On Mon, Oct 19, 2009 at 09:39:37AM -0700, Mark A. Greer wrote:
> On Mon, Oct 19, 2009 at 03:07:52PM +0100, Russell King - ARM Linux wrote:
> > On Fri, Oct 16, 2009 at 12:09:20PM -0700, Kevin Hilman wrote:

> > > +static int da830_evm_mmc_get_ro(int index)
> > > +{
> > > + int val, status, gpio_num = 33;
> > > +
> > > + status = gpio_request(gpio_num, "MMC WP\n");
> > > + if (status < 0) {
> > > +         pr_warning("%s can not open GPIO %d\n", __func__, gpio_num);
> > > +         return 0;
> > > + }
> > > + gpio_direction_input(gpio_num);
> > > + val = gpio_get_value(gpio_num);
> > > + gpio_free(gpio_num);
> > 
> > Should this really be requesting the gpio every time it wants to be
> > read?  This approach sounds wrong.  Why not just request the GPIO in
> > the platform initialisation code, setting its direction, and merely
> > reading it in this function?
> 
> This routine doesn't get called on every read.  AFAICT, its only called
> by mmc_sd_init_card() which is called during init and resume.  That seems
> like the logical time to check if the card is set to read-only since cards
> can be inserted and removed while the system is running.

Nevermind, I see what you're saying now.  Yes, that would work.

Mark
--

_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to