All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@us.ibm.com>
To: Josh Triplett <josht@vnet.ibm.com>
Cc: linux-kernel@vger.kernel.org,
	Dipkanar Sarma <dipankar@in.ibm.com>,
	Andrew Morton <akpm@osdl.org>, Samuel Ortiz <samuel@sortiz.org>
Subject: Re: [PATCH] irda: Fix RCU lock pairing on error path
Date: Thu, 29 Jun 2006 07:27:41 -0700	[thread overview]
Message-ID: <20060629142741.GA1294@us.ibm.com> (raw)
In-Reply-To: <1151542602.18723.19.camel@josh-work.beaverton.ibm.com>

On Wed, Jun 28, 2006 at 05:56:41PM -0700, Josh Triplett wrote:
> irlan_client_discovery_indication calls rcu_read_lock and rcu_read_unlock, but
> returns without unlocking in an error case.  Fix that by replacing the return
> with a goto so that the rcu_read_unlock always gets executed.

Good catch!!!  Looks good from an RCU viewpoint, in fact, looks absolutely
necessary to permit IRDA to work in debug mode, particularly in either
a CONFIG_PREEMPT or a -rt kernel.

One question below, but up to the maintainer.  ;-)

							Thanx, Paul

Acked-by: Paul E. McKenney <paulmck@us.ibm.com>
> Signed-off-by: Josh Triplett <josh@freedesktop.org>
> 
> ---
> 
>  net/irda/irlan/irlan_client.c |    3 ++-
>  1 files changed, 2 insertions(+), 1 deletions(-)
> 
> e20ab96944814489277c3bfd4e69133854ab01e9
> diff --git a/net/irda/irlan/irlan_client.c b/net/irda/irlan/irlan_client.c
> index f8e6cb0..5ce7d2e 100644
> --- a/net/irda/irlan/irlan_client.c
> +++ b/net/irda/irlan/irlan_client.c
> @@ -173,13 +173,14 @@ void irlan_client_discovery_indication(d
>  	rcu_read_lock();
>  	self = irlan_get_any();
>  	if (self) {
> -		IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
> +		IRDA_ASSERT(self->magic == IRLAN_MAGIC, goto out;);
>  
>  		IRDA_DEBUG(1, "%s(), Found instance (%08x)!\n", __FUNCTION__ ,
>  		      daddr);
>  		
>  		irlan_client_wakeup(self, saddr, daddr);
>  	}
> +out:

Should the above label instead be as follows?

	IRDA_ASSERT_LABEL(out:)

Just in case some variant of gcc, sparse, or whatever complains about
labels that don't have a corresponding goto (in CONFIG_IRDA_DEBUG=n
builds).

>  	rcu_read_unlock();
>  }
>  	
> 
> 

  reply	other threads:[~2006-06-29 14:27 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-06-29  0:56 [PATCH] irda: Fix RCU lock pairing on error path Josh Triplett
2006-06-29 14:27 ` Paul E. McKenney [this message]
2006-06-29 14:59   ` Samuel Ortiz
2006-06-30  6:55   ` [PATCH 1/2] [IrDA] " Samuel Ortiz
2006-06-30  6:55     ` Samuel Ortiz
2006-06-30  0:02     ` David Miller
2006-06-30  0:02       ` David Miller
2006-06-29 19:48 ` [PATCH] irda: " David 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=20060629142741.GA1294@us.ibm.com \
    --to=paulmck@us.ibm.com \
    --cc=akpm@osdl.org \
    --cc=dipankar@in.ibm.com \
    --cc=josht@vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=samuel@sortiz.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.