On Thursday 21 June 2012 12:50 PM, Tony Lindgren wrote:

* Shubhrajyoti [off-list ref] [120621 00:08]:

quoted

On Wednesday 20 June 2012 03:59 PM, Tony Lindgren wrote:

quoted

See the comments regarding driver specific resets in hwmod code.

you mean omap_hwmod.c

quoted

The way to set this up is to have a shared inline function in
i2c-omap.h that both the driver and hwmod code can use.

hwmod reset function uses oh (omap_hwmod).

How could the driver also pass oh ?
Could we keep a local copy in driver data?

quoted

Eventually hwmod code will do the reset only in late initcall
if no driver is loaded for the device in question.

There's no need for the driver to know anything about oh.
The driver just needs to know the iobase.

I will rework to make the hwmod and driver use the same function.
In the meanwhile I will send a minimal/ remaining cleanups/ fixes so that
it might get some time to bake in the next branch.

Tony