On Mon, May 23, 2011 at 03:53:43PM -0700, Randy Dunlap wrote:
On Tue, 24 May 2011 06:47:28 +0800 Mark Brown wrote:
case, please talk to the architecture maintainers about this - it's an issue in the architecture GPIO support (or lack thereof) rather than a driver problem.
except that a driver should not assume that defines like GPIOF_OUT_INIT_LOW are always available.
No, really we should. The GPIO APIs are stubbed out when not in use for a very good reason, think about the usability here. The goal here isn't to litter the code with ifdefs - if architectures aren't able to keep up with API changes they should convert to using gpiolib so this stuff happens automatically (indeed, I can't think of any good reason for an architecture to not be using gpiolib at this point).
Also adding Dmitry who submitted the driver - Randy, please try to remember to CC relevant people.
Which driver did Dmitry submit? how would I know that? I don't download every linux-next git tree -- just linux-next tarballs.
I *strongly* suggest looking at git if you want to find relevant people to mail; the internal documentation in the code really isn't a terribly useful guide, the authors listed in the code often bear no relation to who's actually working on it at the current time.
and wm8915.c says: MODULE_AUTHOR("Mark Brown broonie@opensource.wolfsonmicro.com");
You've clearly not looked at MAINTAINERS for this one.