From: Jakub Sitnicki <jakub@cloudflare.com>
To: Liu Jian <liujian56@huawei.com>, john.fastabend@gmail.com
Cc: edumazet@google.com, davem@davemloft.net,
yoshfuji@linux-ipv6.org, dsahern@kernel.org, kuba@kernel.org,
pabeni@redhat.com, andrii@kernel.org, mykolal@fb.com,
ast@kernel.org, daniel@iogearbox.net, martin.lau@linux.dev,
song@kernel.org, yhs@fb.com, kpsingh@kernel.org, sdf@google.com,
haoluo@google.com, jolsa@kernel.org, shuah@kernel.org,
bpf@vger.kernel.org
Subject: Re: [PATCH bpf-next v2 0/2] If the sock is dead, do not access sock's sk_wq in sk_stream_wait_memory
Date: Thu, 27 Oct 2022 17:30:51 +0200 [thread overview]
Message-ID: <87zgdhgu0p.fsf@cloudflare.com> (raw)
In-Reply-To: <874jvpim37.fsf@cloudflare.com>
On Thu, Oct 27, 2022 at 12:36 PM +02, Jakub Sitnicki wrote:
[...]
> While testing Cong's fix for the dead lock in sk_psock_backlog [1], I've
> noticed that this change introduces a memory accounting issue. See
> warnings below.
>
> So what I've proposed is not completely sound. We will need to revisit
> it.
>
> [1] https://lore.kernel.org/bpf/Y0xJUc%2FLRu8K%2FAf8@pop-os.localdomain/
>
> --8<--
> bash-5.1# uname -r
> 6.0.0-rc3-00892-g3f8ef65af927
> bash-5.1# ./test_sockmap
> # 1/ 6 sockmap::txmsg test passthrough:OK
> # 2/ 6 sockmap::txmsg test redirect:OK
> # 3/ 1 sockmap::txmsg test redirect wait send mem:OK
> # 4/ 6 sockmap::txmsg test drop:OK
> # 5/ 6 sockmap::txmsg test ingress redirect:OK
> # 6/ 7 sockmap::txmsg test skb:OK
> # 7/ 8 sockmap::txmsg test apply:OK
> # 8/12 sockmap::txmsg test cork:OK
> [ 46.324023] ------------[ cut here ]------------
> [ 46.325114] WARNING: CPU: 3 PID: 199 at net/core/stream.c:206
> sk_stream_kill_queues+0xd6/0xf0
> [ 46.326573] Modules linked in:
> [ 46.327105] CPU: 3 PID: 199 Comm: test_sockmap Not tainted
> 6.0.0-rc3-00892-g3f8ef65af927 #36
> [ 46.328406] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> 1.15.0-1.fc35 04/01/2014
> [ 46.330000] RIP: 0010:sk_stream_kill_queues+0xd6/0xf0
> [ 46.330816] Code: 29 5b 5d 31 c0 89 c2 89 c6 89 c7 c3 48 89 df e8 10 14 ff ff
> 8b 83 70 02 00 00 8b b3 28 02 00 00 85 c0 74 d9 0f 0b 85 f6 74 d7 <0f> 0b 5b 5d
> 31 c0 89 c2 89 c6 89 c7 c3 0f 0b eb 92 66 0f 1f 84 00
> [ 46.331889] RSP: 0018:ffffc90000bc7d48 EFLAGS: 00010206
> [ 46.332186] RAX: 0000000000000000 RBX: ffff88810567dc00 RCX: 0000000000000000
> [ 46.332583] RDX: 0000000000000000 RSI: 0000000000000fc0 RDI: ffff88810567ddb8
> [ 46.332991] RBP: ffff88810567ddb8 R08: 0000000000000000 R09: 0000000000000000
> [ 46.333321] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88810567dc00
> [ 46.333661] R13: ffff888103894500 R14: ffff888101fdf8e0 R15: ffff88810567dd30
> [ 46.334074] FS: 00007f420a4d8b80(0000) GS:ffff88813bd80000(0000) knlGS:0000000000000000
> [ 46.334532] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 46.334881] CR2: 00007f420a4d8af8 CR3: 0000000102e17001 CR4: 0000000000370ee0
> [ 46.335300] Call Trace:
> [ 46.335444] <TASK>
> [ 46.335573] inet_csk_destroy_sock+0x4f/0x110
> [ 46.335832] tcp_rcv_state_process+0xdcf/0x1140
> [ 46.336076] ? tcp_v4_do_rcv+0x77/0x2a0
> [ 46.336299] tcp_v4_do_rcv+0x77/0x2a0
> [ 46.336512] __release_sock+0x58/0xb0
> [ 46.336721] __tcp_close+0x186/0x450
> [ 46.336883] tcp_close+0x20/0x70
> [ 46.337026] inet_release+0x39/0x80
> [ 46.337177] __sock_release+0x37/0xa0
> [ 46.337341] sock_close+0x14/0x20
> [ 46.337486] __fput+0xa2/0x260
> [ 46.337622] task_work_run+0x59/0xa0
> [ 46.337785] exit_to_user_mode_prepare+0x185/0x190
> [ 46.337992] syscall_exit_to_user_mode+0x19/0x40
> [ 46.338189] do_syscall_64+0x42/0x90
> [ 46.338350] entry_SYSCALL_64_after_hwframe+0x46/0xb0
> [ 46.338565] RIP: 0033:0x7f420a618eb7
> [ 46.338741] Code: ff e8 7d e2 01 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3
> 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 03 00 00 00 0f 05 <48> 3d 00 f0
> ff ff 77 41 c3 48 83 ec 18 89 7c 24 0c e8 43 cd f5 ff
> [ 46.339596] RSP: 002b:00007ffd54748df8 EFLAGS: 00000246 ORIG_RAX: 0000000000000003
> [ 46.339944] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 00007f420a618eb7
> [ 46.340349] RDX: 0000000000000018 RSI: 00007ffd54748d50 RDI: 0000000000000019
> [ 46.340753] RBP: 00007ffd54748e40 R08: 00007ffd54748d50 R09: 00007ffd54748d50
> [ 46.341162] R10: 00007ffd54748d50 R11: 0000000000000246 R12: 00007ffd54749118
> [ 46.341574] R13: 000000000040d0ad R14: 0000000000474df8 R15: 00007f420a769000
> [ 46.341927] </TASK>
> [ 46.342025] irq event stamp: 206385
> [ 46.342175] hardirqs last enabled at (206393): [<ffffffff810f1f82>]
> __up_console_sem+0x52/0x60
> [ 46.342546] hardirqs last disabled at (206400): [<ffffffff810f1f67>]
> __up_console_sem+0x37/0x60
> [ 46.342965] softirqs last enabled at (206414): [<ffffffff8107bcc5>]
> __irq_exit_rcu+0xc5/0x120
> [ 46.343454] softirqs last disabled at (206409): [<ffffffff8107bcc5>]
> __irq_exit_rcu+0xc5/0x120
> [ 46.343951] ---[ end trace 0000000000000000 ]---
> [ 46.344199] ------------[ cut here ]------------
> [ 46.344421] WARNING: CPU: 3 PID: 199 at net/ipv4/af_inet.c:154
> inet_sock_destruct+0x1a0/0x1d0
> [ 46.344897] Modules linked in:
> [ 46.345074] CPU: 3 PID: 199 Comm: test_sockmap Tainted: G W
> 6.0.0-rc3-00892-g3f8ef65af927 #36
> [ 46.345549] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> 1.15.0-1.fc35 04/01/2014
> [ 46.346028] RIP: 0010:inet_sock_destruct+0x1a0/0x1d0
> [ 46.346313] Code: ff 49 8b bc 24 60 02 00 00 e8 cc 1e e9 ff 49 8b bc 24 88 00
> 00 00 5b 41 5c e9 bc 1e e9 ff 41 8b 84 24 28 02 00 00 85 c0 74 ca <0f> 0b eb c6
> 4c 89 e7 e8 a4 2d e6 ff e9 50 ff ff ff 0f 0b 41 8b 84
> [ 46.347370] RSP: 0018:ffffc90000bc7e40 EFLAGS: 00010206
> [ 46.347670] RAX: 0000000000000fc0 RBX: ffff88810567dd60 RCX: 0000000000000000
> [ 46.348090] RDX: 0000000000000303 RSI: 0000000000000fc0 RDI: ffff88810567dd60
> [ 46.348495] RBP: ffff88810567dc00 R08: 0000000000000000 R09: 0000000000000000
> [ 46.348921] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88810567dc00
> [ 46.349306] R13: ffff8881001d48e0 R14: ffff88810257a3a8 R15: 0000000000000000
> [ 46.349717] FS: 00007f420a4d8b80(0000) GS:ffff88813bd80000(0000) knlGS:0000000000000000
> [ 46.350191] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 46.350531] CR2: 00007f420a4d8af8 CR3: 0000000102e17001 CR4: 0000000000370ee0
> [ 46.350947] Call Trace:
> [ 46.351098] <TASK>
> [ 46.351246] __sk_destruct+0x23/0x250
> [ 46.351599] inet_release+0x39/0x80
> [ 46.351835] __sock_release+0x37/0xa0
> [ 46.352059] sock_close+0x14/0x20
> [ 46.352206] __fput+0xa2/0x260
> [ 46.352347] task_work_run+0x59/0xa0
> [ 46.352515] exit_to_user_mode_prepare+0x185/0x190
> [ 46.352728] syscall_exit_to_user_mode+0x19/0x40
> [ 46.352954] do_syscall_64+0x42/0x90
> [ 46.353119] entry_SYSCALL_64_after_hwframe+0x46/0xb0
> [ 46.353342] RIP: 0033:0x7f420a618eb7
> [ 46.353498] Code: ff e8 7d e2 01 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3
> 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 03 00 00 00 0f 05 <48> 3d 00 f0
> ff ff 77 41 c3 48 83 ec 18 89 7c 24 0c e8 43 cd f5 ff
> [ 46.354284] RSP: 002b:00007ffd54748df8 EFLAGS: 00000246 ORIG_RAX: 0000000000000003
> [ 46.354601] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 00007f420a618eb7
> [ 46.354908] RDX: 0000000000000018 RSI: 00007ffd54748d50 RDI: 0000000000000019
> [ 46.355308] RBP: 00007ffd54748e40 R08: 00007ffd54748d50 R09: 00007ffd54748d50
> [ 46.355710] R10: 00007ffd54748d50 R11: 0000000000000246 R12: 00007ffd54749118
> [ 46.356117] R13: 000000000040d0ad R14: 0000000000474df8 R15: 00007f420a769000
> [ 46.356532] </TASK>
> [ 46.356664] irq event stamp: 206837
> [ 46.356873] hardirqs last enabled at (206847): [<ffffffff810f1f82>]
> __up_console_sem+0x52/0x60
> [ 46.357369] hardirqs last disabled at (206854): [<ffffffff810f1f67>]
> __up_console_sem+0x37/0x60
> [ 46.357864] softirqs last enabled at (206604): [<ffffffff8107bcc5>]
> __irq_exit_rcu+0xc5/0x120
> [ 46.358358] softirqs last disabled at (206591): [<ffffffff8107bcc5>]
> __irq_exit_rcu+0xc5/0x120
> [ 46.358865] ---[ end trace 0000000000000000 ]---
> # 9/ 3 sockmap::txmsg test hanging corks:OK
[...]
Actually, we had a fix for these warnings in 9c34e38c4a87 ("bpf,
sockmap: Fix memleak in tcp_bpf_sendmsg while sk msg is full") [1].
But they reappear for me in v5.18-rc1. Trying to bisect what went
wrong...
[1] https://lore.kernel.org/bpf/20220304081145.2037182-3-wangyufen@huawei.com
next prev parent reply other threads:[~2022-10-27 15:34 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-23 13:37 [PATCH bpf-next v2 0/2] If the sock is dead, do not access sock's sk_wq in sk_stream_wait_memory Liu Jian
2022-08-23 13:37 ` [PATCH bpf-next v2 1/2] net: " Liu Jian
2022-08-30 2:33 ` John Fastabend
2022-08-23 13:37 ` [PATCH bpf-next v2 2/2] selftests/bpf: Add wait send memory test for sockmap redirect Liu Jian
2022-09-26 16:00 ` [PATCH bpf-next v2 0/2] If the sock is dead, do not access sock's sk_wq in sk_stream_wait_memory patchwork-bot+netdevbpf
2022-10-27 10:36 ` Jakub Sitnicki
2022-10-27 15:30 ` Jakub Sitnicki [this message]
2022-10-27 21:24 ` Jakub Sitnicki
2022-10-28 2:56 ` 答复: " liujian (CE)
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87zgdhgu0p.fsf@cloudflare.com \
--to=jakub@cloudflare.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=edumazet@google.com \
--cc=haoluo@google.com \
--cc=john.fastabend@gmail.com \
--cc=jolsa@kernel.org \
--cc=kpsingh@kernel.org \
--cc=kuba@kernel.org \
--cc=liujian56@huawei.com \
--cc=martin.lau@linux.dev \
--cc=mykolal@fb.com \
--cc=pabeni@redhat.com \
--cc=sdf@google.com \
--cc=shuah@kernel.org \
--cc=song@kernel.org \
--cc=yhs@fb.com \
--cc=yoshfuji@linux-ipv6.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.