netdev - Re: PPPoE performance regression

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: <1339581421.11011.18.camel@shinybook.infradead.org>
Date:	Wed, 13 Jun 2012 10:57:01 +0100
From:	David Woodhouse <dwmw2@...radead.org>
To:	Nathan Williams <nathan@...verse.com.au>
Cc:	Karl Hiramoto <karl@...amoto.org>,
	"David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org,
	Paul Mackerras <paulus@...ba.org>
Subject: Re: PPPoE performance regression

On Sun, 2012-06-10 at 09:32 +0100, David Woodhouse wrote:
> I was looking to see if we could (ab)use the skb->destructor somehow so
> that we get *notified* when the packet is actually sent (or dropped),
> and then that would allow us to manage the queue 'downstream' of PPP
> more sanely. But I haven't really got very far with that yet.

This doesn't look *so* evil... if the basic concept of using
skb_orphan() and then setting our own destructor is OK, then I'll work
out the rest of the details and do it for l2tp too.

If the end-goal is to do BQL for PPP and honour the netdev queue limits,
perhaps the best option is to add a 'done sending skb' callback to the
generic ppp code and let it manage the queues (and the BQL callbacks)
*there*.

-- 
David Woodhouse                            Open Source Technology Centre
David.Woodhouse@...el.com                              Intel Corporation



Download attachment "smime.p7s" of type "application/x-pkcs7-signature" (6171 bytes)