linux-kernel - Re: [PATCH 1/5] iio: buffer-dmaengine: add dev-managed calls for buffer alloc/free

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:   Tue, 25 Feb 2020 13:33:59 +0000
From:   "Ardelean, Alexandru" <alexandru.Ardelean@...log.com>
To:     "jic23@...nel.org" <jic23@...nel.org>
CC:     "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-iio@...r.kernel.org" <linux-iio@...r.kernel.org>,
        "robh+dt@...nel.org" <robh+dt@...nel.org>
Subject: Re: [PATCH 1/5] iio: buffer-dmaengine: add dev-managed calls for
 buffer alloc/free

On Fri, 2020-02-21 at 12:21 +0000, Jonathan Cameron wrote:
> [External]
> 
> On Thu, 20 Feb 2020 17:03:13 +0200
> Alexandru Ardelean <alexandru.ardelean@...log.com> wrote:
> 
> > Currently, when using a 'iio_dmaengine_buffer_alloc()', an matching call to
> > 'iio_dmaengine_buffer_free()' must be made.
> > 
> > With this change, this can be avoided by using
> > 'devm_iio_dmaengine_buffer_alloc()'. The buffer will get free'd via the
> > device's devres handling.
> > 
> > Signed-off-by: Alexandru Ardelean <alexandru.ardelean@...log.com>
> > ---
> >  .../buffer/industrialio-buffer-dmaengine.c    | 70 +++++++++++++++++++
> >  include/linux/iio/buffer-dmaengine.h          |  5 ++
> >  2 files changed, 75 insertions(+)
> > 
> > diff --git a/drivers/iio/buffer/industrialio-buffer-dmaengine.c
> > b/drivers/iio/buffer/industrialio-buffer-dmaengine.c
> > index b129693af0fd..eff89037e3f5 100644
> > --- a/drivers/iio/buffer/industrialio-buffer-dmaengine.c
> > +++ b/drivers/iio/buffer/industrialio-buffer-dmaengine.c
> > @@ -229,6 +229,76 @@ void iio_dmaengine_buffer_free(struct iio_buffer
> > *buffer)
> >  }
> >  EXPORT_SYMBOL_GPL(iio_dmaengine_buffer_free);
> >  
> > +static void __devm_iio_dmaengine_buffer_free(struct device *dev, void *res)
> > +{
> > +	iio_dmaengine_buffer_free(*(struct iio_buffer **)res);
> > +}
> > +
> > +/**
> > + * devm_iio_dmaengine_buffer_alloc() - Resource-managed
> > iio_dmaengine_buffer_alloc()
> > + * @dev: Parent device for the buffer
> > + * @channel: DMA channel name, typically "rx".
> > + *
> > + * This allocates a new IIO buffer which internally uses the DMAengine
> > framework
> > + * to perform its transfers. The parent device will be used to request the
> > DMA
> > + * channel.
> > + *
> > + * Once done using the buffer iio_dmaengine_buffer_free() should be used to
> > + * release it.
> 
> Umm.  It really shouldn't!

Yeah.
Copy+paste.
My bad.


> > + */
> > +struct iio_buffer *devm_iio_dmaengine_buffer_alloc(struct device *dev,
> > +	const char *channel)
> > +{
> > +	struct iio_buffer **bufferp, *buffer;
> > +
> > +	bufferp = devres_alloc(__devm_iio_dmaengine_buffer_free,
> > +			       sizeof(*bufferp), GFP_KERNEL);
> > +	if (!bufferp)
> > +		return ERR_PTR(-ENOMEM);
> > +
> > +	buffer = iio_dmaengine_buffer_alloc(dev, channel);
> > +	if (!IS_ERR(buffer)) {
> > +		*bufferp = buffer;
> > +		devres_add(dev, bufferp);
> 
> From a flow point of view I'd prefer.
> 
> 	if (IS_ERR(buffer) {
> 		devres_free(buferp)
> 		return buffer;
> 	}
> 
> 	*bufferp = buffer;
> 	devres_add(dev, bufferp);
> 
> 	return buffer;
> 
> 
> > +	} else {
> > +		devres_free(bufferp);
> > +	}
> > +
> > +	return buffer;
> > +}
> > +EXPORT_SYMBOL_GPL(devm_iio_dmaengine_buffer_alloc);
> > +
> > +static int devm_iio_dmaengine_buffer_match(struct device *dev, void *res,
> > +	void *data)
> > +{
> > +	struct iio_buffer **r = res;
> > +
> > +	if (!r || !*r) {
> > +		WARN_ON(!r || !*r);
> > +		return 0;
> > +	}
> > +
> > +	return *r == data;
> > +}
> > +
> > +/**
> > + * devm_iio_dmaengine_buffer_free - iio_dmaengine_buffer_free
> > + * @dev: Device this iio_buffer belongs to
> > + * @buffer: The iio_buffer associated with the device
> > + *
> > + * Free buffer allocated with devm_iio_dmaengine_buffer_alloc().
> > + */
> > +void devm_iio_dmaengine_buffer_free(struct device *dev,
> > +	struct iio_buffer *buffer)
> > +{
> > +	int rc;
> > +
> > +	rc = devres_release(dev, __devm_iio_dmaengine_buffer_free,
> > +			    devm_iio_dmaengine_buffer_match, buffer);
> > +	WARN_ON(rc);
> > +}
> > +EXPORT_SYMBOL_GPL(devm_iio_dmaengine_buffer_free);

Should I remove devm_iio_dmaengine_buffer_free() ?
There was a comment on the AXI ADC that may apply here as well, about maybe
removing it.

> > +
> >  MODULE_AUTHOR("Lars-Peter Clausen <lars@...afoo.de>");
> >  MODULE_DESCRIPTION("DMA buffer for the IIO framework");
> >  MODULE_LICENSE("GPL");
> > diff --git a/include/linux/iio/buffer-dmaengine.h
> > b/include/linux/iio/buffer-dmaengine.h
> > index b3a57444a886..8dcd973d76c1 100644
> > --- a/include/linux/iio/buffer-dmaengine.h
> > +++ b/include/linux/iio/buffer-dmaengine.h
> > @@ -14,4 +14,9 @@ struct iio_buffer *iio_dmaengine_buffer_alloc(struct
> > device *dev,
> >  	const char *channel);
> >  void iio_dmaengine_buffer_free(struct iio_buffer *buffer);
> >  
> > +struct iio_buffer *devm_iio_dmaengine_buffer_alloc(struct device *dev,
> > +	const char *channel);
> > +void devm_iio_dmaengine_buffer_free(struct device *dev,
> > +	struct iio_buffer *buffer);
> Please align parameters with opening bracket where possible.
> 

ack

> Thanks,
> 
> Jonathan
> 
> > +
> >  #endif