netdev - Re: Killing sk->sk_callback_lock (was Re: [PATCH] net/core/sock.c remove extra wakeup)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <4858E838020000C7000397CF@lucius.provo.novell.com>
Date:	Wed, 18 Jun 2008 10:49:28 -0600
From:	"Patrick Mullaney" <pmullaney@...ell.com>
To:	<herbert@...dor.apana.org.au>
Cc:	<davem@...emloft.net>, "Gregory Haskins" <GHaskins@...ell.com>,
	<chuck.lever@...cle.com>, <netdev@...r.kernel.org>
Subject: Re: Killing sk->sk_callback_lock (was Re: [PATCH]
	net/core/sock.c remove extra wakeup)

On Wed, 2008-06-18 at 09:33 +1000, Herbert Xu wrote:

> This is buggy.  You're relying on the caller to not call this
> on the same socket concurrently.  While this may be true for
> UDP right now (which is a less than ideal situation in itself),
> it isn't true in general.  In particular, raw sockets still
> allow parallel writes to the same socket.
> 
Point taken, I wasn't aware of the raw sockets usage. In that case,
there would be a race introduced that can't be solved with bit flags.
As a side note, it also seems the setting of the NOSPACE flag is
useless here as well.
> So forget about the flags and just do the two queues as you proposed
> yourself :)
> 
I will give it a shot - I still have reservations about changing the
socket structures - at first glance, it seems like sock and socket will
have to change.
> Cheers,

Thanks.
Pat


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html