Re: [PATCH v2 3/7] test_firmware: add partial read support for request_firmware_into_buf

[Scott Branden]

Hi Dan,

Thanks for your review and valuable comments.
Will have to investigate fully and correct anything wrong.

On 2020-02-20 12:42 a.m., Dan Carpenter wrote:
> On Wed, Feb 19, 2020 at 04:48:21PM -0800, Scott Branden wrote:
> > +static int test_dev_config_update_size_t(const char *buf,
> > +					 size_t size,
> > +					 size_t *cfg)
> > +{
> > +	int ret;
> > +	long new;
> > +
> > +	ret = kstrtol(buf, 10, &new);
> > +	if (ret)
> > +		return ret;
> > +
> > +	if (new > SIZE_MAX)
> This "new" variable is long and SIZE_MAX is ULONG_MAX so the condition
> can't be true.
>
> > +		return -EINVAL;
> > +
> > +	mutex_lock(&test_fw_mutex);
> > +	*(size_t *)cfg = new;
> > +	mutex_unlock(&test_fw_mutex);
> > +
> > +	/* Always return full write size even if we didn't consume all */
> > +	return size;
> > +}
> > +
> > +static ssize_t test_dev_config_show_size_t(char *buf, int cfg)
> > +{
> > +	size_t val;
> > +
> > +	mutex_lock(&test_fw_mutex);
> > +	val = cfg;
> > +	mutex_unlock(&test_fw_mutex);
> Both val and cfg are stack variables so there is no need for locking.
> Probably you meant to pass a pointer to cfg?
>
> > +
> > +	return snprintf(buf, PAGE_SIZE, "%zu\n", val);
> > +}
> > +
> >   static ssize_t test_dev_config_show_int(char *buf, int cfg)
> >   {
> >   	int val;
> regards,
> dan carpenter