netdev - Re: [PATCH v2 net] wireguard: use DEV_STATS

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: <ZVswP0vvfCm0FlWO@Laptop-X1>
Date: Mon, 20 Nov 2023 18:09:03 +0800
From: Hangbin Liu <liuhangbin@...il.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: "Jason A. Donenfeld" <Jason@...c4.com>,
	patchwork-bot+netdevbpf@...nel.org, davem@...emloft.net,
	kuba@...nel.org, pabeni@...hat.com, netdev@...r.kernel.org,
	eric.dumazet@...il.com, syzkaller@...glegroups.com
Subject: Re: [PATCH v2 net] wireguard: use DEV_STATS_INC()

On Mon, Nov 20, 2023 at 08:56:02AM +0100, Eric Dumazet wrote:
> > I thought that, given my concerns, if this was to be committed, at
> > least Eric (or you?) would expand on the rationale in the context of
> > my concerns while (or before) doing so, rather than just applying this
> > without further discussion. As I mentioned, this is fine with me if
> > you feel strongly about it, but I would appreciate some expanded
> > explanation, just for my own understanding of the matter.
> >
> > Jason
> 
> Jason, I was in week end mode, so could not reply to your message.
> 
> This fix is rather obvious to me. I do not want to spend too much time on it,
> and you gave an ACK if I am not mistaken.
> 
> If you prefer not letting syzbot find other bugs in wireguard (because
> hitting this issue first),
> this is fine by me. We can ask syzbot team to not include wireguard in
> their kernels.

Some performance test data may helps.

As I don't have good and same configure test machines. I just did a rough test.

Client: RHEL9.2, CPU  Intel E5-2620, Memory 4096 MB, NIC I350
Server: 6.7.0-rc1, CPU Intel Xeon Silver 4216, 196608 MB, NIC I350

Before patch:
=== 4in4 TCP_STREAM: 901.05 Mbits/sec ===
=== 4in4 MPTCP_STREAM: 885.22 Mbits/sec ===
=== 4in4 UDP_STREAM: 919.91 Mbits/sec ===
=== 4in4 SCTP_STREAM: 903.12 Mbits/sec ===

After patch:
=== 4in4 TCP_STREAM: 901.07 Mbits/sec ===
=== 4in4 MPTCP_STREAM: 885.24 Mbits/sec ===
=== 4in4 UDP_STREAM: 919.91 Mbits/sec ===
=== 4in4 SCTP_STREAM: 903.14 Mbits/sec ===

Exchange the client/server role:

Before patch:
=== 4in4 TCP_STREAM: 901.08 Mbits/sec ===
=== 4in4 MPTCP_STREAM: 885.12 Mbits/sec ===
=== 4in4 UDP_STREAM: 919.94 Mbits/sec ===
=== 4in4 SCTP_STREAM: 903.09 Mbits/sec ===

After patch:
=== 4in4 TCP_STREAM: 901.04 Mbits/sec ===
=== 4in4 MPTCP_STREAM: 885.24 Mbits/sec ===
=== 4in4 UDP_STREAM: 919.91 Mbits/sec ===
=== 4in4 SCTP_STREAM: 903.11 Mbits/sec ===

The result looks good to me.

Thanks
Hangbin