netdev - Re: NAT performance regression caused by vlan GRO support

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]

Date:   Fri, 5 Apr 2019 04:00:20 -0700
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Florian Westphal <fw@...len.de>,
        Toke Høiland-Jørgensen 
        <toke@...hat.com>
Cc:     Toshiaki Makita <makita.toshiaki@....ntt.co.jp>,
        Felix Fietkau <nbd@....name>,
        Rafał Miłecki <zajec5@...il.com>,
        Toshiaki Makita <toshiaki.makita1@...il.com>,
        netdev@...r.kernel.org, "David S. Miller" <davem@...emloft.net>,
        Stefano Brivio <sbrivio@...hat.com>,
        Sabrina Dubroca <sd@...asysnail.net>,
        David Ahern <dsahern@...il.com>, Jo-Philipp Wich <jo@...n.io>,
        Koen Vandeputte <koen.vandeputte@...ntric.com>
Subject: Re: NAT performance regression caused by vlan GRO support



On 04/05/2019 03:51 AM, Florian Westphal wrote:
> Toke Høiland-Jørgensen <toke@...hat.com> wrote:
>> As a first approximation, maybe just:
>>
>> if (!has_hardware_cksum_offload(netdev) && link_rate(netdev) <= 1Gbps)
>>   disable_gro();
> 
> I don't think its a good idea.  For local delivery case, there is no
> way to avoid the checksum cost, so might as well have GRO enabled.
> 

We might add a sysctl or a way to tell GRO layer :

Do not attempt checksumming if forwarding is enabled on the host.

Basically GRO if NIC has provided checksum offload.