linux-kernel - Re: [PATCH 07/11] mfd: menelaus: use devm_request_irq() and devm

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

Message-ID: <20130220081014.GA4268@core.coreip.homeip.net>
Date:	Wed, 20 Feb 2013 00:10:14 -0800
From:	Dmitry Torokhov <dmitry.torokhov@...il.com>
To:	Jingoo Han <jg1.han@...sung.com>
Cc:	'Samuel Ortiz' <sameo@...ux.intel.com>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 07/11] mfd: menelaus: use devm_request_irq() and
 devm_kzalloc()

On Wed, Feb 20, 2013 at 05:05:10PM +0900, Jingoo Han wrote:
> On Wednesday, February 20, 2013 4:31 PM, Dmitry Torokhov wrote:
> > 
> > Hi Jongoo,
> > 
> > On Wed, Feb 20, 2013 at 03:12:38PM +0900, Jingoo Han wrote:
> > > Use devm_request_irq() and devm_kzalloc() to make cleanup paths
> > > more simple.
> > >
> > 
> > ...
> > 
> > > @@ -1269,9 +1266,7 @@ static int __exit menelaus_remove(struct i2c_client *client)
> > >  {
> > >  	struct menelaus_chip	*menelaus = i2c_get_clientdata(client);
> > >
> > > -	free_irq(client->irq, menelaus);
> > >  	flush_work(&menelaus->work);
> > > -	kfree(menelaus);
> > >  	the_menelaus = NULL;
> > >  	return 0;
> > 
> > This conversion is certainly wrong - you really want to disable IRQ and
> > then wait for the scheduled work to finish before freeing memory. Here
> > you flush work but nothing stops IRQ from firing and scheduling that
> > work again.
> 
> Yes, you're right. 
> I will use devm_free_irq() before flush_work().

Why change it at all if you have to call it manually in both error
unwinding and menelaus_remove() cases?

BTW, that __exit markup on menelaus_remove() is surprising... I am
pretty sure it can be unbound via sysfs and so there will be a nasty
oops.

-- 
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/