linux-kernel - Re: [PATCH] leds-lp3944: properly handle lp3944_configure fail in lp3944

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]

Date:	Mon, 17 May 2010 11:24:57 +0200
From:	Wolfram Sang <w.sang@...gutronix.de>
To:	Antonio Ospite <ospite@...denti.unina.it>
Cc:	Axel Lin <axel.lin@...il.com>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	Richard Purdie <rpurdie@...ux.intel.com>
Subject: Re: [PATCH] leds-lp3944: properly handle lp3944_configure fail in
	lp3944_probe

On Mon, May 17, 2010 at 11:15:13AM +0200, Antonio Ospite wrote:
> On Mon, 17 May 2010 09:34:02 +0800
> Axel Lin <axel.lin@...il.com> wrote:
> 
> > In current implementation, lp3944_probe return 0 even if lp3944_configure fail.
> > Therefore, led_classdev_unregister will be executed twice ( in error handling of lp3944_configure and lp3944_remove ).
> > This patch properly handles lp3944_configure fail in lp3944_probe.
> >
> 
> Hi Axel, thanks for the fix, I agree it's needed.
> There are some minor comments inlined below.
> 
> Plus, when possible, I prefer commit messages to be wrapped to 72/80
> characters per line. Please, consider this if you are sending a v2.
> 
> Ah, may I ask where you are using this driver? Just curious.
> 
> > Signed-off-by: Axel Lin <axel.lin@...il.com>
> > ---
> >  drivers/leds/leds-lp3944.c |   10 +++++++++-
> >  1 files changed, 9 insertions(+), 1 deletions(-)
> > 
> > diff --git a/drivers/leds/leds-lp3944.c b/drivers/leds/leds-lp3944.c
> > index 8d5ecce..03a24d0 100644
> > --- a/drivers/leds/leds-lp3944.c
> > +++ b/drivers/leds/leds-lp3944.c
> > @@ -379,6 +379,7 @@ static int __devinit lp3944_probe(struct i2c_client *client,
> >  {
> >  	struct lp3944_platform_data *lp3944_pdata = client->dev.platform_data;
> >  	struct lp3944_data *data;
> > +	int err;
> >  
> >  	if (lp3944_pdata == NULL) {
> >  		dev_err(&client->dev, "no platform data\n");
> > @@ -403,8 +404,15 @@ static int __devinit lp3944_probe(struct i2c_client *client,
> >  
> >  	dev_info(&client->dev, "lp3944 enabled\n");
> >  
> > -	lp3944_configure(client, data, lp3944_pdata);
> > +	err = lp3944_configure(client, data, lp3944_pdata);
> > +	if (err < 0)
> > +		goto err_configure;
> > +
> 
> The dev_info(&client->dev, "lp3944 enabled\n"); could now go right
> here, before the return 0, so we don't report the driver as enabled
> even in the case its probe fails.
> 
> >  	return 0;
> > +
> > +err_configure:
> > +	kfree(data);
> 
> add i2c_set_clientdata(client, NULL) here just like in lp3944_remove()

Just to note:
This is not needed anymore, the i2c-core does this meanwhile for the drivers.
I am preparing a cleanup patch fixing this in all drivers for -rc1.

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

Download attachment "signature.asc" of type "application/pgp-signature" (198 bytes)