mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>,
	rusty@rustcorp.com.au, virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	Michael Dalton <mwdalton@google.com>,
	Eric Dumazet <edumazet@google.com>
Subject: Re: [PATCH net] virtio-net: fix page refcnt leaking when fail to allocate frag skb
Date: Wed, 20 Nov 2013 07:16:33 -0800	[thread overview]
Message-ID: <1384960593.8604.147.camel@edumazet-glaptop2.roam.corp.google.com> (raw)
In-Reply-To: <20131120085835.GB19341@redhat.com>

On Wed, 2013-11-20 at 10:58 +0200, Michael S. Tsirkin wrote:
> On Tue, Nov 19, 2013 at 02:00:11PM -0800, Eric Dumazet wrote:
> > On Tue, 2013-11-19 at 23:53 +0200, Michael S. Tsirkin wrote:
> > 
> > > Which NIC? Virtio? Prior to 2613af0ed18a11d5c566a81f9a6510b73180660a
> > > it didn't drop packets received from host as far as I can tell.
> > > virtio is more like a pipe than a real NIC in this respect.
> > 
> > Prior/after to this patch, you were not posting buffers, so if packets
> > were received on a physical NIC, you were dropping the packets anyway.
> >
> > It makes no difference at all, adding a cushion might make you feel
> > better, but its really not worth it.
> > 
> > Under memory stress, it makes better sense to drop a super big GRO
> > packet (The one needing frag_list extension ...)
> > 
> > It gives a better signal to the sender to reduce its pressure, and gives
> > opportunity to free more of your memory.
> > 
> 
> OK, but in that case one wonders whether we should do more to free memory?
> 
> E.g. imagine that we dropped a packet of a specific TCP flow
> because we couldn't allocate a new packet.
> 
> What happens now is that the old packet is freed as well.
> 
> So quite likely the next packet in queue will get processed
> since it will reuse the memory we have just freed.
> 
> The next packet and the next after it etc all will have to go through
> the net stack until they get at the socket and are dropped then
> because we missed a segment.  Even worse, GRO gets disabled so the load
> on receiver goes up instead of down.
> 
> Sounds like a problem doesn't it?

I see no problem at all. GRO is a hint for high rates (and obviously
when there is enough memory)

> 
> GRO actually detects it's the same flow and can see packet is
> out of sequence. Why doesn't it drop the packet then?
> Alternatively, we could (for example using the pre-allocated skb
> like I suggested) notify GRO that it should start dropping packets
> of this flow.
> 
> What do you think?
> 

I think we disagree a lot on memory management on networking stacks.

We did a lot of work in TCP stack and Qdisc layers to lower memory
pressure (and bufferbloat), an you seem to try hard to introduce yet
another layer of buffer bloat in virtio_net.

So add whatever you want to proudly state to your management :

"Look how smart we are : we drop no packets in our layer"




  reply	other threads:[~2013-11-20 15:16 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-19  8:05 Jason Wang
2013-11-19 14:03 ` Eric Dumazet
2013-11-19 18:44   ` Michael S. Tsirkin
2013-11-19 20:49   ` Michael S. Tsirkin
2013-11-19 21:36     ` Eric Dumazet
2013-11-19 21:53       ` Michael S. Tsirkin
2013-11-19 22:00         ` Eric Dumazet
2013-11-20  1:34           ` Michael Dalton
2013-11-20  3:17             ` Jason Wang
2013-11-20  9:00             ` Michael S. Tsirkin
2013-11-20  8:58           ` Michael S. Tsirkin
2013-11-20 15:16             ` Eric Dumazet [this message]
2013-11-20 16:06               ` Michael S. Tsirkin
2013-11-20 16:14                 ` Eric Dumazet
2013-11-20 17:03                   ` Michael S. Tsirkin
2013-11-19 21:38     ` Michael Dalton
2013-11-20  9:06       ` Michael S. Tsirkin
2013-11-20  3:05     ` Jason Wang
2013-11-20  3:00   ` Jason Wang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1384960593.8604.147.camel@edumazet-glaptop2.roam.corp.google.com \
    --to=eric.dumazet@gmail.com \
    --cc=edumazet@google.com \
    --cc=jasowang@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=mwdalton@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=rusty@rustcorp.com.au \
    --cc=virtualization@lists.linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®