Re: [PATCH v5] cpufreq: fix governor start/stop race condition

From: Xiaoguang Chen
Date: Tue Jun 18 2013 - 23:26:20 EST


2013/6/19 Viresh Kumar <viresh.kumar@xxxxxxxxxx>:
> On 19 June 2013 08:43, Viresh Kumar <viresh.kumar@xxxxxxxxxx> wrote:
>> On 19 June 2013 06:50, Xiaoguang Chen <chenxg.marvell@xxxxxxxxx> wrote:
>>> 2013/6/19 Rafael J. Wysocki <rjw@xxxxxxx>:
>>
>>>>> 2) Current governor is userspace, now cpu0 hotplugs in cpu3, it will
>>>>
>>>> Can you please tell me what the above is supposed to mean? Is it supposed to
>>>> mean "the online of cpu3 is being run on cpu0" or something different? If
>>>> something different, then what?

Sorry I missed this question, Let me explain it in detail
Suppose we are in such condtition, current cpufreq goveror is
userspace governor. cpu3 is offline.
and two things happen as above two cases, first thing is application
tries to change current governor to ondemand governor,
second thing is cpu0 tries to make cpu3 online which is off line
before. both of these two cases will try to stop current governor and
start a governor. if above two things interleave, unexpected behavior
will happen.


>>
>> Please read all the questions carefully. You missed this one.
>>
>> Actually you should write: Current governor is userspace, now cpu0 hot-unplugs
>> cpu3, it will **
>
> Ahh I am mistaken, you are actually bringing cpu3 back to the system. Then you
> must have mentioned earlier that cpu3 wasn't online.
--
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/