[Skiboot] [PATCH] hw/fsp/rtc: read/write cached rtc tod on fsp hir.

[Stewart Smith]

Vasant Hegde <hegdevasant at linux.vnet.ibm.com> writes:
> On 07/23/2018 02:52 PM, Stewart Smith wrote:
>> Vasant Hegde <hegdevasant at linux.vnet.ibm.com> writes:
>>> On 04/03/2017 07:27 AM, Pridhiviraj Paidipeddi wrote:
>>>> Currently fsp-rtc reads/writes the cached RTC TOD on an fsp
>>>> reset. Use latest fsp_in_rr() function to properly read the cached rtc
>>>> value when fsp reset initiated by the hir.
>>>>
>>>
>>> .../...
>>>
>>> Looks like we are not triggering fsp_start_rr() for host initiated FSP R/R.
>>> While this patch works for
>>> RTC we may hit issues in some other place.
>>>
>>> Can you try with below patch?
>>>
>>> diff --git a/hw/fsp/fsp.c b/hw/fsp/fsp.c
>>> index a0c5a78..7321e63 100644
>>> --- a/hw/fsp/fsp.c
>>> +++ b/hw/fsp/fsp.c
>>> @@ -1571,6 +1571,7 @@ static void __fsp_poll(bool interrupt)
>>>           /* Handle host initiated resets */
>>>           if (fsp_in_hir(fsp)) {
>>>                   fsp_hir_poll(fsp, iop->psi);
>>> +               fsp_start_rr(fsp);
>>>                   return;
>>>           }
>> 
>> So, it looks like I pretty successfully ignored non-POWER9 OpenPOWER
>> things for err... way too long and really should have merged this in by
>> now.
>> 
>> The patch no longer cleanly applies just because things have moved, but
>> it looks like it should be easy to rebase, and with the acked-by's and
>> some testing, we should be good to bring it in.
>> 
>> Are you able to rebase and post?
>
> Stewart,
>
> We did merge original Pridhiviraj patch  to master (commit : 447ccc4d).
> Also we added subsequent cleanup patch (commit :  a3436963 ).
>
> Am I missing something?

I was wondering about the fsp_start_rr() call in that if(fsp_in_hir)
block... I didn't see it when I looked briefly at current skiboot and
was wondering if we still need it.

-- 
Stewart Smith
OPAL Architect, IBM.