linux-kernel - Re: [PATCH v3 2/4] iio: mpl3115: use guards from cleanup.h

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: <20250928094148.yid75l5jwywpvfei@antoni-VivoBook-ASUSLaptop-X512FAY-K512FA>
Date: Sun, 28 Sep 2025 11:41:48 +0200
From: Antoni Pokusinski <apokusinski01@...il.com>
To: Jonathan Cameron <jic23@...nel.org>, dlechner@...libre.com,
	nuno.sa@...log.com, andy@...nel.org, robh@...nel.org,
	krzk+dt@...nel.org
Cc: conor+dt@...nel.org, linux-kernel@...r.kernel.org,
	devicetree@...r.kernel.org, linux-iio@...r.kernel.org,
	linux@...ck-us.net, rodrigo.gobbi.7@...il.com,
	naresh.solanki@...ements.com, michal.simek@....com,
	grantpeltier93@...il.com, farouk.bouabid@...rry.de,
	marcelo.schmitt1@...il.com
Subject: Re: [PATCH v3 2/4] iio: mpl3115: use guards from cleanup.h

On Sat, Sep 27, 2025 at 05:36:21PM +0100, Jonathan Cameron wrote:
> On Sat, 27 Sep 2025 00:01:48 +0200
> Antoni Pokusinski <apokusinski01@...il.com> wrote:
> 
> > Include linux/cleanup.h and use the scoped_guard() to simplify the code.
> See below. I'm not sure this is in general a good idea in this driver, but
> see the comments below.  I think more traditional factoring out of the code
> under the lock into a helper function should be the main change here.
> That might or might not make sense combined with a scoped_guard().
> 
> 
> > 
> > Signed-off-by: Antoni Pokusinski <apokusinski01@...il.com>
> > ---
> >  drivers/iio/pressure/mpl3115.c | 42 +++++++++++++++-------------------
> >  1 file changed, 19 insertions(+), 23 deletions(-)
> > 
> > diff --git a/drivers/iio/pressure/mpl3115.c b/drivers/iio/pressure/mpl3115.c
> > index 579da60ef441..80af672f65c6 100644
> > --- a/drivers/iio/pressure/mpl3115.c
> > +++ b/drivers/iio/pressure/mpl3115.c
> > @@ -10,14 +10,16 @@
> >   * interrupts, user offset correction, raw mode
> >   */
> >  
> > -#include <linux/module.h>
> > +#include <linux/cleanup.h>
> > +#include <linux/delay.h>
> >  #include <linux/i2c.h>
> > +#include <linux/module.h>
> > +
> >  #include <linux/iio/iio.h>
> >  #include <linux/iio/sysfs.h>
> >  #include <linux/iio/trigger_consumer.h>
> >  #include <linux/iio/buffer.h>
> >  #include <linux/iio/triggered_buffer.h>
> > -#include <linux/delay.h>
> >  
> >  #define MPL3115_STATUS 0x00
> >  #define MPL3115_OUT_PRESS 0x01 /* MSB first, 20 bit */
> > @@ -163,32 +165,26 @@ static irqreturn_t mpl3115_trigger_handler(int irq, void *p)
> >  	u8 buffer[16] __aligned(8) = { };
> >  	int ret, pos = 0;
> >  
> > -	mutex_lock(&data->lock);
> > -	ret = mpl3115_request(data);
> > -	if (ret < 0) {
> > -		mutex_unlock(&data->lock);
> > -		goto done;
> > -	}
> > -
> > -	if (test_bit(0, indio_dev->active_scan_mask)) {
> > -		ret = i2c_smbus_read_i2c_block_data(data->client,
> > -			MPL3115_OUT_PRESS, 3, &buffer[pos]);
> > -		if (ret < 0) {
> > -			mutex_unlock(&data->lock);
> > +	scoped_guard(mutex, &data->lock) {
> > +		ret = mpl3115_request(data);
> > +		if (ret < 0)
> >  			goto done;
> Read the guidance in cleanup.h.  Whilst what you have here is actually not
> a bug, it is considered fragile to combine gotos and scoped cleanup in a function.
> Sometimes that means that if we are using guards() we need to also duplicate
> some error handling.
> 
> So, the way to avoid that is to factor out the stuff under the goto to a helper
> function.  That function than then return directly on errors like this.
> 
> Looks something like
> 
> 	scoped_guard(mutex, &data->lock) {
> 		ret = mpl3115_fill_buffer(data, buffer);
> 		if (ret) {
> 			iio_trigger_notify_done(indio_dev->trig);
> 			return IRQ_HANDLED;
> 		}
> 	}
> 
> 	iio_push_to_buffers_with_ts...
> 	iio_trigger_notify_done(indio_dev->trig);
> 	return IRQ_HANDLED;
> 
> 
> However, it is also worth keeping in mind that sometimes scoped cleanup
> of which guards are a special case is not the right solution for a whole
> driver. I'm not sure if it is worth while in this case, but try the approach
> mentioned above and see how it looks.
> 
> Alternative would still be to factor out the helper, but instead just have
> 	mutex_lock(&data->lock);
> 	ret = mpl3115_fill_buffer(data, buffer);
> 	mutex_unlock(&data->lock);
> 	if (ret)
> 		goto...
> 
> 
> Jonathan
>
Thanks for the explanation, both approaches look quite neat to me.
However, if we use scoped_guard() then the iio_trigger_notify_done and
return IRQ_HANDLED are duplicated, so I'd lean slightly towards the
mutex_lock/mutex_unlock solution.

> > +
> > +		if (test_bit(0, indio_dev->active_scan_mask)) {
> > +			ret = i2c_smbus_read_i2c_block_data(data->client,
> > +				MPL3115_OUT_PRESS, 3, &buffer[pos]);
> > +			if (ret < 0)
> > +				goto done;
> > +			pos += 4;
> >  		}
> > -		pos += 4;
> > -	}
> >  
> > -	if (test_bit(1, indio_dev->active_scan_mask)) {
> > -		ret = i2c_smbus_read_i2c_block_data(data->client,
> > -			MPL3115_OUT_TEMP, 2, &buffer[pos]);
> > -		if (ret < 0) {
> > -			mutex_unlock(&data->lock);
> > -			goto done;
> > +		if (test_bit(1, indio_dev->active_scan_mask)) {
> > +			ret = i2c_smbus_read_i2c_block_data(data->client,
> > +				MPL3115_OUT_TEMP, 2, &buffer[pos]);
> > +			if (ret < 0)
> > +				goto done;
> >  		}
> >  	}
> > -	mutex_unlock(&data->lock);
> >  
> >  	iio_push_to_buffers_with_ts(indio_dev, buffer, sizeof(buffer),
> >  				    iio_get_time_ns(indio_dev));
>