Encap mode is related to switchdev mode only. Move the init of
the encap mode to eswitch_offloads. Before this change, we reported
that eswitch supports encap, even tough the device was in non
SRIOV mode.

Fixes: 7768d197 ('net/mlx5: E-Switch, Add control for encapsulation')
Signed-off-by: NMaor Gottlieb <maorg@mellanox.com>
Reviewed-by: NRoi Dayan <roid@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>

If an app is playing tricks to reuse a socket via tcp_disconnect(),
bytes_acked/received needs to be reset to 0. Otherwise tcp_info will
report the sum of the current and the old connection..

Cc: Eric Dumazet <edumazet@google.com>
Fixes: 0df48c26 ("tcp: add tcpi_bytes_acked to tcp_info")
Fixes: bdd1f9ed ("tcp: add tcpi_bytes_received to tcp_info")
Signed-off-by: NChristoph Paasch <cpaasch@apple.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If mmap() fails it returns MAP_FAILED, which is defined as ((void *) -1).
The current if-statement incorrectly tests if *ring is NULL.

Fixes: 358be656 ("selftests/net: add txring_overwrite")
Signed-off-by: NFrank de Brabander <debrabander@gmail.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The frags_q is not properly initialized, it may result in illegal memory
access when conn_info is NULL.
The "goto free_exit" should be replaced by "goto exit".
Signed-off-by: Yang Wei <albin_yang@163.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Ben Hutchings says:
	"This is the wrong place to change the queue mapping.
	stmmac_xmit() is called with a specific TX queue locked,
	and accessing a different TX queue results in a data race
	for all of that queue's state.

	I think this commit should be reverted upstream and in all
	stable branches.  Instead, the driver should implement the
	ndo_select_queue operation and override the queue mapping there."

Fixes: c5acdbee ("net: stmmac: Send TSO packets always from Queue 0")
Suggested-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NJose Abreu <joabreu@synopsys.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

bpfilter_umh currently printed all messages to /dev/console and this
might interfere the user activity(*).

This commit changes the output device to /dev/kmsg so that the messages
from bpfilter_umh won't show on the console directly.

(*) https://bugzilla.suse.com/show_bug.cgi?id=1140221Signed-off-by: NGary Lin <glin@suse.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Probable cut&paste typo - use the correct field size.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Probable cut&paste typo - use the correct field size.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Taehee Yoo says:

====================
gtp: fix several bugs

This patch series fixes several bugs in the gtp module.

First patch fixes suspicious RCU usage.
The problem is to use rcu_dereference_sk_user_data() outside of
RCU read critical section.

Second patch fixes use-after-free.
gtp_encap_destroy() is called twice.
gtp_encap_destroy() use both gtp->sk0 and gtp->sk1u.
these pointers can be freed in gtp_encap_destroy().
So, gtp_encap_destroy() should avoid using freed sk pointer.

Third patch removes duplicate code in gtp_dellink().
gtp_dellink() calls gtp_encap_disable() twice.
So, remove one of them.

Fourth patch fixes usage of GFP_KERNEL.
GFP_KERNEL can not be used in RCU read critical section.
This patch make ipv4_pdp_add() to use GFP_ATOMIC instead of GFP_KERNEL.

Fifth patch fixes use-after-free in gtp_newlink().
gtp_newlink() uses gtp_net which would be destroyed by the __exit_net
routine.
So, gtp_newlink should not be called after the __exit_net routine.

Sixth patch adds missing error handling routine in gtp_encap_enable().
gtp_encap_enable() will fail, if invalid role value is sent from
user-space. if so, gtp_encap_enable() should execute error handling
routine.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If an invalid role is sent from user space, gtp_encap_enable() will fail.
Then, it should call gtp_encap_disable_sock() but current code doesn't.
It makes memory leak.

Fixes: 91ed81f9 ("gtp: support SGSN-side tunnels")
Signed-off-by: NTaehee Yoo <ap420073@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Current gtp_newlink() could be called after unregister_pernet_subsys().
gtp_newlink() uses gtp_net but it can be destroyed by
unregister_pernet_subsys().
So unregister_pernet_subsys() should be called after
rtnl_link_unregister().

Test commands:
   #SHELL 1
   while :
   do
	   for i in {1..5}
	   do
		./gtp-link add gtp$i &
	   done
	   killall gtp-link
   done

   #SHELL 2
   while :
   do
	modprobe -rv gtp
   done

Splat looks like:
[  753.176631] BUG: KASAN: use-after-free in gtp_newlink+0x9b4/0xa5c [gtp]
[  753.177722] Read of size 8 at addr ffff8880d48f2458 by task gtp-link/7126
[  753.179082] CPU: 0 PID: 7126 Comm: gtp-link Tainted: G        W         5.2.0-rc6+ #50
[  753.185801] Call Trace:
[  753.186264]  dump_stack+0x7c/0xbb
[  753.186863]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.187583]  print_address_description+0xc7/0x240
[  753.188382]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.189097]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.189846]  __kasan_report+0x12a/0x16f
[  753.190542]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.191298]  kasan_report+0xe/0x20
[  753.191893]  gtp_newlink+0x9b4/0xa5c [gtp]
[  753.192580]  ? __netlink_ns_capable+0xc3/0xf0
[  753.193370]  __rtnl_newlink+0xb9f/0x11b0
[ ... ]
[  753.241201] Allocated by task 7186:
[  753.241844]  save_stack+0x19/0x80
[  753.242399]  __kasan_kmalloc.constprop.3+0xa0/0xd0
[  753.243192]  __kmalloc+0x13e/0x300
[  753.243764]  ops_init+0xd6/0x350
[  753.244314]  register_pernet_operations+0x249/0x6f0
[ ... ]
[  753.251770] Freed by task 7178:
[  753.252288]  save_stack+0x19/0x80
[  753.252833]  __kasan_slab_free+0x111/0x150
[  753.253962]  kfree+0xc7/0x280
[  753.254509]  ops_free_list.part.11+0x1c4/0x2d0
[  753.255241]  unregister_pernet_operations+0x262/0x390
[ ... ]
[  753.285883] list_add corruption. next->prev should be prev (ffff8880d48f2458), but was ffff8880d497d878. (next.
[  753.287241] ------------[ cut here ]------------
[  753.287794] kernel BUG at lib/list_debug.c:25!
[  753.288364] invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC KASAN PTI
[  753.289099] CPU: 0 PID: 7126 Comm: gtp-link Tainted: G    B   W         5.2.0-rc6+ #50
[  753.291036] RIP: 0010:__list_add_valid+0x74/0xd0
[  753.291589] Code: 48 39 da 75 27 48 39 f5 74 36 48 39 dd 74 31 48 83 c4 08 b8 01 00 00 00 5b 5d c3 48 89 d9 48b
[  753.293779] RSP: 0018:ffff8880cae8f398 EFLAGS: 00010286
[  753.294401] RAX: 0000000000000075 RBX: ffff8880d497d878 RCX: 0000000000000000
[  753.296260] RDX: 0000000000000075 RSI: 0000000000000008 RDI: ffffed10195d1e69
[  753.297070] RBP: ffff8880cd250ae0 R08: ffffed101b4bff21 R09: ffffed101b4bff21
[  753.297899] R10: 0000000000000001 R11: ffffed101b4bff20 R12: ffff8880d497d878
[  753.298703] R13: 0000000000000000 R14: ffff8880cd250ae0 R15: ffff8880d48f2458
[  753.299564] FS:  00007f5f79805740(0000) GS:ffff8880da400000(0000) knlGS:0000000000000000
[  753.300533] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  753.301231] CR2: 00007fe8c7ef4f10 CR3: 00000000b71a6006 CR4: 00000000000606f0
[  753.302183] Call Trace:
[  753.302530]  gtp_newlink+0x5f6/0xa5c [gtp]
[  753.303037]  ? __netlink_ns_capable+0xc3/0xf0
[  753.303576]  __rtnl_newlink+0xb9f/0x11b0
[  753.304092]  ? rtnl_link_unregister+0x230/0x230

Fixes: 459aa660 ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
Signed-off-by: NTaehee Yoo <ap420073@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ipv4_pdp_add() is called in RCU read-side critical section.
So GFP_KERNEL should not be used in the function.
This patch make ipv4_pdp_add() to use GFP_ATOMIC instead of GFP_KERNEL.

Test commands:
gtp-link add gtp1 &
gtp-tunnel add gtp1 v1 100 200 1.1.1.1 2.2.2.2

Splat looks like:
[  130.618881] =============================
[  130.626382] WARNING: suspicious RCU usage
[  130.626994] 5.2.0-rc6+ #50 Not tainted
[  130.627622] -----------------------------
[  130.628223] ./include/linux/rcupdate.h:266 Illegal context switch in RCU read-side critical section!
[  130.629684]
[  130.629684] other info that might help us debug this:
[  130.629684]
[  130.631022]
[  130.631022] rcu_scheduler_active = 2, debug_locks = 1
[  130.632136] 4 locks held by gtp-tunnel/1025:
[  130.632925]  #0: 000000002b93c8b7 (cb_lock){++++}, at: genl_rcv+0x15/0x40
[  130.634159]  #1: 00000000f17bc999 (genl_mutex){+.+.}, at: genl_rcv_msg+0xfb/0x130
[  130.635487]  #2: 00000000c644ed8e (rtnl_mutex){+.+.}, at: gtp_genl_new_pdp+0x18c/0x1150 [gtp]
[  130.636936]  #3: 0000000007a1cde7 (rcu_read_lock){....}, at: gtp_genl_new_pdp+0x187/0x1150 [gtp]
[  130.638348]
[  130.638348] stack backtrace:
[  130.639062] CPU: 1 PID: 1025 Comm: gtp-tunnel Not tainted 5.2.0-rc6+ #50
[  130.641318] Call Trace:
[  130.641707]  dump_stack+0x7c/0xbb
[  130.642252]  ___might_sleep+0x2c0/0x3b0
[  130.642862]  kmem_cache_alloc_trace+0x1cd/0x2b0
[  130.643591]  gtp_genl_new_pdp+0x6c5/0x1150 [gtp]
[  130.644371]  genl_family_rcv_msg+0x63a/0x1030
[  130.645074]  ? mutex_lock_io_nested+0x1090/0x1090
[  130.645845]  ? genl_unregister_family+0x630/0x630
[  130.646592]  ? debug_show_all_locks+0x2d0/0x2d0
[  130.647293]  ? check_flags.part.40+0x440/0x440
[  130.648099]  genl_rcv_msg+0xa3/0x130
[ ... ]

Fixes: 459aa660 ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
Signed-off-by: NTaehee Yoo <ap420073@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

gtp_encap_disable() in gtp_dellink() is unnecessary because it will be
called by unregister_netdevice().
unregister_netdevice() internally calls gtp_dev_uninit() by ->ndo_uninit().
And gtp_dev_uninit() calls gtp_encap_disable().
Signed-off-by: NTaehee Yoo <ap420073@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

gtp_encap_destroy() is called twice.
1. When interface is deleted.
2. When udp socket is destroyed.
either gtp->sk0 or gtp->sk1u could be freed by sock_put() in
gtp_encap_destroy(). so, when gtp_encap_destroy() is called again,
it would uses freed sk pointer.

patch makes gtp_encap_destroy() to set either gtp->sk0 or gtp->sk1u to
null. in addition, both gtp->sk0 and gtp->sk1u pointer are protected
by rtnl_lock. so, rtnl_lock() is added.

Test command:
   gtp-link add gtp1 &
   killall gtp-link
   ip link del gtp1

Splat looks like:
[   83.182767] BUG: KASAN: use-after-free in __lock_acquire+0x3a20/0x46a0
[   83.184128] Read of size 8 at addr ffff8880cc7d5360 by task ip/1008
[   83.185567] CPU: 1 PID: 1008 Comm: ip Not tainted 5.2.0-rc6+ #50
[   83.188469] Call Trace:
[ ... ]
[   83.200126]  lock_acquire+0x141/0x380
[   83.200575]  ? lock_sock_nested+0x3a/0xf0
[   83.201069]  _raw_spin_lock_bh+0x38/0x70
[   83.201551]  ? lock_sock_nested+0x3a/0xf0
[   83.202044]  lock_sock_nested+0x3a/0xf0
[   83.202520]  gtp_encap_destroy+0x18/0xe0 [gtp]
[   83.203065]  gtp_encap_disable.isra.14+0x13/0x50 [gtp]
[   83.203687]  gtp_dellink+0x56/0x170 [gtp]
[   83.204190]  rtnl_delete_link+0xb4/0x100
[ ... ]
[   83.236513] Allocated by task 976:
[   83.236925]  save_stack+0x19/0x80
[   83.237332]  __kasan_kmalloc.constprop.3+0xa0/0xd0
[   83.237894]  kmem_cache_alloc+0xd8/0x280
[   83.238360]  sk_prot_alloc.isra.42+0x50/0x200
[   83.238874]  sk_alloc+0x32/0x940
[   83.239264]  inet_create+0x283/0xc20
[   83.239684]  __sock_create+0x2dd/0x540
[   83.240136]  __sys_socket+0xca/0x1a0
[   83.240550]  __x64_sys_socket+0x6f/0xb0
[   83.240998]  do_syscall_64+0x9c/0x450
[   83.241466]  entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   83.242061]
[   83.242249] Freed by task 0:
[   83.242616]  save_stack+0x19/0x80
[   83.243013]  __kasan_slab_free+0x111/0x150
[   83.243498]  kmem_cache_free+0x89/0x250
[   83.244444]  __sk_destruct+0x38f/0x5a0
[   83.245366]  rcu_core+0x7e9/0x1c20
[   83.245766]  __do_softirq+0x213/0x8fa

Fixes: 1e3a3abd ("gtp: make GTP sockets in gtp_newlink optional")
Signed-off-by: NTaehee Yoo <ap420073@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

gtp_encap_enable_socket() and gtp_encap_destroy() are not protected
by rcu_read_lock(). and it's not safe to write sk->sk_user_data.
This patch make these functions to use lock_sock() instead of
rcu_dereference_sk_user_data().

Test commands:
    gtp-link add gtp1

Splat looks like:
[   83.238315] =============================
[   83.239127] WARNING: suspicious RCU usage
[   83.239702] 5.2.0-rc6+ #49 Not tainted
[   83.240268] -----------------------------
[   83.241205] drivers/net/gtp.c:799 suspicious rcu_dereference_check() usage!
[   83.243828]
[   83.243828] other info that might help us debug this:
[   83.243828]
[   83.246325]
[   83.246325] rcu_scheduler_active = 2, debug_locks = 1
[   83.247314] 1 lock held by gtp-link/1008:
[   83.248523]  #0: 0000000017772c7f (rtnl_mutex){+.+.}, at: __rtnl_newlink+0x5f5/0x11b0
[   83.251503]
[   83.251503] stack backtrace:
[   83.252173] CPU: 0 PID: 1008 Comm: gtp-link Not tainted 5.2.0-rc6+ #49
[   83.253271] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[   83.254562] Call Trace:
[   83.254995]  dump_stack+0x7c/0xbb
[   83.255567]  gtp_encap_enable_socket+0x2df/0x360 [gtp]
[   83.256415]  ? gtp_find_dev+0x1a0/0x1a0 [gtp]
[   83.257161]  ? memset+0x1f/0x40
[   83.257843]  gtp_newlink+0x90/0xa21 [gtp]
[   83.258497]  ? __netlink_ns_capable+0xc3/0xf0
[   83.259260]  __rtnl_newlink+0xb9f/0x11b0
[   83.260022]  ? rtnl_link_unregister+0x230/0x230
[ ... ]

Fixes: 1e3a3abd ("gtp: make GTP sockets in gtp_newlink optional")
Signed-off-by: NTaehee Yoo <ap420073@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jakub Kicinski says:

====================
net/tls: fix poll() wake up

This small fix + selftest series is very similar to the previous
commit 04b25a54 ("net/tls: fix no wakeup on partial reads").
This time instead of recvmsg we're fixing poll wake up.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a test which checks if leftover record data in TLS
layer correctly wakes up poll().
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NDirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

David reports that RPC applications which use epoll() occasionally
get stuck, and that TLS ULP causes the kernel to not wake applications,
even though read() will return data.

This is indeed true. The ctx->rx_list which holds partially copied
records is not consulted when deciding whether socket is readable.

Note that SO_RCVLOWAT with epoll() is and has always been broken for
kernel TLS. We'd need to parse all records from the TCP layer, instead
of just the first one.

Fixes: 692d7b5d ("tls: Fix recvmsg() to be able to peek across multiple records")
Reported-by: NDavid Beckett <david.beckett@netronome.com>
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NDirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Both ip_neigh_gw4() and ip_neigh_gw6() can return either a valid pointer
or an error pointer, but the code currently checks that the pointer is
not NULL.

Fix this by checking that the pointer is not an error pointer, as this
can result in a NULL pointer dereference [1]. Specifically, I believe
that what happened is that ip_neigh_gw4() returned '-EINVAL'
(0xffffffffffffffea) to which the offset of 'refcnt' (0x70) was added,
which resulted in the address 0x000000000000005a.

[1]
 BUG: KASAN: null-ptr-deref in refcount_inc_not_zero_checked+0x6e/0x180
 Read of size 4 at addr 000000000000005a by task swapper/2/0

 CPU: 2 PID: 0 Comm: swapper/2 Not tainted 5.2.0-rc6-custom-reg-179657-gaa32d89 #396
 Hardware name: Mellanox Technologies Ltd. MSN2010/SA002610, BIOS 5.6.5 08/24/2017
 Call Trace:
 <IRQ>
 dump_stack+0x73/0xbb
 __kasan_report+0x188/0x1ea
 kasan_report+0xe/0x20
 refcount_inc_not_zero_checked+0x6e/0x180
 ipv4_neigh_lookup+0x365/0x12c0
 __neigh_update+0x1467/0x22f0
 arp_process.constprop.6+0x82e/0x1f00
 __netif_receive_skb_one_core+0xee/0x170
 process_backlog+0xe3/0x640
 net_rx_action+0x755/0xd90
 __do_softirq+0x29b/0xae7
 irq_exit+0x177/0x1c0
 smp_apic_timer_interrupt+0x164/0x5e0
 apic_timer_interrupt+0xf/0x20
 </IRQ>

Fixes: 5c9f7c1d ("ipv4: Add helpers for neigh lookup for nexthop")
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Reported-by: NShalom Toledo <shalomt@mellanox.com>
Reviewed-by: NJiri Pirko <jiri@mellanox.com>
Reviewed-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Set the flag of RTL8152_UNPLUG if and only if the device is unplugged.
Some error codes sometimes don't mean the real disconnection of usb device.
For those situations, set the flag of RTL8152_UNPLUG causes the driver skips
some flows of disabling the device, and it let the device stay at incorrect
state.
Signed-off-by: NHayes Wang <hayeswang@realtek.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Cong Wang says:

====================
hsr: a few bug fixes

This patchset contains 3 bug fixes for hsr triggered by a syzbot
reproducer, please check each patch for details.
====================
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>

hsr_port_get_hsr() could return NULL and kernel
could crash:

 BUG: kernel NULL pointer dereference, address: 0000000000000010
 #PF: supervisor read access in kernel mode
 #PF: error_code(0x0000) - not-present page
 PGD 8000000074b84067 P4D 8000000074b84067 PUD 7057d067 PMD 0
 Oops: 0000 [#1] SMP PTI
 CPU: 0 PID: 754 Comm: a.out Not tainted 5.2.0-rc6+ #718
 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-2.fc30 04/01/2014
 RIP: 0010:hsr_dev_xmit+0x20/0x31
 Code: 48 8b 1b eb e0 5b 5d 41 5c c3 66 66 66 66 90 55 48 89 fd 48 8d be 40 0b 00 00 be 04 00 00 00 e8 ee f2 ff ff 48 89 ef 48 89 c6 <48> 8b 40 10 48 89 45 10 e8 6c 1b 00 00 31 c0 5d c3 66 66 66 66 90
 RSP: 0018:ffffb5b400003c48 EFLAGS: 00010246
 RAX: 0000000000000000 RBX: ffff9821b4509a88 RCX: 0000000000000000
 RDX: ffff9821b4509a88 RSI: 0000000000000000 RDI: ffff9821bc3fc7c0
 RBP: ffff9821bc3fc7c0 R08: 0000000000000000 R09: 00000000000c2019
 R10: 0000000000000000 R11: 0000000000000002 R12: ffff9821bc3fc7c0
 R13: ffff9821b4509a88 R14: 0000000000000000 R15: 000000000000006e
 FS:  00007fee112a1800(0000) GS:ffff9821bd800000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 0000000000000010 CR3: 000000006e9ce000 CR4: 00000000000406f0
 Call Trace:
  <IRQ>
  netdev_start_xmit+0x1b/0x38
  dev_hard_start_xmit+0x121/0x21e
  ? validate_xmit_skb.isra.0+0x19/0x1e3
  __dev_queue_xmit+0x74c/0x823
  ? lockdep_hardirqs_on+0x12b/0x17d
  ip6_finish_output2+0x3d3/0x42c
  ? ip6_mtu+0x55/0x5c
  ? mld_sendpack+0x191/0x229
  mld_sendpack+0x191/0x229
  mld_ifc_timer_expire+0x1f7/0x230
  ? mld_dad_timer_expire+0x58/0x58
  call_timer_fn+0x12e/0x273
  __run_timers.part.0+0x174/0x1b5
  ? mld_dad_timer_expire+0x58/0x58
  ? sched_clock_cpu+0x10/0xad
  ? mark_lock+0x26/0x1f2
  ? __lock_is_held+0x40/0x71
  run_timer_softirq+0x26/0x48
  __do_softirq+0x1af/0x392
  irq_exit+0x53/0xa2
  smp_apic_timer_interrupt+0x1c4/0x1d9
  apic_timer_interrupt+0xf/0x20
  </IRQ>

Cc: Arvid Brodin <arvid.brodin@alten.se>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

hsr_link_ops implements ->newlink() but not ->dellink(),
which leads that resources not released after removing the device,
particularly the entries in self_node_db and node_db.

So add ->dellink() implementation to replace the priv_destructor.
This also makes the code slightly easier to understand.

Reported-by: syzbot+c6167ec3de7def23d1e8@syzkaller.appspotmail.com
Cc: Arvid Brodin <arvid.brodin@alten.se>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

hsr_del_port() should release all the resources allocated
in hsr_add_port().

As a consequence of this change, hsr_for_each_port() is no
longer safe to work with hsr_del_port(), switch to
list_for_each_entry_safe() as we always hold RTNL lock.

Cc: Arvid Brodin <arvid.brodin@alten.se>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Steffen Klassert says:

====================
pull request (net): ipsec 2019-07-05

1)  Fix xfrm selector prefix length validation for
    inter address family tunneling.
    From Anirudh Gupta.

2) Fix a memleak in pfkey.
   From Jeremy Sowden.

3) Fix SA selector validation to allow empty selectors again.
   From Nicolas Dichtel.

4) Select crypto ciphers for xfrm_algo, this fixes some
   randconfig builds. From Arnd Bergmann.

5) Remove a duplicated assignment in xfrm_bydst_resize.
   From Cong Wang.

6) Fix a hlist corruption on hash rebuild.
   From Florian Westphal.

7) Fix a memory leak when creating xfrm interfaces.
   From Nicolas Dichtel.

Please pull or let me know if there are problems.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

bond_xmit_roundrobin() checks for IGMP packets but it parses
the IP header even before checking skb->protocol.

We should validate the IP header with pskb_may_pull() before
using iph->protocol.

Reported-and-tested-by: syzbot+e5be16aa39ad6e755391@syzkaller.appspotmail.com
Fixes: a2fd940f ("bonding: fix broken multicast with round-robin mode")
Cc: Jay Vosburgh <j.vosburgh@gmail.com>
Cc: Veaceslav Falico <vfalico@gmail.com>
Cc: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Daniel Borkmann says:

====================
pull-request: bpf 2019-07-03

The following pull-request contains BPF updates for your *net* tree.

The main changes are:

1) Fix the interpreter to properly handle BPF_ALU32 | BPF_ARSH
   on BE architectures, from Jiong.

2) Fix several bugs in the x32 BPF JIT for handling shifts by 0,
   from Luke and Xi.

3) Fix NULL pointer deref in btf_type_is_resolve_source_only(),
   from Stanislav.

4) Properly handle the check that forwarding is enabled on the device
   in bpf_ipv6_fib_lookup() helper code, from Anton.

5) Fix UAPI bpf_prog_info fields alignment for archs that have 16 bit
   alignment such as m68k, from Baruch.

6) Fix kernel hanging in unregister_netdevice loop while unregistering
   device bound to XDP socket, from Ilya.

7) Properly terminate tail update in xskq_produce_flush_desc(), from Nathan.

8) Fix broken always_inline handling in test_lwt_seg6local, from Jiri.

9) Fix bpftool to use correct argument in cgroup errors, from Jakub.

10) Fix detaching dummy prog in XDP redirect sample code, from Prashant.

11) Add Jonathan to AF_XDP reviewers, from Björn.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The hip07 chip support vlan TSO, this patch adds NETIF_F_TSO
and NETIF_F_TSO6 flags to vlan_features to improve the
performance after adding vlan to the net ports.
Signed-off-by: NYonglong Liu <liuyonglong@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now all ctrl chunks are counted for asoc stats.octrlchunks and net
SCTP_MIB_OUTCTRLCHUNKS either after queuing up or bundling, other
than the chunk maked and bundled in sctp_packet_bundle_sack, which
caused 'outctrlchunks' not consistent with 'inctrlchunks' in peer.

This issue exists since very beginning, here to fix it by increasing
both net SCTP_MIB_OUTCTRLCHUNKS and asoc stats.octrlchunks when sack
chunk is maked and bundled in sctp_packet_bundle_sack.
Reported-by: NJa Ram Jeon <jajeon@redhat.com>
Signed-off-by: NXin Long <lucien.xin@gmail.com>
Acked-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

r8153b_rx_agg_chg_indicate() needs to be called after enabling TX/RX and
before calling rxdy_gated_en(tp, false). Otherwise, the change of the
settings of RX aggregation wouldn't work.

Besides, adjust rtl8152_set_coalesce() for the same reason. If
rx_coalesce_usecs is changed, restart TX/RX to let the setting work.
Signed-off-by: NHayes Wang <hayeswang@realtek.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If IPV6 was disabled, then ss command would cause a kernel warning
because the command was attempting to dump IPV6 socket information.
The fix is to just remove the warning.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=202249
Fixes: 432490f9 ("net: ip, diag -- Add diag interface for raw sockets")
Signed-off-by: NStephen Hemminger <stephen@networkplumber.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Device that bound to XDP socket will not have zero refcount until the
userspace application will not close it. This leads to hang inside
'netdev_wait_allrefs()' if device unregistering requested:

  # ip link del p1
  < hang on recvmsg on netlink socket >

  # ps -x | grep ip
  5126  pts/0    D+   0:00 ip link del p1

  # journalctl -b

  Jun 05 07:19:16 kernel:
  unregister_netdevice: waiting for p1 to become free. Usage count = 1

  Jun 05 07:19:27 kernel:
  unregister_netdevice: waiting for p1 to become free. Usage count = 1
  ...

Fix that by implementing NETDEV_UNREGISTER event notification handler
to properly clean up all the resources and unref device.

This should also allow socket killing via ss(8) utility.

Fixes: 965a9909 ("xsk: add support for bind for Rx")
Signed-off-by: NIlya Maximets <i.maximets@samsung.com>
Acked-by: NJonathan Lemon <jonathan.lemon@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

Device pointer stored in umem regardless of zero-copy mode,
so we heed to hold the device in all cases.

Fixes: c9b47cc1 ("xsk: fix bug when trying to use both copy and zero-copy on one queue id")
Signed-off-by: NIlya Maximets <i.maximets@samsung.com>
Acked-by: NJonathan Lemon <jonathan.lemon@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

Selftests are reporting this failure in test_lwt_seg6local.sh:

+ ip netns exec ns2 ip -6 route add fb00::6 encap bpf in obj test_lwt_seg6local.o sec encap_srh dev veth2
Error fetching program/map!
Failed to parse eBPF program: Operation not permitted

The problem is __attribute__((always_inline)) alone is not enough to prevent
clang from inserting those functions in .text. In that case, .text is not
marked as relocateable.

See the output of objdump -h test_lwt_seg6local.o:

Idx Name          Size      VMA               LMA               File off  Algn
  0 .text         00003530  0000000000000000  0000000000000000  00000040  2**3
                  CONTENTS, ALLOC, LOAD, READONLY, CODE

This causes the iproute bpf loader to fail in bpf_fetch_prog_sec:
bpf_has_call_data returns true but bpf_fetch_prog_relo fails as there's no
relocateable .text section in the file.

To fix this, convert to 'static __always_inline'.

v2: Use 'static __always_inline' instead of 'static inline
    __attribute__((always_inline))'

Fixes: c99a84ea ("selftests/bpf: test for seg6local End.BPF action")
Signed-off-by: NJiri Benc <jbenc@redhat.com>
Acked-by: NYonghong Song <yhs@fb.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

There are currently no tests for ALU64 shift operations when the shift
amount is 0. This adds 6 new tests to make sure they are equivalent
to a no-op. The x32 JIT had such bugs that could have been caught by
these tests.

Cc: Xi Wang <xi.wang@gmail.com>
Signed-off-by: NLuke Nelson <luke.r.nels@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

The current x32 BPF JIT does not correctly compile shift operations when
the immediate shift amount is 0. The expected behavior is for this to
be a no-op.

The following program demonstrates the bug. The expexceted result is 1,
but the current JITed code returns 2.

  r0 = 1
  r1 = 1
  r1 <<= 0
  if r1 == 1 goto end
  r0 = 2
end:
  exit

This patch simplifies the code and fixes the bug.

Fixes: 03f5781b ("bpf, x86_32: add eBPF JIT compiler for ia32")
Co-developed-by: NXi Wang <xi.wang@gmail.com>
Signed-off-by: NXi Wang <xi.wang@gmail.com>
Signed-off-by: NLuke Nelson <luke.r.nels@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

The current x32 BPF JIT for shift operations is not correct when the
shift amount in a register is 0. The expected behavior is a no-op, whereas
the current implementation changes bits in the destination register.

The following example demonstrates the bug. The expected result of this
program is 1, but the current JITed code returns 2.

  r0 = 1
  r1 = 1
  r2 = 0
  r1 <<= r2
  if r1 == 1 goto end
  r0 = 2
end:
  exit

The bug is caused by an incorrect assumption by the JIT that a shift by
32 clear the register. On x32 however, shifts use the lower 5 bits of
the source, making a shift by 32 equivalent to a shift by 0.

This patch fixes the bug using double-precision shifts, which also
simplifies the code.

Fixes: 03f5781b ("bpf, x86_32: add eBPF JIT compiler for ia32")
Co-developed-by: NXi Wang <xi.wang@gmail.com>
Signed-off-by: NXi Wang <xi.wang@gmail.com>
Signed-off-by: NLuke Nelson <luke.r.nels@gmail.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

The following commands produce a backtrace and return an error but the xfrm
interface is created (in the wrong netns):
$ ip netns add foo
$ ip netns add bar
$ ip -n foo netns set bar 0
$ ip -n foo link add xfrmi0 link-netnsid 0 type xfrm dev lo if_id 23
RTNETLINK answers: Invalid argument
$ ip -n bar link ls xfrmi0
2: xfrmi0@lo: <NOARP,M-DOWN> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/none 00:00:00:00:00:00 brd 00:00:00:00:00:00

Here is the backtrace:
[   79.879174] WARNING: CPU: 0 PID: 1178 at net/core/dev.c:8172 rollback_registered_many+0x86/0x3c1
[   79.880260] Modules linked in: xfrm_interface nfsv3 nfs_acl auth_rpcgss nfsv4 nfs lockd grace sunrpc fscache button parport_pc parport serio_raw evdev pcspkr loop ext4 crc16 mbcache jbd2 crc32c_generic ide_cd_mod ide_gd_mod cdrom ata_$
eneric ata_piix libata scsi_mod 8139too piix psmouse i2c_piix4 ide_core 8139cp mii i2c_core floppy
[   79.883698] CPU: 0 PID: 1178 Comm: ip Not tainted 5.2.0-rc6+ #106
[   79.884462] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[   79.885447] RIP: 0010:rollback_registered_many+0x86/0x3c1
[   79.886120] Code: 01 e8 d7 7d c6 ff 0f 0b 48 8b 45 00 4c 8b 20 48 8d 58 90 49 83 ec 70 48 8d 7b 70 48 39 ef 74 44 8a 83 d0 04 00 00 84 c0 75 1f <0f> 0b e8 61 cd ff ff 48 b8 00 01 00 00 00 00 ad de 48 89 43 70 66
[   79.888667] RSP: 0018:ffffc900015ab740 EFLAGS: 00010246
[   79.889339] RAX: ffff8882353e5700 RBX: ffff8882353e56a0 RCX: ffff8882353e5710
[   79.890174] RDX: ffffc900015ab7e0 RSI: ffffc900015ab7e0 RDI: ffff8882353e5710
[   79.891029] RBP: ffffc900015ab7e0 R08: ffffc900015ab7e0 R09: ffffc900015ab7e0
[   79.891866] R10: ffffc900015ab7a0 R11: ffffffff82233fec R12: ffffc900015ab770
[   79.892728] R13: ffffffff81eb7ec0 R14: ffff88822ed6cf00 R15: 00000000ffffffea
[   79.893557] FS:  00007ff350f31740(0000) GS:ffff888237a00000(0000) knlGS:0000000000000000
[   79.894581] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   79.895317] CR2: 00000000006c8580 CR3: 000000022c272000 CR4: 00000000000006f0
[   79.896137] Call Trace:
[   79.896464]  unregister_netdevice_many+0x12/0x6c
[   79.896998]  __rtnl_newlink+0x6e2/0x73b
[   79.897446]  ? __kmalloc_node_track_caller+0x15e/0x185
[   79.898039]  ? pskb_expand_head+0x5f/0x1fe
[   79.898556]  ? stack_access_ok+0xd/0x2c
[   79.899009]  ? deref_stack_reg+0x12/0x20
[   79.899462]  ? stack_access_ok+0xd/0x2c
[   79.899927]  ? stack_access_ok+0xd/0x2c
[   79.900404]  ? __module_text_address+0x9/0x4f
[   79.900910]  ? is_bpf_text_address+0x5/0xc
[   79.901390]  ? kernel_text_address+0x67/0x7b
[   79.901884]  ? __kernel_text_address+0x1a/0x25
[   79.902397]  ? unwind_get_return_address+0x12/0x23
[   79.903122]  ? __cmpxchg_double_slab.isra.37+0x46/0x77
[   79.903772]  rtnl_newlink+0x43/0x56
[   79.904217]  rtnetlink_rcv_msg+0x200/0x24c

In fact, each time a xfrm interface was created, a netdev was allocated
by __rtnl_newlink()/rtnl_create_link() and then another one by
xfrmi_newlink()/xfrmi_create(). Only the second one was registered, it's
why the previous commands produce a backtrace: dev_change_net_namespace()
was called on a netdev with reg_state set to NETREG_UNINITIALIZED (the
first one).

CC: Lorenzo Colitti <lorenzo@google.com>
CC: Benedict Wong <benedictwong@google.com>
CC: Steffen Klassert <steffen.klassert@secunet.com>
CC: Shannon Nelson <shannon.nelson@oracle.com>
CC: Antony Antony <antony@phenome.org>
CC: Eyal Birger <eyal.birger@gmail.com>
Fixes: f203b76d ("xfrm: Add virtual xfrm interfaces")
Reported-by: NJulien Floret <julien.floret@6wind.com>
Signed-off-by: NNicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

syzbot reported following spat:

BUG: KASAN: use-after-free in __write_once_size include/linux/compiler.h:221
BUG: KASAN: use-after-free in hlist_del_rcu include/linux/rculist.h:455
BUG: KASAN: use-after-free in xfrm_hash_rebuild+0xa0d/0x1000 net/xfrm/xfrm_policy.c:1318
Write of size 8 at addr ffff888095e79c00 by task kworker/1:3/8066
Workqueue: events xfrm_hash_rebuild
Call Trace:
 __write_once_size include/linux/compiler.h:221 [inline]
 hlist_del_rcu include/linux/rculist.h:455 [inline]
 xfrm_hash_rebuild+0xa0d/0x1000 net/xfrm/xfrm_policy.c:1318
 process_one_work+0x814/0x1130 kernel/workqueue.c:2269
Allocated by task 8064:
 __kmalloc+0x23c/0x310 mm/slab.c:3669
 kzalloc include/linux/slab.h:742 [inline]
 xfrm_hash_alloc+0x38/0xe0 net/xfrm/xfrm_hash.c:21
 xfrm_policy_init net/xfrm/xfrm_policy.c:4036 [inline]
 xfrm_net_init+0x269/0xd60 net/xfrm/xfrm_policy.c:4120
 ops_init+0x336/0x420 net/core/net_namespace.c:130
 setup_net+0x212/0x690 net/core/net_namespace.c:316

The faulting address is the address of the old chain head,
free'd by xfrm_hash_resize().

In xfrm_hash_rehash(), chain heads get re-initialized without
any hlist_del_rcu:

 for (i = hmask; i >= 0; i--)
    INIT_HLIST_HEAD(odst + i);

Then, hlist_del_rcu() gets called on the about to-be-reinserted policy
when iterating the per-net list of policies.

hlist_del_rcu() will then make chain->first be nonzero again:

static inline void __hlist_del(struct hlist_node *n)
{
   struct hlist_node *next = n->next;   // address of next element in list
   struct hlist_node **pprev = n->pprev;// location of previous elem, this
                                        // can point at chain->first
        WRITE_ONCE(*pprev, next);       // chain->first points to next elem
        if (next)
                next->pprev = pprev;

Then, when we walk chainlist to find insertion point, we may find a
non-empty list even though we're supposedly reinserting the first
policy to an empty chain.

To fix this first unlink all exact and inexact policies instead of
zeroing the list heads.

Add the commands equivalent to the syzbot reproducer to xfrm_policy.sh,
without fix KASAN catches the corruption as it happens, SLUB poisoning
detects it a bit later.

Reported-by: syzbot+0165480d4ef07360eeda@syzkaller.appspotmail.com
Fixes: 1548bc4e ("xfrm: policy: delete inexact policies from inexact list on hash rebuild")
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

The psock_tpacket test will need to access /proc/kallsyms, this would
require the kernel config CONFIG_KALLSYMS to be enabled first.

Apart from adding CONFIG_KALLSYMS to the net/config file here, check the
file existence to determine if we can run this test will be helpful to
avoid a false-positive test result when testing it directly with the
following commad against a kernel that have CONFIG_KALLSYMS disabled:
    make -C tools/testing/selftests TARGETS=net run_tests
Signed-off-by: NPo-Hsu Lin <po-hsu.lin@canonical.com>
Acked-by: NShuah Khan <skhan@linuxfoundation.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>