Hi,
On Fri, Apr 25, 2014 at 12:03:31AM +0900, Chanwoo Choi wrote:
> On Thu, Apr 24, 2014 at 11:44 PM, Felipe Balbi <ba...@ti.com> wrote:
> > On Thu, Apr 24, 2014 at 09:34:50PM +0900, Chanwoo Choi wrote:
> >> This patch add device managed devm_extcon_dev_{allocate,free} to
> >> automatically
> >> free the memory of extcon_dev structure without handling free operation.
> >>
> >> Signed-off-by: Chanwoo Choi <cw00.c...@samsung.com>
> >> ---
> >> drivers/extcon/extcon-class.c | 70
> >> +++++++++++++++++++++++++++++++++++--------
> >> include/linux/extcon.h | 11 +++++++
> >> 2 files changed, 69 insertions(+), 12 deletions(-)
> >>
> >> diff --git a/drivers/extcon/extcon-class.c b/drivers/extcon/extcon-class.c
> >> index bec66d4..f369168 100644
> >> --- a/drivers/extcon/extcon-class.c
> >> +++ b/drivers/extcon/extcon-class.c
> >> @@ -604,6 +604,64 @@ void extcon_dev_free(struct extcon_dev *edev)
> >> }
> >> EXPORT_SYMBOL_GPL(extcon_dev_free);
> >>
> >> +static int devm_extcon_dev_match(struct device *dev, void *res, void
> >> *data)
> >> +{
> >> + struct extcon_dev **r = res;
> >> +
> >> + if (!r || !*r) {
> >> + WARN_ON(!r || !*r);
> >> + return 0;
> >> + }
> >> +
> >> + return *r == data;
> >> +}
> >> +
> >> +static void devm_extcon_dev_release(struct device *dev, void *res)
> >> +{
> >> + extcon_dev_free(*(struct extcon_dev **)res);
> >> +}
> >> +
> >> +/**
> >> + * devm_extcon_dev_allocate - Allocate managed extcon device
> >> + * @dev: device owning the extcon device being created
> >> + * @supported_cable: Array of supported cable names ending with NULL.
> >> + * If supported_cable is NULL, cable name related APIs
> >> + * are disabled.
> >> + *
> >> + * This function manages automatically the memory of extcon device using
> >> device
> >> + * resource management and simplify the control of freeing the memory of
> >> extcon
> >> + * device.
> >> + *
> >> + * Returns the pointer memory of allocated extcon_dev if success or NULL
> >> if fail
> >> + */
> >> +struct extcon_dev *devm_extcon_dev_allocate(struct device *dev,
> >> + const char **supported_cable)
> >> +{
> >> + struct extcon_dev **ptr, *edev;
> >> +
> >> + ptr = devres_alloc(devm_extcon_dev_release, sizeof(*ptr),
> >> GFP_KERNEL);
> >> + if (!ptr)
> >> + return NULL;
> >
> > if this fails you return NULL;
> >
> >> +
> >> + edev = extcon_dev_allocate(supported_cable);
> >
> > if this fails, you return -ENOMEM. Not very nice, if you ask me.
>
> I'll fix it as following:
>
> If both extcon_dev_allocate() and devm_excon_dev_allocate() failed to
> allocate memory,
> two functions would return NULL pointer instead of ERR_PTR(-ENOMEM).your call, I'd rather see them both returning ERR_PTR(-ENOMEM), but no strong feelings. cheers -- balbi
signature.asc
Description: Digital signature
