LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Jarek Poplawski <jarkao2@o2.pl>
To: Thomas Graf <tgraf@suug.ch>
Cc: "Kok\, Auke" <auke-jan.h.kok@intel.com>,
	David Miller <davem@davemloft.net>, "Garzik\,
	Jeff" <jgarzik@pobox.com>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	Peter Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>,
	"Brandeburg\, Jesse" <jesse.brandeburg@intel.com>, "Kok\,
	Auke" <auke@foo-projects.org>, "Ronciak\,
	John" <john.ronciak@intel.com>
Subject: Re: [PATCH 1/2] NET: Multiple queue network device support
Date: Mon, 12 Mar 2007 09:58:02 +0100	[thread overview]
Message-ID: <20070312085802.GA1664@ff.dom.local> (raw)
In-Reply-To: <20070309134037.GH521@postel.suug.ch>

On 09-03-2007 14:40, Thomas Graf wrote:
> * Kok, Auke <auke-jan.h.kok@intel.com> 2007-02-08 16:09
>> diff --git a/net/core/dev.c b/net/core/dev.c
>> index 455d589..42b635c 100644
>> --- a/net/core/dev.c
>> +++ b/net/core/dev.c
>> @@ -1477,6 +1477,49 @@ gso:
>>  	skb->tc_verd = SET_TC_AT(skb->tc_verd,AT_EGRESS);
>>  #endif
>>  	if (q->enqueue) {
>> +#ifdef CONFIG_NET_MULTI_QUEUE_DEVICE
>> +		int queue_index;
>> +		/* If we're a multi-queue device, get a queue index to lock */
>> +		if (netif_is_multiqueue(dev))
>> +		{
>> +			/* Get the queue index and lock it. */
>> +			if (likely(q->ops->map_queue)) {
>> +				queue_index = q->ops->map_queue(skb, q);
>> +				spin_lock(&dev->egress_subqueue[queue_index].queue_lock);
>> +				rc = q->enqueue(skb, q);

I'm not sure Dave Miller thought about this place, when he proposed
to save the mapping, but I think this could be not enough. This place
is racy: ->map_queue() is called 2 times and with some filters (and
policies/actions) results could differ. And of course the subqueue
lock doesn't prevent any filter from a config change in the meantime.

After second reading of this patch I have doubts it's the proper way
to solve the problem: there are many subqueues but we need a top queue
(prio here) to mange them, anyway. So, why not to build this
functionality directly into the queue? There is no difference for a
device if skbs are going from the subqueue or a class, it is only
interested in the mapping result and a possibility to stop and start
a subqueue and to query its status. All this could be done by adding
the callbacks directly to any classful scheduler or, if not enough,
to write some specialized qdisc based on prio. The possibility to lock
only a subqueue instead of a queue could be analized independently -
current proposal doesn't solve this anyway.

Regards,
Jarek P.

  parent reply	other threads:[~2007-03-12  8:53 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-02-09  0:09 [PATCH 0/2 REVIEW] Multiple transmit/receive queue kernel Kok, Auke
2007-02-09  0:09 ` [PATCH 1/2] NET: Multiple queue network device support Kok, Auke
2007-02-27  1:03   ` David Miller
2007-02-27 19:38     ` Waskiewicz Jr, Peter P
2007-03-07 22:18     ` Waskiewicz Jr, Peter P
2007-03-07 22:42       ` David Miller
2007-03-09  7:26         ` Jarek Poplawski
2007-03-09 13:40   ` Thomas Graf
2007-03-09 19:25     ` Waskiewicz Jr, Peter P
2007-03-09 23:01       ` Thomas Graf
2007-03-09 23:27         ` Waskiewicz Jr, Peter P
2007-03-10  2:34           ` Thomas Graf
2007-03-10 20:37             ` Waskiewicz Jr, Peter P
2007-03-12  8:58     ` Jarek Poplawski [this message]
2007-03-12 20:21       ` Waskiewicz Jr, Peter P
2007-02-09  0:09 ` [PATCH 2/2] e1000: Implement the new kernel API for multiqueue TX support Kok, Auke
2007-03-09 13:11   ` Thomas Graf
2007-02-23  9:00 [PATCH 1/2] NET: Multiple queue network device support Sreenivasa Honnur
2007-02-23 19:05 ` Waskiewicz Jr, Peter P
2007-02-23 19:19 ` Stephen Hemminger
2007-02-23 19:23   ` Kok, Auke
2007-02-23  9:02 Sreenivasa Honnur

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=20070312085802.GA1664@ff.dom.local \
    --to=jarkao2@o2.pl \
    --cc=auke-jan.h.kok@intel.com \
    --cc=auke@foo-projects.org \
    --cc=davem@davemloft.net \
    --cc=jesse.brandeburg@intel.com \
    --cc=jgarzik@pobox.com \
    --cc=john.ronciak@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=peter.p.waskiewicz.jr@intel.com \
    --cc=tgraf@suug.ch \
    --subject='Re: [PATCH 1/2] NET: Multiple queue network device support' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).