0x14/0x20 arch/x86/kernel/apic/hw_nmi.c:38
 trigger_single_cpu_backtrace include/linux/nmi.h:164 [inline]
 rcu_dump_cpu_stacks+0x183/0x1cf kernel/rcu/tree.c:1211
 print_cpu_stall kernel/rcu/tree.c:1348 [inline]
 check_cpu_stall kernel/rcu/tree.c:1422 [inline]
 rcu_pending kernel/rcu/tree.c:3018 [inline]
 rcu_check_callbacks.cold+0x500/0xa4a kernel/rcu/tree.c:2521
 update_process_times+0x32/0x80 kernel/time/timer.c:1635
 tick_sched_handle+0xa2/0x190 kernel/time/tick-sched.c:161
 tick_sched_timer+0x47/0x130 kernel/time/tick-sched.c:1271
 __run_hrtimer kernel/time/hrtimer.c:1389 [inline]
 __hrtimer_run_queues+0x33e/0xde0 kernel/time/hrtimer.c:1451
 hrtimer_interrupt+0x314/0x770 kernel/time/hrtimer.c:1509
 local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1035 [inline]
 smp_apic_timer_interrupt+0x120/0x570 arch/x86/kernel/apic/apic.c:1060
 apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:807
 </IRQ>
RIP: 0010:__read_once_size include/linux/compiler.h:193 [inline]
RIP: 0010:queued_write_lock_slowpath+0x13e/0x290 kernel/locking/qrwlock.c:86
Code: 00 00 fc ff df 4c 8d 2c 01 41 83 c7 03 41 0f b6 45 00 41 38 c7 7c 08 84 c0 0f 85 0c 01 00 00 8b 03 3d 00 01 00 00 74 1a f3 90 <41> 0f b6 55 00 41 38 d7 7c eb 84 d2 74 e7 48 89 df e8 6c 0f 4f 00
RSP: 0018:ffff88805f117bd8 EFLAGS: 00000206 ORIG_RAX: ffffffffffffff13
RAX: 0000000000000300 RBX: ffffffff89413ba0 RCX: 1ffffffff1282774
RDX: 0000000000000000 RSI: 0000000000000004 RDI: ffffffff89413ba0
RBP: ffff88805f117c70 R08: 1ffffffff1282774 R09: fffffbfff1282775
R10: fffffbfff1282774 R11: ffffffff89413ba3 R12: 00000000000000ff
R13: fffffbfff1282774 R14: 1ffff1100be22f7d R15: 0000000000000003
 queued_write_lock include/asm-generic/qrwlock.h:104 [inline]
 do_raw_write_lock+0x1d6/0x290 kernel/locking/spinlock_debug.c:203
 __raw_write_lock_bh include/linux/rwlock_api_smp.h:204 [inline]
 _raw_write_lock_bh+0x3b/0x50 kernel/locking/spinlock.c:312
 x25_insert_socket+0x21/0xe0 net/x25/af_x25.c:267
 x25_bind+0x273/0x340 net/x25/af_x25.c:705
 __sys_bind+0x23f/0x290 net/socket.c:1505
 __do_sys_bind net/socket.c:1516 [inline]
 __se_sys_bind net/socket.c:1514 [inline]
 __x64_sys_bind+0x73/0xb0 net/socket.c:1514
 do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x457e39
Code: ad b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 7b b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007fafccd0dc78 EFLAGS: 00000246 ORIG_RAX: 0000000000000031
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000000457e39
RDX: 0000000000000012 RSI: 0000000020000240 RDI: 0000000000000004
RBP: 000000000073bf00 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007fafccd0e6d4
R13: 00000000004bdf8b R14: 00000000004ce4b8 R15: 00000000ffffffff
Sending NMI from CPU 0 to CPUs 1:
NMI backtrace for cpu 1
CPU: 1 PID: 8752 Comm: syz-executor4 Not tainted 5.0.0-rc4+ #51
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:__x25_find_socket+0x78/0x120 net/x25/af_x25.c:328
Code: 89 f8 48 c1 e8 03 80 3c 18 00 0f 85 a6 00 00 00 4d 8b 64 24 68 4d 85 e4 74 7f e8 03 97 3d fb 49 83 ec 68 74 74 e8 f8 96 3d fb <49> 8d bc 24 88 04 00 00 48 89 f8 48 c1 e8 03 0f b6 04 18 84 c0 74
RSP: 0018:ffff8880639efc58 EFLAGS: 00000246
RAX: 0000000000040000 RBX: dffffc0000000000 RCX: ffffc9000e677000
RDX: 0000000000040000 RSI: ffffffff863244b8 RDI: ffff88806a764628
RBP: ffff8880639efc80 R08: ffff8880a80d05c0 R09: fffffbfff1282775
R10: fffffbfff1282774 R11: ffffffff89413ba3 R12: ffff88806a7645c0
R13: 0000000000000001 R14: ffff88809f29ac00 R15: 0000000000000000
FS:  00007fe8d0c58700(0000) GS:ffff8880ae900000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b32823000 CR3: 00000000672eb000 CR4: 00000000001406e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 x25_new_lci net/x25/af_x25.c:357 [inline]
 x25_connect+0x374/0xdf0 net/x25/af_x25.c:786
 __sys_connect+0x266/0x330 net/socket.c:1686
 __do_sys_connect net/socket.c:1697 [inline]
 __se_sys_connect net/socket.c:1694 [inline]
 __x64_sys_connect+0x73/0xb0 net/socket.c:1694
 do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x457e39
Code: ad b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 7b b8 fb ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007fe8d0c57c78 EFLAGS: 00000246 ORIG_RAX: 000000000000002a
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000000457e39
RDX: 0000000000000012 RSI: 0000000020000200 RDI: 0000000000000004
RBP: 000000000073bf00 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007fe8d0c586d4
R13: 00000000004be378 R14: 00000000004ceb00 R15: 00000000ffffffff

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Andrew Hendry <andrew.hendry@gmail.com>
Cc: linux-x25@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agovxlan: test dev->flags & IFF_UP before calling netif_rx()
Eric Dumazet [Thu, 7 Feb 2019 20:27:38 +0000 (12:27 -0800)]
vxlan: test dev->flags & IFF_UP before calling netif_rx()

netif_rx() must be called under a strict contract.

At device dismantle phase, core networking clears IFF_UP
and flush_all_backlogs() is called after rcu grace period
to make sure no incoming packet might be in a cpu backlog
and still referencing the device.

Most drivers call netif_rx() from their interrupt handler,
and since the interrupts are disabled at device dismantle,
netif_rx() does not have to check dev->flags & IFF_UP

Virtual drivers do not have this guarantee, and must
therefore make the check themselves.

Otherwise we risk use-after-free and/or crashes.

Note this patch also fixes a small issue that came
with commit ce6502a8f957 ("vxlan: fix a use after free
in vxlan_encap_bypass"), since the dev->stats.rx_dropped
change was done on the wrong device.

Fixes: d342894c5d2f ("vxlan: virtual extensible lan")
Fixes: ce6502a8f957 ("vxlan: fix a use after free in vxlan_encap_bypass")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Petr Machata <petrm@mellanox.com>
Cc: Ido Schimmel <idosch@mellanox.com>
Cc: Roopa Prabhu <roopa@cumulusnetworks.com>
Cc: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoDocumentation: bring operstate documentation up-to-date
Jouke Witteveen [Thu, 7 Feb 2019 16:14:32 +0000 (17:14 +0100)]
Documentation: bring operstate documentation up-to-date

Netlink has moved from bitmasks to group numbers long ago.

Signed-off-by: Jouke Witteveen <j.witteveen@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf
David S. Miller [Mon, 11 Feb 2019 18:42:50 +0000 (10:42 -0800)]
Merge git://git./pub/scm/linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

The following patchset contains Netfilter fixes for net:

1) Out-of-bound access to packet data from the snmp nat helper,
   from Jann Horn.

2) ICMP(v6) error packets are set as related traffic by conntrack,
   update protocol number before calling nf_nat_ipv4_manip_pkt()
   to use ICMP(v6) rather than the original protocol number,
   from Florian Westphal.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonetfilter: nat: fix spurious connection timeouts
Florian Westphal [Fri, 8 Feb 2019 15:39:52 +0000 (16:39 +0100)]
netfilter: nat: fix spurious connection timeouts

Sander Eikelenboom bisected a NAT related regression down
to the l4proto->manip_pkt indirection removal.

I forgot that ICMP(v6) errors (e.g. PKTTOOBIG) can be set as related
to the existing conntrack entry.

Therefore, when passing the skb to nf_nat_ipv4/6_manip_pkt(), that
ended up calling the wrong l4 manip function, as tuple->dst.protonum
is the original flows l4 protocol (TCP, UDP, etc).

Set the dst protocol field to ICMP(v6), we already have a private copy
of the tuple due to the inversion of src/dst.

Reported-by: Sander Eikelenboom <linux@eikelenboom.it>
Tested-by: Sander Eikelenboom <linux@eikelenboom.it>
Fixes: faec18dbb0405 ("netfilter: nat: remove l4proto->manip_pkt")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonetfilter: nf_nat_snmp_basic: add missing length checks in ASN.1 cbs
Jann Horn [Wed, 6 Feb 2019 21:56:15 +0000 (22:56 +0100)]
netfilter: nf_nat_snmp_basic: add missing length checks in ASN.1 cbs

The generic ASN.1 decoder infrastructure doesn't guarantee that callbacks
will get as much data as they expect; callbacks have to check the `datalen`
parameter before looking at `data`. Make sure that snmp_version() and
snmp_helper() don't read/write beyond the end of the packet data.

(Also move the assignment to `pdata` down below the check to make it clear
that it isn't necessarily a pointer we can use before the `datalen` check.)

Fixes: cc2d58634e0f ("netfilter: nf_nat_snmp_basic: use asn1 decoder library")
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agomac80211: Fix Tx aggregation session tear down with ITXQs
Ilan Peer [Wed, 6 Feb 2019 11:17:21 +0000 (13:17 +0200)]
mac80211: Fix Tx aggregation session tear down with ITXQs

When mac80211 requests the low level driver to stop an ongoing
Tx aggregation, the low level driver is expected to call
ieee80211_stop_tx_ba_cb_irqsafe() to indicate that it is ready
to stop the session. The callback in turn schedules a worker
to complete the session tear down, which in turn also handles
the relevant state for the intermediate Tx queue.

However, as this flow in asynchronous, the intermediate queue
should be stopped and not continue servicing frames, as in
such a case frames that are dequeued would be marked as part
of an aggregation, although the aggregation is already been
stopped.

Fix this by stopping the intermediate Tx queue, before
calling the low level driver to stop the Tx aggregation.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
5 years agocfg80211: prevent speculation on cfg80211_classify8021d() return
Johannes Berg [Wed, 6 Feb 2019 11:17:14 +0000 (13:17 +0200)]
cfg80211: prevent speculation on cfg80211_classify8021d() return

It's possible that the caller of cfg80211_classify8021d() uses the
value to index an array, like mac80211 in ieee80211_downgrade_queue().
Prevent speculation on the return value.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
5 years agocfg80211: pmsr: record netlink port ID
Johannes Berg [Wed, 6 Feb 2019 11:17:07 +0000 (13:17 +0200)]
cfg80211: pmsr: record netlink port ID

Without recording the netlink port ID, we cannot return the
results or complete messages to userspace, nor will we be
able to abort if the socket is closed, so clearly we need
to fill the value.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
5 years agonl80211: Fix FTM per burst maximum value
Aviya Erenfeld [Wed, 6 Feb 2019 11:17:08 +0000 (13:17 +0200)]
nl80211: Fix FTM per burst maximum value

Fix FTM per burst maximum value from 15 to 31
(The maximal bits that represents that number in the frame
is 5 hence a maximal value of 31)

Signed-off-by: Aviya Erenfeld <aviya.erenfeld@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
5 years agomac80211: call drv_ibss_join() on restart
Johannes Berg [Wed, 6 Feb 2019 11:17:12 +0000 (13:17 +0200)]
mac80211: call drv_ibss_join() on restart

If a driver does any significant activity in its ibss_join method,
then it will very well expect that to be called during restart,
before any stations are added. Do that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
5 years agoMerge branch 'r8169-revert-two-commits-due-to-a-regression'
David S. Miller [Sun, 10 Feb 2019 20:54:49 +0000 (12:54 -0800)]
Merge branch 'r8169-revert-two-commits-due-to-a-regression'

Heiner Kallweit says:

====================
r8169: revert two commits due to a regression

Sander reported a regression (kernel panic, see[1]), therefore let's
revert these commits. Removal of the barriers doesn't seem to
contribute to the issue, the patch just overlaps with the problematic
one and only reverting both patches was tested.

[1] https://marc.info/?t=154965066400001&r=1&w=2

v2:
- improve commit message
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoRevert "r8169: make use of xmit_more and __netdev_sent_queue"
Heiner Kallweit [Sun, 10 Feb 2019 14:28:04 +0000 (15:28 +0100)]
Revert "r8169: make use of xmit_more and __netdev_sent_queue"

This reverts commit 2e6eedb4813e34d8d84ac0eb3afb668966f3f356.

Sander reported a regression causing a kernel panic[1],
therefore let's revert this commit.

[1] https://marc.info/?t=154965066400001&r=1&w=2

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoRevert "r8169: remove unneeded mmiowb barriers"
Heiner Kallweit [Sun, 10 Feb 2019 14:26:07 +0000 (15:26 +0100)]
Revert "r8169: remove unneeded mmiowb barriers"

This reverts commit bd7153bd83b806bfcc2e79b7a6f43aa653d06ef3.

There doesn't seem to be anything wrong with this patch,
it's just reverted to get a stable baseline again.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/smc: fix byte_order for rx_curs_confirmed
Ursula Braun [Thu, 7 Feb 2019 13:52:54 +0000 (14:52 +0100)]
net/smc: fix byte_order for rx_curs_confirmed

The recent change in the rx_curs_confirmed assignment disregards
byte order, which causes problems on little endian architectures.
This patch fixes it.

Fixes: b8649efad879 ("net/smc: fix sender_free computation") (net-tree)
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agovsock: cope with memory allocation failure at socket creation time
Paolo Abeni [Thu, 7 Feb 2019 13:13:18 +0000 (14:13 +0100)]
vsock: cope with memory allocation failure at socket creation time

In the unlikely event that the kmalloc call in vmci_transport_socket_init()
fails, we end-up calling vmci_transport_destruct() with a NULL vmci_trans()
and oopsing.

This change addresses the above explicitly checking for zero vmci_trans()
at destruction time.

Reported-by: Xiumei Mu <xmu@redhat.com>
Fixes: d021c344051a ("VSOCK: Introduce VM Sockets")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Jorgen Hansen <jhansen@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: ipv4: use a dedicated counter for icmp_v4 redirect packets
Lorenzo Bianconi [Wed, 6 Feb 2019 18:18:04 +0000 (19:18 +0100)]
net: ipv4: use a dedicated counter for icmp_v4 redirect packets

According to the algorithm described in the comment block at the
beginning of ip_rt_send_redirect, the host should try to send
'ip_rt_redirect_number' ICMP redirect packets with an exponential
backoff and then stop sending them at all assuming that the destination
ignores redirects.
If the device has previously sent some ICMP error packets that are
rate-limited (e.g TTL expired) and continues to receive traffic,
the redirect packets will never be transmitted. This happens since
peer->rate_tokens will be typically greater than 'ip_rt_redirect_number'
and so it will never be reset even if the redirect silence timeout
(ip_rt_redirect_silence) has elapsed without receiving any packet
requiring redirects.

Fix it by using a dedicated counter for the number of ICMP redirect
packets that has been sent by the host

I have not been able to identify a given commit that introduced the
issue since ip_rt_send_redirect implements the same rate-limiting
algorithm from commit 1da177e4c3f4 ("Linux-2.6.12-rc2")

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: sfp: do not probe SFP module before we're attached
Russell King [Wed, 6 Feb 2019 10:52:30 +0000 (10:52 +0000)]
net: sfp: do not probe SFP module before we're attached

When we probe a SFP module, we expect to be able to call the upstream
device's module_insert() function so that the upstream link can be
configured.  However, when the upstream device is delayed, we currently
may end up probing the module before the upstream device is available,
and lose the module_insert() call.

Avoid this by holding off probing the module until the SFP bus is
properly connected to both the SFP socket driver and the upstream
driver.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Fri, 8 Feb 2019 19:21:54 +0000 (11:21 -0800)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:
 "This pull request is dedicated to the upcoming snowpocalypse parts 2
  and 3 in the Pacific Northwest:

   1) Drop profiles are broken because some drivers use dev_kfree_skb*
      instead of dev_consume_skb*, from Yang Wei.

   2) Fix IWLWIFI kconfig deps, from Luca Coelho.

   3) Fix percpu maps updating in bpftool, from Paolo Abeni.

   4) Missing station release in batman-adv, from Felix Fietkau.

   5) Fix some networking compat ioctl bugs, from Johannes Berg.

   6) ucc_geth must reset the BQL queue state when stopping the device,
      from Mathias Thore.

   7) Several XDP bug fixes in virtio_net from Toshiaki Makita.

   8) TSO packets must be sent always on queue 0 in stmmac, from Jose
      Abreu.

   9) Fix socket refcounting bug in RDS, from Eric Dumazet.

  10) Handle sparse cpu allocations in bpf selftests, from Martynas
      Pumputis.

  11) Make sure mgmt frames have enough tailroom in mac80211, from Felix
      Feitkau.

  12) Use safe list walking in sctp_sendmsg() asoc list traversal, from
      Greg Kroah-Hartman.

  13) Make DCCP's ccid_hc_[rt]x_parse_options always check for NULL
      ccid, from Eric Dumazet.

  14) Need to reload WoL password into bcmsysport device after deep
      sleeps, from Florian Fainelli.

  15) Remove filter from mask before freeing in cls_flower, from Petr
      Machata.

  16) Missing release and use after free in error paths of s390 qeth
      code, from Julian Wiedmann.

  17) Fix lockdep false positive in dsa code, from Marc Zyngier.

  18) Fix counting of ATU violations in mv88e6xxx, from Andrew Lunn.

  19) Fix EQ firmware assert in qed driver, from Manish Chopra.

  20) Don't default Caivum PTP to Y in kconfig, from Bjorn Helgaas"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (116 commits)
  net: dsa: b53: Fix for failure when irq is not defined in dt
  sit: check if IPv6 enabled before calling ip6_err_gen_icmpv6_unreach()
  geneve: should not call rt6_lookup() when ipv6 was disabled
  net: Don't default Cavium PTP driver to 'y'
  net: broadcom: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: via-velocity: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: tehuti: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: sun: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: fsl_ucc_hdlc: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: fec_mpc52xx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: smsc: epic100: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: dscc4: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: tulip: de2104x: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net: defxx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
  net/mlx5e: Don't overwrite pedit action when multiple pedit used
  net/mlx5e: Update hw flows when encap source mac changed
  qed*: Advance drivers version to 8.37.0.20
  qed: Change verbosity for coalescing message.
  qede: Fix system crash on configuring channels.
  qed: Consider TX tcs while deriving the max num_queues for PF.
  ...

5 years agoMerge tag 'char-misc-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Fri, 8 Feb 2019 18:56:31 +0000 (10:56 -0800)]
Merge tag 'char-misc-5.0-rc6' of git://git./linux/kernel/git/gregkh/char-misc

Pull char/misc fixes from Greg KH:
 "Here are some small char and misc driver fixes for 5.0-rc6.

  Nothing huge here, some more binderfs fixups found as people use it,
  and there is a "large" selftest added to validate the binderfs code,
  which makes up the majority of this pull request.

  There's also some small mei and mic fixes to resolve some reported
  issues.

  All of these have been in linux-next for over a week with no reported
  issues"

* tag 'char-misc-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  mic: vop: Fix crash on remove
  mic: vop: Fix use-after-free on remove
  binderfs: remove separate device_initcall()
  fpga: stratix10-soc: fix wrong of_node_put() in init function
  mic: vop: Fix broken virtqueues
  mei: free read cb on ctrl_wr list flush
  samples: mei: use /dev/mei0 instead of /dev/mei
  mei: me: add ice lake point device id.
  binderfs: respect limit on binder control creation
  binder: fix CONFIG_ANDROID_BINDER_DEVICES
  selftests: add binderfs selftests

5 years agoMerge tag 'driver-core-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 8 Feb 2019 18:53:44 +0000 (10:53 -0800)]
Merge tag 'driver-core-5.0-rc6' of git://git./linux/kernel/git/gregkh/driver-core

Pull driver core fixes from Greg KH:
 "Here are some driver core fixes for 5.0-rc6.

  Well, not so much "driver core" as "debugfs". There's a lot of
  outstanding debugfs cleanup patches coming in through different
  subsystem trees, and in that process the debugfs core was found that
  it really should return errors when something bad happens, to prevent
  random files from showing up in the root of debugfs afterward. So
  debugfs was fixed up to handle this properly, and then two fixes for
  the relay and blk-mq code was needed as it was making invalid
  assumptions about debugfs return values.

  There's also a cacheinfo fix in here that resolves a tiny issue.

  All of these have been in linux-next for over a week with no reported
  problems"

* tag 'driver-core-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  blk-mq: protect debugfs_create_files() from failures
  relay: check return of create_buf_file() properly
  debugfs: debugfs_lookup() should return NULL if not found
  debugfs: return error values, not NULL
  debugfs: fix debugfs_rename parameter checking
  cacheinfo: Keep the old value if of_property_read_u32 fails

5 years agoMerge tag 'staging-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Fri, 8 Feb 2019 18:51:59 +0000 (10:51 -0800)]
Merge tag 'staging-5.0-rc6' of git://git./linux/kernel/git/gregkh/staging

Pull staging/IIO driver fixes from Greg KH:
 "Here are some small iio and staging driver fixes for 5.0-rc6.

  Nothing big, just resolve some reported IIO driver issues, and one
  staging driver bug. One staging driver patch was added and then
  reverted as well.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'staging-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  Revert "staging: erofs: keep corrupted fs from crashing kernel in erofs_namei()"
  staging: erofs: keep corrupted fs from crashing kernel in erofs_namei()
  staging: octeon: fix broken phylib usage
  iio: ti-ads8688: Update buffer allocation for timestamps
  tools: iio: iio_generic_buffer: make num_loops signed
  iio: adc: axp288: Fix TS-pin handling
  iio: chemical: atlas-ph-sensor: correct IIO_TEMP values to millicelsius

5 years agoMerge tag 'tty-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Linus Torvalds [Fri, 8 Feb 2019 18:49:55 +0000 (10:49 -0800)]
Merge tag 'tty-5.0-rc6' of git://git./linux/kernel/git/gregkh/tty

Pull tty/serial fixes from Greg KH:
 "Here are some small tty and serial fixes for 5.0-rc6.

  Nothing huge, just a few small fixes for reported issues. The speakup
  fix is in here as it is a tty operation issue.

  All of these have been in linux-next for a while with no reported
  problems"

* tag 'tty-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  serial: fix race between flush_to_ldisc and tty_open
  staging: speakup: fix tty-operation NULL derefs
  serial: sh-sci: Do not free irqs that have already been freed
  serial: 8250_pci: Make PCI class test non fatal
  tty: serial: 8250_mtk: Fix potential NULL pointer dereference

5 years agoMerge tag 'usb-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Fri, 8 Feb 2019 18:48:26 +0000 (10:48 -0800)]
Merge tag 'usb-5.0-rc6' of git://git./linux/kernel/git/gregkh/usb

Pull USB fixes from Grek KH:
 "Here are some small USB fixes for 5.0-rc6.

  Nothing huge, the normal amount of USB gadget fixes as well as some
  USB phy fixes. There's also a typec fix as well. Full details are in
  the shortlog.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'usb-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: typec: tcpm: Correct the PPS out_volt calculation
  usb: gadget: musb: fix short isoc packets with inventra dma
  usb: phy: am335x: fix race condition in _probe
  usb: dwc3: exynos: Fix error handling of clk_prepare_enable
  usb: phy: fix link errors
  usb: gadget: udc: net2272: Fix bitwise and boolean operations
  usb: dwc3: gadget: Handle 0 xfer length for OUT EP

5 years agoMerge tag 'xfs-5.0-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Linus Torvalds [Fri, 8 Feb 2019 18:46:14 +0000 (10:46 -0800)]
Merge tag 'xfs-5.0-fixes-1' of git://git./fs/xfs/xfs-linux

Pull xfs fixes from Darrick Wong:
 "Here are a handful of XFS fixes to fix a data corruption problem, a
  crasher bug, and a deadlock.

  Summary:

   - Fix cache coherency problem with writeback mappings

   - Fix buffer deadlock when shutting fs down

   - Fix a null pointer dereference when running online repair"

* tag 'xfs-5.0-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: set buffer ops when repair probes for btree type
  xfs: end sync buffer I/O properly on shutdown error
  xfs: eof trim writeback mapping as soon as it is cached

5 years agoMerge tag 'drm-fixes-2019-02-08' of git://anongit.freedesktop.org/drm/drm
Linus Torvalds [Fri, 8 Feb 2019 18:43:15 +0000 (10:43 -0800)]
Merge tag 'drm-fixes-2019-02-08' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Missed fixes last week as had nothing until amdgpu showed up on
  Saturday. Other stuff has since rolled in along with some more amdgpu
  fixes, so we have two weeks of those, and some i915, vmwgfx, sun4i,
  rockchip and omap fixes.

  amdgpu/radeon:
   - fix crash on passthrough for SI
   - fencing fix for shared buffers
   - APU hwmon fix
   - API powerplay fix
   - eDP freesync fix
   - PASID mgr locking fix
   - KFD warning fix
   - DC/powerplay fix
   - raven revision ids fix
   - vega20 doorbell fix

  i915:
   - SNB display fix
   - SKL srckey mask fix
   - ICL DDI clock selection fix

  vmwgfx:
   - DMA API fix
   - IOMMU detection fix
   - display fixes

  sun4i:
   - tcon clock fix

  rockchip:
   - SPDX identifier fix

  omap:
   - DSI fixes"

* tag 'drm-fixes-2019-02-08' of git://anongit.freedesktop.org/drm/drm: (28 commits)
  drm/omap: dsi: Hack-fix DSI bus flags
  drm/omap: dsi: Fix OF platform depopulate
  drm/omap: dsi: Fix crash in DSI debug dumps
  drm/i915: Try to sanitize bogus DPLL state left over by broken SNB BIOSen
  drm/amd/display: Attach VRR properties for eDP connectors
  drm/amdkfd: Fix if preprocessor statement above kfd_fill_iolink_info_for_cpu
  drm/amdgpu: use spin_lock_irqsave to protect vm_manager.pasid_idr
  drm/i915: always return something on DDI clock selection
  drm/i915: Fix skl srckey mask bits
  drm/vmwgfx: Improve on IOMMU detection
  drm/vmwgfx: Fix setting of dma masks
  drm/vmwgfx: Also check for crtc status while checking for DU active
  drm/vmwgfx: Fix an uninitialized fence handle value
  drm/vmwgfx: Return error code from vmw_execbuf_copy_fence_user
  drm/sun4i: tcon: Prepare and enable TCON channel 0 clock at init
  drm/amdgpu: fix the incorrect external id for raven series
  drm/amdgpu: Implement doorbell self-ring for NBIO 7.4
  drm/amd/display: Fix fclk idle state
  drm/amdgpu: Transfer fences to dmabuf importer
  drm/amd/powerplay: Fix missing break in switch
  ...

5 years agonet: dsa: b53: Fix for failure when irq is not defined in dt
Arun Parameswaran [Fri, 8 Feb 2019 00:01:18 +0000 (16:01 -0800)]
net: dsa: b53: Fix for failure when irq is not defined in dt

Fixes the issues with non BCM58XX chips in the b53 driver
failing, when the irq is not specified in the device tree.

Removed the check for BCM58XX in b53_srab_prepare_irq(),
so the 'port->irq' will be set to '-EXIO' if the irq is not
specified in the device tree.

Fixes: 16994374a6fc ("net: dsa: b53: Make SRAB driver manage port interrupts")
Fixes: b2ddc48a81b5 ("net: dsa: b53: Do not fail when IRQ are not initialized")
Signed-off-by: Arun Parameswaran <arun.parameswaran@broadcom.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'drm-misc-fixes-2019-02-07' of git://anongit.freedesktop.org/drm/drm-misc...
Dave Airlie [Fri, 8 Feb 2019 00:32:47 +0000 (10:32 +1000)]
Merge tag 'drm-misc-fixes-2019-02-07' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes

drm-misc-fixes for v5.0-rc6:
- Fixes to omap/dsi encoder.
- Clock fix for sun4i.
- Licensing header fix for rockchip.
- Fix division by zero in the mode when trying to set a mode on
  i915 with GVT-g enabled.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/84462cef-609f-e2af-084a-f9fe2b05c53e@linux.intel.com
5 years agoMerge tag 'drm-intel-fixes-2019-02-07-1' of git://anongit.freedesktop.org/drm/drm...
Dave Airlie [Fri, 8 Feb 2019 00:30:29 +0000 (10:30 +1000)]
Merge tag 'drm-intel-fixes-2019-02-07-1' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes

drm/i915 fixes for v5.0-rc6:
- SNB DPLL sanitize
- ICL DDI clock selection
- SLK srckey mask

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/87lg2s6nur.fsf@intel.com
5 years agoMerge tag 'platform-drivers-x86-v5.0-2' of git://git.infradead.org/linux-platform...
Linus Torvalds [Thu, 7 Feb 2019 22:54:45 +0000 (15:54 -0700)]
Merge tag 'platform-drivers-x86-v5.0-2' of git://git.infradead.org/linux-platform-drivers-x86

Pull x86 platform driver fixlet from Darren Hart:
 "Correct Documentation/ABI 4.21 KernelVersion to 5.0"

* tag 'platform-drivers-x86-v5.0-2' of git://git.infradead.org/linux-platform-drivers-x86:
  Documentation/ABI: Correct mlxreg-io KernelVersion for 5.0

5 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Thu, 7 Feb 2019 22:53:26 +0000 (15:53 -0700)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull KVM fixes from Paolo Bonzini:
 "Three security fixes"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: nVMX: unconditionally cancel preemption timer in free_nested (CVE-2019-7221)
  KVM: x86: work around leak of uninitialized stack contents (CVE-2019-7222)
  kvm: fix kvm_ioctl_create_device() reference counting (CVE-2019-6974)

5 years agoMerge tag 'nfsd-5.0-1' of git://linux-nfs.org/~bfields/linux
Linus Torvalds [Thu, 7 Feb 2019 22:44:45 +0000 (15:44 -0700)]
Merge tag 'nfsd-5.0-1' of git://linux-nfs.org/~bfields/linux

Pull nfsd fixes from Bruce Fields:
 "Two small nfsd bugfixes for 5.0, for an RDMA bug and a file clone bug"

* tag 'nfsd-5.0-1' of git://linux-nfs.org/~bfields/linux:
  svcrdma: Remove max_sge check at connect time
  nfsd: Fix error return values for nfsd4_clone_file_range()

5 years agoMerge tag 'for-5.0/dm-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/devic...
Linus Torvalds [Thu, 7 Feb 2019 22:42:43 +0000 (15:42 -0700)]
Merge tag 'for-5.0/dm-fixes-2' of git://git./linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mike Snitzer:
 "Both of these fixes address issues in changes merged for 5.0-rc4:

   - Fix DM core's missing memory barrier before waitqueue_active()
     calls.

   - Fix DM core's clone_bio() to work when cloning a subset of a bio
     with an integrity payload; bio_integrity_trim() wasn't getting
     called due to bio_trim()'s early return"

* tag 'for-5.0/dm-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm: don't use bio_trim() afterall
  dm: add memory barrier before waitqueue_active

5 years agoMerge branch 'ipv6-fixes'
David S. Miller [Thu, 7 Feb 2019 18:48:42 +0000 (10:48 -0800)]
Merge branch 'ipv6-fixes'

Hangbin Liu says:

====================
fix two kernel panics when disabled IPv6 on boot up

When disabled IPv6 on boot up, since there is no ipv6 route tables, we should
not call rt6_lookup. Fix them by checking if we have inet6_dev pointer on
netdevice.

v2: Fix idev reference leak, declarations and code mixing as Stefano,
    Eric pointed. Since we only want to check if idev exists and not
    reference it, use __in6_dev_get() insteand of in6_dev_get().
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agosit: check if IPv6 enabled before calling ip6_err_gen_icmpv6_unreach()
Hangbin Liu [Thu, 7 Feb 2019 10:36:11 +0000 (18:36 +0800)]
sit: check if IPv6 enabled before calling ip6_err_gen_icmpv6_unreach()

If we disabled IPv6 from the kernel command line (ipv6.disable=1), we should
not call ip6_err_gen_icmpv6_unreach(). This:

  ip link add sit1 type sit local 192.0.2.1 remote 192.0.2.2 ttl 1
  ip link set sit1 up
  ip addr add 198.51.100.1/24 dev sit1
  ping 198.51.100.2

if IPv6 is disabled at boot time, will crash the kernel.

v2: there's no need to use in6_dev_get(), use __in6_dev_get() instead,
    as we only need to check that idev exists and we are under
    rcu_read_lock() (from netif_receive_skb_internal()).

Reported-by: Jianlin Shi <jishi@redhat.com>
Fixes: ca15a078bd90 ("sit: generate icmpv6 error when receiving icmpv4 error")
Cc: Oussama Ghorbel <ghorbel@pivasoftware.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agogeneve: should not call rt6_lookup() when ipv6 was disabled
Hangbin Liu [Thu, 7 Feb 2019 10:36:10 +0000 (18:36 +0800)]
geneve: should not call rt6_lookup() when ipv6 was disabled

When we add a new GENEVE device with IPv6 remote, checking only for
IS_ENABLED(CONFIG_IPV6) is not enough as we may disable IPv6 in the
kernel command line (ipv6.disable=1), and calling rt6_lookup() would
cause a NULL pointer dereference.

v2:
- don't mix declarations and code (reported by Stefano Brivio, Eric Dumazet)
- there's no need to use in6_dev_get() as we only need to check that
  idev exists (reported by David Ahern). This is under RTNL, so we can
  simply use __in6_dev_get() instead (Stefano, Eric).

Reported-by: Jianlin Shi <jishi@redhat.com>
Fixes: c40e89fd358e9 ("geneve: configure MTU based on a lower device")
Cc: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoKVM: nVMX: unconditionally cancel preemption timer in free_nested (CVE-2019-7221)
Peter Shier [Thu, 11 Oct 2018 18:46:46 +0000 (11:46 -0700)]
KVM: nVMX: unconditionally cancel preemption timer in free_nested (CVE-2019-7221)

Bugzilla: 1671904

There are multiple code paths where an hrtimer may have been started to
emulate an L1 VMX preemption timer that can result in a call to free_nested
without an intervening L2 exit where the hrtimer is normally
cancelled. Unconditionally cancel in free_nested to cover all cases.

Embargoed until Feb 7th 2019.

Signed-off-by: Peter Shier <pshier@google.com>
Reported-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Reported-by: Felix Wilhelm <fwilhelm@google.com>
Cc: stable@kernel.org
Message-Id: <20181011184646.154065-1-pshier@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
5 years agoKVM: x86: work around leak of uninitialized stack contents (CVE-2019-7222)
Paolo Bonzini [Tue, 29 Jan 2019 17:41:16 +0000 (18:41 +0100)]
KVM: x86: work around leak of uninitialized stack contents (CVE-2019-7222)

Bugzilla: 1671930

Emulation of certain instructions (VMXON, VMCLEAR, VMPTRLD, VMWRITE with
memory operand, INVEPT, INVVPID) can incorrectly inject a page fault
when passed an operand that points to an MMIO address.  The page fault
will use uninitialized kernel stack memory as the CR2 and error code.

The right behavior would be to abort the VM with a KVM_EXIT_INTERNAL_ERROR
exit to userspace; however, it is not an easy fix, so for now just
ensure that the error code and CR2 are zero.

Embargoed until Feb 7th 2019.

Reported-by: Felix Wilhelm <fwilhelm@google.com>
Cc: stable@kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
5 years agokvm: fix kvm_ioctl_create_device() reference counting (CVE-2019-6974)
Jann Horn [Sat, 26 Jan 2019 00:54:33 +0000 (01:54 +0100)]
kvm: fix kvm_ioctl_create_device() reference counting (CVE-2019-6974)

kvm_ioctl_create_device() does the following:

1. creates a device that holds a reference to the VM object (with a borrowed
   reference, the VM's refcount has not been bumped yet)
2. initializes the device
3. transfers the reference to the device to the caller's file descriptor table
4. calls kvm_get_kvm() to turn the borrowed reference to the VM into a real
   reference

The ownership transfer in step 3 must not happen before the reference to the VM
becomes a proper, non-borrowed reference, which only happens in step 4.
After step 3, an attacker can close the file descriptor and drop the borrowed
reference, which can cause the refcount of the kvm object to drop to zero.

This means that we need to grab a reference for the device before
anon_inode_getfd(), otherwise the VM can disappear from under us.

Fixes: 852b6d57dc7f ("kvm: add device control API")
Cc: stable@kernel.org
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
5 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid
Linus Torvalds [Thu, 7 Feb 2019 11:51:31 +0000 (11:51 +0000)]
Merge branch 'for-linus' of git://git./linux/kernel/git/hid/hid

Pull HID fix from Jiri Kosina:
 "A fix for a bug in hid-debug that can lock up the kernel in infinite
  loop (CVE-2019-3819), from Vladis Dronov"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:
  HID: debug: fix the ring buffer implementation

5 years agoMerge tag 'sound-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Thu, 7 Feb 2019 08:33:56 +0000 (08:33 +0000)]
Merge tag 'sound-5.0-rc6' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "A collection of a few small fixes.

  The most significant one is the fix for the possible race at loading
  HD-audio drivers. This has been present for long time and surfaced
  only in a rare occasion, but finally spotted out"

* tag 'sound-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda/ca0132 - Fix build error without CONFIG_PCI
  ALSA: compress: Fix stop handling on compressed capture streams
  ALSA: usb-audio: Add support for new T+A USB DAC
  ALSA: hda - Serialize codec registrations
  ALSA: hda/realtek - Use a common helper for hp pin reference
  ALSA: hda/realtek - Fix lose hp_pins for disable auto mute
  ALSA: hda/realtek - Headset microphone support for System76 darp5

5 years agoMerge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
Linus Torvalds [Thu, 7 Feb 2019 08:05:28 +0000 (08:05 +0000)]
Merge tag 'for_linus' of git://git./linux/kernel/git/mst/vhost

Pull virtio fixes from Michael Tsirkin:
 "A small fix for a uapi header, and a fix for VDPA for non-x86 guests"

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  virtio: drop internal struct from UAPI
  virtio: support VIRTIO_F_ORDER_PLATFORM

5 years agoMerge tag 'trace-v5.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt...
Linus Torvalds [Thu, 7 Feb 2019 07:59:01 +0000 (07:59 +0000)]
Merge tag 'trace-v5.0-rc3' of git://git./linux/kernel/git/rostedt/linux-trace

Pull tracing fixes from Steven Rostedt:
 "This has two fixes for uprobe code.

   - Cut and paste fix to have uprobe printks say "uprobe" and not
     "kprobe"

   - Add terminating '\0' byte when copying function arguments"

* tag 'trace-v5.0-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing/uprobes: Fix output for multiple string arguments
  tracing: uprobes: Fix typo in pr_fmt string

5 years agoMerge tag 'fuse-fixes-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mszer...
Linus Torvalds [Thu, 7 Feb 2019 07:52:08 +0000 (07:52 +0000)]
Merge tag 'fuse-fixes-5.0-rc6' of git://git./linux/kernel/git/mszeredi/fuse

Pull fuse fixes from Miklos Szeredi:
 "A fix for a CUSE regression introduced in v4.20, as well as fixes for
  a couple of old bugs"

* tag 'fuse-fixes-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: decrement NR_WRITEBACK_TEMP on the right page
  fuse: call pipe_buf_release() under pipe lock
  cuse: fix ioctl
  fuse: handle zero sized retrieve correctly

5 years agoMerge tag 'pinctrl-v5.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw...
Linus Torvalds [Thu, 7 Feb 2019 07:47:08 +0000 (07:47 +0000)]
Merge tag 'pinctrl-v5.0-2' of git://git./linux/kernel/git/linusw/linux-pinctrl

Pull pin control fixes from Linus Walleij:

 - Mediatek Kconfig fix

 - Sunxi regulator, IRQ banks and pin base fixup

 - Intel Cherryview Strago DMI workaround

 - Potential regmap problem on mcp23s08

* tag 'pinctrl-v5.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
  pinctrl: sunxi: Correct number of IRQ banks on H6 main pin controller
  pinctrl: mcp23s08: spi: Fix regmap allocation for mcp23s18
  pinctrl: cherryview: fix Strago DMI workaround
  pinctrl: sunxi: Consider pin_base when calculating regulator array index
  pinctrl: sunxi: Fix and simplify pin bank regulator handling
  pinctrl: mediatek: fix Kconfig build errors for moore core

5 years agonet: Don't default Cavium PTP driver to 'y'
Bjorn Helgaas [Tue, 5 Feb 2019 20:47:21 +0000 (14:47 -0600)]
net: Don't default Cavium PTP driver to 'y'

8c56df372bc1 ("net: add support for Cavium PTP coprocessor") added the
Cavium PTP coprocessor driver and enabled it by default.  Remove the
"default y" because the driver only applies to Cavium ThunderX processors.

Fixes: 8c56df372bc1 ("net: add support for Cavium PTP coprocessor")
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: broadcom: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:25:54 +0000 (00:25 +0800)]
net: broadcom: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in sbdma_tx_process() when
skb xmit done. It makes drop profiles(dropwatch, perf) more
friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: via-velocity: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:22:54 +0000 (00:22 +0800)]
net: via-velocity: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in velocity_free_tx_buf()
when skb xmit done. It makes drop profiles(dropwatch, perf) more
friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: tehuti: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:21:31 +0000 (00:21 +0800)]
net: tehuti: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in bdx_tx_cleanup() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: sun: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:19:44 +0000 (00:19 +0800)]
net: sun: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called when skb xmit done. It makes
drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: fsl_ucc_hdlc: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:14:51 +0000 (00:14 +0800)]
net: fsl_ucc_hdlc: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in hdlc_tx_done() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: fec_mpc52xx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:12:04 +0000 (00:12 +0800)]
net: fec_mpc52xx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in mpc52xx_fec_tx_interrupt()
when skb xmit done. It makes drop profiles(dropwatch, perf) more
friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: smsc: epic100: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:09:41 +0000 (00:09 +0800)]
net: smsc: epic100: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in epic_tx() when skb xmit
done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: dscc4: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:07:03 +0000 (00:07 +0800)]
net: dscc4: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in dscc4_tx_irq() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: tulip: de2104x: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:03:51 +0000 (00:03 +0800)]
net: tulip: de2104x: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in de_tx() when skb xmit
done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: defxx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles
Yang Wei [Tue, 5 Feb 2019 16:01:04 +0000 (00:01 +0800)]
net: defxx: replace dev_kfree_skb_irq by dev_consume_skb_irq for drop profiles

dev_consume_skb_irq() should be called in dfx_xmt_done() when skb
xmit done. It makes drop profiles(dropwatch, perf) more friendly.

Signed-off-by: Yang Wei <yang.wei9@zte.com.cn>
Reviewed-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx5e: Don't overwrite pedit action when multiple pedit used
Tonghao Zhang [Mon, 28 Jan 2019 23:28:06 +0000 (15:28 -0800)]
net/mlx5e: Don't overwrite pedit action when multiple pedit used

In some case, we may use multiple pedit actions to modify packets.
The command shown as below: the last pedit action is effective.

$ tc filter add dev netdev_rep parent ffff: protocol ip prio 1    \
flower skip_sw ip_proto icmp dst_ip 3.3.3.3        \
action pedit ex munge ip dst set 192.168.1.100 pipe    \
action pedit ex munge eth src set 00:00:00:00:00:01 pipe    \
action pedit ex munge eth dst set 00:00:00:00:00:02 pipe    \
action csum ip pipe    \
action tunnel_key set src_ip 1.1.1.100 dst_ip 1.1.1.200 dst_port 4789 id 100 \
action mirred egress redirect dev vxlan0

To fix it, we add max_mod_hdr_actions to mlx5e_tc_flow_parse_attr struction,
max_mod_hdr_actions will store the max pedit action number we support and
num_mod_hdr_actions indicates how many pedit action we used, and store all
pedit action to mod_hdr_actions.

Fixes: d79b6df6b10a ("net/mlx5e: Add parsing of TC pedit actions to HW format")
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>