platform/kernel/linux-starfive.git
5 years agonet: sched: don't use tc_action->order during action dump
Vlad Buslov [Thu, 23 May 2019 06:32:31 +0000 (09:32 +0300)]
net: sched: don't use tc_action->order during action dump

Function tcf_action_dump() relies on tc_action->order field when starting
nested nla to send action data to userspace. This approach breaks in
several cases:

- When multiple filters point to same shared action, tc_action->order field
  is overwritten each time it is attached to filter. This causes filter
  dump to output action with incorrect attribute for all filters that have
  the action in different position (different order) from the last set
  tc_action->order value.

- When action data is displayed using tc action API (RTM_GETACTION), action
  order is overwritten by tca_action_gd() according to its position in
  resulting array of nl attributes, which will break filter dump for all
  filters attached to that shared action that expect it to have different
  order value.

Don't rely on tc_action->order when dumping actions. Set nla according to
action position in resulting array of actions instead.

Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agocxgb4: Revert "cxgb4: Remove SGE_HOST_PAGE_SIZE dependency on page size"
Vishal Kulkarni [Thu, 23 May 2019 02:37:21 +0000 (08:07 +0530)]
cxgb4: Revert "cxgb4: Remove SGE_HOST_PAGE_SIZE dependency on page size"

This reverts commit 2391b0030e241386d710df10e53e2cfc3c5d4fc1 which has
introduced regression. Now SGE's BAR2 Doorbell/GTS Page Size is
interpreted correctly in the firmware itself by using actual host
page size. Hence previous commit needs to be reverted.

Signed-off-by: Vishal Kulkarni <vishal@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: fec: fix the clk mismatch in failed_reset path
Andy Duan [Thu, 23 May 2019 01:55:28 +0000 (01:55 +0000)]
net: fec: fix the clk mismatch in failed_reset path

Fix the clk mismatch in the error path "failed_reset" because
below error path will disable clk_ahb and clk_ipg directly, it
should use pm_runtime_put_noidle() instead of pm_runtime_put()
to avoid to call runtime resume callback.

Reported-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
Tested-by: Baruch Siach <baruch@tkos.co.il>
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, 23 May 2019 21:45:36 +0000 (14:45 -0700)]
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 your net tree:

1) Fix crash when dumping rules after conversion to RCU,
   from Florian Westphal.

2) Fix incorrect hook reinjection from nf_queue in case NF_REPEAT,
   from Jagdish Motwani.

3) Fix check for route existence in fib extension, from Phil Sutter.

4) Fix use after free in ip_vs_in() hook, from YueHaibing.

5) Check for veth existence from netfilter selftests,
   from Jeffrin Jose T.

6) Checksum corruption in UDP NAT helpers due to typo,
   from Florian Westphal.

7) Pass up packets to classic forwarding path regardless of
   IPv4 DF bit, patch for the flowtable infrastructure from Florian.

8) Set liberal TCP tracking for flows that are placed in the
   flowtable, in case they need to go back to classic forwarding path,
   also from Florian.

9) Don't add flow with sequence adjustment to flowtable, from Florian.

10) Skip IPv4 options from IPv6 datapath in flowtable, from Florian.

11) Add selftest for the flowtable infrastructure, from Florian.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agocxgb4: offload VLAN flows regardless of VLAN ethtype
Raju Rangoju [Thu, 23 May 2019 15:11:44 +0000 (20:41 +0530)]
cxgb4: offload VLAN flows regardless of VLAN ethtype

VLAN flows never get offloaded unless ivlan_vld is set in filter spec.
It's not compulsory for vlan_ethtype to be set.

So, always enable ivlan_vld bit for offloading VLAN flows regardless of
vlan_ethtype is set or not.

Fixes: ad9af3e09c (cxgb4: add tc flower match support for vlan)
Signed-off-by: Raju Rangoju <rajur@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agohsr: fix don't prune the master node from the node_db
Andreas Oetken [Thu, 23 May 2019 11:57:14 +0000 (13:57 +0200)]
hsr: fix don't prune the master node from the node_db

Don't prune the master node in the hsr_prune_nodes function.
Neither time_in[HSR_PT_SLAVE_A] nor time_in[HSR_PT_SLAVE_B]
will ever be updated by hsr_register_frame_in for the master port.
Thus, the master node will be repeatedly pruned leading to
repeated packet loss.
This bug never appeared because the hsr_prune_nodes function
was only called once. Since commit 5150b45fd355
("net: hsr: Fix node prune function for forget time expiry") this issue
is fixed unveiling the issue described above.

Fixes: 5150b45fd355 ("net: hsr: Fix node prune function for forget time expiry")
Signed-off-by: Andreas Oetken <andreas.oetken@siemens.com>
Tested-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: mvpp2: cls: Fix leaked ethtool_rx_flow_rule
Maxime Chevallier [Thu, 23 May 2019 08:47:24 +0000 (10:47 +0200)]
net: mvpp2: cls: Fix leaked ethtool_rx_flow_rule

The flow_rule is only used when configuring the classification tables,
and should be free'd once we're done using it. The current code only
frees it in the error path.

Fixes: 90b509b39ac9 ("net: mvpp2: cls: Add Classification offload support")
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4/igmp: fix build error if !CONFIG_IP_MULTICAST
Eric Dumazet [Thu, 23 May 2019 01:35:16 +0000 (18:35 -0700)]
ipv4/igmp: fix build error if !CONFIG_IP_MULTICAST

ip_sf_list_clear_all() needs to be defined even if !CONFIG_IP_MULTICAST

Fixes: 3580d04aa674 ("ipv4/igmp: fix another memory leak in igmpv3_del_delrec()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4/igmp: fix another memory leak in igmpv3_del_delrec()
Eric Dumazet [Wed, 22 May 2019 23:51:22 +0000 (16:51 -0700)]
ipv4/igmp: fix another memory leak in igmpv3_del_delrec()

syzbot reported memory leaks [1] that I have back tracked to
a missing cleanup from igmpv3_del_delrec() when
(im->sfmode != MCAST_INCLUDE)

Add ip_sf_list_clear_all() and kfree_pmc() helpers to explicitely
handle the cleanups before freeing.

[1]

BUG: memory leak
unreferenced object 0xffff888123e32b00 (size 64):
  comm "softirq", pid 0, jiffies 4294942968 (age 8.010s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 e0 00 00 01 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<000000006105011b>] kmemleak_alloc_recursive include/linux/kmemleak.h:55 [inline]
    [<000000006105011b>] slab_post_alloc_hook mm/slab.h:439 [inline]
    [<000000006105011b>] slab_alloc mm/slab.c:3326 [inline]
    [<000000006105011b>] kmem_cache_alloc_trace+0x13d/0x280 mm/slab.c:3553
    [<000000004bba8073>] kmalloc include/linux/slab.h:547 [inline]
    [<000000004bba8073>] kzalloc include/linux/slab.h:742 [inline]
    [<000000004bba8073>] ip_mc_add1_src net/ipv4/igmp.c:1961 [inline]
    [<000000004bba8073>] ip_mc_add_src+0x36b/0x400 net/ipv4/igmp.c:2085
    [<00000000a46a65a0>] ip_mc_msfilter+0x22d/0x310 net/ipv4/igmp.c:2475
    [<000000005956ca89>] do_ip_setsockopt.isra.0+0x1795/0x1930 net/ipv4/ip_sockglue.c:957
    [<00000000848e2d2f>] ip_setsockopt+0x3b/0xb0 net/ipv4/ip_sockglue.c:1246
    [<00000000b9db185c>] udp_setsockopt+0x4e/0x90 net/ipv4/udp.c:2616
    [<000000003028e438>] sock_common_setsockopt+0x38/0x50 net/core/sock.c:3130
    [<0000000015b65589>] __sys_setsockopt+0x98/0x120 net/socket.c:2078
    [<00000000ac198ef0>] __do_sys_setsockopt net/socket.c:2089 [inline]
    [<00000000ac198ef0>] __se_sys_setsockopt net/socket.c:2086 [inline]
    [<00000000ac198ef0>] __x64_sys_setsockopt+0x26/0x30 net/socket.c:2086
    [<000000000a770437>] do_syscall_64+0x76/0x1a0 arch/x86/entry/common.c:301
    [<00000000d3adb93b>] entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fixes: 9c8bb163ae78 ("igmp, mld: Fix memory leak in igmpv3/mld_del_delrec()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Hangbin Liu <liuhangbin@gmail.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'bnxt_en-Bug-fixes'
David S. Miller [Thu, 23 May 2019 01:02:14 +0000 (18:02 -0700)]
Merge branch 'bnxt_en-Bug-fixes'

Michael Chan says:

===================
bnxt_en: Bug fixes.

There are 4 driver fixes in this series:

1. Fix RX buffer leak during OOM condition.
2. Call pci_disable_msix() under correct conditions to prevent hitting BUG.
3. Reduce unneeded mmeory allocation in kdump kernel to prevent OOM.
4. Don't read device serial number on VFs because it is not supported.

Please queue #1, #2, #3 for -stable as well.  Thanks.
===================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobnxt_en: Device serial number is supported only for PFs.
Vasundhara Volam [Wed, 22 May 2019 23:12:57 +0000 (19:12 -0400)]
bnxt_en: Device serial number is supported only for PFs.

Don't read DSN on VFs that do not have the PCI capability.

Fixes: 03213a996531 ("bnxt: move bp->switch_id initialization to PF probe")
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobnxt_en: Reduce memory usage when running in kdump kernel.
Michael Chan [Wed, 22 May 2019 23:12:56 +0000 (19:12 -0400)]
bnxt_en: Reduce memory usage when running in kdump kernel.

Skip RDMA context memory allocations, reduce to 1 ring, and disable
TPA when running in the kdump kernel.  Without this patch, the driver
fails to initialize with memory allocation errors when running in a
typical kdump kernel.

Fixes: cf6daed098d1 ("bnxt_en: Increase context memory allocations on 57500 chips for RDMA.")
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobnxt_en: Fix possible BUG() condition when calling pci_disable_msix().
Michael Chan [Wed, 22 May 2019 23:12:55 +0000 (19:12 -0400)]
bnxt_en: Fix possible BUG() condition when calling pci_disable_msix().

When making configuration changes, the driver calls bnxt_close_nic()
and then bnxt_open_nic() for the changes to take effect.  A parameter
irq_re_init is passed to the call sequence to indicate if IRQ
should be re-initialized.  This irq_re_init parameter needs to
be included in the bnxt_reserve_rings() call.  bnxt_reserve_rings()
can only call pci_disable_msix() if the irq_re_init parameter is
true, otherwise it may hit BUG() because some IRQs may not have been
freed yet.

Fixes: 41e8d7983752 ("bnxt_en: Modify the ring reservation functions for 57500 series chips.")
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobnxt_en: Fix aggregation buffer leak under OOM condition.
Michael Chan [Wed, 22 May 2019 23:12:54 +0000 (19:12 -0400)]
bnxt_en: Fix aggregation buffer leak under OOM condition.

For every RX packet, the driver replenishes all buffers used for that
packet and puts them back into the RX ring and RX aggregation ring.
In one code path where the RX packet has one RX buffer and one or more
aggregation buffers, we missed recycling the aggregation buffer(s) if
we are unable to allocate a new SKB buffer.  This leads to the
aggregation ring slowly running out of buffers over time.  Fix it
by properly recycling the aggregation buffers.

Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.")
Reported-by: Rakesh Hemnani <rhemnani@fb.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv6: Fix redirect with VRF
David Ahern [Wed, 22 May 2019 22:12:18 +0000 (15:12 -0700)]
ipv6: Fix redirect with VRF

IPv6 redirect is broken for VRF. __ip6_route_redirect walks the FIB
entries looking for an exact match on ifindex. With VRF the flowi6_oif
is updated by l3mdev_update_flow to the l3mdev index and the
FLOWI_FLAG_SKIP_NH_OIF set in the flags to tell the lookup to skip the
device match. For redirects the device match is requires so use that
flag to know when the oif needs to be reset to the skb device index.

Fixes: ca254490c8df ("net: Add VRF support to IPv6 stack")
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: stmmac: fix reset gpio free missing
Jisheng Zhang [Wed, 22 May 2019 10:05:09 +0000 (10:05 +0000)]
net: stmmac: fix reset gpio free missing

Commit 984203ceff27 ("net: stmmac: mdio: remove reset gpio free")
removed the reset gpio free, when the driver is unbinded or rmmod,
we miss the gpio free.

This patch uses managed API to request the reset gpio, so that the
gpio could be freed properly.

Fixes: 984203ceff27 ("net: stmmac: mdio: remove reset gpio free")
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomISDN: make sure device name is NUL terminated
Dan Carpenter [Wed, 22 May 2019 08:45:13 +0000 (11:45 +0300)]
mISDN: make sure device name is NUL terminated

The user can change the device_name with the IMSETDEVNAME ioctl, but we
need to ensure that the user's name is NUL terminated.  Otherwise it
could result in a buffer overflow when we copy the name back to the user
with IMGETDEVINFO ioctl.

I also changed two strcpy() calls which handle the name to strscpy().
Hopefully, there aren't any other ways to create a too long name, but
it's nice to do this as a kernel hardening measure.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: macb: save/restore the remaining registers and features
Claudiu Beznea [Wed, 22 May 2019 08:24:43 +0000 (08:24 +0000)]
net: macb: save/restore the remaining registers and features

SAMA5D2 SoC has a suspend mode where SoC's power is cut off. Due to this
the registers content is lost after a suspend/resume cycle. The current
suspend/resume implementation covers some of these registers. However
there are few which were not treated (e.g. SCRT2 and USRIO). Apart
from this, netdev features are not restored. Treat these issues.

Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'net-tls-fix-device-surprise-removal-with-offload'
David S. Miller [Wed, 22 May 2019 19:21:44 +0000 (12:21 -0700)]
Merge branch 'net-tls-fix-device-surprise-removal-with-offload'

Jakub Kicinski says:

====================
net/tls: fix device surprise removal with offload

This series fixes two issues with device surprise removal.
First we need to take a read lock around resync, otherwise
netdev notifier handler may clear the structures from under
our feet.

Secondly we need to be careful about the interpretation
of device features.  Offload has to be properly cleaned
up even if the TLS device features got cleared after
connection state was installed.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/tls: don't ignore netdev notifications if no TLS features
Jakub Kicinski [Wed, 22 May 2019 02:02:02 +0000 (19:02 -0700)]
net/tls: don't ignore netdev notifications if no TLS features

On device surprise removal path (the notifier) we can't
bail just because the features are disabled.  They may
have been enabled during the lifetime of the device.
This bug leads to leaking netdev references and
use-after-frees if there are active connections while
device features are cleared.

Fixes: e8f69799810c ("net/tls: Add generic NIC offload infrastructure")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/tls: fix state removal with feature flags off
Jakub Kicinski [Wed, 22 May 2019 02:02:01 +0000 (19:02 -0700)]
net/tls: fix state removal with feature flags off

TLS offload drivers shouldn't (and currently don't) block
the TLS offload feature changes based on whether there are
active offloaded connections or not.

This seems to be a good idea, because we want the admin to
be able to disable the TLS offload at any time, and there
is no clean way of disabling it for active connections
(TX side is quite problematic).  So if features are cleared
existing connections will stay offloaded until they close,
and new connections will not attempt offload to a given
device.

However, the offload state removal handling is currently
broken if feature flags get cleared while there are
active TLS offloads.

RX side will completely bail from cleanup, even on normal
remove path, leaving device state dangling, potentially
causing issues when the 5-tuple is reused.  It will also
fail to release the netdev reference.

Remove the RX-side warning message, in next release cycle
it should be printed when features are disabled, rather
than when connection dies, but for that we need a more
efficient method of finding connection of a given netdev
(a'la BPF offload code).

Fixes: 4799ac81e52a ("tls: Add rx inline crypto offload")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/tls: avoid NULL-deref on resync during device removal
Jakub Kicinski [Wed, 22 May 2019 02:02:00 +0000 (19:02 -0700)]
net/tls: avoid NULL-deref on resync during device removal

When netdev with active kTLS sockets in unregistered
notifier callback walks the offloaded sockets and
cleans up offload state.  RX data may still be processed,
however, and if resync was requested prior to device
removal we would hit a NULL pointer dereference on
ctx->netdev use.

Make sure resync is under the device offload lock
and NULL-check the netdev pointer.

This should be safe, because the pointer is set to
NULL either in the netdev notifier (under said lock)
or when socket is completely dead and no resync can
happen.

The other access to ctx->netdev in tls_validate_xmit_skb()
does not dereference the pointer, it just checks it against
other device pointer, so it should be pretty safe (perhaps
we can add a READ_ONCE/WRITE_ONCE there, if paranoid).

Fixes: 4799ac81e52a ("tls: Add rx inline crypto offload")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'Documentation-tls--add-offload-documentation'
David S. Miller [Wed, 22 May 2019 19:18:20 +0000 (12:18 -0700)]
Merge branch 'Documentation-tls--add-offload-documentation'

Jakub Kicinski says:

====================
Documentation: tls: add offload documentation

This set adds documentation for TLS offload. It starts
by making the networking documentation a little easier
to navigate by hiding driver docs a little deeper.
It then RSTifys the existing Kernel TLS documentation.
Last but not least TLS offload documentation is added.
This should help vendors navigate the TLS offload, and
help ensure different implementations stay aligned from
user perspective.

v2:
 - address Alexei's and Boris'es commands on patch 3.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoDocumentation: add TLS offload documentation
Jakub Kicinski [Wed, 22 May 2019 01:57:14 +0000 (18:57 -0700)]
Documentation: add TLS offload documentation

Describe existing kernel TLS offload (added back in Linux 4.19) -
the mechanism, the expected behavior and the notable corner cases.

This documentation is mostly targeting hardware vendors who want
to implement offload, to ensure consistency between implementations.

v2:
 - add emphasis around TLS_SW/TLS_HW/TLS_HW_RECORD;
 - remove mentions of ongoing work (Boris);
 - split the flow of data in SW vs. HW cases in TX overview
   (Boris);
 - call out which fields are updated by the device and which
   are filled by the stack (Boris);
 - move error handling into it's own section (Boris);
 - add more words about fallback (Boris);
 - note that checksum validation is required (Alexei);
 - note that drivers shouldn't pay attention to the TLS
   device features.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoDocumentation: tls: RSTify the ktls documentation
Jakub Kicinski [Wed, 22 May 2019 01:57:13 +0000 (18:57 -0700)]
Documentation: tls: RSTify the ktls documentation

Convert the TLS doc to RST.  Use C code blocks for the code
samples, and mark hyperlinks.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoDocumentation: net: move device drivers docs to a submenu
Jakub Kicinski [Wed, 22 May 2019 01:57:12 +0000 (18:57 -0700)]
Documentation: net: move device drivers docs to a submenu

Some of the device drivers have really long document titles
making the networking table of contents hard to look through.
Place vendor drivers under a submenu.

Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomISDN: Fix indenting in dsp_cmx.c
Dan Carpenter [Tue, 21 May 2019 09:42:56 +0000 (12:42 +0300)]
mISDN: Fix indenting in dsp_cmx.c

We used a script to indent this code back in 2012, but I guess it got
confused by the ifdefs and added some extra tabs.  This patch removes
them.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoocelot: Dont allocate another multicast list, use __dev_mc_sync
Claudiu Manoil [Tue, 21 May 2019 16:52:55 +0000 (19:52 +0300)]
ocelot: Dont allocate another multicast list, use __dev_mc_sync

Doing kmalloc in atomic context is always an issue,
more so for a list that can grow significantly.
Turns out that the driver only uses the duplicated
list of multicast mac addresses to keep track of
what addresses to delete from h/w before committing
the new list from kernel to h/w back again via set_rx_mode,
every time this list gets updated by the kernel.
Given that the h/w knows how to add and delete mac addresses
based on the mac address value alone, __dev_mc_sync should be
the much better choice of kernel API for these operations
avoiding the considerable overhead of maintaining a duplicated
list in the driver.

Signed-off-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Tested-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoValidate required parameters in inet6_validate_link_af
Maxim Mikityanskiy [Tue, 21 May 2019 06:40:04 +0000 (06:40 +0000)]
Validate required parameters in inet6_validate_link_af

inet6_set_link_af requires that at least one of IFLA_INET6_TOKEN or
IFLA_INET6_ADDR_GET_MODE is passed. If none of them is passed, it
returns -EINVAL, which may cause do_setlink() to fail in the middle of
processing other commands and give the following warning message:

  A link change request failed with some changes committed already.
  Interface eth0 may have been left with an inconsistent configuration,
  please check.

Check the presence of at least one of them in inet6_validate_link_af to
detect invalid parameters at an early stage, before do_setlink does
anything. Also validate the address generation mode at an early stage.

Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Linus Torvalds [Wed, 22 May 2019 15:36:16 +0000 (08:36 -0700)]
Merge tag 'arm64-fixes' of git://git./linux/kernel/git/arm64/linux

Pull arm64 fixes from Will Deacon:

 - Fix SPE probe failure when backing auxbuf with high-order pages

 - Fix handling of DMA allocations from outside of the vmalloc area

 - Fix generation of build-id ELF section for vDSO object

 - Disable huge I/O mappings if kernel page table dumping is enabled

 - A few other minor fixes (comments, kconfig etc)

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: vdso: Explicitly add build-id option
  arm64/mm: Inhibit huge-vmap with ptdump
  arm64: Print physical address of page table base in show_pte()
  arm64: don't trash config with compat symbol if COMPAT is disabled
  arm64: assembler: Update comment above cond_yield_neon() macro
  drivers/perf: arm_spe: Don't error on high-order pages for aux buf
  arm64/iommu: handle non-remapped addresses in ->mmap and ->get_sgtable

5 years agoMerge tag 'gfs2-5.1.fixes2' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2...
Linus Torvalds [Wed, 22 May 2019 15:31:09 +0000 (08:31 -0700)]
Merge tag 'gfs2-5.1.fixes2' of git://git./linux/kernel/git/gfs2/linux-gfs2

Pull gfs2 fix from Andreas Gruenbacher:
 "Fix a gfs2 sign extension bug introduced in v4.3"

* tag 'gfs2-5.1.fixes2' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
  gfs2: Fix sign extension bug in gfs2_update_stats

5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Wed, 22 May 2019 15:28:16 +0000 (08:28 -0700)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) Clear up some recent tipc regressions because of registration
    ordering. Fix from Junwei Hu.

 2) tipc's TLV_SET() can read past the end of the supplied buffer during
    the copy. From Chris Packham.

 3) ptp example program doesn't match the kernel, from Richard Cochran.

 4) Outgoing message type fix in qrtr, from Bjorn Andersson.

 5) Flow control regression in stmmac, from Tan Tee Min.

 6) Fix inband autonegotiation in phylink, from Russell King.

 7) Fix sk_bound_dev_if handling in rawv6_bind(), from Mike Manning.

 8) Fix usbnet crash after disconnect, from Kloetzke Jan.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (21 commits)
  usbnet: fix kernel crash after disconnect
  selftests: fib_rule_tests: use pre-defined DEV_ADDR
  net-next: net: Fix typos in ip-sysctl.txt
  ipv6: Consider sk_bound_dev_if when binding a raw socket to an address
  net: phylink: ensure inband AN works correctly
  usbnet: ipheth: fix racing condition
  net: stmmac: dma channel control register need to be init first
  net: stmmac: fix ethtool flow control not able to get/set
  net: qrtr: Fix message type of outgoing packets
  networking: : fix typos in code comments
  ptp: Fix example program to match kernel.
  fddi: fix typos in code comments
  selftests: fib_rule_tests: enable forwarding before ipv4 from/iif test
  selftests: fib_rule_tests: fix local IPv4 address typo
  tipc: Avoid copying bytes beyond the supplied data
  2/2] net: xilinx_emaclite: use readx_poll_timeout() in mdio wait function
  1/2] net: axienet: use readx_poll_timeout() in mdio wait function
  vlan: Mark expected switch fall-through
  macvlan: Mark expected switch fall-through
  net/mlx4_en: ethtool, Remove unsupported SFP EEPROM high pages query
  ...

5 years agoMerge tag 'for-5.2/dm-fix-1' of git://git.kernel.org/pub/scm/linux/kernel/git/device...
Linus Torvalds [Wed, 22 May 2019 15:10:35 +0000 (08:10 -0700)]
Merge tag 'for-5.2/dm-fix-1' of git://git./linux/kernel/git/device-mapper/linux-dm

Pull device mapper fix from Mike Snitzer:
 "Fix a particularly glaring oversight in a DM core commit from 5.1 that
  doesn't properly trim special IOs (e.g. discards) relative to
  corresponding target's max_io_len_target_boundary()"

* tag 'for-5.2/dm-fix-1' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm: make sure to obey max_io_len_target_boundary

5 years agogfs2: Fix sign extension bug in gfs2_update_stats
Andreas Gruenbacher [Fri, 17 May 2019 18:18:43 +0000 (19:18 +0100)]
gfs2: Fix sign extension bug in gfs2_update_stats

Commit 4d207133e9c3 changed the types of the statistic values in struct
gfs2_lkstats from s64 to u64.  Because of that, what should be a signed
value in gfs2_update_stats turned into an unsigned value.  When shifted
right, we end up with a large positive value instead of a small negative
value, which results in an incorrect variance estimate.

Fixes: 4d207133e9c3 ("gfs2: Make statistics unsigned, suitable for use with do_div()")
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Cc: stable@vger.kernel.org # v4.4+
5 years agoselftests: netfilter: add flowtable test script
Florian Westphal [Tue, 21 May 2019 11:24:34 +0000 (13:24 +0200)]
selftests: netfilter: add flowtable test script

Exercises 3 cases:

1. no pmtu discovery (need to frag)
2. no PMTUd + NAT (don't flag packets as invalid from conntrack)
3. PMTU + NAT (need to send icmp error)

The first two cases make sure we handle fragments correctly, i.e.
pass them to classic forwarding path.

Third case checks we offload everything (in the test case,
PMTUd will kick in so all packets should be within link mtu).

Nftables rules will filter packets that are supposed to be
handled by the fast-path.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonetfilter: nft_flow_offload: IPCB is only valid for ipv4 family
Florian Westphal [Tue, 21 May 2019 11:24:33 +0000 (13:24 +0200)]
netfilter: nft_flow_offload: IPCB is only valid for ipv4 family

Guard this with a check vs. ipv4, IPCB isn't valid in ipv6 case.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonetfilter: nft_flow_offload: don't offload when sequence numbers need adjustment
Florian Westphal [Tue, 21 May 2019 11:24:32 +0000 (13:24 +0200)]
netfilter: nft_flow_offload: don't offload when sequence numbers need adjustment

We can't deal with tcp sequence number rewrite in flow_offload.
While at it, simplify helper check, we only need to know if the extension
is present, we don't need the helper data.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonetfilter: nft_flow_offload: set liberal tracking mode for tcp
Florian Westphal [Tue, 21 May 2019 11:24:31 +0000 (13:24 +0200)]
netfilter: nft_flow_offload: set liberal tracking mode for tcp

Without it, whenever a packet has to be pushed up the stack (e.g. because
of mtu mismatch), then conntrack will flag packets as invalid, which in
turn breaks NAT.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonetfilter: nf_flow_table: ignore DF bit setting
Florian Westphal [Tue, 21 May 2019 11:24:30 +0000 (13:24 +0200)]
netfilter: nf_flow_table: ignore DF bit setting

Its irrelevant if the DF bit is set or not, we must pass packet to
stack in either case.

If the DF bit is set, we must pass it to stack so the appropriate
ICMP error can be generated.

If the DF is not set, we must pass it to stack for fragmentation.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agodm: make sure to obey max_io_len_target_boundary
Michael Lass [Tue, 21 May 2019 19:58:07 +0000 (21:58 +0200)]
dm: make sure to obey max_io_len_target_boundary

Commit 61697a6abd24 ("dm: eliminate 'split_discard_bios' flag from DM
target interface") incorrectly removed code from
__send_changing_extent_only() that is required to impose a per-target IO
boundary on IO that exceeds max_io_len_target_boundary().  Otherwise
"special" IO (e.g. DISCARD, WRITE SAME, WRITE ZEROES) can write beyond
where allowed.

Fix this by restoring the max_io_len_target_boundary() limit in
__send_changing_extent_only()

Fixes: 61697a6abd24 ("dm: eliminate 'split_discard_bios' flag from DM target interface")
Cc: stable@vger.kernel.org # 5.1+
Signed-off-by: Michael Lass <bevan@bi-co.net>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
5 years agousbnet: fix kernel crash after disconnect
Kloetzke Jan [Tue, 21 May 2019 13:18:40 +0000 (13:18 +0000)]
usbnet: fix kernel crash after disconnect

When disconnecting cdc_ncm the kernel sporadically crashes shortly
after the disconnect:

  [   57.868812] Unable to handle kernel NULL pointer dereference at virtual address 00000000
  ...
  [   58.006653] PC is at 0x0
  [   58.009202] LR is at call_timer_fn+0xec/0x1b4
  [   58.013567] pc : [<0000000000000000>] lr : [<ffffff80080f5130>] pstate: 00000145
  [   58.020976] sp : ffffff8008003da0
  [   58.024295] x29: ffffff8008003da0 x28: 0000000000000001
  [   58.029618] x27: 000000000000000a x26: 0000000000000100
  [   58.034941] x25: 0000000000000000 x24: ffffff8008003e68
  [   58.040263] x23: 0000000000000000 x22: 0000000000000000
  [   58.045587] x21: 0000000000000000 x20: ffffffc68fac1808
  [   58.050910] x19: 0000000000000100 x18: 0000000000000000
  [   58.056232] x17: 0000007f885aff8c x16: 0000007f883a9f10
  [   58.061556] x15: 0000000000000001 x14: 000000000000006e
  [   58.066878] x13: 0000000000000000 x12: 00000000000000ba
  [   58.072201] x11: ffffffc69ff1db30 x10: 0000000000000020
  [   58.077524] x9 : 8000100008001000 x8 : 0000000000000001
  [   58.082847] x7 : 0000000000000800 x6 : ffffff8008003e70
  [   58.088169] x5 : ffffffc69ff17a28 x4 : 00000000ffff138b
  [   58.093492] x3 : 0000000000000000 x2 : 0000000000000000
  [   58.098814] x1 : 0000000000000000 x0 : 0000000000000000
  ...
  [   58.205800] [<          (null)>]           (null)
  [   58.210521] [<ffffff80080f5298>] expire_timers+0xa0/0x14c
  [   58.215937] [<ffffff80080f542c>] run_timer_softirq+0xe8/0x128
  [   58.221702] [<ffffff8008081120>] __do_softirq+0x298/0x348
  [   58.227118] [<ffffff80080a6304>] irq_exit+0x74/0xbc
  [   58.232009] [<ffffff80080e17dc>] __handle_domain_irq+0x78/0xac
  [   58.237857] [<ffffff8008080cf4>] gic_handle_irq+0x80/0xac
  ...

The crash happens roughly 125..130ms after the disconnect. This
correlates with the 'delay' timer that is started on certain USB tx/rx
errors in the URB completion handler.

The problem is a race of usbnet_stop() with usbnet_start_xmit(). In
usbnet_stop() we call usbnet_terminate_urbs() to cancel all URBs in
flight. This only makes sense if no new URBs are submitted
concurrently, though. But the usbnet_start_xmit() can run at the same
time on another CPU which almost unconditionally submits an URB. The
error callback of the new URB will then schedule the timer after it was
already stopped.

The fix adds a check if the tx queue is stopped after the tx list lock
has been taken. This should reliably prevent the submission of new URBs
while usbnet_terminate_urbs() does its job. The same thing is done on
the rx side even though it might be safe due to other flags that are
checked there.

Signed-off-by: Jan Klötzke <Jan.Kloetzke@preh.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: fib_rule_tests: use pre-defined DEV_ADDR
Hangbin Liu [Tue, 21 May 2019 06:40:47 +0000 (14:40 +0800)]
selftests: fib_rule_tests: use pre-defined DEV_ADDR

DEV_ADDR is defined but not used. Use it in address setting.
Do the same with IPv6 for consistency.

Reported-by: David Ahern <dsahern@gmail.com>
Fixes: fc82d93e57e3 ("selftests: fib_rule_tests: fix local IPv4 address typo")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet-next: net: Fix typos in ip-sysctl.txt
Masanari Iida [Tue, 21 May 2019 03:41:15 +0000 (12:41 +0900)]
net-next: net: Fix typos in ip-sysctl.txt

This patch fixes some spelling typos found in ip-sysctl.txt

Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv6: Consider sk_bound_dev_if when binding a raw socket to an address
Mike Manning [Mon, 20 May 2019 18:57:17 +0000 (19:57 +0100)]
ipv6: Consider sk_bound_dev_if when binding a raw socket to an address

IPv6 does not consider if the socket is bound to a device when binding
to an address. The result is that a socket can be bound to eth0 and
then bound to the address of eth1. If the device is a VRF, the result
is that a socket can only be bound to an address in the default VRF.

Resolve by considering the device if sk_bound_dev_if is set.

Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Tested-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phylink: ensure inband AN works correctly
Russell King [Mon, 20 May 2019 15:07:20 +0000 (16:07 +0100)]
net: phylink: ensure inband AN works correctly

Do not update the link interface mode while the link is down to avoid
spurious link interface changes.

Always call mac_config if we have a PHY to propagate the pause mode
settings to the MAC.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agousbnet: ipheth: fix racing condition
Bernd Eckstein [Mon, 20 May 2019 15:31:09 +0000 (17:31 +0200)]
usbnet: ipheth: fix racing condition

Fix a racing condition in ipheth.c that can lead to slow performance.

Bug: In ipheth_tx(), netif_wake_queue() may be called on the callback
ipheth_sndbulk_callback(), _before_ netif_stop_queue() is called.
When this happens, the queue is stopped longer than it needs to be,
thus reducing network performance.

Fix: Move netif_stop_queue() in front of usb_submit_urb(). Now the order
is always correct. In case, usb_submit_urb() fails, the queue is woken up
again as callback will not fire.

Testing: This racing condition is usually not noticeable, as it has to
occur very frequently to slowdown the network. The callback from the USB
is usually triggered slow enough, so the situation does not appear.
However, on a Ubuntu Linux on VMWare Workstation, running on Windows 10,
the we loose the race quite often and the following speedup can be noticed:

Without this patch: Download:  4.10 Mbit/s, Upload:  4.01 Mbit/s
With this patch:    Download: 36.23 Mbit/s, Upload: 17.61 Mbit/s

Signed-off-by: Oliver Zweigle <Oliver.Zweigle@faro.com>
Signed-off-by: Bernd Eckstein <3ernd.Eckstein@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'selinux-pr-20190521' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Tue, 21 May 2019 19:51:20 +0000 (12:51 -0700)]
Merge tag 'selinux-pr-20190521' of git://git./linux/kernel/git/pcmoore/selinux

Pull SELinux fix from Paul Moore:
 "One small SELinux patch to fix a problem when disconnecting a SCTP
  socket with connect(AF_UNSPEC)"

* tag 'selinux-pr-20190521' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux:
  selinux: do not report error on connect(AF_UNSPEC)

5 years agoMerge tag 'spdx-5.2-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Tue, 21 May 2019 19:33:38 +0000 (12:33 -0700)]
Merge tag 'spdx-5.2-rc2' of git://git./linux/kernel/git/gregkh/driver-core

Pull SPDX update from Greg KH:
 "Here is a series of patches that add SPDX tags to different kernel
  files, based on two different things:

   - SPDX entries are added to a bunch of files that we missed a year
     ago that do not have any license information at all.

     These were either missed because the tool saw the MODULE_LICENSE()
     tag, or some EXPORT_SYMBOL tags, and got confused and thought the
     file had a real license, or the files have been added since the
     last big sweep, or they were Makefile/Kconfig files, which we
     didn't touch last time.

   - Add GPL-2.0-only or GPL-2.0-or-later tags to files where our scan
     tools can determine the license text in the file itself. Where this
     happens, the license text is removed, in order to cut down on the
     700+ different ways we have in the kernel today, in a quest to get
     rid of all of these.

  These patches have been out for review on the linux-spdx@vger mailing
  list, and while they were created by automatic tools, they were
  hand-verified by a bunch of different people, all whom names are on
  the patches are reviewers.

  The reason for these "large" patches is if we were to continue to
  progress at the current rate of change in the kernel, adding license
  tags to individual files in different subsystems, we would be finished
  in about 10 years at the earliest.

  There will be more series of these types of patches coming over the
  next few weeks as the tools and reviewers crunch through the more
  "odd" variants of how to say "GPLv2" that developers have come up with
  over the years, combined with other fun oddities (GPL + a BSD
  disclaimer?) that are being unearthed, with the goal for the whole
  kernel to be cleaned up.

  These diffstats are not small, 3840 files are touched, over 10k lines
  removed in just 24 patches"

* tag 'spdx-5.2-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (24 commits)
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 25
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 24
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 23
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 22
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 21
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 20
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 19
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 18
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 17
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 15
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 14
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 13
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 12
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 11
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 10
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 9
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 7
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 5
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 4
  treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 3
  ...

5 years agoMerge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Linus Torvalds [Tue, 21 May 2019 19:24:24 +0000 (12:24 -0700)]
Merge branch 'linus' of git://git./linux/kernel/git/herbert/crypto-2.6

Pull crypto fixes from Herbert Xu:

 - Two long-standing bugs in the powerpc assembly of vmx

 - Stack overrun caused by HASH_MAX_DESCSIZE being too small

 - Regression in caam

* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
  crypto: vmx - ghash: do nosimd fallback manually
  crypto: vmx - CTR: always increment IV as quadword
  crypto: hash - fix incorrect HASH_MAX_DESCSIZE
  crypto: caam - fix typo in i.MX6 devices list for errata

5 years agonetfilter: nat: fix udp checksum corruption
Florian Westphal [Mon, 20 May 2019 11:48:10 +0000 (13:48 +0200)]
netfilter: nat: fix udp checksum corruption

Due to copy&paste error nf_nat_mangle_udp_packet passes IPPROTO_TCP,
resulting in incorrect udp checksum when payload had to be mangled.

Fixes: dac3fe72596f9 ("netfilter: nat: remove csum_recalc hook")
Reported-by: Marc Haber <mh+netdev@zugschlus.de>
Tested-by: Marc Haber <mh+netdev@zugschlus.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agoselftests: netfilter: missing error check when setting up veth interface
Jeffrin Jose T [Wed, 15 May 2019 06:44:04 +0000 (12:14 +0530)]
selftests: netfilter: missing error check when setting up veth interface

A test for the basic NAT functionality uses ip command which needs veth
device. There is a condition where the kernel support for veth is not
compiled into the kernel and the test script breaks. This patch contains
code for reasonable error display and correct code exit.

Signed-off-by: Jeffrin Jose T <jeffrin@rajagiritech.edu.in>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agoipvs: Fix use-after-free in ip_vs_in
YueHaibing [Fri, 17 May 2019 14:31:49 +0000 (22:31 +0800)]
ipvs: Fix use-after-free in ip_vs_in

BUG: KASAN: use-after-free in ip_vs_in.part.29+0xe8/0xd20 [ip_vs]
Read of size 4 at addr ffff8881e9b26e2c by task sshd/5603

CPU: 0 PID: 5603 Comm: sshd Not tainted 4.19.39+ #30
Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011
Call Trace:
 dump_stack+0x71/0xab
 print_address_description+0x6a/0x270
 kasan_report+0x179/0x2c0
 ip_vs_in.part.29+0xe8/0xd20 [ip_vs]
 ip_vs_in+0xd8/0x170 [ip_vs]
 nf_hook_slow+0x5f/0xe0
 __ip_local_out+0x1d5/0x250
 ip_local_out+0x19/0x60
 __tcp_transmit_skb+0xba1/0x14f0
 tcp_write_xmit+0x41f/0x1ed0
 ? _copy_from_iter_full+0xca/0x340
 __tcp_push_pending_frames+0x52/0x140
 tcp_sendmsg_locked+0x787/0x1600
 ? tcp_sendpage+0x60/0x60
 ? inet_sk_set_state+0xb0/0xb0
 tcp_sendmsg+0x27/0x40
 sock_sendmsg+0x6d/0x80
 sock_write_iter+0x121/0x1c0
 ? sock_sendmsg+0x80/0x80
 __vfs_write+0x23e/0x370
 vfs_write+0xe7/0x230
 ksys_write+0xa1/0x120
 ? __ia32_sys_read+0x50/0x50
 ? __audit_syscall_exit+0x3ce/0x450
 do_syscall_64+0x73/0x200
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x7ff6f6147c60
Code: 73 01 c3 48 8b 0d 28 12 2d 00 f7 d8 64 89 01 48 83 c8 ff c3 66 0f 1f 44 00 00 83 3d 5d 73 2d 00 00 75 10 b8 01 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 31 c3 48 83
RSP: 002b:00007ffd772ead18 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 0000000000000034 RCX: 00007ff6f6147c60
RDX: 0000000000000034 RSI: 000055df30a31270 RDI: 0000000000000003
RBP: 000055df30a31270 R08: 0000000000000000 R09: 0000000000000000
R10: 00007ffd772ead70 R11: 0000000000000246 R12: 00007ffd772ead74
R13: 00007ffd772eae20 R14: 00007ffd772eae24 R15: 000055df2f12ddc0

Allocated by task 6052:
 kasan_kmalloc+0xa0/0xd0
 __kmalloc+0x10a/0x220
 ops_init+0x97/0x190
 register_pernet_operations+0x1ac/0x360
 register_pernet_subsys+0x24/0x40
 0xffffffffc0ea016d
 do_one_initcall+0x8b/0x253
 do_init_module+0xe3/0x335
 load_module+0x2fc0/0x3890
 __do_sys_finit_module+0x192/0x1c0
 do_syscall_64+0x73/0x200
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

Freed by task 6067:
 __kasan_slab_free+0x130/0x180
 kfree+0x90/0x1a0
 ops_free_list.part.7+0xa6/0xc0
 unregister_pernet_operations+0x18b/0x1f0
 unregister_pernet_subsys+0x1d/0x30
 ip_vs_cleanup+0x1d/0xd2f [ip_vs]
 __x64_sys_delete_module+0x20c/0x300
 do_syscall_64+0x73/0x200
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

The buggy address belongs to the object at ffff8881e9b26600 which belongs to the cache kmalloc-4096 of size 4096
The buggy address is located 2092 bytes inside of 4096-byte region [ffff8881e9b26600ffff8881e9b27600)
The buggy address belongs to the page:
page:ffffea0007a6c800 count:1 mapcount:0 mapping:ffff888107c0e600 index:0x0 compound_mapcount: 0
flags: 0x17ffffc0008100(slab|head)
raw: 0017ffffc0008100 dead000000000100 dead000000000200 ffff888107c0e600
raw: 0000000000000000 0000000080070007 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

while unregistering ipvs module, ops_free_list calls
__ip_vs_cleanup, then nf_unregister_net_hooks be called to
do remove nf hook entries. It need a RCU period to finish,
however net->ipvs is set to NULL immediately, which will
trigger NULL pointer dereference when a packet is hooked
and handled by ip_vs_in where net->ipvs is dereferenced.

Another scene is ops_free_list call ops_free to free the
net_generic directly while __ip_vs_cleanup finished, then
calling ip_vs_in will triggers use-after-free.

This patch moves nf_unregister_net_hooks from __ip_vs_cleanup()
to __ip_vs_dev_cleanup(),  where rcu_barrier() is called by
unregister_pernet_device -> unregister_pernet_operations,
that will do the needed grace period.

Reported-by: Hulk Robot <hulkci@huawei.com>
Fixes: efe41606184e ("ipvs: convert to use pernet nf_hook api")
Suggested-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonetfilter: nft_fib: Fix existence check support
Phil Sutter [Wed, 15 May 2019 18:15:32 +0000 (20:15 +0200)]
netfilter: nft_fib: Fix existence check support

NFTA_FIB_F_PRESENT flag was not always honored since eval functions did
not call nft_fib_store_result in all cases.

Given that in all callsites there is a struct net_device pointer
available which holds the interface data to be stored in destination
register, simplify nft_fib_store_result() to just accept that pointer
instead of the nft_pktinfo pointer and interface index. This also
allows to drop the index to interface lookup previously needed to get
the name associated with given index.

Fixes: 055c4b34b94f6 ("netfilter: nft_fib: Support existence check")
Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agonetfilter: nf_queue: fix reinject verdict handling
Jagdish Motwani [Mon, 13 May 2019 18:17:40 +0000 (23:47 +0530)]
netfilter: nf_queue: fix reinject verdict handling

This patch fixes netfilter hook traversal when there are more than 1 hooks
returning NF_QUEUE verdict. When the first queue reinjects the packet,
'nf_reinject' starts traversing hooks with a proper hook_index. However,
if it again receives a NF_QUEUE verdict (by some other netfilter hook), it
queues the packet with a wrong hook_index. So, when the second queue
reinjects the packet, it re-executes hooks in between.

Fixes: 960632ece694 ("netfilter: convert hook list to an array")
Signed-off-by: Jagdish Motwani <jagdish.motwani@sophos.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 25
Thomas Gleixner [Sun, 19 May 2019 13:51:55 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 25

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it would be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 6 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154043.007767574@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 24
Thomas Gleixner [Sun, 19 May 2019 13:51:54 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 24

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or any
  later version this program is distributed in the hope that it will
  be useful but without any warranty without even the implied warranty
  of merchantability or fitness for a particular purpose see the gnu
  general public license for more details

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 50 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.917228456@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 23
Thomas Gleixner [Sun, 19 May 2019 13:51:53 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 23

Based on 1 normalized pattern(s):

  this software may be redistributed and or modified under the terms
  of the gnu general public license as published by the free software
  foundation either version 2 of the license or any later version this
  program is distributed in the hope that it will be useful but
  without any warranty without even the implied warranty of
  merchantability or fitness for a particular purpose see the gnu
  general public license for more details

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 2 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.800979546@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 22
Thomas Gleixner [Sun, 19 May 2019 13:51:52 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 22

Based on 1 normalized pattern(s):

  the code contained herein is licensed under the gnu general public
  license you may obtain a copy of the gnu general public license
  version 2 or later at the following locations

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 4 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.707528683@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 21
Thomas Gleixner [Sun, 19 May 2019 13:51:51 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 21

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program if not write to the free software foundation

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 2 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.615184352@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 20
Thomas Gleixner [Sun, 19 May 2019 13:51:50 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 20

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program see the file copying if not write to the free
  software foundation inc 51 franklin steet fifth floor boston ma
  02110 1301 usa

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 41 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.524645346@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 19
Thomas Gleixner [Sun, 19 May 2019 13:51:49 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 19

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program if not write to the free software foundation 51
  franklin street fifth floor boston ma 02110 1301 usa

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 2 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.432790911@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 18
Thomas Gleixner [Sun, 19 May 2019 13:51:48 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 18

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 or at your option any
  later version this program is distributed in the hope that it will
  be useful but without any warranty without even the implied warranty
  of merchantability or fitness for a particular purpose see the gnu
  general public license for more details you should have received a
  copy of the gnu general public license along with this program see
  the file copying if not write to the free software foundation 675
  mass ave cambridge ma 02139 usa

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 52 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.342335923@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 17
Thomas Gleixner [Sun, 19 May 2019 13:51:47 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 17

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 or at your option any
  later version this program is distributed in the hope that it will
  be useful but without any warranty without even the implied warranty
  of merchantability or fitness for a particular purpose see the gnu
  general public license for more details you should have received a
  copy of the gnu general public license along with this program if
  not see http www gnu org licenses

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 13 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.236620792@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 15
Thomas Gleixner [Sun, 19 May 2019 13:51:45 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 15

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program if not see http www gnu org licenses the full gnu
  general public license is included in this distribution in the file
  called copying

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 9 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154042.045106120@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 14
Thomas Gleixner [Sun, 19 May 2019 13:51:44 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 14

Based on 1 normalized pattern(s):

  gnu cc is free software you can redistribute it and or modify it
  under the terms of the gnu general public license as published by
  the free software foundation either version 2 or at your option any
  later version gnu cc is distributed in the hope that it will be
  useful but without any warranty without even the implied warranty of
  merchantability or fitness for a particular purpose see the gnu
  general public license for more details you should have received a
  copy of the gnu general public license along with gnu cc see the
  file copying if not write to the free software foundation 59 temple
  place suite 330 boston ma 02111 1307 usa

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 3 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.940713800@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 13
Thomas Gleixner [Sun, 19 May 2019 13:51:43 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 13

Based on 2 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program if not see http www gnu org licenses

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details [based]
  [from] [clk] [highbank] [c] you should have received a copy of the
  gnu general public license along with this program if not see http
  www gnu org licenses

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 355 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.837383322@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 12
Thomas Gleixner [Sun, 19 May 2019 13:51:42 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 12

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose good title or non infringement see the gnu general public
  license for more details

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 7 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.727898173@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 11
Thomas Gleixner [Sun, 19 May 2019 13:51:41 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 11

Based on 1 normalized pattern(s):

  this program is free software you can distribute it and or modify it
  under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 1 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.622608495@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 10
Thomas Gleixner [Sun, 19 May 2019 13:51:40 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 10

Based on 1 normalized pattern(s):

  licensed under the fsf s gnu public license v2 or later

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 2 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.526489261@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 9
Thomas Gleixner [Sun, 19 May 2019 13:51:39 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 9

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program if not you can access it online at http www gnu
  org licenses gpl 2 0 html

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 1 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.430943677@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 7
Thomas Gleixner [Sun, 19 May 2019 13:51:37 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 7

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details the full
  gnu general public license is included in this distribution in the
  file called copying

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 9 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.244154651@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 5
Thomas Gleixner [Sun, 19 May 2019 13:51:35 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 5

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program if not see http www gnu org licenses the full gnu
  general public license is included in this distribution in the file
  called license

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 3 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154041.052102771@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 4
Thomas Gleixner [Sun, 19 May 2019 13:51:34 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 4

Based on 1 normalized pattern(s):

  licensed under gplv2 or later

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 118 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154040.961286471@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 3
Thomas Gleixner [Sun, 19 May 2019 13:51:33 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 3

Based on 1 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license version 2 or
  later as published by the free software foundation

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 9 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154040.848507137@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Replace GPLv2 boilerplate/reference with SPDX - rule 1
Thomas Gleixner [Sun, 19 May 2019 13:51:31 +0000 (15:51 +0200)]
treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 1

Based on 2 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option any later version this program is distributed in the
  hope that it will be useful but without any warranty without even
  the implied warranty of merchantability or fitness for a particular
  purpose see the gnu general public license for more details you
  should have received a copy of the gnu general public license along
  with this program if not write to the free software foundation inc
  51 franklin street fifth floor boston ma 02110 1301 usa

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license as published by
  the free software foundation either version 2 of the license or at
  your option [no]_[pad]_[ctrl] any later version this program is
  distributed in the hope that it will be useful but without any
  warranty without even the implied warranty of merchantability or
  fitness for a particular purpose see the gnu general public license
  for more details you should have received a copy of the gnu general
  public license along with this program if not write to the free
  software foundation inc 51 franklin street fifth floor boston ma
  02110 1301 usa

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-or-later

has been chosen to replace the boilerplate/reference in 176 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190519154040.652910950@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Add SPDX license identifier - Makefile/Kconfig
Thomas Gleixner [Sun, 19 May 2019 12:07:45 +0000 (13:07 +0100)]
treewide: Add SPDX license identifier - Makefile/Kconfig

Add SPDX license identifiers to all Make/Kconfig files which:

 - Have no license information of any form

These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:

  GPL-2.0-only

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Add SPDX license identifier for more missed files
Thomas Gleixner [Sun, 19 May 2019 12:08:20 +0000 (13:08 +0100)]
treewide: Add SPDX license identifier for more missed files

Add SPDX license identifiers to all files which:

 - Have no license information of any form

 - Have MODULE_LICENCE("GPL*") inside which was used in the initial
   scan/conversion to ignore the file

These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:

  GPL-2.0-only

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agotreewide: Add SPDX license identifier for missed files
Thomas Gleixner [Sun, 19 May 2019 12:08:55 +0000 (13:08 +0100)]
treewide: Add SPDX license identifier for missed files

Add SPDX license identifiers to all files which:

 - Have no license information of any form

 - Have EXPORT_.*_SYMBOL_GPL inside which was used in the
   initial scan/conversion to ignore the file

These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:

  GPL-2.0-only

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agoselinux: do not report error on connect(AF_UNSPEC)
Paolo Abeni [Fri, 10 May 2019 17:12:33 +0000 (19:12 +0200)]
selinux: do not report error on connect(AF_UNSPEC)

calling connect(AF_UNSPEC) on an already connected TCP socket is an
established way to disconnect() such socket. After commit 68741a8adab9
("selinux: Fix ltp test connect-syscall failure") it no longer works
and, in the above scenario connect() fails with EAFNOSUPPORT.

Fix the above explicitly early checking for AF_UNSPEC family, and
returning success in that case.

Reported-by: Tom Deseyn <tdeseyn@redhat.com>
Cc: stable@vger.kernel.org
Fixes: 68741a8adab9 ("selinux: Fix ltp test connect-syscall failure")
Suggested-by: Paul Moore <paul@paul-moore.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
5 years agonet: stmmac: dma channel control register need to be init first
Weifeng Voon [Tue, 21 May 2019 05:38:38 +0000 (13:38 +0800)]
net: stmmac: dma channel control register need to be init first

stmmac_init_chan() needs to be called before stmmac_init_rx_chan() and
stmmac_init_tx_chan(). This is because if PBLx8 is to be used,
"DMA_CH(#i)_Control.PBLx8" needs to be set before programming
"DMA_CH(#i)_TX_Control.TxPBL" and "DMA_CH(#i)_RX_Control.RxPBL".

Fixes: 47f2a9ce527a ("net: stmmac: dma channel init prepared for multiple queues")
Reviewed-by: Zhang, Baoli <baoli.zhang@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: Weifeng Voon <weifeng.voon@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: stmmac: fix ethtool flow control not able to get/set
Tan, Tee Min [Tue, 21 May 2019 04:55:42 +0000 (12:55 +0800)]
net: stmmac: fix ethtool flow control not able to get/set

Currently ethtool was not able to get/set the flow control due to a
missing "!". It will always return -EOPNOTSUPP even the device is
flow control supported.

This patch fixes the condition check for ethtool flow control get/set
function for ETHTOOL_LINK_MODE_Asym_Pause_BIT.

Fixes: 3c1bcc8614db (“net: ethernet: Convert phydev advertize and supported from u32 to link mode”)
Signed-off-by: Tan, Tee Min <tee.min.tan@intel.com>
Reviewed-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: Voon, Weifeng <weifeng.voon@intel.com@intel.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: qrtr: Fix message type of outgoing packets
Bjorn Andersson [Mon, 20 May 2019 23:51:56 +0000 (16:51 -0700)]
net: qrtr: Fix message type of outgoing packets

QRTR packets has a message type in the header, which is repeated in the
control header. For control packets we therefor copy the type from
beginning of the outgoing payload and use that as message type.

For non-control messages an endianness fix introduced in v5.2-rc1 caused the
type to be 0, rather than QRTR_TYPE_DATA, causing all messages to be dropped by
the receiver. Fix this by converting and using qrtr_type, which will remain
QRTR_TYPE_DATA for non-control messages.

Fixes: 8f5e24514cbd ("net: qrtr: use protocol endiannes variable")
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonetworking: : fix typos in code comments
Weitao Hou [Mon, 20 May 2019 05:23:17 +0000 (13:23 +0800)]
networking: : fix typos in code comments

fix accelleration to acceleration

Signed-off-by: Weitao Hou <houweitaoo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoptp: Fix example program to match kernel.
Richard Cochran [Mon, 20 May 2019 05:15:05 +0000 (22:15 -0700)]
ptp: Fix example program to match kernel.

Ever since commit 3a06c7ac24f9 ("posix-clocks: Remove interval timer
facility and mmap/fasync callbacks") the possibility of PHC based
posix timers has been removed.  In addition it will probably never
make sense to implement this functionality.

This patch removes the misleading example code which seems to suggest
that posix timers for PHC devices will ever be a thing.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'kbuild-fixes-v5.2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahi...
Linus Torvalds [Tue, 21 May 2019 00:22:17 +0000 (17:22 -0700)]
Merge tag 'kbuild-fixes-v5.2' of git://git./linux/kernel/git/masahiroy/linux-kbuild

Pull Kbuild fixes from Masahiro Yamada:

 - remove unused cc-ldoption

 - do not check the name uniquness of builtin modules to avoid false
   positives

* tag 'kbuild-fixes-v5.2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  kbuild: do not check name uniqueness of builtin modules
  kbuild: drop support for cc-ldoption

5 years agofddi: fix typos in code comments
Weitao Hou [Mon, 20 May 2019 04:49:38 +0000 (12:49 +0800)]
fddi: fix typos in code comments

fix abord to abort

Signed-off-by: Weitao Hou <houweitaoo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'kselftests-fib_rule_tests-fix'
David S. Miller [Tue, 21 May 2019 00:20:20 +0000 (20:20 -0400)]
Merge branch 'kselftests-fib_rule_tests-fix'

Hangbin Liu says:

====================
kselftests: fib_rule_tests: fix "from $SRC_IP iif $DEV" match testing

As all the IPv4 testing addresses are in the same subnet and egress device ==
ingress device, to pass "from $SRC_IP iif $DEV" match test, we need enable
forwarding to get the route entry.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: fib_rule_tests: enable forwarding before ipv4 from/iif test
Hangbin Liu [Mon, 20 May 2019 04:36:55 +0000 (12:36 +0800)]
selftests: fib_rule_tests: enable forwarding before ipv4 from/iif test

As all the testing addresses are in the same subnet and egress device ==
ingress device. We need enable forwarding to get the route entry.

Also disable rp_filer separately as some distributions enable it in
startup scripts.

Fixes: 65b2b4939a64 ("selftests: net: initial fib rule tests")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: fib_rule_tests: fix local IPv4 address typo
Hangbin Liu [Mon, 20 May 2019 04:36:54 +0000 (12:36 +0800)]
selftests: fib_rule_tests: fix local IPv4 address typo

The IPv4 testing address are all in 192.51.100.0 subnet. It doesn't make
sense to set a 198.51.100.1 local address. Should be a typo.

Fixes: 65b2b4939a64 ("selftests: net: initial fib rule tests")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agotipc: Avoid copying bytes beyond the supplied data
Chris Packham [Mon, 20 May 2019 03:45:36 +0000 (15:45 +1200)]
tipc: Avoid copying bytes beyond the supplied data

TLV_SET is called with a data pointer and a len parameter that tells us
how many bytes are pointed to by data. When invoking memcpy() we need
to careful to only copy len bytes.

Previously we would copy TLV_LENGTH(len) bytes which would copy an extra
4 bytes past the end of the data pointer which newer GCC versions
complain about.

 In file included from test.c:17:
 In function 'TLV_SET',
     inlined from 'test' at test.c:186:5:
 /usr/include/linux/tipc_config.h:317:3:
 warning: 'memcpy' forming offset [33, 36] is out of the bounds [0, 32]
 of object 'bearer_name' with type 'char[32]' [-Warray-bounds]
     memcpy(TLV_DATA(tlv_ptr), data, tlv_len);
     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 test.c: In function 'test':
 test.c::161:10: note:
 'bearer_name' declared here
     char bearer_name[TIPC_MAX_BEARER_NAME];
          ^~~~~~~~~~~

We still want to ensure any padding bytes at the end are initialised, do
this with a explicit memset() rather than copy bytes past the end of
data. Apply the same logic to TCM_SET.

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'net-readx_poll_timeout'
David S. Miller [Tue, 21 May 2019 00:00:47 +0000 (20:00 -0400)]
Merge branch 'net-readx_poll_timeout'

Benedikt Spranger says:

====================
Convert mdio wait function to use readx_poll_timeout()

On loaded systems with a preemptible kernel both functions
axienet_mdio_wait_until_ready() and xemaclite_mdio_wait() may report a
false positive error return.
Convert both functions to use readx_poll_timeout() to handle the
situation in a safe manner.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years ago2/2] net: xilinx_emaclite: use readx_poll_timeout() in mdio wait function
Kurt Kanzenbach [Sun, 19 May 2019 17:59:37 +0000 (19:59 +0200)]
2/2] net: xilinx_emaclite: use readx_poll_timeout() in mdio wait function

On loaded systems with a preemptible kernel the mdio_wait() function may
report an error while everything is working fine:

xemaclite_mdio_wait():
  xemaclite_readl() -> chip not ready
  --> interrupt here (other work for some time / chip become ready)
  if (time_before_eq(end, jiffies))
    --> false positive error report

Replace the current code with readx_poll_timeout() which takes care
of the situation.

Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years ago1/2] net: axienet: use readx_poll_timeout() in mdio wait function
Kurt Kanzenbach [Sun, 19 May 2019 17:59:36 +0000 (19:59 +0200)]
1/2] net: axienet: use readx_poll_timeout() in mdio wait function

On loaded systems with a preemptible kernel the mdio_wait() function may
report an error while everything is working fine:

axienet_mdio_wait_until_ready():
  axienet_ior() -> chip not ready
  --> interrupt here (other work for some time / chip become ready)
  if (time_before_eq(end, jiffies))
    --> false positive error report

Replace the current code with readx_poll_timeout() which take care
of the situation.

Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agokbuild: do not check name uniqueness of builtin modules
Masahiro Yamada [Mon, 20 May 2019 02:54:37 +0000 (11:54 +0900)]
kbuild: do not check name uniqueness of builtin modules

I just thought it was a good idea to scan builtin.modules in the name
uniqueness checking, but a couple of false positives were found.

Stephen reported a false positive for ppc64_defconfig:

  warning: same basename if the following are built as modules:
    arch/powerpc/platforms/powermac/nvram.ko
    drivers/char/nvram.ko

The former is never built as a module as you see in
arch/powerpc/platforms/powermac/Makefile:

  # CONFIG_NVRAM is an arch. independent tristate symbol, for pmac32 we really
  # need this to be a bool.  Cheat here and pretend CONFIG_NVRAM=m is really
  # CONFIG_NVRAM=y
  obj-$(CONFIG_NVRAM:m=y)         += nvram.o

Another example of false positive is arm64 defconfig:

  warning: same basename if the following are built as modules:
    arch/arm64/lib/crc32.ko
    lib/crc32.ko

It is true CONFIG_CRC32 is a tristate option but it is always 'y' since
it is select'ed by ARM64. Hence, neither of them is built as a module
for the arm64 build.

From the above, modules.builtin essentially contains false positives.
I do not think it is a big deal as far as kmod is concerned, but false
positive warnings in the kernel build make people upset. It is better
to not check it.

Even without builtin.modules checked, we have enough (and more solid)
test coverage with allmodconfig.

While I touched this part, I replaced the sed code with neater one
provided by Stephen.

Link: https://lkml.org/lkml/2019/5/19/120
Link: https://lkml.org/lkml/2019/5/19/123
Fixes: 3a48a91901c5 ("kbuild: check uniqueness of module names")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years agovlan: Mark expected switch fall-through
Gustavo A. R. Silva [Mon, 20 May 2019 14:51:05 +0000 (09:51 -0500)]
vlan: Mark expected switch fall-through

In preparation to enabling -Wimplicit-fallthrough, mark switch
cases where we are expecting to fall through.

This patch fixes the following warning:

net/8021q/vlan_dev.c: In function ‘vlan_dev_ioctl’:
net/8021q/vlan_dev.c:374:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
   if (!net_eq(dev_net(dev), &init_net))
      ^
net/8021q/vlan_dev.c:376:2: note: here
  case SIOCGMIIPHY:
  ^~~~

Warning level 3 was used: -Wimplicit-fallthrough=3

This patch is part of the ongoing efforts to enable
-Wimplicit-fallthrough.

Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomacvlan: Mark expected switch fall-through
Gustavo A. R. Silva [Mon, 20 May 2019 14:44:49 +0000 (09:44 -0500)]
macvlan: Mark expected switch fall-through

In preparation to enabling -Wimplicit-fallthrough, mark switch
cases where we are expecting to fall through.

This patch fixes the following warning:

drivers/net/macvlan.c: In function ‘macvlan_do_ioctl’:
drivers/net/macvlan.c:839:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
   if (!net_eq(dev_net(dev), &init_net))
      ^
drivers/net/macvlan.c:841:2: note: here
  case SIOCGHWTSTAMP:
  ^~~~

Warning level 3 was used: -Wimplicit-fallthrough=3

This patch is part of the ongoing efforts to enable
-Wimplicit-fallthrough.

Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx4_en: ethtool, Remove unsupported SFP EEPROM high pages query
Erez Alfasi [Mon, 20 May 2019 14:42:52 +0000 (17:42 +0300)]
net/mlx4_en: ethtool, Remove unsupported SFP EEPROM high pages query

Querying EEPROM high pages data for SFP module is currently
not supported by our driver but is still tried, resulting in
invalid FW queries.

Set the EEPROM ethtool data length to 256 for SFP module to
limit the reading for page 0 only and prevent invalid FW queries.

Fixes: 7202da8b7f71 ("ethtool, net/mlx4_en: Cable info, get_module_info/eeprom ethtool support")
Signed-off-by: Erez Alfasi <ereza@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agotipc: fix modprobe tipc failed after switch order of device registration
Junwei Hu [Mon, 20 May 2019 06:43:59 +0000 (14:43 +0800)]
tipc: fix modprobe tipc failed after switch order of device registration

Error message printed:
modprobe: ERROR: could not insert 'tipc': Address family not
supported by protocol.
when modprobe tipc after the following patch: switch order of
device registration, commit 7e27e8d6130c
("tipc: switch order of device registration to fix a crash")

Because sock_create_kern(net, AF_TIPC, ...) called by
tipc_topsrv_create_listener() in the initialization process
of tipc_init_net(), so tipc_socket_init() must be execute before that.
Meanwhile, tipc_net_id need to be initialized when sock_create()
called, and tipc_socket_init() is no need to be called for each namespace.

I add a variable tipc_topsrv_net_ops, and split the
register_pernet_subsys() of tipc into two parts, and split
tipc_socket_init() with initialization of pernet params.

By the way, I fixed resources rollback error when tipc_bcast_init()
failed in tipc_init_net().

Fixes: 7e27e8d6130c ("tipc: switch order of device registration to fix a crash")
Signed-off-by: Junwei Hu <hujunwei4@huawei.com>
Reported-by: Wang Wang <wangwang2@huawei.com>
Reported-by: syzbot+1e8114b61079bfe9cbc5@syzkaller.appspotmail.com
Reviewed-by: Kang Zhou <zhoukang7@huawei.com>
Reviewed-by: Suanming Mou <mousuanming@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonetfilter: nf_tables: fix oops during rule dump
Florian Westphal [Tue, 30 Apr 2019 12:53:11 +0000 (14:53 +0200)]
netfilter: nf_tables: fix oops during rule dump

We can oops in nf_tables_fill_rule_info().

Its not possible to fetch previous element in rcu-protected lists
when deletions are not prevented somehow: list_del_rcu poisons
the ->prev pointer value.

Before rcu-conversion this was safe as dump operations did hold
nfnetlink mutex.

Pass previous rule as argument, obtained by keeping a pointer to
the previous rule during traversal.

Fixes: d9adf22a291883 ("netfilter: nf_tables: use call_rcu in netlink dumps")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
5 years agoMerge tag 'for-5.2-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave...
Linus Torvalds [Mon, 20 May 2019 16:52:35 +0000 (09:52 -0700)]
Merge tag 'for-5.2-rc1-tag' of git://git./linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "Notable highlights:

   - fixes for some long-standing bugs in fsync that were quite hard to
     catch but now finaly fixed

   - some fixups to error handling paths that did not properly clean up
     (locking, memory)

   - fix to space reservation for inheriting properties"

* tag 'for-5.2-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  Btrfs: tree-checker: detect file extent items with overlapping ranges
  Btrfs: fix race between ranged fsync and writeback of adjacent ranges
  Btrfs: avoid fallback to transaction commit during fsync of files with holes
  btrfs: extent-tree: Fix a bug that btrfs is unable to add pinned bytes
  btrfs: sysfs: don't leak memory when failing add fsid
  btrfs: sysfs: Fix error path kobject memory leak
  Btrfs: do not abort transaction at btrfs_update_root() after failure to COW path
  btrfs: use the existing reserved items for our first prop for inheritance
  btrfs: don't double unlock on error in btrfs_punch_hole
  btrfs: Check the compression level before getting a workspace