netdev - Re: [PATCH v2] rfs: Receive Flow Steering

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:	Tue, 06 Apr 2010 17:10:52 +0200
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Tom Herbert <therbert@...gle.com>
Cc:	davem@...emloft.net, netdev@...r.kernel.org
Subject: Re: [PATCH v2] rfs: Receive Flow Steering

Le mardi 06 avril 2010 à 07:25 -0700, Tom Herbert a écrit :

> > 2) inet_rps_save_rxhash(sk, skb->rxhash);
> >
> >        It should have a check to make sure some part of the stack doesnt feed
> > many different rxhash for a given socket (Make sure we dont pollute flow
> > table with pseudo random values)
> >
> If packets for a connection are always received on the same device, is
> it reasonable to assume the rxhash is constant for that connection?
> 
> I suppose it's possible that packets for a same sockets are being
> constantly received on two different devices that are giving different
> rxhashes.  This would already be bad in that OOO is probably happening
> anyway.  I don't know if thrashing the sock_flow_table is going to
> aggravate this scenario much.
> 
> Are there any other degenerative cases you're worried about?

No, I was only considering this mostly as a debugging aid, before RPS is
stable, because mismatches could be unnoticed and performance not
optimal.
 


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html