[Skiboot] [RFC] phb4: Wait for PRD to reset the CAPP Fir during recovery

[Andrew Donnellan]

On 17/10/18 2:43 am, Frederic Barrat wrote:
> 
> 
> Le 16/10/2018 à 13:32, Vaibhav Jain a écrit :
>> During CAPP recovery do_capp_recovery_scoms() will reset the CAPP Fir
>> register just after CAPP recovery is completed. This has an
>> unintentional side effect of preventing PRD from analyzing and
>> reporting this error. If PRD tries to read the CAPP FIR after opal has
>> already reset it, then it logs a critical error complaining "No active
>> error bits found".
>>
>> To prevent this from happening we update do_capp_recovery_scoms() to
>> wait for CAPP Fir to be reset by PRD just after CAPP recovery
>> completes and before we proceed with rest of the CAPP recovery
>> sequence. A timeout of 5ms is used to wait for CAPP-Fir reset before
>> we reset the register on our own. This is to guard against the
>> possibility of Opal PRD daemon crashing/not-running.
>>
>> Signed-off-by: Vaibhav Jain <vaibhav at linux.ibm.com>
>> ---
> 
> This looks really odd to me. I think we need to understand why the PRD 
> is messing with the CAPP FIR.

Spinning in skiboot for 5ms is also not great.

How does PRD cope with FIRs in other parts of the system which get 
cleared as part of recovery? (I'm thinking PHBs etc)

-- 
Andrew Donnellan              OzLabs, ADL Canberra
andrew.donnellan at au1.ibm.com  IBM Australia Limited