Re: [PATCH v2] intel_idle: add idle values for Cherrytrail/Braswell

From: Li, Aubrey
Date: Tue Aug 26 2014 - 20:04:29 EST


On 2014/8/26 17:00, Mika Westerberg wrote:
> On Tue, Aug 26, 2014 at 10:38:32AM +0800, Li, Aubrey wrote:
>> On 2014/8/25 18:12, Mika Westerberg wrote:
>>> On Fri, Aug 22, 2014 at 10:06:21PM +0800, Li, Aubrey wrote:
>>>> On 2014/8/22 19:39, Mika Westerberg wrote:
>>>>> From: Mahesh Kumar P <mahesh.kumar.p@xxxxxxxxx>
>>>>>
>>>>> Cherrytrail/Braswell is a successor of Intel Baytrail but has slighly
>>>>> different CPU idle values and latencies.
>>>>>
>>>>> Signed-off-by: Kumar P Mahesh <mahesh.kumar.p@xxxxxxxxx>
>>>>> Signed-off-by: Alan Cox <alan@xxxxxxxxxxxxxxx>
>>>>> Signed-off-by: Mika Westerberg <mika.westerberg@xxxxxxxxxxxxxxx>
>>>>> ---
>>>>> I learned from Mahesh that C1e, C4 and S0i2 states are deprecated so those
>>>>> are now removed.
>>>>
>>>> If the platform has ACPI supported, I believe Len wants to see what
>>>> C-states number exported by ACPI.
>>>>
>>>> Please boot the machine with "intel_idle.max_cstate=0" to disable
>>>> intel_idle, then post the output of the following commands
>>>
>>> OK, here goes:
>>
>> ACPI exposes 0x58 as C6 mwait hint while the patch proposes 0x52 for C6.
>> Can you please remove "intel_idle.max_cstate=0" and run "dmesg | grep
>> idle" again to see how many C6 supports in CPUID enumeration?
>
> [ 16.685769] intel_idle: MWAIT substates: 0x33000020
> [ 16.691236] intel_idle: v0.4 model 0x4C
> [ 16.695539] intel_idle: lapic_timer_reliable_states 0xffffffff
>

enough to turn over to Len to finalize, thanks Mika.

-Aubrey

--
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/