Re: [PATCH B 04/10] OMAP3 PRCM: add DPLL1-5 powerdomains, clockdomains; mark clocks

[Russell King - ARM Linux]

On Tue, Jan 27, 2009 at 07:44:18PM -0700, Paul Walmsley wrote:
> Each DPLL exists in its own powerdomain (cf 34xx TRM figure 4-18) and
> clockdomain; so, create powerdomain and clockdomain structures for them.
> Mark each DPLL clock as belonging to their respective DPLL clockdomain.
> cf. 34xx TRM Table 4-27 (among other references).

I don't really see the need to mark all these clocks as having a domain.
Yes, it makes sense to mark the actual DPLL clocks themselves with the
relevent DPLL domain.  However, it makes no sense to mark the children
as well.

Why?  When we enable a clock, we walk up the tree enabling the domains
first.  So, even if we're enabling a child clock, we will walk the tree
up to the DPLL and enable the clock domain before we do anything at all.

Sure, if a clock has a multiplexer and that multiplexer is in a different
clock domain from its parent, it makes sense to.  But otherwise it doesn't.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/