All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Thomas De Schampheleire <patrickdepinguin+linuxppc@gmail.com>
Cc: David Miller <davem@davemloft.net>,
	ronny.meeus@gmail.com, linuxppc-dev@ozlabs.org,
	netdev@vger.kernel.org, afleming@freescale.com
Subject: Re: softirqs are invoked while bottom halves are masked
Date: Wed, 13 Jul 2011 09:38:53 +0200	[thread overview]
Message-ID: <1310542733.2634.17.camel@edumazet-laptop> (raw)
In-Reply-To: <CAAXf6LXddZUHEz738DDTi-XqENRYT7dTODw+1C4RTYsCdyro7g@mail.gmail.com>

Le mercredi 13 juillet 2011 à 09:20 +0200, Thomas De Schampheleire a
écrit :

> I just want to mention that, although you seem to think that we
> shouldn't have created this thread in the first place, your and Eric's
> remarks have actually helped us in identifying the problem and
> increasing our understanding.
> 

Thread was fine IMHO, but should have mentioned the context of "non
standard linux-2.6 or net-next-2.6 trees"

I would like to mention NETIF_F_LLTX existence.

While being marked deprecated, it allows a lockless tx path :

The device transmit handler can then perform custom actions, including
its own locking if needed.

If the user wants a specific Qdisc setup (instead of a no qdisc one), it
still can.

veth, loopback, dummy, macvlan & bond devices use NETIF_F_LLTX for
example.

But ndo_start_xmit() must always be called with BH disabled.




WARNING: multiple messages have this Message-ID (diff)
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Thomas De Schampheleire <patrickdepinguin+linuxppc@gmail.com>
Cc: linuxppc-dev@ozlabs.org, netdev@vger.kernel.org,
	afleming@freescale.com, David Miller <davem@davemloft.net>,
	ronny.meeus@gmail.com
Subject: Re: softirqs are invoked while bottom halves are masked
Date: Wed, 13 Jul 2011 09:38:53 +0200	[thread overview]
Message-ID: <1310542733.2634.17.camel@edumazet-laptop> (raw)
In-Reply-To: <CAAXf6LXddZUHEz738DDTi-XqENRYT7dTODw+1C4RTYsCdyro7g@mail.gmail.com>

Le mercredi 13 juillet 2011 à 09:20 +0200, Thomas De Schampheleire a
écrit :

> I just want to mention that, although you seem to think that we
> shouldn't have created this thread in the first place, your and Eric's
> remarks have actually helped us in identifying the problem and
> increasing our understanding.
> 

Thread was fine IMHO, but should have mentioned the context of "non
standard linux-2.6 or net-next-2.6 trees"

I would like to mention NETIF_F_LLTX existence.

While being marked deprecated, it allows a lockless tx path :

The device transmit handler can then perform custom actions, including
its own locking if needed.

If the user wants a specific Qdisc setup (instead of a no qdisc one), it
still can.

veth, loopback, dummy, macvlan & bond devices use NETIF_F_LLTX for
example.

But ndo_start_xmit() must always be called with BH disabled.

  reply	other threads:[~2011-07-13  7:38 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-07-12  9:23 softirqs are invoked while bottom halves are masked (was: Re: [PATCH] [PATCH] Fix deadlock in af_packet while stressing raw ethernet socket interface) Thomas De Schampheleire
2011-07-12  9:23 ` Thomas De Schampheleire
2011-07-12 10:01 ` softirqs are invoked while bottom halves are masked David Miller
2011-07-12 10:01   ` David Miller
2011-07-12 10:10 ` softirqs are invoked while bottom halves are masked (was: Re: [PATCH] [PATCH] Fix deadlock in af_packet while stressing raw ethernet socket interface) Eric Dumazet
2011-07-12 10:10   ` Eric Dumazet
2011-07-12 12:03   ` Ronny Meeus
2011-07-12 12:03     ` Ronny Meeus
2011-07-12 12:08     ` softirqs are invoked while bottom halves are masked David Miller
2011-07-12 12:08       ` David Miller
2011-07-12 12:13       ` David Miller
2011-07-12 12:13         ` David Miller
2011-07-13  7:20         ` Thomas De Schampheleire
2011-07-13  7:20           ` Thomas De Schampheleire
2011-07-13  7:38           ` Eric Dumazet [this message]
2011-07-13  7:38             ` Eric Dumazet
2011-07-12 15:27     ` softirqs are invoked while bottom halves are masked (was: Re: [PATCH] [PATCH] Fix deadlock in af_packet while stressing raw ethernet socket interface) Eric Dumazet
2011-07-12 15:27       ` Eric Dumazet
2011-07-12 19:52       ` Ronny Meeus
2011-07-12 19:52         ` Ronny Meeus

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=1310542733.2634.17.camel@edumazet-laptop \
    --to=eric.dumazet@gmail.com \
    --cc=afleming@freescale.com \
    --cc=davem@davemloft.net \
    --cc=linuxppc-dev@ozlabs.org \
    --cc=netdev@vger.kernel.org \
    --cc=patrickdepinguin+linuxppc@gmail.com \
    --cc=ronny.meeus@gmail.com \
    /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.