All of lore.kernel.org
 help / color / mirror / Atom feed
From: Patrick McHardy <kaber@trash.net>
To: Eric Lemoine <eric.lemoine@gmail.com>
Cc: netdev@oss.sgi.com, hadi@cyberus.ca,
	"David S. Miller" <davem@davemloft.net>,
	openib-general@openib.org
Subject: Re: LLTX and netif_stop_queue
Date: Thu, 23 Dec 2004 17:37:24 +0100	[thread overview]
Message-ID: <41CAF444.3000305@trash.net> (raw)
In-Reply-To: <5cac192f0412230110628749e3@mail.gmail.com>

Eric Lemoine wrote:
> I still have one concern with the LLTX code (and it may be that the
> correct patch is Jamal's) :
> 
> Without LLTX we do : lock(queue_lock), lock(xmit_lock),
> release(queue_lock), release(xmit_lock). With LLTX (without Jamal's
> patch) we do : lock(queue_lock), release(queue_lock), lock(tx_lock),
> release(tx_lock). LLTX doesn't look correct because it creates a race
> condition window between the the two lock-protected sections. So you
> may want to reconsider Jamal's patch or pull out LLTX...

You're right, it can cause packet reordering if something like this
happens:

CPU1			CPU2
lock(queue_lock)	
dequeue
unlock(queue_lock)
			lock(queue_lock)
			dequeue
			unlock(queue_lock)
			lock(xmit_lock)
			hard_start_xmit
			unlock(xmit_lock)
lock(xmit_lock)
hard_start_xmit
unlock(xmit_lock)

Jamal's patch should fix this.

Regards
Patrick

  reply	other threads:[~2004-12-23 16:37 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-12-17 21:57 LLTX and netif_stop_queue Roland Dreier
2004-12-18  5:44 ` David S. Miller
2004-12-18 15:35   ` Roland Dreier
2004-12-18 17:58     ` Roland Dreier
2004-12-18 18:26       ` Roland Dreier
2004-12-19 19:33     ` jamal
2004-12-19 19:31   ` jamal
2004-12-19 19:54     ` jamal
2004-12-19 20:02       ` Jamal Hadi Salim
2004-12-19 22:35     ` Roland Dreier
2004-12-19 23:06       ` jamal
2004-12-19 23:16         ` Roland Dreier
2004-12-22 18:49     ` Eric Lemoine
2004-12-23  4:29       ` David S. Miller
2004-12-23  4:38         ` Roland Dreier
2004-12-23  9:10         ` Eric Lemoine
2004-12-23 16:37           ` Patrick McHardy [this message]
2004-12-23 18:11             ` Eric Lemoine
2004-12-24 16:10             ` Eric Lemoine
2004-12-28 13:31               ` jamal
2005-01-02 23:30                 ` Eric Lemoine
2005-01-03  7:41                   ` Eric Lemoine
2005-01-03 15:04                   ` jamal
2005-01-03 15:48                     ` Eric Lemoine
2005-01-03 15:57                     ` Roland Dreier
2005-01-03 16:41                       ` Eric Lemoine
2005-01-03 16:54                         ` Roland Dreier
2005-01-03 17:07                           ` Eric Lemoine
2005-01-03 17:12                             ` Grant Grundler
2005-01-04  4:18                               ` jamal
2005-01-19 22:47                                 ` David S. Miller
2005-01-19 23:18                                   ` Stephen Hemminger
2005-01-19 23:41                                     ` David S. Miller
2005-01-20  0:02                                       ` [openib-general] " Jeff Garzik
2005-01-20  0:46                                         ` Stephen Hemminger
2005-01-20  0:47                                           ` David S. Miller
2005-01-20  0:47                                       ` Francois Romieu
2005-01-20  0:52                                         ` David S. Miller
2005-01-20  1:17                                           ` Francois Romieu
2005-01-20  0:46                                     ` [PATCH]: was " David S. Miller
2005-01-20  3:14                                       ` Andi Kleen
2005-01-20  7:05                                         ` David S. Miller
2005-01-20  3:43                                       ` Roland Dreier
2005-01-20  7:05                                         ` David S. Miller
2005-01-20 13:51                                           ` Tommy Christensen
2005-01-20 21:34                                             ` David S. Miller
2005-01-20 21:56                                               ` Grant Grundler
2005-01-21  1:01                                                 ` David S. Miller
2005-01-22  3:17                                                   ` Roland Dreier
2005-01-22  3:53                                                     ` David S. Miller
2005-01-20 21:41                                             ` David S. Miller
2005-01-20 16:56                                           ` Stephen Hemminger
2005-01-21 10:54                                             ` Lennert Buytenhek
2005-01-26  6:27                                               ` David S. Miller
2005-01-26 13:25                                                 ` Lennert Buytenhek
2005-01-27  6:32                                                   ` David S. Miller
2005-01-27  7:16                                                     ` Andi Kleen
2005-01-27  7:22                                                       ` David S. Miller
2005-01-27  8:26                                                         ` Andi Kleen
2005-01-20 19:16                                           ` Jeff Garzik
2005-01-20  4:01                                       ` jamal
2005-01-20  5:18                                         ` David S. Miller

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=41CAF444.3000305@trash.net \
    --to=kaber@trash.net \
    --cc=davem@davemloft.net \
    --cc=eric.lemoine@gmail.com \
    --cc=hadi@cyberus.ca \
    --cc=netdev@oss.sgi.com \
    --cc=openib-general@openib.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.