[PATCH v4 2/3] ARM: mx28evk: add platform data for saif

[Sascha Hauer]

On Thu, Nov 10, 2011 at 09:23:18AM +0000, Dong Aisheng-B29396 wrote:
> > > +
> > > +static const struct mxs_saif_platform_data
> > > +			mx28evk_mxs_saif_pdata __initconst = {
> > > +	.init = mx28evk_mxs_saif_pinit,
> > > +	.get_master_id = mxs_get_saif_clk_master_id,
> > 
> > This looks *very* suspicious. .init sets the mux register
> > and .get_master_id reads the very same information back. Why not simply
> > put a bool is_master into platform data?
> 
> Originally it's used to avoid introduce a global variable to save clkmux
> Since we can directly read it from register.
> 
> If put a bool is_master into pdata, we are assuming there are only two saif
> Instances. As I was suggested by Wolfram before that how about if there are
> more than two saifs in the future SoCs?

I don't understand. A is_master variable in platform_data does not make
any assumption about the number of interfaces in the system.

> So the original saif driver design is following this rule that it only needs
> to know its master id.

If that's the design then put the master id into platform_data. This
information is purely static and the board knows it. No need to put a
function in platform_data. Something like this:

	struct saif_pdata {
		bool master_mode; /* if true use master mode */
		int master_id; /* id of the master if in slave mode */
	};

> If we decide to change here, I may also have a lot to change in saif driver.

Some lines in the probe code, not more.

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |