netdev - Re: RPS arp processing

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 for Android: free password hash cracker in your pocket

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

Message-Id: <911391576102085@myt2-c3952fd46804.qloud-c.yandex.net>
Date:   Thu, 12 Dec 2019 01:08:05 +0300
From:   Aleksei Zakharov <zakharov.a.g@...dex.ru>
To:     Willem de Bruijn <willemdebruijn.kernel@...il.com>
Cc:     Network Development <netdev@...r.kernel.org>
Subject: Re: RPS arp processing



11.12.2019, 22:00, "Willem de Bruijn" <willemdebruijn.kernel@...il.com>:
> On Wed, Dec 11, 2019 at 10:34 AM Aleksei Zakharov
> <zakharov.a.g@...dex.ru> wrote:
>>  Hi, everyone
>>  Is it possible to balance ARP across CPUs with RPS?
>>  I don't clearly understand how hash is calulated for ARP packets, but it seems that it should consider source and target IPs.
>
> The hash is derived by flow dissection:
>
>     get_rps_cpus
>       ___skb_get_hash
>           skb_flow_dissect_flow_keys
>
> This calls __skb_flow_dissector with the flow_keys_dissector
> dissection program, which is initialized in
> init_default_flow_dissectors from flow_keys_dissector_keys.
>
> That program incorporates IPV4_ADDRS and IPV6_ADDRS. But that does not
> apply to ARP packets. Contrast case ETH_P_IPV6 with case ETH_P_ARP in
> __skb_flow_dissect.
>
> The flow dissector calls __skb_flow_dissect_arp() for deeper
> dissection, from which you could extract entropy for RPS. But the
> flow_keys_dissector program does not have FLOW_DISSECTOR_KEY_ARP
> enabled.
Thank you very much for the explanation!

>
>>  In our current setup we have one l2 segment between external hardware routers and namespaces on linux server.
>>  When router sends ARP request, it is passed through server's physical port, then via openvswitch bridge it is copied to every namespace.
>>  We've found that all ARPs (for different destination ips and few source ips) are processed on one CPU inside namespaces. We use RPS, and most packets are balanced between all CPUs.
>
> I suggest looking at the newer BPF flow dissector, which allows tuning
> dissection to specific use cases, like yours.
Thanks, I'll take a look at bpf dissector.

>
>>  Kernel 4.15.0-65 from ubuntu 18.04.
>>
>>  Might this issue be related to namespaces somehow?
>>
>>  --
>>  Regards,
>>  Aleksei Zakharov

-- 
Regards,
Aleksei Zakharov