[PATCH] ARM: OMAP2+: Fix populating the hwmod data from device tree

[tony@atomide.com (Tony Lindgren)]

* Paul Walmsley <paul@pwsan.com> [131205 16:09]:
> On Thu, 5 Dec 2013, Tony Lindgren wrote:
> 
> > * Tony Lindgren <tony@atomide.com> [131121 12:49]:
> > > * Tony Lindgren <tony@atomide.com> [131120 17:46]:
> > > > * Tony Lindgren <tony@atomide.com> [131120 16:06]:
> > > > > 
> > > > > They at least had interrupts listed looking at commit 3b9b10. Probably
> > > > > the thing to do for now is to revert those changes, and see if we can
> > > > > just remove the L3 entries from the .dtsi files.
> > > > 
> > > > Actually the patch I posted should be able to handle also the
> > > > ti,hwmods = "l3_main_1", "l3_main_2", "l3_main_3" property in omap4.dtsi,
> > > > but we're not currently parsing that as we only look at the children
> > > > and not the properties of the OCP bus. Should be fixable, will take a look
> > > > tomorrow if this approach makes sense to you.
> > > 
> > > OK this one seems to do the right thing for me.
> > 
> > No comments? I'll queue the patch below to the fixes, please yell
> > if you see any issues with that.
> 
> Looks reasonable to me based on a quick glance:
> 
> Acked-by: Paul Walmsley <paul@pwsan.com>
> 
> Have not tested it though.

OK thanks for looking. I've used it with my mach-omap2 DT patches for past
few weeks on various boards without issues.
 
> I like the warning message for the bad data.

:)

Tony