Re: [PATCH 2/2 V7] intel_pstate: add kernel parameter to force loading on Sun X86 servers.

[Linda Knippers]

On 12/4/2014 9:05 PM, Rafael J. Wysocki wrote:
> On Thursday, December 04, 2014 06:03:05 PM Linda Knippers wrote:
>> On 12/4/2014 5:38 PM, Kristen Carlson Accardi wrote:
>>> On Thu, 04 Dec 2014 23:10:58 +0100
>>> "Rafael J. Wysocki" <rjw@xxxxxxxxxxxxx> wrote:
>>>
>>>> On Thursday, December 04, 2014 11:07:31 AM Ethan Zhao wrote:
>>>>> To force loading on Oracle Sun X86 servers, provide one kernel command line
>>>>> parameter
>>>>>
>>>>>   intel_pstate = ora_force
>>>>
>>>> I would suggest to change the name of the option to "oracle_force" or "sun_force"
>>>> for clarity.
>>>>
>>>> Anyway, I need an ACK from Kristen if this patch is to be applied.
>>>>
>>>>> For those who be aware of the risk of no power capping capabily working and
>>>>> try to get better performance with this driver.
>>>>>
>>>>> Signed-off-by: Ethan Zhao <ethan.zhao@xxxxxxxxxx>
>>>>> ---
>>>>>  v2: change to hardware vendor specific naming parameter.
>>>>>  v4: refine code and doc.
>>>>>  v5&v6: fix a typo in doc.
>>>>>  v7: change enum PCC to PPC.
>>>>>
>>>>>  Documentation/kernel-parameters.txt | 5 +++++
>>>>>  drivers/cpufreq/intel_pstate.c      | 6 +++++-
>>>>>  2 files changed, 10 insertions(+), 1 deletion(-)
>>>>>
>>>>> diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
>>>>> index 479f332..7d0983e 100644
>>>>> --- a/Documentation/kernel-parameters.txt
>>>>> +++ b/Documentation/kernel-parameters.txt
>>>>> @@ -1446,6 +1446,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
>>>>>  		       disable
>>>>>  		         Do not enable intel_pstate as the default
>>>>>  		         scaling driver for the supported processors
>>>>> +		       ora_force
>>>>> +			 Force loading intel_pstate on Oracle Sun Servers(X86).
>>>>> +			 only for those who be aware of the risk of no power capping
>>>>> +			 capability working and try to get better performance with this
>>>>> +			 driver.
>>>>
>>>> That is not sufficiently clear.  What does "risk of no power capping capability
>>>> working" mean, in particular?
>>>>
>>>>>  
>>>>>  	intremap=	[X86-64, Intel-IOMMU]
>>>>>  			on	enable Interrupt Remapping (default)
>>>>> diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
>>>>> index 1bb62ca..2654e13 100644
>>>>> --- a/drivers/cpufreq/intel_pstate.c
>>>>> +++ b/drivers/cpufreq/intel_pstate.c
>>>>> @@ -866,6 +866,7 @@ static struct cpufreq_driver intel_pstate_driver = {
>>>>>  };
>>>>>  
>>>>>  static int __initdata no_load;
>>>>> +static unsigned int  ora_force;
>>>>>  
>>>>>  static int intel_pstate_msrs_not_valid(void)
>>>>>  {
>>>>> @@ -1003,7 +1004,8 @@ static bool intel_pstate_platform_pwr_mgmt_exists(void)
>>>>>  			case PSS:
>>>>>  				return intel_pstate_no_acpi_pss();
>>>>>  			case PPC:
>>>>> -				return intel_pstate_has_acpi_ppc();
>>>>> +				return intel_pstate_has_acpi_ppc() &&
>>>>> +					(!ora_force);
>>>>>  			}
>>>>>  	}
>>>>>  
>>>>> @@ -1078,6 +1080,8 @@ static int __init intel_pstate_setup(char *str)
>>>>>  
>>>>>  	if (!strcmp(str, "disable"))
>>>>>  		no_load = 1;
>>>>> +	if (!strcmp(str, "ora_force"))
>>>>> +		ora_force = 1;
>>>>>  	return 0;
>>>>>  }
>>>>>  early_param("intel_pstate", intel_pstate_setup);
>>>>
>>>> And can anyone please remind me what was wrong with a "force" option that would
>>>> work for everyone, not just Oracle/Sun?
>>>>
>>>
>>> That was my suggestion as well (i.e. a parameter to bypass the vendor
>>> checks), but Linda didn't like it.  My personal opinion is that unless
>>> it's generic, I don't really feel like having a force option solely for
>>> oracle.  I'm not convinced you want this for production machines, and I
>>> think for debug purposes I don't want a vendor specific param.
>>
>> I'd be happy with it if it somehow disabled what the platform is doing,
>> but it doesn't.  I don't see the point of forcing intel_pstate if you
>> can't force the platform to stop doing power management at the same time.
>> Even if it's for test/debug purposes, I'm not sure what you're testing
>> when you have dueling power management.
>>
>> The description would need to be different too since I think on
>> ProLiant, power capping can happen at any time, even if the
>> system is in OS control mode and the intel_pstate driver is
>> loaded.
>>
>> Can anyone suggest a description for a force option that would
>> make sense generically?
> 
> What about:
> 
> 	force
> 		Enable intel_pstate on systems where it may cause problems to
> 		happen due to conflicts with platform firmware attempting to
> 		drive P-states by itself in certain situations (for thermal
> 		control or power capping in general or other purposes).

Except in the case of HP, it's not just for "certain situations" like for power
capping for thermal control.  If the BIOS is configured to manage the power,
it's going to constantly managing the power, just like the intel_pstate  driver
does.  It would be like running intel_pstate while also running
apci_cpufreq.   Is there ever a case where that makes sense?

I still don't understand the Oracle case.  Ethan seems to want to not load
the intel_state driver normally because it will conflict with power capping,
so I understand why one might (maybe) want to disable power capping.  But
how do the Oracle platforms do steady-state (not power capped) power
management?  How does that work without the intel_pstate driver or some
other cpufreq driver?  Is the platform firmware managing p-states?  If so, then
what happens if you load the driver anyway?

-- ljk
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/