netdev - Re: net: tso: add UDP segmentation support: adds regression for ax200 upload

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

Message-ID: <765f370d-ce2d-b75a-2dde-87f69ae7c185@candelatech.com>
Date:   Wed, 16 Dec 2020 15:59:54 -0800
From:   Ben Greear <greearb@...delatech.com>
To:     netdev <netdev@...r.kernel.org>, edumazet@...gle.com
Subject: Re: net: tso: add UDP segmentation support: adds regression for ax200
 upload

On 12/16/20 3:09 PM, Ben Greear wrote:
> Hello Eric,
> 
> The patch below evidently causes TCP throughput to be about 50Mbps instead of 700Mbps
> when using ax200 to upload tcp traffic.
> 
> When I disable TSO, performance goes back up to around 700Mbps.

As a followup, when I revert the patch, upload speed goes to ~900Mbps,
so even better than just disabling TSO (I left TSO enabled after reverting the patch).

Thanks,
Ben

> 
> I recall ~5 years ago we had similar TCP related performance issues with ath10k.
> I vaguely recall that there might be some driver-level socket pacing tuning value, but I cannot
> find the right thing to search for.  Is this really a thing?  If so, maybe it will
> be a way to resolve this issue?
> 
> See this more thorough bug report:
> 
> https://bugzilla.kernel.org/show_bug.cgi?id=209913
> 
> Patch description:
> net: tso: add UDP segmentation support
> Note that like TCP, we do not support additional encapsulations,
> and that checksums must be offloaded to the NIC.
> 
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> Signed-off-by: David S. Miller <davem@...emloft.net>
> 
> Thanks,
> Ben
>