From: Brad Chapman <kakadu_croc@yahoo.com>
To: "David S. Miller" <davem@redhat.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: brlock_is_locked()?
Date: Wed, 22 Aug 2001 12:08:20 -0700 (PDT) [thread overview]
Message-ID: <20010822190820.57208.qmail@web10904.mail.yahoo.com> (raw)
In-Reply-To: <20010822.120051.25423285.davem@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 2022 bytes --]
Mr. Miller,
--- "David S. Miller" <davem@redhat.com> wrote:
> From: Brad Chapman <kakadu_croc@yahoo.com>
> Date: Wed, 22 Aug 2001 11:53:51 -0700 (PDT)
>
> It's not really a deficiency. Rusty apparently decided that in
> order to be SMP-compliant and to prevent Oopses, that the unregistration
> function should grab the brlock so that all the packets would pass through
> the protocol-handling functions.
>
> So arrange you code such that you aren't holding the netproto
> lock when you call the unregistration function.
>
> It is possible to shut down all references to whatever you
> are unregistering, safely drop the lock, then call the
> netfilter unregister routine.
I understand that. What I'm afraid of is someone who writes a third-party
protocol module for ip_conntrack (or ip6_conntrack) and tries to call the wrapper
with BR_NETPROTO_LOCK held. LISB, IMHO the primary difficulty is not protecting
the protocol linked-list from packets inside the conntrack; it's protecting the
protocol linked-list from packets in the netfilter stack. FWICT, it should
be almost impossible for packets to move around in conntrack with the rwlock
but not the brlock; but strange things can happen......
>
> (I checked the brlock code and didn't find any schedule()s; there's
> probably a reason for that).
>
> Ummm, this is SMP 101, you can't sleep with a lock held.
> The global kernel lock is special in this regard, but all
> other SMP locking primitives may not sleep.
Grrr....I read Rusty's Unreliable Guide to Kernel Locking (twice) and
still didn't remember that. Guess you have to schedule() yourself.
>
> Later,
> David S. Miller
> davem@redhat.com
Brad
=====
Brad Chapman
Permanent e-mail: kakadu_croc@yahoo.com
Current e-mail: kakadu@adelphia.net
Reply to the address I used in the message to you,
please!
__________________________________________________
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger
http://phonecard.yahoo.com/
next prev parent reply other threads:[~2001-08-22 19:08 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-08-22 13:04 brlock_is_locked()? Brad Chapman
2001-08-22 16:33 ` brlock_is_locked()? Ben LaHaise
2001-08-22 18:00 ` brlock_is_locked()? Brad Chapman
2001-08-22 18:03 ` brlock_is_locked()? David S. Miller
2001-08-22 18:17 ` brlock_is_locked()? Brad Chapman
2001-08-22 18:26 ` brlock_is_locked()? David S. Miller
2001-08-22 18:33 ` brlock_is_locked()? Brad Chapman
2001-08-22 18:47 ` brlock_is_locked()? David S. Miller
2001-08-22 18:53 ` brlock_is_locked()? Brad Chapman
2001-08-22 19:00 ` brlock_is_locked()? David S. Miller
2001-08-22 19:08 ` Brad Chapman [this message]
2001-08-24 6:49 ` brlock_is_locked()? Jens Axboe
2001-08-24 14:11 ` brlock_is_locked()? Brad Chapman
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=20010822190820.57208.qmail@web10904.mail.yahoo.com \
--to=kakadu_croc@yahoo.com \
--cc=davem@redhat.com \
--cc=linux-kernel@vger.kernel.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