On Wed, Jul 03, 2024 at 11:47:46AM +0800, Wen Gu wrote:
From: Wang Yufen wangyufen@huawei.com
[ Upstream commit d8616ee2affcff37c5d315310da557a694a3303d ]
During TCP sockmap redirect pressure test, the following warning is triggered:
WARNING: CPU: 3 PID: 2145 at net/core/stream.c:205 sk_stream_kill_queues+0xbc/0xd0 CPU: 3 PID: 2145 Comm: iperf Kdump: loaded Tainted: G W 5.10.0+ #9 Call Trace: inet_csk_destroy_sock+0x55/0x110 inet_csk_listen_stop+0xbb/0x380 tcp_close+0x41b/0x480 inet_release+0x42/0x80 __sock_release+0x3d/0xa0 sock_close+0x11/0x20 __fput+0x9d/0x240 task_work_run+0x62/0x90 exit_to_user_mode_prepare+0x110/0x120 syscall_exit_to_user_mode+0x27/0x190 entry_SYSCALL_64_after_hwframe+0x44/0xa9
The reason we observed is that:
When the listener is closing, a connection may have completed the three-way handshake but not accepted, and the client has sent some packets. The child sks in accept queue release by inet_child_forget()->inet_csk_destroy_sock(), but psocks of child sks have not released.
To fix, add sock_map_destroy to release psocks.
Signed-off-by: Wang Yufen wangyufen@huawei.com Signed-off-by: Daniel Borkmann daniel@iogearbox.net Signed-off-by: Andrii Nakryiko andrii@kernel.org Acked-by: Jakub Sitnicki jakub@cloudflare.com Acked-by: John Fastabend john.fastabend@gmail.com Link: https://lore.kernel.org/bpf/20220524075311.649153-1-wangyufen@huawei.com Stable-dep-of: 8bbabb3fddcd ("bpf, sock_map: Move cancel_work_sync() out of sock lock") Signed-off-by: Sasha Levin sashal@kernel.org [Conflict in include/linux/bpf.h due to function declaration position and remove non-existed sk_psock_stop helper from sock_map_destroy.] Signed-off-by: Wen Gu guwen@linux.alibaba.com
background: Link: https://lore.kernel.org/stable/d11bc7e6-a2c7-445a-8561-3599eafb07b0@linux.al...
@stable team: This backport has 2 changes compared to the original patch:
- fix conflict due to sock_map_destroy declaration position in include/linux/bpf.h;
- remove the non-existed sk_psock_stop helper from sock_map_destroy. This helper is introduced by 799aa7f98d53 ("skmsg: Avoid lock_sock() in sk_psock_backlog()") after v5.10, it is not a fix and hard to backport. Considering that what did in sk_psock_stop is done in sk_psock_drop and neither sock_map_close nor sock_map_unhash in v5.10 introduces sk_psock_stop, I removed it from sock_map_destroy too.
I tested it in my environment, the regression was gone.
Now queued up, thanks.
greg k-h