Re: [PATCH] move get_idle_time , get_iowait_time to sched/core.c [message #44704] |
Fri, 23 December 2011 10:08  |
Glauber Costa
Messages: 916 Registered: October 2011
|
Senior Member |
|
|
On 12/12/2011 01:31 PM, Glauber Costa wrote:
> In commit a25cac5198 we changed the way we calculate idle and iowait
> for /proc/stat displaying purposes. However, this same information
> is also displayed in /proc/uptime. These values can now be inconsistent.
>
> So I propose we draw both idle values from the same place. In theory,
> we only need to do it for get_idle_time(). get_iowait_time() is moved
> as well for consistency only.
>
> I moved the functions to sched/core.c so it can live among its other
> friends like nr_iowait(), etc.
>
You guys have any opinions on this ?
I see that Michal got another fix for this in stat.c already, so I'd
have to respin this one anyway. Just let me know if this is wanted
Thanks
|
|
|
|
Re: [PATCH] move get_idle_time , get_iowait_time to sched/core.c [message #44709 is a reply to message #44708] |
Fri, 23 December 2011 13:40  |
Glauber Costa
Messages: 916 Registered: October 2011
|
Senior Member |
|
|
On 12/23/2011 05:39 PM, Michal Hocko wrote:
> On Fri 23-12-11 14:08:09, Glauber Costa wrote:
>> On 12/12/2011 01:31 PM, Glauber Costa wrote:
>>> In commit a25cac5198 we changed the way we calculate idle and iowait
>>> for /proc/stat displaying purposes. However, this same information
>>> is also displayed in /proc/uptime. These values can now be inconsistent.
>>>
>>> So I propose we draw both idle values from the same place. In theory,
>>> we only need to do it for get_idle_time(). get_iowait_time() is moved
>>> as well for consistency only.
>>>
>>> I moved the functions to sched/core.c so it can live among its other
>>> friends like nr_iowait(), etc.
>>>
>>
>> You guys have any opinions on this ?
>> I see that Michal got another fix for this in stat.c already, so I'd
>> have to respin this one anyway. Just let me know if this is wanted
>
> The last follow up fix is in Andrew's tree at the moment: procfs: do not
> confuse jiffies with cputime64_t.
>
> I don't mind moving those functions outside stat.c if it is reused.
Well, they are not (currently). The point of this patch is precisely
this question: Should it be?
I think the values we show in /proc/uptime should be consistent with the
ones in /proc/stat, so yes.
|
|
|