linux-kernel - Re: [PATCH 1/6] media: ar0521: Refactor ar0521_power

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: <ce0ed949-5e57-4193-8190-9e4aa9cb62be@stanley.mountain>
Date: Tue, 13 Aug 2024 15:54:35 +0300
From: Dan Carpenter <dan.carpenter@...aro.org>
To: Ricardo Ribalda <ribalda@...omium.org>
Cc: Sakari Ailus <sakari.ailus@...ux.intel.com>,
	Krzysztof Hałasa <khalasa@...p.pl>,
	Mauro Carvalho Chehab <mchehab@...nel.org>,
	Sylwester Nawrocki <s.nawrocki@...sung.com>,
	Andrzej Hajda <andrzej.hajda@...el.com>,
	Neil Armstrong <neil.armstrong@...aro.org>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Kevin Hilman <khilman@...libre.com>,
	Jerome Brunet <jbrunet@...libre.com>,
	Martin Blumenstingl <martin.blumenstingl@...glemail.com>,
	Hans Verkuil <hverkuil-cisco@...all.nl>,
	linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-amlogic@...ts.infradead.org, linux-staging@...ts.linux.dev,
	linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH 1/6] media: ar0521: Refactor ar0521_power_off()

On Tue, Aug 13, 2024 at 12:13:48PM +0000, Ricardo Ribalda wrote:
> Factor out all the power off logic, except the clk_disable_unprepare(),
> to a new function __ar0521_power_off().
> 
> This allows ar0521_power_on() to explicitly clean-out the clock during
> the error-path.
> 
> The following smatch warning is fixed:
> drivers/media/i2c/ar0521.c:912 ar0521_power_on() warn: 'sensor->extclk' from clk_prepare_enable() not released on lines: 912.
> 
> Signed-off-by: Ricardo Ribalda <ribalda@...omium.org>
> ---

It's better to just ignore false positives...  The problem here is that
Smatch can't track that to_ar0521_dev(dev_get_drvdata(dev))->sensor is the same
as sensor.  What I could do is say that "this frees *something* unknown" so
let's silence the warning."

The problem is that check_unwind.c is not very granular.  It just marks things
as allocated and freed.  I could make it more granular so the free and the
alloc have to match.  Or we could match based on the type.  This frees a
"struct ar0521_dev" so mark all those as freed in the caller.

>  drivers/media/i2c/ar0521.c | 15 +++++++++++++--
>  1 file changed, 13 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/media/i2c/ar0521.c b/drivers/media/i2c/ar0521.c
> index 09331cf95c62..2c528db31ba6 100644
> --- a/drivers/media/i2c/ar0521.c
> +++ b/drivers/media/i2c/ar0521.c
> @@ -835,7 +835,7 @@ static const struct initial_reg {
>  	     be(0x0707)), /* 3F44: couple k factor 2 */
>  };
>  
> -static int ar0521_power_off(struct device *dev)
> +static void __ar0521_power_off(struct device *dev)
>  {
>  	struct v4l2_subdev *sd = dev_get_drvdata(dev);
>  	struct ar0521_dev *sensor = to_ar0521_dev(sd);
> @@ -850,6 +850,16 @@ static int ar0521_power_off(struct device *dev)
>  		if (sensor->supplies[i])
>  			regulator_disable(sensor->supplies[i]);
>  	}
> +}
> +
> +static int ar0521_power_off(struct device *dev)
> +{
> +	struct v4l2_subdev *sd = dev_get_drvdata(dev);
> +	struct ar0521_dev *sensor = to_ar0521_dev(sd);
> +
> +	clk_disable_unprepare(sensor->extclk);
> +	__ar0521_power_off(dev);

You had intended to remove the clk_disable_unprepare() from __ar0521_power_off()
but forgot so these are double unprepares.

regards,
dan carpenter

> +
>  	return 0;
>  }
>  
> @@ -908,7 +918,8 @@ static int ar0521_power_on(struct device *dev)
>  
>  	return 0;
>  off:
> -	ar0521_power_off(dev);
> +	clk_disable_unprepare(sensor->extclk);
> +	__ar0521_power_off(dev);
>  	return ret;
>  }
>  
> 
> -- 
> 2.46.0.76.ge559c4bf1a-goog
>