======================================================
WARNING: possible circular locking dependency detected
4.14.157-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor.0/28252 is trying to acquire lock:
 (pmus_lock){+.+.}, at: [<0000000085b48352>] swevent_hlist_get kernel/events/core.c:7917 [inline]
 (pmus_lock){+.+.}, at: [<0000000085b48352>] perf_swevent_init+0x123/0x4f0 kernel/events/core.c:7977

but task is already holding lock:
 (&cpuctx_mutex/1){+.+.}, at: [<0000000044582a4c>] perf_event_ctx_lock_nested+0x15a/0x2d0 kernel/events/core.c:1241

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #2 (&cpuctx_mutex/1){+.+.}:
       __mutex_lock_common kernel/locking/mutex.c:756 [inline]
       __mutex_lock+0xf7/0x13e0 kernel/locking/mutex.c:893
       mutex_lock_double kernel/events/core.c:9939 [inline]
       __perf_event_ctx_lock_double kernel/events/core.c:9998 [inline]
       SYSC_perf_event_open kernel/events/core.c:10263 [inline]
       SyS_perf_event_open+0x12cf/0x2560 kernel/events/core.c:10018
       do_syscall_64+0x19b/0x520 arch/x86/entry/common.c:292
       entry_SYSCALL_64_after_hwframe+0x42/0xb7

-> #1 (&cpuctx_mutex){+.+.}:
       __mutex_lock_common kernel/locking/mutex.c:756 [inline]
       __mutex_lock+0xf7/0x13e0 kernel/locking/mutex.c:893
       perf_event_init_cpu+0xa8/0x150 kernel/events/core.c:11261
       perf_event_init+0x289/0x2c5 kernel/events/core.c:11308
       start_kernel+0x583/0x890 init/main.c:645
       secondary_startup_64+0xa5/0xb0 arch/x86/kernel/head_64.S:240

-> #0 (pmus_lock){+.+.}:
       lock_acquire+0x12b/0x360 kernel/locking/lockdep.c:3994
       __mutex_lock_common kernel/locking/mutex.c:756 [inline]
       __mutex_lock+0xf7/0x13e0 kernel/locking/mutex.c:893
       swevent_hlist_get kernel/events/core.c:7917 [inline]
       perf_swevent_init+0x123/0x4f0 kernel/events/core.c:7977
       perf_try_init_event+0xe5/0x200 kernel/events/core.c:9363
       perf_init_event kernel/events/core.c:9401 [inline]
       perf_event_alloc.part.0+0xd1e/0x21d0 kernel/events/core.c:9661
       perf_event_alloc kernel/events/core.c:10028 [inline]
       SYSC_perf_event_open kernel/events/core.c:10135 [inline]
       SyS_perf_event_open+0x6eb/0x2560 kernel/events/core.c:10018
       do_syscall_64+0x19b/0x520 arch/x86/entry/common.c:292
       entry_SYSCALL_64_after_hwframe+0x42/0xb7

other info that might help us debug this:

Chain exists of:
  pmus_lock --> &cpuctx_mutex --> &cpuctx_mutex/1

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&cpuctx_mutex/1);
                               lock(&cpuctx_mutex);
                               lock(&cpuctx_mutex/1);
  lock(pmus_lock);

 *** DEADLOCK ***

2 locks held by syz-executor.0/28252:
 #0:  (&pmus_srcu){....}, at: [<000000001ef1181a>] perf_event_alloc.part.0+0xb39/0x21d0 kernel/events/core.c:9655
 #1:  (&cpuctx_mutex/1){+.+.}, at: [<0000000044582a4c>] perf_event_ctx_lock_nested+0x15a/0x2d0 kernel/events/core.c:1241

stack backtrace:
CPU: 0 PID: 28252 Comm: syz-executor.0 Not tainted 4.14.157-syzkaller #0
Call Trace:
 __dump_stack lib/dump_stack.c:17 [inline]
 dump_stack+0xe5/0x154 lib/dump_stack.c:58
 print_circular_bug.isra.0.cold+0x2dc/0x425 kernel/locking/lockdep.c:1258
 check_prev_add kernel/locking/lockdep.c:1901 [inline]
 check_prevs_add kernel/locking/lockdep.c:2018 [inline]
 validate_chain kernel/locking/lockdep.c:2460 [inline]
 __lock_acquire+0x2f5f/0x4320 kernel/locking/lockdep.c:3487
 lock_acquire+0x12b/0x360 kernel/locking/lockdep.c:3994
 __mutex_lock_common kernel/locking/mutex.c:756 [inline]
 __mutex_lock+0xf7/0x13e0 kernel/locking/mutex.c:893
 swevent_hlist_get kernel/events/core.c:7917 [inline]
 perf_swevent_init+0x123/0x4f0 kernel/events/core.c:7977
 perf_try_init_event+0xe5/0x200 kernel/events/core.c:9363
 perf_init_event kernel/events/core.c:9401 [inline]
 perf_event_alloc.part.0+0xd1e/0x21d0 kernel/events/core.c:9661
 perf_event_alloc kernel/events/core.c:10028 [inline]
 SYSC_perf_event_open kernel/events/core.c:10135 [inline]
 SyS_perf_event_open+0x6eb/0x2560 kernel/events/core.c:10018
 do_syscall_64+0x19b/0x520 arch/x86/entry/common.c:292
 entry_SYSCALL_64_after_hwframe+0x42/0xb7
RIP: 0033:0x45a679
RSP: 002b:00007f2fa3cf6c78 EFLAGS: 00000246 ORIG_RAX: 000000000000012a
RAX: ffffffffffffffda RBX: 0000000000000005 RCX: 000000000045a679
RDX: 0000000000000000 RSI: ffffffffffffffff RDI: 0000000020000200
RBP: 000000000075bf20 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000003 R11: 0000000000000246 R12: 00007f2fa3cf76d4
R13: 00000000004c834f R14: 00000000004df530 R15: 00000000ffffffff
ip6_tunnel:  xmit: Local address not yet configured!
ip6_tunnel: ³ xmit: Local address not yet configured!
netlink: 20 bytes leftover after parsing attributes in process `syz-executor.3'.
netlink: 20 bytes leftover after parsing attributes in process `syz-executor.3'.
netlink: 228 bytes leftover after parsing attributes in process `syz-executor.5'.
selinux_nlmsg_perm: 7 callbacks suppressed
SELinux: unrecognized netlink message: protocol=0 nlmsg_type=65535 sclass=netlink_route_socket pig=28699 comm=syz-executor.2
ip6_tunnel: f xmit: Local address not yet configured!
ip6_tunnel: ë xmit: Local address not yet configured!
ip6_tunnel:  xmit: Local address not yet configured!
ip6_tunnel: a xmit: Local address not yet configured!
ip6_tunnel:  xmit: Local address not yet configured!
ip6_tunnel:  xmit: Local address not yet configured!
ip6_tunnel: þ xmit: Local address not yet configured!
SELinux: unrecognized netlink message: protocol=0 nlmsg_type=65535 sclass=netlink_route_socket pig=28734 comm=syz-executor.2
SELinux: unrecognized netlink message: protocol=0 nlmsg_type=65535 sclass=netlink_route_socket pig=28747 comm=syz-executor.2
netlink: 16 bytes leftover after parsing attributes in process `syz-executor.0'.