Messages in this thread | | | From | Dong Aisheng-B29396 <> | Subject | RE: [PATCH 1/1] gpio: introduce gpio_export_array to ease export for gpio arrays | Date | Wed, 16 Nov 2011 13:43:27 +0000 |
| |
> -----Original Message----- > From: linux-kernel-owner@vger.kernel.org [mailto:linux-kernel- > owner@vger.kernel.org] On Behalf Of Stephen Warren > Sent: Wednesday, November 16, 2011 2:37 AM > To: Dong Aisheng; linux-kernel@vger.kernel.org > Cc: grant.likely@secretlab.ca; arnd@arndb.de > Subject: RE: [PATCH 1/1] gpio: introduce gpio_export_array to ease export > for gpio arrays > > Dong Aisheng wrote at Tuesday, November 15, 2011 2:54 AM: > > Introduce gpio_export_array function to ease the export operation for > > gpio arrays. > > This can cooperate with the exist gpio_request_array function. > ... > > +/** > > + * gpio_export_array - export multiple GPIOs in a single call > > + * @array: array of the 'struct gpio' > > + * @num: how many GPIOs in the array > > + */ > > +int gpio_export_array(const struct gpio *array, size_t num, > > + bool direction_may_change) > > Should direction_may_change be added to "struct gpio" so it can vary per > pin? > I just thougt that after adding this member(or flag) into struct gpio, we may not need gpio_export_array function. It can be done in gpio_request_one by checking the flag just like wolfram's patch does (I just found his patch). It seems his patch is more light to implement this function.
> I might this new function to create its own parameter type instead of re- > using struct gpio, but I guess the main value here is re-using the data > passed to both request_array and export_array. > Agree. :-)
Regards Dong Aisheng
| |