netdev - Re: [PATCH v2] net: batch skb dequeueing from softnet input_pkt

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, 13 Apr 2010 15:21:34 +0200
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Changli Gao <xiaosuo@...il.com>
Cc:	"David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org
Subject: Re: [PATCH v2] net: batch skb dequeueing from softnet
 input_pkt_queue

Le mardi 13 avril 2010 à 20:53 +0800, Changli Gao a écrit :
> OK. If we make processing_queue is a stack variable. When quota or
> jiffies limit is reached, we have to splice processing_queue back to
> input_pkt_queue. If flush_backlog() is called before the
> processing_queue is spliced, there will still packets which refer to
> the NIC going. Then these packets are queued to input_pkt_queue. When
> process_backlog() is called again, the dev field of these skbs are
> wild...
> 

This is a problem of cooperation between flush_backlog() and
process_backlog(). Dont allow flush_backlog() to return if
process_backlog() is running. Exactly as before, but lock acquisition
done in flush_backlog() should be a bit smarter.

> Oh, my GOD. When RPS is enabled, if flush_backlog(eth0) is called on
> CPU1 when a skb0(eth0) is dequeued from CPU0's softnet and isn't
> queued to CPU1's softnet, what will happen?
> 

I am a bit lost here. flush_backlog() drops skbs, not requeue them.

> >
> > Absolutely not. You missed something apparently.
> >
> > You pay the price at each packet enqueue, because you have to compute
> > the sum of two lengthes, and guess what, if you do this you have a cache
> > line miss in one of the operand. Your patch as is is suboptimal.
> >
> > Remember : this batch mode should not change packet queueing at all,
> > only speed it because of less cache line misses.
> >
> 
> WoW, is it really so expensive?
> 

Yes. Whole point of your idea is to remove cache line misses.

They cost much more than a spinlock/unlock pair

--
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