[Skiboot] [PATCH 59/61] xive2: Add NCU_SPEC_BAR to stop engine for restore

[Pratik Sampat]

On 22/07/21 3:01 pm, Vasant Hegde wrote:
> On 7/21/21 7:05 PM, Cédric Le Goater wrote:
>> On 7/19/21 3:20 PM, Vasant Hegde wrote:
>>> From: Vaidyanathan Srinivasan <svaidy at linux.ibm.com>
>>>
>>> P10 Stop engines have apis similar to P9 to set xscom restores
>>> after wakeup from deep-sleep states.
>>>
>>> This xscom restore will be used to support STOP11 on P10.
>>>
>>> Signed-off-by: Vaidyanathan Srinivasan <svaidy at linux.ibm.com>
>>> Signed-off-by: Pratik Rajesh Sampat <psampat at linux.ibm.com>
>>> Signed-off-by: Vasant Hegde <hegdevasant at linux.vnet.ibm.com>
>>> ---
>>>   hw/xive2.c | 31 ++++++++++++++++++++++++++-----
>>>   1 file changed, 26 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/hw/xive2.c b/hw/xive2.c
>>> index f559b0f9a..d2fc2ad69 100644
>>> --- a/hw/xive2.c
>>> +++ b/hw/xive2.c
>>> @@ -20,8 +20,8 @@
>>>   #include <bitmap.h>
>>>   #include <buddy.h>
>>>   #include <phys-map.h>
>>> -#include <p9_stop_api.H> /* TODO (p10): need P10 stop state engine */
>>> -
>>> +#include <p9_stop_api.H>
>>> +#include <p10_stop_api.H>
>>
>> Why do we need both ?
>>
>
> Because p10_stop_api.H uses definitions from p9_stop_api.H. May be we 
> should move common
> definitions to separate header file (say stop_api.h ) and then include 
> that in P[9/10]_stop_api.H file?
>
I agree that there are a lot of common definitions around both P9 and P10 stop
API header, and could definitely be cleaned up and merged. However the headers
are direct mirror from hostboot and they maintain different code-bases for
P9 and P10.

My cause of concern stems from the fact that if we refactor the libraries here
and if hostboot updates their libraries we would manually each time have
to code those changes up in our refactored library too.

Thanks,
Pratik

> -Vasant
>