Re: [syzbot] [net?] WARNING: refcount bug in inet_twsk_kill

[Kuniyuki Iwashima <kuniyu@amazon.com>]

From: Florian Westphal <fw@strlen.de>
Date: Sun, 11 Aug 2024 15:32:00 +0200
> > > ------------[ cut here ]------------
> > > refcount_t: decrement hit 0; leaking memory.
> > > WARNING: CPU: 3 PID: 1396 at lib/refcount.c:31 refcount_warn_saturate+0x1ed/0x210 lib/refcount.c:31
> > 
> > Eric, this is the weird report I was talking about at netdevconf :)
> > 
> > It seems refcount_dec(&tw->tw_dr->tw_refcount) is somehow done earlier
> > than refcount_inc().
> > 
> > I started to see the same splat at a very low rate after consuming
> > commit b334b924c9b7 ("net: tcp/dccp: prepare for tw_timer un-pinning").
> 
> Could you share a splat you saw?
> 
> Was it also for the fallback tcp_hashinfo and/or from error
> unwindinding after a ops->init() failure on netns creation?
> 
> I don't think it makes sense following this specific report given
> b334b924c9b7 made changes in this area, but a post-b334b924c9b7
> would be more relevant I think.

Ah, I missed the initial syzbot's report was created before the commit.

Here's splats I saw:

[ Note both splats were mixed in the log as with the syzbot's report,
  so two threads were calling tcp_sk_exit_batch() concurrently. ]

---8<---
refcount_t: decrement hit 0; leaking memory.
WARNING: CPU: 0 PID: 1760 at lib/refcount.c:31 refcount_warn_saturate (lib/refcount.c:31 (discriminator 3))
Modules linked in:
CPU: 0 PID: 1760 Comm: syz-executor.4 Not tainted 6.10.0-rc3-00688-g934c29999b57-dirty #20
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
RIP: 0010:refcount_warn_saturate (lib/refcount.c:31 (discriminator 3))
Code: 05 cd 8d aa 04 01 e8 48 e4 e9 fe 0f 0b e9 d3 fe ff ff e8 1c 36 1c ff 48 c7 c7 a0 4a 80 85 c6 05 aa 8d aa 04 01 e8 29 e4 e9 fe <0f> 0b e9 b4 fe ff ff 48 89 ef e8 fa 56 69 ff e9 5c fe ff ff 0f 1f
RSP: 0018:ffffc9000118fa78 EFLAGS: 00010286
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffc9000b13b000
RDX: 0000000000040000 RSI: ffffffff8118dedf RDI: 0000000000000001
RBP: ffff88800bb59440 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000000 R11: 000000002d2d2d2d R12: ffff88800bb59440
R13: ffff88810130f3d0 R14: ffff8881013ee820 R15: ffff888102001198
FS:  00007f5252dbd640(0000) GS:ffff88811ae00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00000000004a6128 CR3: 00000001038b8003 CR4: 0000000000770ef0
PKRU: 55555554
Call Trace:
 <TASK>
 inet_twsk_kill (./include/linux/refcount.h:336 ./include/linux/refcount.h:351 net/ipv4/inet_timewait_sock.c:70)
 inet_twsk_deschedule_put (net/ipv4/inet_timewait_sock.c:266)
 inet_twsk_purge (net/ipv4/inet_timewait_sock.c:349)
 tcp_twsk_purge (net/ipv4/tcp_minisocks.c:402)
 tcp_sk_exit_batch (net/ipv4/tcp_ipv4.c:3512)
 ops_exit_list (net/core/net_namespace.c:179)
 setup_net (net/core/net_namespace.c:374 (discriminator 3))
 copy_net_ns (net/core/net_namespace.c:510)
 create_new_namespaces (kernel/nsproxy.c:110)
 unshare_nsproxy_namespaces (kernel/nsproxy.c:228 (discriminator 4))
 ksys_unshare (kernel/fork.c:3325)
 __x64_sys_unshare (kernel/fork.c:3392)
 do_syscall_64 (arch/x86/entry/common.c:52 arch/x86/entry/common.c:83)
 entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
RIP: 0033:0x7f5253a29e5d
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 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 8b 0d 73 9f 1b 00 f7 d8 64 89 01 48
RSP: 002b:00007f5252dbcc88 EFLAGS: 00000246 ORIG_RAX: 0000000000000110
RAX: ffffffffffffffda RBX: 00000000004d4070 RCX: 00007f5253a29e5d
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000040000000
RBP: 00000000004d4070 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 000000000000006e R14: 00007f5253a8a530 R15: 0000000000000000
 </TASK>

WARNING: CPU: 1 PID: 671 at net/ipv4/tcp_ipv4.c:3514 tcp_sk_exit_batch (net/ipv4/tcp_ipv4.c:3514)
Modules linked in:
CPU: 1 PID: 671 Comm: kworker/u8:15 Not tainted 6.10.0-rc3-00688-g934c29999b57-dirty #20
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
Workqueue: netns cleanup_net
RIP: 0010:tcp_sk_exit_batch (net/ipv4/tcp_ipv4.c:3514)
Code: 89 ee e8 9f 1b 28 fd 85 ed 0f 88 b3 00 00 00 e8 92 22 28 fd 31 ff 89 ee e8 89 1b 28 fd 85 ed 0f 84 9d 00 00 00 e8 7c 22 28 fd <0f> 0b e8 75 22 28 fd 48 89 df e8 ed 23 02 00 48 8d 7b 28 48 89 f8
RSP: 0018:ffffc90000ebfc60 EFLAGS: 00010293
RAX: 0000000000000000 RBX: ffff88800bb59140 RCX: ffffffff84230cb7
RDX: ffff888103871180 RSI: ffffffff84230cc4 RDI: 0000000000000005
RBP: 0000000000000002 R08: 0000000000000005 R09: 0000000000000000
R10: 0000000000000002 R11: 00000000000002bd R12: ffff88800bb59440
R13: dffffc0000000000 R14: ffffc90000ebfd08 R15: fffffbfff0d7fb5c
FS:  0000000000000000(0000) GS:ffff88811af00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f274114ddf8 CR3: 000000010a11c006 CR4: 0000000000770ef0
PKRU: 55555554
Call Trace:
 <TASK>
 ops_exit_list (net/core/net_namespace.c:179)
 cleanup_net (net/core/net_namespace.c:639 (discriminator 3))
 process_one_work (kernel/workqueue.c:3236)
 worker_thread (kernel/workqueue.c:3306 kernel/workqueue.c:3393)
 kthread (kernel/kthread.c:389)
 ret_from_fork (arch/x86/kernel/process.c:153)
 ret_from_fork_asm (arch/x86/entry/entry_64.S:257)
 </TASK>
---8<---