Re: [PATCH] powerpc/pseries: export timebase register sample in lparcfg

[Michael Ellerman <mpe@ellerman.id.au>]

Tyrel Datwyler <tyreld@linux.vnet.ibm.com> writes:
> Ping? Any objections to this patch? A fix is already upstream in powerpc-utils
> to utilize the timebase value if available.

I'm slightly wary of adding things to lparcfg as it's world-readable and
hard to remove things once they're added (because tools rely on it).

But I guess this is fine.

If we ever want to restrict mftb access then we need to remember to also
lock down this file, or reduce the precision it returns.

cheers

> On 12/08/2018 03:48 PM, Tyrel Datwyler wrote:
>> The Processor Utilzation of Resource Registers (PURR) provide an estimate of
>> resources used by a cpu thread. Section 7.6 in Book III of the ISA outlines
>> how to calculate the percentage of shared resources for threads using the
>> ratio of the PURR delta and Timebase Register delta for a sampled period.
>> 
>> This calculation is currently done erroneously by the lparstat tool from the
>> powerpc-utils package. This patch exports the current timebase value after
>> we sample the PURRs and exposes it to userspace accounting tools via
>> /proc/ppc64/lparcfg.
>> 
>> Signed-off-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
>> ---
>>  arch/powerpc/platforms/pseries/lparcfg.c | 1 +
>>  1 file changed, 1 insertion(+)
>> 
>> diff --git a/arch/powerpc/platforms/pseries/lparcfg.c b/arch/powerpc/platforms/pseries/lparcfg.c
>> index 7944873..1ea73ec 100644
>> --- a/arch/powerpc/platforms/pseries/lparcfg.c
>> +++ b/arch/powerpc/platforms/pseries/lparcfg.c
>> @@ -475,6 +475,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
>>  		splpar_dispatch_data(m);
>> 
>>  		seq_printf(m, "purr=%ld\n", get_purr());
>> +        seq_printf(m, "tbr=%ld\n", mftb());
>>  	} else {		/* non SPLPAR case */
>> 
>>  		seq_printf(m, "system_active_processors=%d\n",
>>