Skip to content

L
linux
Switch branch/tag
  1. 11 Apr, 2024 18 commits
  3. 10 Apr, 2024 7 commits
    • Heiner Kallweit's avatar
      r8169: fix LED-related deadlock on module removal · 19fa4f2a
      Heiner Kallweit authored 
      Binding devm_led_classdev_register() to the netdev is problematic
because on module removal we get a RTNL-related deadlock. Fix this
by avoiding the device-managed LED functions.

Note: We can safely call led_classdev_unregister() for a LED even
if registering it failed, because led_classdev_unregister() detects
this and is a no-op in this case.

Fixes: 18764b88 ("r8169: add support for LED's on RTL8168/RTL8101")
Cc: stable@vger.kernel.org
Reported-by: default avatarLukas Wunner <lukas@wunner.de>
Signed-off-by: default avatarHeiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: default avatarAndrew Lunn <andrew@lunn.ch>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      19fa4f2a
    • Brett Creeley's avatar
      pds_core: Fix pdsc_check_pci_health function to use work thread · 81665adf
      Brett Creeley authored 
      When the driver notices fw_status == 0xff it tries to perform a PCI
reset on itself via pci_reset_function() in the context of the driver's
health thread. However, pdsc_reset_prepare calls
pdsc_stop_health_thread(), which attempts to stop/flush the health
thread. This results in a deadlock because the stop/flush will never
complete since the driver called pci_reset_function() from the health
thread context. Fix by changing the pdsc_check_pci_health_function()
to queue a newly introduced pdsc_pci_reset_thread() on the pdsc's
work queue.

Unloading the driver in the fw_down/dead state uncovered another issue,
which can be seen in the following trace:

WARNING: CPU: 51 PID: 6914 at kernel/workqueue.c:1450 __queue_work+0x358/0x440
[...]
RIP: 0010:__queue_work+0x358/0x440
[...]
Call Trace:
 <TASK>
 ? __warn+0x85/0x140
 ? __queue_work+0x358/0x440
 ? report_bug+0xfc/0x1e0
 ? handle_bug+0x3f/0x70
 ? exc_invalid_op+0x17/0x70
 ? asm_exc_invalid_op+0x1a/0x20
 ? __queue_work+0x358/0x440
 queue_work_on+0x28/0x30
 pdsc_devcmd_locked+0x96/0xe0 [pds_core]
 pdsc_devcmd_reset+0x71/0xb0 [pds_core]
 pdsc_teardown+0x51/0xe0 [pds_core]
 pdsc_remove+0x106/0x200 [pds_core]
 pci_device_remove+0x37/0xc0
 device_release_driver_internal+0xae/0x140
 driver_detach+0x48/0x90
 bus_remove_driver+0x6d/0xf0
 pci_unregister_driver+0x2e/0xa0
 pdsc_cleanup_module+0x10/0x780 [pds_core]
 __x64_sys_delete_module+0x142/0x2b0
 ? syscall_trace_enter.isra.18+0x126/0x1a0
 do_syscall_64+0x3b/0x90
 entry_SYSCALL_64_after_hwframe+0x72/0xdc
RIP: 0033:0x7fbd9d03a14b
[...]

Fix this by preventing the devcmd reset if the FW is not running.

Fixes: d9407ff1 ("pds_core: Prevent health thread from running during reset/remove")
Reviewed-by: default avatarShannon Nelson <shannon.nelson@amd.com>
Signed-off-by: default avatarBrett Creeley <brett.creeley@amd.com>
Reviewed-by: default avatarJacob Keller <jacob.e.keller@intel.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      81665adf
    • Jiri Benc's avatar
      ipv6: fix race condition between ipv6_get_ifaddr and ipv6_del_addr · 7633c4da
      Jiri Benc authored 
      Although ipv6_get_ifaddr walks inet6_addr_lst under the RCU lock, it
still means hlist_for_each_entry_rcu can return an item that got removed
from the list. The memory itself of such item is not freed thanks to RCU
but nothing guarantees the actual content of the memory is sane.

In particular, the reference count can be zero. This can happen if
ipv6_del_addr is called in parallel. ipv6_del_addr removes the entry
from inet6_addr_lst (hlist_del_init_rcu(&ifp->addr_lst)) and drops all
references (__in6_ifa_put(ifp) + in6_ifa_put(ifp)). With bad enough
timing, this can happen:

1. In ipv6_get_ifaddr, hlist_for_each_entry_rcu returns an entry.

2. Then, the whole ipv6_del_addr is executed for the given entry. The
   reference count drops to zero and kfree_rcu is scheduled.

3. ipv6_get_ifaddr continues and tries to increments the reference count
   (in6_ifa_hold).

4. The rcu is unlocked and the entry is freed.

5. The freed entry is returned.

Prevent increasing of the reference count in such case. The name
in6_ifa_hold_safe is chosen to mimic the existing fib6_info_hold_safe.

[   41.506330] refcount_t: addition on 0; use-after-free.
[   41.506760] WARNING: CPU: 0 PID: 595 at lib/refcount.c:25 refcount_warn_saturate+0xa5/0x130
[   41.507413] Modules linked in: veth bridge stp llc
[   41.507821] CPU: 0 PID: 595 Comm: python3 Not tainted 6.9.0-rc2.main-00208-g49563be8 #14
[   41.508479] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
[   41.509163] RIP: 0010:refcount_warn_saturate+0xa5/0x130
[   41.509586] Code: ad ff 90 0f 0b 90 90 c3 cc cc cc cc 80 3d c0 30 ad 01 00 75 a0 c6 05 b7 30 ad 01 01 90 48 c7 c7 38 cc 7a 8c e8 cc 18 ad ff 90 <0f> 0b 90 90 c3 cc cc cc cc 80 3d 98 30 ad 01 00 0f 85 75 ff ff ff
[   41.510956] RSP: 0018:ffffbda3c026baf0 EFLAGS: 00010282
[   41.511368] RAX: 0000000000000000 RBX: ffff9e9c46914800 RCX: 0000000000000000
[   41.511910] RDX: ffff9e9c7ec29c00 RSI: ffff9e9c7ec1c900 RDI: ffff9e9c7ec1c900
[   41.512445] RBP: ffff9e9c43660c9c R08: 0000000000009ffb R09: 00000000ffffdfff
[   41.512998] R10: 00000000ffffdfff R11: ffffffff8ca58a40 R12: ffff9e9c4339a000
[   41.513534] R13: 0000000000000001 R14: ffff9e9c438a0000 R15: ffffbda3c026bb48
[   41.514086] FS:  00007fbc4cda1740(0000) GS:ffff9e9c7ec00000(0000) knlGS:0000000000000000
[   41.514726] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   41.515176] CR2: 000056233b337d88 CR3: 000000000376e006 CR4: 0000000000370ef0
[   41.515713] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   41.516252] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[   41.516799] Call Trace:
[   41.517037]  <TASK>
[   41.517249]  ? __warn+0x7b/0x120
[   41.517535]  ? refcount_warn_saturate+0xa5/0x130
[   41.517923]  ? report_bug+0x164/0x190
[   41.518240]  ? handle_bug+0x3d/0x70
[   41.518541]  ? exc_invalid_op+0x17/0x70
[   41.520972]  ? asm_exc_invalid_op+0x1a/0x20
[   41.521325]  ? refcount_warn_saturate+0xa5/0x130
[   41.521708]  ipv6_get_ifaddr+0xda/0xe0
[   41.522035]  inet6_rtm_getaddr+0x342/0x3f0
[   41.522376]  ? __pfx_inet6_rtm_getaddr+0x10/0x10
[   41.522758]  rtnetlink_rcv_msg+0x334/0x3d0
[   41.523102]  ? netlink_unicast+0x30f/0x390
[   41.523445]  ? __pfx_rtnetlink_rcv_msg+0x10/0x10
[   41.523832]  netlink_rcv_skb+0x53/0x100
[   41.524157]  netlink_unicast+0x23b/0x390
[   41.524484]  netlink_sendmsg+0x1f2/0x440
[   41.524826]  __sys_sendto+0x1d8/0x1f0
[   41.525145]  __x64_sys_sendto+0x1f/0x30
[   41.525467]  do_syscall_64+0xa5/0x1b0
[   41.525794]  entry_SYSCALL_64_after_hwframe+0x72/0x7a
[   41.526213] RIP: 0033:0x7fbc4cfcea9a
[   41.526528] Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f 1e fa 41 89 ca 64 8b 04 25 18 00 00 00 85 c0 75 15 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 7e c3 0f 1f 44 00 00 41 54 48 83 ec 30 44 89
[   41.527942] RSP: 002b:00007ffcf54012a8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
[   41.528593] RAX: ffffffffffffffda RBX: 00007ffcf5401368 RCX: 00007fbc4cfcea9a
[   41.529173] RDX: 000000000000002c RSI: 00007fbc4b9d9bd0 RDI: 0000000000000005
[   41.529786] RBP: 00007fbc4bafb040 R08: 00007ffcf54013e0 R09: 000000000000000c
[   41.530375] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
[   41.530977] R13: ffffffffc4653600 R14: 0000000000000001 R15: 00007fbc4ca85d1b
[   41.531573]  </TASK>

Fixes: 5c578aed ("IPv6: convert addrconf hash list to RCU")
Reviewed-by: default avatarEric Dumazet <edumazet@google.com>
Reviewed-by: default avatarDavid Ahern <dsahern@kernel.org>
Signed-off-by: default avatarJiri Benc <jbenc@redhat.com>
Link: https://lore.kernel.org/r/8ab821e36073a4a406c50ec83c9e8dc586c539e4.1712585809.git.jbenc@redhat.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      7633c4da
    • Jakub Kicinski's avatar
      Merge branch 'net-start-to-replace-copy_from_sockptr' · 7b6575c6
      Jakub Kicinski authored 
      Eric Dumazet says:

====================
net: start to replace copy_from_sockptr()

We got several syzbot reports about unsafe copy_from_sockptr()
calls. After fixing some of them, it appears that we could
use a new helper to factorize all the checks in one place.

This series targets net tree, we can later start converting
many call sites in net-next.
====================

Link: https://lore.kernel.org/r/20240408082845.3957374-1-edumazet@google.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      7b6575c6
    • Eric Dumazet's avatar
      nfc: llcp: fix nfc_llcp_setsockopt() unsafe copies · 7a87441c
      Eric Dumazet authored 
      syzbot reported unsafe calls to copy_from_sockptr() [1]

Use copy_safe_from_sockptr() instead.

[1]

BUG: KASAN: slab-out-of-bounds in copy_from_sockptr_offset include/linux/sockptr.h:49 [inline]
 BUG: KASAN: slab-out-of-bounds in copy_from_sockptr include/linux/sockptr.h:55 [inline]
 BUG: KASAN: slab-out-of-bounds in nfc_llcp_setsockopt+0x6c2/0x850 net/nfc/llcp_sock.c:255
Read of size 4 at addr ffff88801caa1ec3 by task syz-executor459/5078

CPU: 0 PID: 5078 Comm: syz-executor459 Not tainted 6.8.0-syzkaller-08951-gfe46a7dd #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024
Call Trace:
 <TASK>
  __dump_stack lib/dump_stack.c:88 [inline]
  dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114
  print_address_description mm/kasan/report.c:377 [inline]
  print_report+0x169/0x550 mm/kasan/report.c:488
  kasan_report+0x143/0x180 mm/kasan/report.c:601
  copy_from_sockptr_offset include/linux/sockptr.h:49 [inline]
  copy_from_sockptr include/linux/sockptr.h:55 [inline]
  nfc_llcp_setsockopt+0x6c2/0x850 net/nfc/llcp_sock.c:255
  do_sock_setsockopt+0x3b1/0x720 net/socket.c:2311
  __sys_setsockopt+0x1ae/0x250 net/socket.c:2334
  __do_sys_setsockopt net/socket.c:2343 [inline]
  __se_sys_setsockopt net/socket.c:2340 [inline]
  __x64_sys_setsockopt+0xb5/0xd0 net/socket.c:2340
 do_syscall_64+0xfd/0x240
 entry_SYSCALL_64_after_hwframe+0x6d/0x75
RIP: 0033:0x7f7fac07fd89
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 91 18 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fff660eb788 EFLAGS: 00000246 ORIG_RAX: 0000000000000036
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f7fac07fd89
RDX: 0000000000000000 RSI: 0000000000000118 RDI: 0000000000000004
RBP: 0000000000000000 R08: 0000000000000002 R09: 0000000000000000
R10: 0000000020000a80 R11: 0000000000000246 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Reported-by: default avatarsyzbot <syzkaller@googlegroups.com>
Reviewed-by: default avatarKrzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240408082845.3957374-4-edumazet@google.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      7a87441c
    • Eric Dumazet's avatar
      mISDN: fix MISDN_TIME_STAMP handling · 138b7878
      Eric Dumazet authored 
      syzbot reports one unsafe call to copy_from_sockptr() [1]

Use copy_safe_from_sockptr() instead.

[1]

 BUG: KASAN: slab-out-of-bounds in copy_from_sockptr_offset include/linux/sockptr.h:49 [inline]
 BUG: KASAN: slab-out-of-bounds in copy_from_sockptr include/linux/sockptr.h:55 [inline]
 BUG: KASAN: slab-out-of-bounds in data_sock_setsockopt+0x46c/0x4cc drivers/isdn/mISDN/socket.c:417
Read of size 4 at addr ffff0000c6d54083 by task syz-executor406/6167

CPU: 1 PID: 6167 Comm: syz-executor406 Not tainted 6.8.0-rc7-syzkaller-g707081b61156 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024
Call trace:
  dump_backtrace+0x1b8/0x1e4 arch/arm64/kernel/stacktrace.c:291
  show_stack+0x2c/0x3c arch/arm64/kernel/stacktrace.c:298
  __dump_stack lib/dump_stack.c:88 [inline]
  dump_stack_lvl+0xd0/0x124 lib/dump_stack.c:106
  print_address_description mm/kasan/report.c:377 [inline]
  print_report+0x178/0x518 mm/kasan/report.c:488
  kasan_report+0xd8/0x138 mm/kasan/report.c:601
  __asan_report_load_n_noabort+0x1c/0x28 mm/kasan/report_generic.c:391
  copy_from_sockptr_offset include/linux/sockptr.h:49 [inline]
  copy_from_sockptr include/linux/sockptr.h:55 [inline]
  data_sock_setsockopt+0x46c/0x4cc drivers/isdn/mISDN/socket.c:417
  do_sock_setsockopt+0x2a0/0x4e0 net/socket.c:2311
  __sys_setsockopt+0x128/0x1a8 net/socket.c:2334
  __do_sys_setsockopt net/socket.c:2343 [inline]
  __se_sys_setsockopt net/socket.c:2340 [inline]
  __arm64_sys_setsockopt+0xb8/0xd4 net/socket.c:2340
  __invoke_syscall arch/arm64/kernel/syscall.c:34 [inline]
  invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:48
  el0_svc_common+0x130/0x23c arch/arm64/kernel/syscall.c:133
  do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:152
  el0_svc+0x54/0x168 arch/arm64/kernel/entry-common.c:712
  el0t_64_sync_handler+0x84/0xfc arch/arm64/kernel/entry-common.c:730
  el0t_64_sync+0x190/0x194 arch/arm64/kernel/entry.S:598

Fixes: 1b2b03f8 ("Add mISDN core files")
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Reported-by: default avatarsyzbot <syzkaller@googlegroups.com>
Cc: Karsten Keil <isdn@linux-pingi.de>
Link: https://lore.kernel.org/r/20240408082845.3957374-3-edumazet@google.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      138b7878
    • Eric Dumazet's avatar
      net: add copy_safe_from_sockptr() helper · 6309863b
      Eric Dumazet authored 
      copy_from_sockptr() helper is unsafe, unless callers
did the prior check against user provided optlen.

Too many callers get this wrong, lets add a helper to
fix them and avoid future copy/paste bugs.

Instead of :

   if (optlen < sizeof(opt)) {
       err = -EINVAL;
       break;
   }
   if (copy_from_sockptr(&opt, optval, sizeof(opt)) {
       err = -EFAULT;
       break;
   }

Use :

   err = copy_safe_from_sockptr(&opt, sizeof(opt),
                                optval, optlen);
   if (err)
       break;
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240408082845.3957374-2-edumazet@google.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      6309863b
  5. 09 Apr, 2024 6 commits
    • Arnd Bergmann's avatar
      ipv4/route: avoid unused-but-set-variable warning · cf1b7201
      Arnd Bergmann authored 
      The log_martians variable is only used in an #ifdef, causing a 'make W=1'
warning with gcc:

net/ipv4/route.c: In function 'ip_rt_send_redirect':
net/ipv4/route.c:880:13: error: variable 'log_martians' set but not used [-Werror=unused-but-set-variable]

Change the #ifdef to an equivalent IS_ENABLED() to let the compiler
see where the variable is used.

Fixes: 30038fc6 ("net: ip_rt_send_redirect() optimization")
Reviewed-by: default avatarDavid Ahern <dsahern@kernel.org>
Signed-off-by: default avatarArnd Bergmann <arnd@arndb.de>
Reviewed-by: default avatarEric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240408074219.3030256-2-arnd@kernel.orgSigned-off-by: default avatarPaolo Abeni <pabeni@redhat.com>
      cf1b7201
    • Arnd Bergmann's avatar
      ipv6: fib: hide unused 'pn' variable · 74043489
      Arnd Bergmann authored 
      When CONFIG_IPV6_SUBTREES is disabled, the only user is hidden, causing
a 'make W=1' warning:

net/ipv6/ip6_fib.c: In function 'fib6_add':
net/ipv6/ip6_fib.c:1388:32: error: variable 'pn' set but not used [-Werror=unused-but-set-variable]

Add another #ifdef around the variable declaration, matching the other
uses in this file.

Fixes: 66729e18 ("[IPV6] ROUTE: Make sure we have fn->leaf when adding a node on subtree.")
Link: https://lore.kernel.org/netdev/20240322131746.904943-1-arnd@kernel.org/Reviewed-by: default avatarDavid Ahern <dsahern@kernel.org>
Signed-off-by: default avatarArnd Bergmann <arnd@arndb.de>
Reviewed-by: default avatarEric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240408074219.3030256-1-arnd@kernel.orgSigned-off-by: default avatarPaolo Abeni <pabeni@redhat.com>
      74043489
    • Geetha sowjanya's avatar
      octeontx2-af: Fix NIX SQ mode and BP config · faf23006
      Geetha sowjanya authored 
      NIX SQ mode and link backpressure configuration is required for
all platforms. But in current driver this code is wrongly placed
under specific platform check. This patch fixes the issue by
moving the code out of platform check.

Fixes: 5d9b976d ("octeontx2-af: Support fixed transmit scheduler topology")
Signed-off-by: default avatarGeetha sowjanya <gakula@marvell.com>
Link: https://lore.kernel.org/r/20240408063643.26288-1-gakula@marvell.comSigned-off-by: default avatarPaolo Abeni <pabeni@redhat.com>
      faf23006
    • Kuniyuki Iwashima's avatar
      af_unix: Clear stale u->oob_skb. · b46f4eaa
      Kuniyuki Iwashima authored 
      syzkaller started to report deadlock of unix_gc_lock after commit
4090fa37 ("af_unix: Replace garbage collection algorithm."), but
it just uncovers the bug that has been there since commit 314001f0
("af_unix: Add OOB support").

The repro basically does the following.

  from socket import *
  from array import array

  c1, c2 = socketpair(AF_UNIX, SOCK_STREAM)
  c1.sendmsg([b'a'], [(SOL_SOCKET, SCM_RIGHTS, array("i", [c2.fileno()]))], MSG_OOB)
  c2.recv(1)  # blocked as no normal data in recv queue

  c2.close()  # done async and unblock recv()
  c1.close()  # done async and trigger GC

A socket sends its file descriptor to itself as OOB data and tries to
receive normal data, but finally recv() fails due to async close().

The problem here is wrong handling of OOB skb in manage_oob().  When
recvmsg() is called without MSG_OOB, manage_oob() is called to check
if the peeked skb is OOB skb.  In such a case, manage_oob() pops it
out of the receive queue but does not clear unix_sock(sk)->oob_skb.
This is wrong in terms of uAPI.

Let's say we send "hello" with MSG_OOB, and "world" without MSG_OOB.
The 'o' is handled as OOB data.  When recv() is called twice without
MSG_OOB, the OOB data should be lost.

  >>> from socket import *
  >>> c1, c2 = socketpair(AF_UNIX, SOCK_STREAM, 0)
  >>> c1.send(b'hello', MSG_OOB)  # 'o' is OOB data
  5
  >>> c1.send(b'world')
  5
  >>> c2.recv(5)  # OOB data is not received
  b'hell'
  >>> c2.recv(5)  # OOB date is skipped
  b'world'
  >>> c2.recv(5, MSG_OOB)  # This should return an error
  b'o'

In the same situation, TCP actually returns -EINVAL for the last
recv().

Also, if we do not clear unix_sk(sk)->oob_skb, unix_poll() always set
EPOLLPRI even though the data has passed through by previous recv().

To avoid these issues, we must clear unix_sk(sk)->oob_skb when dequeuing
it from recv queue.

The reason why the old GC did not trigger the deadlock is because the
old GC relied on the receive queue to detect the loop.

When it is triggered, the socket with OOB data is marked as GC candidate
because file refcount == inflight count (1).  However, after traversing
all inflight sockets, the socket still has a positive inflight count (1),
thus the socket is excluded from candidates.  Then, the old GC lose the
chance to garbage-collect the socket.

With the old GC, the repro continues to create true garbage that will
never be freed nor detected by kmemleak as it's linked to the global
inflight list.  That's why we couldn't even notice the issue.

Fixes: 314001f0 ("af_unix: Add OOB support")
Reported-by: syzbot+7f7f201cc2668a8fd169@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=7f7f201cc2668a8fd169Signed-off-by: default avatarKuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: default avatarEric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240405221057.2406-1-kuniyu@amazon.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      b46f4eaa
    • Marek Vasut's avatar
      net: ks8851: Handle softirqs at the end of IRQ thread to fix hang · be0384bf
      Marek Vasut authored 
      The ks8851_irq() thread may call ks8851_rx_pkts() in case there are
any packets in the MAC FIFO, which calls netif_rx(). This netif_rx()
implementation is guarded by local_bh_disable() and local_bh_enable().
The local_bh_enable() may call do_softirq() to run softirqs in case
any are pending. One of the softirqs is net_rx_action, which ultimately
reaches the driver .start_xmit callback. If that happens, the system
hangs. The entire call chain is below:

ks8851_start_xmit_par from netdev_start_xmit
netdev_start_xmit from dev_hard_start_xmit
dev_hard_start_xmit from sch_direct_xmit
sch_direct_xmit from __dev_queue_xmit
__dev_queue_xmit from __neigh_update
__neigh_update from neigh_update
neigh_update from arp_process.constprop.0
arp_process.constprop.0 from __netif_receive_skb_one_core
__netif_receive_skb_one_core from process_backlog
process_backlog from __napi_poll.constprop.0
__napi_poll.constprop.0 from net_rx_action
net_rx_action from __do_softirq
__do_softirq from call_with_stack
call_with_stack from do_softirq
do_softirq from __local_bh_enable_ip
__local_bh_enable_ip from netif_rx
netif_rx from ks8851_irq
ks8851_irq from irq_thread_fn
irq_thread_fn from irq_thread
irq_thread from kthread
kthread from ret_from_fork

The hang happens because ks8851_irq() first locks a spinlock in
ks8851_par.c ks8851_lock_par() spin_lock_irqsave(&ksp->lock, ...)
and with that spinlock locked, calls netif_rx(). Once the execution
reaches ks8851_start_xmit_par(), it calls ks8851_lock_par() again
which attempts to claim the already locked spinlock again, and the
hang happens.

Move the do_softirq() call outside of the spinlock protected section
of ks8851_irq() by disabling BHs around the entire spinlock protected
section of ks8851_irq() handler. Place local_bh_enable() outside of
the spinlock protected section, so that it can trigger do_softirq()
without the ks8851_par.c ks8851_lock_par() spinlock being held, and
safely call ks8851_start_xmit_par() without attempting to lock the
already locked spinlock.

Since ks8851_irq() is protected by local_bh_disable()/local_bh_enable()
now, replace netif_rx() with __netif_rx() which is not duplicating the
local_bh_disable()/local_bh_enable() calls.

Fixes: 797047f8 ("net: ks8851: Implement Parallel bus operations")
Signed-off-by: default avatarMarek Vasut <marex@denx.de>
Link: https://lore.kernel.org/r/20240405203204.82062-2-marex@denx.deSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      be0384bf
    • Marek Vasut's avatar
      net: ks8851: Inline ks8851_rx_skb() · f96f7004
      Marek Vasut authored 
      Both ks8851_rx_skb_par() and ks8851_rx_skb_spi() call netif_rx(skb),
inline the netif_rx(skb) call directly into ks8851_common.c and drop
the .rx_skb callback and ks8851_rx_skb() wrapper. This removes one
indirect call from the driver, no functional change otherwise.
Signed-off-by: default avatarMarek Vasut <marex@denx.de>
Link: https://lore.kernel.org/r/20240405203204.82062-1-marex@denx.deSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      f96f7004
  7. 08 Apr, 2024 9 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7