linux-kernel - Re: [PATCH v2 2/2] net/tcp: add packetdrill test for FIN-WAIT-1 zero-window fix

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CAAVpQUC7qk_1Dj+fuC-wfesHkUMQhNoVdUY9GXo=vYzmJJ1WdA@mail.gmail.com>
Date: Thu, 23 Oct 2025 12:37:26 -0700
From: Kuniyuki Iwashima <kuniyu@...gle.com>
To: HaiYang Zhong <wokezhong@...il.com>
Cc: edumazet@...gle.com, ncardwell@...gle.com, davem@...emloft.net, 
	dsahern@...nel.org, kuba@...nel.org, pabeni@...hat.com, horms@...nel.org, 
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org, wokezhong@...cent.com
Subject: Re: [PATCH v2 2/2] net/tcp: add packetdrill test for FIN-WAIT-1
 zero-window fix

On Thu, Oct 23, 2025 at 7:48 AM HaiYang Zhong <wokezhong@...il.com> wrote:
>
> Add packetdrill test to reproduce and verify the permanent FIN-WAIT-1
> state issue when continuous zero window packets are received.
>
> The test simulates:
> - TCP connection establishment
> - Peer advertising zero window
> - Local FIN blocked in send buffer due to zero window
> - Continuous zero window ACKs from peer
> - Verification of connection timeout (after fix)
>
> Signed-off-by: HaiYang Zhong <wokezhong@...cent.com>
> ---
>  .../net/tcp_fin_wait1_zero_window.pkt         | 58 +++++++++++++++++++
>  1 file changed, 58 insertions(+)
>  create mode 100644 tools/testing/selftests/net/tcp_fin_wait1_zero_window.pkt
>
> diff --git a/tools/testing/selftests/net/tcp_fin_wait1_zero_window.pkt b/tools/testing/selftests/net/tcp_fin_wait1_zero_window.pkt
> new file mode 100644
> index 000000000000..86ceb95de744
> --- /dev/null
> +++ b/tools/testing/selftests/net/tcp_fin_wait1_zero_window.pkt
> @@ -0,0 +1,58 @@
> +// Test for permanent FIN-WAIT-1 state with continuous zero-window advertisements
> +// Author: HaiYang Zhong <wokezhong@...cent.com>
> +
> +
> +0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
> +0.000 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
> +0.000 bind(3, ..., ...) = 0
> +0.000 listen(3, 1) = 0
> +
> +0.100 < S 0:0(0) win 65535 <mss 1460>
> +0.100 > S. 0:0(0) ack 1 <mss 1460>
> +0.100 < . 1:1(0) ack 1 win 65535
> +0.100 accept(3, ..., ...) = 4
> +
> +// Send data to fill receive window
> +0.200 write(4, ..., 5) = 5
> +0.200 > P. 1:6(5) ack 1
> +
> +// Advertise zero-window
> +0.200 < . 1:1(0) ack 6 win 0
> +
> +// Application closes connection, sends FIN (but blocked by zero window)
> +0.200 close(4) = 0
> +
> +//Send zero-window probe packet
> ++0.200 > . 5:5(0) ack 1
> ++0.400 > . 5:5(0) ack 1
> ++0.800 > . 5:5(0) ack 1
> ++1.600 > . 5:5(0) ack 1
> ++3.200 > . 5:5(0) ack 1
> ++6.400 > . 5:5(0) ack 1
> ++12.800 > . 5:5(0) ack 1
> +
> +// Continuously sending zero-window ACKs
> +30.000 < . 1:1(0) ack 6 win 0
> +
> +// Key verification points
> +// Without fix: waiting for packet timeout due to timer reset
> +// With fix: this probe is sent as scheduled
> ++22.000~+23.000 > . 5:5(0) ack 1
> +
> +// More zero-window ACKs from peer
> +60.000 < . 1:1(0) ack 6 win 0
> +90.000 < . 1:1(0) ack 6 win 0
> ++16.000~+19.000 > . 5:5(0) ack 1
> +120.000 < . 1:1(0) ack 6 win 0
> +150.000 < . 1:1(0) ack 6 win 0
> +180.000 < . 1:1(0) ack 6 win 0
> +210.000 < . 1:1(0) ack 6 win 0
> ++0.000~+5.000  > . 5:5(0) ack 1
> +240.000 < . 1:1(0) ack 6 win 0
> +270.000 < . 1:1(0) ack 6 win 0
> +300.000 < . 1:1(0) ack 6 win 0
> +330.000 < . 1:1(0) ack 6 win 0
> +360.000 < . 1:1(0) ack 6 win 0
> +
> +// Connection reset after zero-window probe timeout
> ++0.000 > R 6:6(0)

I guess this test will time out if you run via selftest ?