Daniel Borkmann says:

====================
pull-request: bpf 2020-02-07

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

We've added 15 non-merge commits during the last 10 day(s) which contain
a total of 12 files changed, 114 insertions(+), 31 deletions(-).

The main changes are:

1) Various BPF sockmap fixes related to RCU handling in the map's tear-
   down code, from Jakub Sitnicki.

2) Fix macro state explosion in BPF sk_storage map when calculating its
   bucket_log on allocation, from Martin KaFai Lau.

3) Fix potential BPF sockmap update race by rechecking socket's established
   state under lock, from Lorenz Bauer.

4) Fix crash in bpftool on missing xlated instructions when kptr_restrict
   sysctl is set, from Toke Høiland-Jørgensen.

5) Fix i40e's XSK wakeup code to return proper error in busy state and
   various misc fixes in xdpsock BPF sample code, from Maciej Fijalkowski.

6) Fix the way modifiers are skipped in BTF in the verifier while walking
   pointers to avoid program rejection, from Alexei Starovoitov.

7) Fix Makefile for runqslower BPF tool to i) rebuild on libbpf changes and
   ii) to fix undefined reference linker errors for older gcc version due to
   order of passed gcc parameters, from Yulia Kartseva and Song Liu.

8) Fix a trampoline_count BPF kselftest warning about missing braces around
   initializer, from Andrii Nakryiko.

9) Fix up redundant "HAVE" prefix from large INSN limit kernel probe in
   bpftool, from Michal Rostecki.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

It was reported that the max_t, ilog2, and roundup_pow_of_two macros have
exponential effects on the number of states in the sparse checker.

This patch breaks them up by calculating the "nbuckets" first so that the
"bucket_log" only needs to take ilog2().

In addition, Linus mentioned:

  Patch looks good, but I'd like to point out that it's not just sparse.

  You can see it with a simple

    make net/core/bpf_sk_storage.i
    grep 'smap->bucket_log = ' net/core/bpf_sk_storage.i | wc

  and see the end result:

      1  365071 2686974

  That's one line (the assignment line) that is 2,686,974 characters in
  length.

  Now, sparse does happen to react particularly badly to that (I didn't
  look to why, but I suspect it's just that evaluating all the types
  that don't actually ever end up getting used ends up being much more
  expensive than it should be), but I bet it's not good for gcc either.

Fixes: 6ac99e8f ("bpf: Introduce bpf sk local storage")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Link: https://lore.kernel.org/bpf/20200207081810.3918919-1-kafai@fb.com

Commit 7e81a353 ("bpf: Sockmap, ensure sock lock held during tear
down") introduced sleeping issues inside RCU critical sections and while
holding a spinlock on sockmap/sockhash tear-down. There has to be at least
one socket in the map for the problem to surface.

This adds a test that triggers the warnings for broken locking rules. Not a
fix per se, but rather tooling to verify the accompanying fixes. Run on a
VM with 1 vCPU to reproduce the warnings.

Fixes: 7e81a353 ("bpf: Sockmap, ensure sock lock held during tear down")
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200206111652.694507-4-jakub@cloudflare.com

We need to have a synchronize_rcu before free'ing the sockhash because any
outstanding psock references will have a pointer to the map and when they
use it, this could trigger a use after free.

This is a sister fix for sockhash, following commit 2bb90e5c ("bpf:
sockmap, synchronize_rcu before free'ing map") which addressed sockmap,
which comes from a manual audit.

Fixes: 604326b4 ("bpf, sockmap: convert to generic sk_msg interface")
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200206111652.694507-3-jakub@cloudflare.com

rcu_read_lock is needed to protect access to psock inside sock_map_unref
when tearing down the map. However, we can't afford to sleep in lock_sock
while in RCU read-side critical section. Grab the RCU lock only after we
have locked the socket.

This fixes RCU warnings triggerable on a VM with 1 vCPU when free'ing a
sockmap/sockhash that contains at least one socket:

| =============================
| WARNING: suspicious RCU usage
| 5.5.0-04005-g8fc91b97 #450 Not tainted
| -----------------------------
| include/linux/rcupdate.h:272 Illegal context switch in RCU read-side critical section!
|
| other info that might help us debug this:
|
|
| rcu_scheduler_active = 2, debug_locks = 1
| 4 locks held by kworker/0:1/62:
|  #0: ffff88813b019748 ((wq_completion)events){+.+.}, at: process_one_work+0x1d7/0x5e0
|  #1: ffffc900000abe50 ((work_completion)(&map->work)){+.+.}, at: process_one_work+0x1d7/0x5e0
|  #2: ffffffff82065d20 (rcu_read_lock){....}, at: sock_map_free+0x5/0x170
|  #3: ffff8881368c5df8 (&stab->lock){+...}, at: sock_map_free+0x64/0x170
|
| stack backtrace:
| CPU: 0 PID: 62 Comm: kworker/0:1 Not tainted 5.5.0-04005-g8fc91b97 #450
| Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20190727_073836-buildvm-ppc64le-16.ppc.fedoraproject.org-3.fc31 04/01/2014
| Workqueue: events bpf_map_free_deferred
| Call Trace:
|  dump_stack+0x71/0xa0
|  ___might_sleep+0x105/0x190
|  lock_sock_nested+0x28/0x90
|  sock_map_free+0x95/0x170
|  bpf_map_free_deferred+0x58/0x80
|  process_one_work+0x260/0x5e0
|  worker_thread+0x4d/0x3e0
|  kthread+0x108/0x140
|  ? process_one_work+0x5e0/0x5e0
|  ? kthread_park+0x90/0x90
|  ret_from_fork+0x3a/0x50

| =============================
| WARNING: suspicious RCU usage
| 5.5.0-04005-g8fc91b97-dirty #452 Not tainted
| -----------------------------
| include/linux/rcupdate.h:272 Illegal context switch in RCU read-side critical section!
|
| other info that might help us debug this:
|
|
| rcu_scheduler_active = 2, debug_locks = 1
| 4 locks held by kworker/0:1/62:
|  #0: ffff88813b019748 ((wq_completion)events){+.+.}, at: process_one_work+0x1d7/0x5e0
|  #1: ffffc900000abe50 ((work_completion)(&map->work)){+.+.}, at: process_one_work+0x1d7/0x5e0
|  #2: ffffffff82065d20 (rcu_read_lock){....}, at: sock_hash_free+0x5/0x1d0
|  #3: ffff888139966e00 (&htab->buckets[i].lock){+...}, at: sock_hash_free+0x92/0x1d0
|
| stack backtrace:
| CPU: 0 PID: 62 Comm: kworker/0:1 Not tainted 5.5.0-04005-g8fc91b97-dirty #452
| Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20190727_073836-buildvm-ppc64le-16.ppc.fedoraproject.org-3.fc31 04/01/2014
| Workqueue: events bpf_map_free_deferred
| Call Trace:
|  dump_stack+0x71/0xa0
|  ___might_sleep+0x105/0x190
|  lock_sock_nested+0x28/0x90
|  sock_hash_free+0xec/0x1d0
|  bpf_map_free_deferred+0x58/0x80
|  process_one_work+0x260/0x5e0
|  worker_thread+0x4d/0x3e0
|  kthread+0x108/0x140
|  ? process_one_work+0x5e0/0x5e0
|  ? kthread_park+0x90/0x90
|  ret_from_fork+0x3a/0x50

Fixes: 7e81a353 ("bpf: Sockmap, ensure sock lock held during tear down")
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200206111652.694507-2-jakub@cloudflare.com

Turns out the xlated program instructions can also be missing if
kptr_restrict sysctl is set. This means that the previous fix to check the
jited_prog_insns pointer was insufficient; add another check of the
xlated_prog_insns pointer as well.

Fixes: 5b79bcdf ("bpftool: Don't crash on missing jited insns or ksyms")
Fixes: cae73f23 ("bpftool: use bpf_program__get_prog_info_linear() in prog.c:do_dump()")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/bpf/20200206102906.112551-1-toke@redhat.com

It's currently possible to insert sockets in unexpected states into
a sockmap, due to a TOCTTOU when updating the map from a syscall.
sock_map_update_elem checks that sk->sk_state == TCP_ESTABLISHED,
locks the socket and then calls sock_map_update_common. At this
point, the socket may have transitioned into another state, and
the earlier assumptions don't hold anymore. Crucially, it's
conceivable (though very unlikely) that a socket has become unhashed.
This breaks the sockmap's assumption that it will get a callback
via sk->sk_prot->unhash.

Fix this by checking the (fixed) sk_type and sk_protocol without the
lock, followed by a locked check of sk_state.

Unfortunately it's not possible to push the check down into
sock_(map|hash)_update_common, since BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB
run before the socket has transitioned from TCP_SYN_RECV into
TCP_ESTABLISHED.

Fixes: 604326b4 ("bpf, sockmap: convert to generic sk_msg interface")
Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Link: https://lore.kernel.org/bpf/20200207103713.28175-1-lmb@cloudflare.com

Drop monitor uses a work item that takes care of constructing and
sending netlink notifications to user space. In case drop monitor never
started to monitor, then the work item is uninitialized and not
associated with a function.

Therefore, a stop command from user space results in canceling an
uninitialized work item which leads to the following warning [1].

Fix this by not processing a stop command if drop monitor is not
currently monitoring.

[1]
[   31.735402] ------------[ cut here ]------------
[   31.736470] WARNING: CPU: 0 PID: 143 at kernel/workqueue.c:3032 __flush_work+0x89f/0x9f0
...
[   31.738120] CPU: 0 PID: 143 Comm: dwdump Not tainted 5.5.0-custom-09491-g16d4077796b8 #727
[   31.741968] RIP: 0010:__flush_work+0x89f/0x9f0
...
[   31.760526] Call Trace:
[   31.771689]  __cancel_work_timer+0x2a6/0x3b0
[   31.776809]  net_dm_cmd_trace+0x300/0xef0
[   31.777549]  genl_rcv_msg+0x5c6/0xd50
[   31.781005]  netlink_rcv_skb+0x13b/0x3a0
[   31.784114]  genl_rcv+0x29/0x40
[   31.784720]  netlink_unicast+0x49f/0x6a0
[   31.787148]  netlink_sendmsg+0x7cf/0xc80
[   31.790426]  ____sys_sendmsg+0x620/0x770
[   31.793458]  ___sys_sendmsg+0xfd/0x170
[   31.802216]  __sys_sendmsg+0xdf/0x1a0
[   31.806195]  do_syscall_64+0xa0/0x540
[   31.806885]  entry_SYSCALL_64_after_hwframe+0x49/0xbe

Fixes: 8e94c3bc ("drop_monitor: Allow user to start monitoring hardware drops")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Ido Schimmel says:

====================
mlxsw: Various fixes

This patch set contains various fixes for the mlxsw driver.

Patch #1 fixes an issue introduced in 5.6 in which a route in the main
table can replace an identical route in the local table despite the
local table having an higher precedence.

Patch #2 contains a test case for the bug fixed in patch #1.

Patch #3 also fixes an issue introduced in 5.6 in which the driver
failed to clear the offload indication from IPv6 nexthops upon abort.

Patch #4 fixes an issue that prevents the driver from loading on
Spectrum-3 systems. The problem and solution are explained in detail in
the commit message.

Patch #5 adds a missing error path. Discovered using smatch.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

In case devlink_dpipe_entry_ctx_prepare() failed, release RTNL that was
previously taken and free the memory allocated by
mlxsw_sp_erif_entry_prepare().

Fixes: 2ba5999f ("mlxsw: spectrum: Add Support for erif table entries access")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When reading the number of gearboxes from the hardware, the driver does
not validate the returned 'device type' field. The driver can therefore
wrongly assume that the queried devices are gearboxes.

On Spectrum-3 systems that support different types of devices, this can
prevent the driver from loading, as it will try to query the
temperature sensors from devices which it assumes are gearboxes and in
fact are not.

For example:
[  218.129230] mlxsw_minimal 2-0048: Reg cmd access status failed (status=7(bad parameter))
[  218.138282] mlxsw_minimal 2-0048: Reg cmd access failed (reg_id=900a(mtmp),type=write)
[  218.147131] mlxsw_minimal 2-0048: Failed to setup temp sensor number 256
[  218.534480] mlxsw_minimal 2-0048: Fail to register core bus
[  218.540714] mlxsw_minimal: probe of 2-0048 failed with error -5

Fix this by validating the 'device type' field.

Fixes: 2e265a8b ("mlxsw: core: Extend hwmon interface with inter-connect temperature attributes")
Fixes: f14f4e62 ("mlxsw: core: Extend thermal core with per inter-connect device thermal zones")
Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Unlike IPv4, in IPv6 there is no unique structure to represent the
nexthop and both the route and nexthop information are squashed to the
same structure ('struct fib6_info'). In order to improve resource
utilization the driver consolidates identical nexthop groups to the same
internal representation of a nexthop group.

Therefore, when the offload indication of a nexthop changes, the driver
needs to iterate over all the linked fib6_info and toggle their offload
flag accordingly.

During abort, all the routes are removed from the device and unlinked
from their nexthop group. The offload indication is cleared just before
the group is destroyed, but by that time no fib6_info is linked to the
group and the offload indication remains set.

Fix this by clearing the offload indication just before dropping the
reference from the nexthop.

Fixes: ee5a0448 ("mlxsw: spectrum_router: Set hardware flags for routes")
Reported-by: Alex Kushnarov <alexanderk@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Tested-by: Alex Kushnarov <alexanderk@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Test that routes in the main table do not replace identical routes in
the local table and that routes in the local table do replace identical
routes in the main table.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The driver uses the same table to represent both the main and local
routing tables. Prevent routes in the main table from replacing routes
in the local table to reflect the fact that the local table is consulted
first during lookup.

Fixes: b6a1d871 ("mlxsw: spectrum_router: Start using new IPv4 route notifications")
Fixes: dacad7b3 ("mlxsw: spectrum_router: Start using new IPv6 route notifications")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This matches /sys/devices/.../spi1.0/modalias content.

Fixes: 9b2d9f05 ("net: dsa: microchip: add ksz9567 to ksz9477 driver")
Fixes: d9033ae9 ("net: dsa: microchip: add KSZ8563 compatibility string")
Fixes: 8c29bebb ("net: dsa: microchip: add KSZ9893 switch support")
Fixes: 45316818 ("net: dsa: add support for ksz9897 ethernet switch")
Fixes: b987e98e ("dsa: add DSA switch driver for Microchip KSZ9477")
Signed-off-by: Razvan Stefanescu <razvan.stefanescu@microchip.com>
Signed-off-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

__in6_dev_get(dev) called from inet6_set_link_af() can return NULL.

The needed check has been recently removed, let's add it back.

While do_setlink() does call validate_linkmsg() :
...
err = validate_linkmsg(dev, tb); /* OK at this point */
...

It is possible that the following call happening before the
->set_link_af() removes IPv6 if MTU is less than 1280 :

if (tb[IFLA_MTU]) {
    err = dev_set_mtu_ext(dev, nla_get_u32(tb[IFLA_MTU]), extack);
    if (err < 0)
          goto errout;
    status |= DO_SETLINK_MODIFIED;
}
...

if (tb[IFLA_AF_SPEC]) {
   ...
   err = af_ops->set_link_af(dev, af);
      ->inet6_set_link_af() // CRASH because idev is NULL

Please note that IPv4 is immune to the bug since inet_set_link_af() does :

struct in_device *in_dev = __in_dev_get_rcu(dev);
if (!in_dev)
    return -EAFNOSUPPORT;

This problem has been mentioned in commit cf7afbfe ("rtnl: make
link af-specific updates atomic") changelog :

    This method is not fail proof, while it is currently sufficient
    to make set_link_af() inerrable and thus 100% atomic, the
    validation function method will not be able to detect all error
    scenarios in the future, there will likely always be errors
    depending on states which are f.e. not protected by rtnl_mutex
    and thus may change between validation and setting.

IPv6: ADDRCONF(NETDEV_CHANGE): lo: link becomes ready
general protection fault, probably for non-canonical address 0xdffffc0000000056: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x00000000000002b0-0x00000000000002b7]
CPU: 0 PID: 9698 Comm: syz-executor712 Not tainted 5.5.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:inet6_set_link_af+0x66e/0xae0 net/ipv6/addrconf.c:5733
Code: 38 d0 7f 08 84 c0 0f 85 20 03 00 00 48 8d bb b0 02 00 00 45 0f b6 64 24 04 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <0f> b6 04 02 84 c0 74 08 3c 03 0f 8e 1a 03 00 00 44 89 a3 b0 02 00
RSP: 0018:ffffc90005b06d40 EFLAGS: 00010206
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff86df39a6
RDX: 0000000000000056 RSI: ffffffff86df3e74 RDI: 00000000000002b0
RBP: ffffc90005b06e70 R08: ffff8880a2ac0380 R09: ffffc90005b06db0
R10: fffff52000b60dbe R11: ffffc90005b06df7 R12: 0000000000000000
R13: 0000000000000000 R14: ffff8880a1fcc424 R15: dffffc0000000000
FS:  0000000000c46880(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000055f0494ca0d0 CR3: 000000009e4ac000 CR4: 00000000001406f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 do_setlink+0x2a9f/0x3720 net/core/rtnetlink.c:2754
 rtnl_group_changelink net/core/rtnetlink.c:3103 [inline]
 __rtnl_newlink+0xdd1/0x1790 net/core/rtnetlink.c:3257
 rtnl_newlink+0x69/0xa0 net/core/rtnetlink.c:3377
 rtnetlink_rcv_msg+0x45e/0xaf0 net/core/rtnetlink.c:5438
 netlink_rcv_skb+0x177/0x450 net/netlink/af_netlink.c:2477
 rtnetlink_rcv+0x1d/0x30 net/core/rtnetlink.c:5456
 netlink_unicast_kernel net/netlink/af_netlink.c:1302 [inline]
 netlink_unicast+0x59e/0x7e0 net/netlink/af_netlink.c:1328
 netlink_sendmsg+0x91c/0xea0 net/netlink/af_netlink.c:1917
 sock_sendmsg_nosec net/socket.c:652 [inline]
 sock_sendmsg+0xd7/0x130 net/socket.c:672
 ____sys_sendmsg+0x753/0x880 net/socket.c:2343
 ___sys_sendmsg+0x100/0x170 net/socket.c:2397
 __sys_sendmsg+0x105/0x1d0 net/socket.c:2430
 __do_sys_sendmsg net/socket.c:2439 [inline]
 __se_sys_sendmsg net/socket.c:2437 [inline]
 __x64_sys_sendmsg+0x78/0xb0 net/socket.c:2437
 do_syscall_64+0xfa/0x790 arch/x86/entry/common.c:294
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x4402e9
Code: 18 89 d0 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 fb 13 fc ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007fffd62fbcf8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 00000000004002c8 RCX: 00000000004402e9
RDX: 0000000000000000 RSI: 0000000020000080 RDI: 0000000000000003
RBP: 00000000006ca018 R08: 0000000000000008 R09: 00000000004002c8
R10: 0000000000000005 R11: 0000000000000246 R12: 0000000000401b70
R13: 0000000000401c00 R14: 0000000000000000 R15: 0000000000000000
Modules linked in:
---[ end trace cfa7664b8fdcdff3 ]---
RIP: 0010:inet6_set_link_af+0x66e/0xae0 net/ipv6/addrconf.c:5733
Code: 38 d0 7f 08 84 c0 0f 85 20 03 00 00 48 8d bb b0 02 00 00 45 0f b6 64 24 04 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <0f> b6 04 02 84 c0 74 08 3c 03 0f 8e 1a 03 00 00 44 89 a3 b0 02 00
RSP: 0018:ffffc90005b06d40 EFLAGS: 00010206
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff86df39a6
RDX: 0000000000000056 RSI: ffffffff86df3e74 RDI: 00000000000002b0
RBP: ffffc90005b06e70 R08: ffff8880a2ac0380 R09: ffffc90005b06db0
R10: fffff52000b60dbe R11: ffffc90005b06df7 R12: 0000000000000000
R13: 0000000000000000 R14: ffff8880a1fcc424 R15: dffffc0000000000
FS:  0000000000c46880(0000) GS:ffff8880ae900000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000020000004 CR3: 000000009e4ac000 CR4: 00000000001406e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400

Fixes: 7dc2bcca ("Validate required parameters in inet6_validate_link_af")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Bisected-and-reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Stop removing modes that are not supported on the system interface
when the connected PHY is capable of rate adaptation. This addresses
an issue with the LS1046ARDB board 10G interface no longer working
with an 1G link partner after autonegotiation support was added
for the Aquantia PHY on board in

commit 09c4c57f ("net: phy: aquantia: add support for auto-negotiation configuration")

Before this commit the values advertised by the PHY were not
influenced by the dpaa_eth driver removal of system-side unsupported
modes as the aqr_config_aneg() was basically a no-op. After this
commit, the modes removed by the dpaa_eth driver were no longer
advertised thus autonegotiation with 1G link partners failed.
Reported-by: Mian Yousaf Kaukab <ykaukab@suse.de>
Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

Ong Boon Leong says:

====================
net: stmmac: general fixes for Ethernet functionality

1/5: It ensures that the previous value of GMAC_VLAN_TAG register is
     read first before for updating the register.

2/5: Similar to 2/6 patch but it is a fix for XGMAC_VLAN_TAG register
     as requested by Jose Abreu.

3/5: It ensures the GMAC IP v4.xx and above behaves correctly to:-
       ip link set <devname> multicast off|on

4/5: Added similar IFF_MULTICAST flag for xgmac2, similar to 4/6.

5/5: It ensures PCI platform data is using plat->phy_interface.

Changes from v4:-
   patch 1/6 - this patch is dropped now and will take the input on
               handling return value from netif_set_real_num_rx|
               tx_queues() in future patch series.

v3:-
   patch 1/6 - add rtnl_lock() and rtnl_unlock() for stmmac_hw_setup()
               called inside stmmac_resume()
   patch 3/6 - Added new patch to fix XGMAC_VLAN_TAG register writting

v2:-
   patch 1/5 - added control for rtnl_lock() & rtnl_unlock() to ensure
               they are used forstmmac_resume()
   patch 4/5 - added IFF_MULTICAST flag check for xgmac to ensure
               multicast works correctly.

v1:-
 - Drop v1 patches (1/7, 3/7 & 4/7) that are not valid.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

The recent patch to support passive mode converter did not take care the
phy interface configuration in PCI platform data. Hence, converting all
the PCI platform data from plat->interface to plat->phy_interface as the
default mode is meant for PHY.

Fixes: 0060c878 ("net: stmmac: implement support for passive mode converters via dt")
Signed-off-by: Voon Weifeng <weifeng.voon@intel.com>
Tested-by: Tan, Tee Min <tee.min.tan@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Without checking for IFF_MULTICAST flag, it is wrong to assume multicast
filtering is always enabled. By checking against IFF_MULTICAST, now
the driver behaves correctly when the multicast support is toggled by below
command:-
  ip link set <devname> multicast off|on

Fixes: 0efedbf1 ("net: stmmac: xgmac: Fix XGMAC selftests")
Signed-off-by: Tan, Tee Min <tee.min.tan@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Without checking for IFF_MULTICAST flag, it is wrong to assume multicast
filtering is always enabled. By checking against IFF_MULTICAST, now
the driver behaves correctly when the multicast support is toggled by below
command:-
  ip link set <devname> multicast off|on

Fixes: 477286b5 ("stmmac: add GMAC4 core support")
Signed-off-by: Verma, Aashish <aashishx.verma@intel.com>
Tested-by: Tan, Tee Min <tee.min.tan@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We should always do a read of current value of XGMAC_VLAN_TAG instead of
directly overwriting the register value.

Fixes: 3cd1cfcb ("net: stmmac: Implement VLAN Hash Filtering in XGMAC")
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

It should always do a read of current value of GMAC_VLAN_TAG instead of
directly overwriting the register value.

Fixes: c1be0022 ("net: stmmac: Add VLAN HASH filtering support in GMAC4+")
Signed-off-by: Tan, Tee Min <tee.min.tan@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The caller of XDP_SETUP_PROG has already incremented refcnt in
__bpf_prog_get(), so drivers should only increment refcnt by
num_queues - 1.

To fix the issue, update netvsc_xdp_set() to add the correct number
to refcnt.

Hold a refcnt in netvsc_xdp_set()’s other caller, netvsc_attach().

And, do the same in netvsc_vf_setxdp(). Otherwise, every time when VF is
removed and added from the host side, the refcnt will be decreased by one,
which may cause page fault when unloading xdp program.

Fixes: 351e1581 ("hv_netvsc: Add XDP support")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Vinicius Costa Gomes says:

====================
taprio: Some fixes

Changes from v3:
  - Replaced ENOTSUPP error code with EOPNOTSUPP (Jakub Kicinski);
  - Added the missing policy validation for the flags netlink argument
    (Jakub Kicinski);
  - Fixed the destroy() flow to also destroy the priority to traffic
    class mapping (David Miller);
  - Fixed dropping packets when taprio offloading is used together
    with ETF offloading (more on this below);

Changes from v2:
  - Squashed commits 2/3 and 3/3 into a single one (I think a single
    commit is going to be easier to review);
  - Removed an "improvement" that was causing changes in user visible
    behavior;

Changes from v1:
  - Fixed ignoring the 'flags' argument when adding a new
    instance (Vladimir Oltean);
  - Changed the order of commits;

Updated cover letter:

One bit that might need some attention is the fix for not dropping all
packets when taprio and ETF offloading are used, patch 5/5. The
behavior when the fix is applied is that packets that have a 'txtime'
that would fall outside of their transmission window are now dropped
by taprio. The question that might be raised is: should taprio be
responsible for dropping these packets, or should it be handled lower
in the stack?

My opinion is: taprio has all the information, and it's able to give
feeback to the user. Lower in the stack, those packets might go into
the void, and the only feedback could be a hard to find counter
increasing.

Patch 1/5: Reported by Po Liu, is more of a improvement of usability for
drivers implementing offloading features, now they can rely on the
value of dev->num_tc, instead of going through some hops to get this
value.

Patch 2/5: Use 'q->flags' as the source of truth for the offloading
flags. Tries to solidify the current behavior, while avoiding going
into invalid states, one of which was causing a "rcu stall" (more
information in the commit message).

Patch 3/5: Adds the missing netlink attribute validation for
TCA_TAPRIO_ATTR_FLAGS.

Patch 4/5: Replaces the usage of netdev_set_num_tc() with
netdev_reset_tc() in taprio_destroy(), taprio_destroy() is called when
applying a configuration fails, making sure that the device traffic
class configuration goes back to the default state.

@Vladimir: If possible, I would appreciate your Ack on patch 2/5. I
have been looking at this code for so long that I might have missed
something obvious (and my growing dislike for the word 'flags' may be
affecting my judgement :-).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

When using taprio offloading together with ETF offloading, configured
like this, for example:

$ tc qdisc replace dev $IFACE parent root handle 100 taprio \
  	num_tc 4 \
        map 2 2 1 0 3 2 2 2 2 2 2 2 2 2 2 2 \
	queues 1@0 1@1 1@2 1@3 \
	base-time $BASE_TIME \
	sched-entry S 01 1000000 \
	sched-entry S 0e 1000000 \
	flags 0x2

$ tc qdisc replace dev $IFACE parent 100:1 etf \
     	offload delta 300000 clockid CLOCK_TAI

During enqueue, it works out that the verification added for the
"txtime" assisted mode is run when using taprio + ETF offloading, the
only thing missing is initializing the 'next_txtime' of all the cycle
entries. (if we don't set 'next_txtime' all packets from SO_TXTIME
sockets are dropped)

Fixes: 4cfd5779 ("taprio: Add support for txtime-assist mode")
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When destroying the current taprio instance, which can happen when the
creation of one fails, we should reset the traffic class configuration
back to the default state.

netdev_reset_tc() is a better way because in addition to setting the
number of traffic classes to zero, it also resets the priority to
traffic classes mapping to the default value.

Fixes: 5a781ccb ("tc: Add support for configuring the taprio scheduler")
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

netlink policy validation for the 'flags' argument was missing.

Fixes: 4cfd5779 ("taprio: Add support for txtime-assist mode")
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Because 'q->flags' starts as zero, and zero is a valid value, we
aren't able to detect the transition from zero to something else
during "runtime".

The solution is to initialize 'q->flags' with an invalid value, so we
can detect if 'q->flags' was set by the user or not.

To better solidify the behavior, 'flags' handling is moved to a
separate function. The behavior is:
 - 'flags' if unspecified by the user, is assumed to be zero;
 - 'flags' cannot change during "runtime" (i.e. a change() request
 cannot modify it);

With this new function we can remove taprio_flags, which should reduce
the risk of future accidents.

Allowing flags to be changed was causing the following RCU stall:

[ 1730.558249] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
[ 1730.558258] rcu: 	  6-...0: (190 ticks this GP) idle=922/0/0x1 softirq=25580/25582 fqs=16250
[ 1730.558264] 		  (detected by 2, t=65002 jiffies, g=33017, q=81)
[ 1730.558269] Sending NMI from CPU 2 to CPUs 6:
[ 1730.559277] NMI backtrace for cpu 6
[ 1730.559277] CPU: 6 PID: 0 Comm: swapper/6 Tainted: G            E     5.5.0-rc6+ #35
[ 1730.559278] Hardware name: Gigabyte Technology Co., Ltd. Z390 AORUS ULTRA/Z390 AORUS ULTRA-CF, BIOS F7 03/14/2019
[ 1730.559278] RIP: 0010:__hrtimer_run_queues+0xe2/0x440
[ 1730.559278] Code: 48 8b 43 28 4c 89 ff 48 8b 75 c0 48 89 45 c8 e8 f4 bb 7c 00 0f 1f 44 00 00 65 8b 05 40 31 f0 68 89 c0 48 0f a3 05 3e 5c 25 01 <0f> 82 fc 01 00 00 48 8b 45 c8 48 89 df ff d0 89 45 c8 0f 1f 44 00
[ 1730.559279] RSP: 0018:ffff9970802d8f10 EFLAGS: 00000083
[ 1730.559279] RAX: 0000000000000006 RBX: ffff8b31645bff38 RCX: 0000000000000000
[ 1730.559280] RDX: 0000000000000000 RSI: ffffffff9710f2ec RDI: ffffffff978daf0e
[ 1730.559280] RBP: ffff9970802d8f68 R08: 0000000000000000 R09: 0000000000000000
[ 1730.559280] R10: 0000018336d7944e R11: 0000000000000001 R12: ffff8b316e39f9c0
[ 1730.559281] R13: ffff8b316e39f940 R14: ffff8b316e39f998 R15: ffff8b316e39f7c0
[ 1730.559281] FS:  0000000000000000(0000) GS:ffff8b316e380000(0000) knlGS:0000000000000000
[ 1730.559281] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1730.559281] CR2: 00007f1105303760 CR3: 0000000227210005 CR4: 00000000003606e0
[ 1730.559282] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1730.559282] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 1730.559282] Call Trace:
[ 1730.559282]  <IRQ>
[ 1730.559283]  ? taprio_dequeue_soft+0x2d0/0x2d0 [sch_taprio]
[ 1730.559283]  hrtimer_interrupt+0x104/0x220
[ 1730.559283]  ? irqtime_account_irq+0x34/0xa0
[ 1730.559283]  smp_apic_timer_interrupt+0x6d/0x230
[ 1730.559284]  apic_timer_interrupt+0xf/0x20
[ 1730.559284]  </IRQ>
[ 1730.559284] RIP: 0010:cpu_idle_poll+0x35/0x1a0
[ 1730.559285] Code: 88 82 ff 65 44 8b 25 12 7d 73 68 0f 1f 44 00 00 e8 90 c3 89 ff fb 65 48 8b 1c 25 c0 7e 01 00 48 8b 03 a8 08 74 0b eb 1c f3 90 <48> 8b 03 a8 08 75 13 8b 05 be a8 a8 00 85 c0 75 ed e8 75 48 84 ff
[ 1730.559285] RSP: 0018:ffff997080137ea8 EFLAGS: 00000202 ORIG_RAX: ffffffffffffff13
[ 1730.559285] RAX: 0000000000000001 RBX: ffff8b316bc3c580 RCX: 0000000000000000
[ 1730.559286] RDX: 0000000000000001 RSI: 000000002819aad9 RDI: ffffffff978da730
[ 1730.559286] RBP: ffff997080137ec0 R08: 0000018324a6d387 R09: 0000000000000000
[ 1730.559286] R10: 0000000000000400 R11: 0000000000000001 R12: 0000000000000006
[ 1730.559286] R13: ffff8b316bc3c580 R14: 0000000000000000 R15: 0000000000000000
[ 1730.559287]  ? cpu_idle_poll+0x20/0x1a0
[ 1730.559287]  ? cpu_idle_poll+0x20/0x1a0
[ 1730.559287]  do_idle+0x4d/0x1f0
[ 1730.559287]  ? complete+0x44/0x50
[ 1730.559288]  cpu_startup_entry+0x1b/0x20
[ 1730.559288]  start_secondary+0x142/0x180
[ 1730.559288]  secondary_startup_64+0xb6/0xc0
[ 1776.686313] nvme nvme0: I/O 96 QID 1 timeout, completion polled

Fixes: 4cfd5779 ("taprio: Add support for txtime-assist mode")
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

If the driver implementing taprio offloading depends on the value of
the network device number of traffic classes (dev->num_tc) for
whatever reason, it was going to receive the value zero. The value was
only set after the offloading function is called.

So, moving setting the number of traffic classes to before the
offloading function is called fixes this issue. This is safe because
this only happens when taprio is instantiated (we don't allow this
configuration to be changed without first removing taprio).

Fixes: 9c66d156 ("taprio: Add support for hardware offloading")
Reported-by: Po Liu <po.liu@nxp.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Acked-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The 7445 switch clocking profiles do not allow us to run the IMP port at
2Gb/sec in a way that it is reliable and consistent. Make sure that the
setting is only applied to the 7278 family.

Fixes: 8f1880cb ("net: dsa: bcm_sf2: Configure IMP port for 2Gb/sec")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

b53_configure_vlan() is called by the bcm_sf2 driver upon setup and
indirectly through resume as well. During the initial setup, we are
guaranteed that dev->vlan_enabled is false, so there is no change in
behavior, however during suspend, we may have enabled VLANs before, so we
do want to restore that setting.

Fixes: dad8d7c6 ("net: dsa: b53: Properly account for VLAN filtering")
Fixes: 967dd82f ("net: dsa: b53: Add support for Broadcom RoboSwitch")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

It forgot to reduce the value of the variable retry in a while loop
in the ethqos_configure() function. It may cause an endless loop and
without timeout.

Fixes: a7c30e62 ("net: stmmac: Add driver for Qualcomm ethqos")
Signed-off-by: Dejin Zheng <zhengdejin5@gmail.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

rxrpc_rcu_destroy_call(), which is called as an RCU callback to clean up a
put call, calls rxrpc_put_connection() which, deep in its bowels, takes a
number of spinlocks in a non-BH-safe way, including rxrpc_conn_id_lock and
local->client_conns_lock.  RCU callbacks, however, are normally called from
softirq context, which can cause lockdep to notice the locking
inconsistency.

To get lockdep to detect this, it's necessary to have the connection
cleaned up on the put at the end of the last of its calls, though normally
the clean up is deferred.  This can be induced, however, by starting a call
on an AF_RXRPC socket and then closing the socket without reading the
reply.

Fix this by having rxrpc_rcu_destroy_call() punt the destruction to a
workqueue if in softirq-mode and defer the destruction to process context.

Note that another way to fix this could be to add a bunch of bh-disable
annotations to the spinlocks concerned - and there might be more than just
those two - but that means spending more time with BHs disabled.

Note also that some of these places were covered by bh-disable spinlocks
belonging to the rxrpc_transport object, but these got removed without the
_bh annotation being retained on the next lock in.

Fixes: 999b69f8 ("rxrpc: Kill the client connection bundle concept")
Reported-by: syzbot+d82f3ac8d87e7ccbb2c9@syzkaller.appspotmail.com
Reported-by: syzbot+3f1fd6b8cbf8702d134e@syzkaller.appspotmail.com
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Hillf Danton <hdanton@sina.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The recent patch that substituted a flag on an rxrpc_call for the
connection pointer being NULL as an indication that a call was disconnected
puts the set_bit in the wrong place for service calls.  This is only a
problem if a call is implicitly terminated by a new call coming in on the
same connection channel instead of a terminating ACK packet.

In such a case, rxrpc_input_implicit_end_call() calls
__rxrpc_disconnect_call(), which is now (incorrectly) setting the
disconnection bit, meaning that when rxrpc_release_call() is later called,
it doesn't call rxrpc_disconnect_call() and so the call isn't removed from
the peer's error distribution list and the list gets corrupted.

KASAN finds the issue as an access after release on a call, but the
position at which it occurs is confusing as it appears to be related to a
different call (the call site is where the latter call is being removed
from the error distribution list and either the next or pprev pointer
points to a previously released call).

Fix this by moving the setting of the flag from __rxrpc_disconnect_call()
to rxrpc_disconnect_call() in the same place that the connection pointer
was being cleared.

Fixes: 5273a191 ("rxrpc: Fix NULL pointer deref due to call->conn being cleared on disconnect")
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Saeed Mahameed says:

====================
Mellanox, mlx5 fixes 2020-02-06

This series introduces some fixes to mlx5 driver.

Please pull and let me know if there is any problem.

For -stable v4.19:
 ('net/mlx5: IPsec, Fix esp modify function attribute')
 ('net/mlx5: IPsec, fix memory leak at mlx5_fpga_ipsec_delete_sa_ctx')

For -stable v5.4:
   ('net/mlx5: Deprecate usage of generic TLS HW capability bit')
   ('net/mlx5: Fix deadlock in fs_core')

For -stable v5.5:
   ('net/mlx5e: TX, Error completion is for last WQE in batch')
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Deprecate the generic TLS cap bit, use the new TX-specific
TLS cap bit instead.

Fixes: a12ff35e ("net/mlx5: Introduce TLS TX offload hardware bits and structures")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

For a cyclic work queue, when not requesting a completion per WQE,
a single CQE might indicate the completion of several WQEs.
However, in case some WQE in the batch causes an error, then an error
completion is issued, breaking the batch, and pointing to the offending
WQE in the wqe_counter field.

Hence, WQE-specific error CQE handling (like printing, breaking, etc...)
should be performed only for the last WQE in batch.

Fixes: 130c7b46 ("net/mlx5e: TX, Dump WQs wqe descriptors on CQE with error events")
Fixes: fd9b4be8 ("net/mlx5e: RX, Support multiple outstanding UMR posts")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

SA context is allocated at mlx5_fpga_ipsec_create_sa_ctx,
however the counterpart mlx5_fpga_ipsec_delete_sa_ctx function
nullifies sa_ctx pointer without freeing the memory allocated,
hence the memory leak.

Fix by free SA context when the SA is released.

Fixes: d6c4f029 ("net/mlx5: Refactor accel IPSec code")
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

The function mlx5_fpga_esp_validate_xfrm_attrs is wrongly used
with negative negation as zero value indicates success but it
used as failure return value instead.

Fix by remove the unary not negation operator.

Fixes: 05564d0a ("net/mlx5: Add flow-steering commands for FPGA IPSec implementation")
Signed-off-by: Raed Salem <raeds@mellanox.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>