[PATCH v5 1/4] ARM: add GPMI support for imx23/imx28

[Shawn Guo]

On Wed, Apr 13, 2011 at 03:51:52PM +0200, Uwe Kleine-König wrote:
> Hi Shawn,
> 
> On Wed, Apr 13, 2011 at 07:43:05PM +0800, Shawn Guo wrote:
> > On Wed, Apr 13, 2011 at 02:24:38PM +0800, Huang Shijie wrote:
> > > +	MX23_PAD_GPMI_D01__GPMI_D01 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_D02__GPMI_D02 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_D03__GPMI_D03 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_D04__GPMI_D04 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_D05__GPMI_D05 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_D06__GPMI_D06 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_D07__GPMI_D07 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_CLE__GPMI_CLE |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_ALE__GPMI_ALE |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_WPN__GPMI_WPN |
> > > +		(MXS_PAD_12MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_WRN__GPMI_WRN |
> > > +		(MXS_PAD_12MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_RDN__GPMI_RDN |
> > > +		(MXS_PAD_12MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_RDY0__GPMI_RDY0 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_RDY1__GPMI_RDY1 |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_CE0N__GPMI_CE0N |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +	MX23_PAD_GPMI_CE1N__GPMI_CE1N |
> > > +		(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
> > > +
> > 
> > mmc (ssp) has pin conflict with gpmi on both mx23evk and mx28evk.
> hmm,
> 
> 	 $ git grep GPMI linus/master -- arch/arm/mach-mxs/mach-mx23evk.c
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D08__LCD_D18 | MXS
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D09__LCD_D19 | MXS
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D10__LCD_D20 | MXS
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D11__LCD_D21 | MXS
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D12__LCD_D22 | MXS
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D13__LCD_D23 | MXS
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D08__SSP1_DATA4 |
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D09__SSP1_DATA5 |
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D10__SSP1_DATA6 |
> 	linus/master:arch/arm/mach-mxs/mach-mx23evk.c:  MX23_PAD_GPMI_D11__SSP1_DATA7 |
> 
> so MX23_PAD_GPMI_D08 - MX23_PAD_GPMI_D13 is used (D08-D11 twice). So the
> problem is already in linus/master. Is it that what you mean?
> 
Yeah, you are right.

> > How do you usually handle such pin conflicts on board?
> I don't understand this question. Maybe: deciding which pad mux to use?
> 
Yes, the question is how you decide which pad mux to use?  The first
thing I think of is to ask CONFIG options to help.  But I'm unsure
if it's the right way.

-- 
Regards,
Shawn