[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <11d6270e-c4c9-4a3a-8d2b-d273031b9d4f@lunn.ch>
Date: Wed, 28 May 2025 13:57:31 +0200
From: Andrew Lunn <andrew@...n.ch>
To: Ricard Bejarano <ricard@...arano.io>
Cc: Mika Westerberg <mika.westerberg@...ux.intel.com>,
netdev@...r.kernel.org, michael.jamet@...el.com,
YehezkelShB@...il.com, andrew+netdev@...n.ch, davem@...emloft.net,
edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com
Subject: Re: Poor thunderbolt-net interface performance when bridged
On Wed, May 28, 2025 at 08:38:29AM +0200, Ricard Bejarano wrote:
> This impacts performance so much that I can't even test it.
>
> iperf3 doesn't even start and ping shows >1s latency:
>
> root@red:~# ping 10.0.0.2
> PING 10.0.0.2 (10.0.0.2) 56(84) bytes of data.
> 64 bytes from 10.0.0.2: icmp_seq=1 ttl=64 time=1041 ms
> 64 bytes from 10.0.0.2: icmp_seq=2 ttl=64 time=1024 ms
> 64 bytes from 10.0.0.2: icmp_seq=3 ttl=64 time=1025 ms
That is something else. The slightly larger than 1000ms, means ping #2
is kicking ping #1 onto the wire. The transmit is not starting for
some reason until the next packet is queued for transmission.
Andrew
Powered by blists - more mailing lists