[RFC PATCH v5 0/1] drivers: mfd: Versatile Express SPC support

Nicolas Pitre nicolas.pitre at linaro.org
Wed Jul 17 11:57:55 EDT 2013


On Wed, 17 Jul 2013, Pawel Moll wrote:

> On Wed, 2013-07-17 at 15:16 +0100, Nicolas Pitre wrote:
> > On Wed, 17 Jul 2013, Pawel Moll wrote:
> > 
> > > On Wed, 2013-07-17 at 13:33 +0100, Nicolas Pitre wrote:
> > > > If this is really miscelaneous code that really doesn't fit 
> > > > anywhere else, it should rather go into drivers/misc/ as a last resort.
> > > 
> > > Interestingly enough drivers/misc was my first choice for all the
> > > vexpress stuff, but it wasn't received well...
> > > 
> > > Anyway, the SPC driver as it is now seem to be a "power management
> > > system driver". Maybe a relevant directory would be in place? Wouldn't
> > > PSCI belong there as well? (there are two psci.c files in arch/arm and
> > > arch/arm64, surprisingly similar ones ;-)
> > > 
> > > The bottom line is: today it is not an MFD driver.
> > 
> > But we know it will, right?  So better  save some churn by storing the 
> > initial code where it would end up anyway once complete.
> 
> Not in that form, no. The code living in mfd will just register
> mfd_cells while "functional" parts are going to live elsewhere. This is
> how I understand what Samuel asked me to do and that's what is happening
> to vexpress-sysreg now.

A drivers/pm/ or drivers/power/ could be created, but that implies sort 
of a more defined subsystem with a common API and the SPC code doesn't 
fit that as it is only providing services to actual drivers on top of 
it.

The sanest location at this point might simply be 
drivers/platform/vexpress_spc.c or drivers/platform/vexpress/spc.c 
depending on whether or not more such driver glue is expected in the 
vexpress future.  No point putting "arm" in the path, especially if this 
is later reused on arm64.


Nicolas



More information about the linux-arm-kernel mailing list