Re: [PATCH v3 3/3] ARM: dts: atlas7: use general dt-binding for hwspinlock

[Suman Anna <s-anna-l0cyMroinI0@public.gmane.org>]

On 06/01/2015 12:31 AM, Barry Song wrote:
> 2015-05-29 23:50 GMT+08:00 Bjorn Andersson <bjorn-UYDU3/A3LUY@public.gmane.org>:
>> On Thu, May 28, 2015 at 2:30 PM, Suman Anna <s-anna-l0cyMroinI0@public.gmane.org> wrote:
>> [..]
>>>>                               reg = <0x13240000 0x00010000>;
>>>
>>> An unrelated question here, why the reg is same for all the child nodes
>>> of the  parent ipc node? If this is partitioned properly, then the
>>> driver can be simplified a bit by using platform_get_resource and
>>> devm_ioremap_resource?
>>>
>>
>> Good catch Suman, I missed that.
>>
>> Barry, if these blocks represents various functionalities of the same
>> hw block then you should consider moving them to be part of a
>> simple-mfd.
> 
> the hwspinlock and the IPC, which works for kicking interrupts between
> multiple cores in the SoC,  are in one bus node but there is no
> overlapping register.
> hwspinlock:
> begin from 0x13240000 + 0x400
> IPC:
> begin from 0x13240000, and end at begin from 0x13240000 + 0x400 - 0x4 ?
> 
> so i guess we can refine the dts memory region to avoid MFD?

Yeah, that should be fine from the hwspinlock perspective, all its
registers are self-contained.

regards
Suman
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html