Re: Measure CPU time accurately

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi...

On Tuesday 14 March 2006 05:06, Mauricio Lin wrote:
>  I am trying to measure the CPU time accurately used by a process in
> the schedule function through sched_info_depart() and
> sched_info_arrive(), but most of time the calculated CPU time is
> zero.

Where do you get the total CPU time spent by the process? from 
t->sched_info.cpu_time ?

>  As in one tick (jiffy increments) many processes are scheduled, it
> is not able to measure accurately the CPU time used by the all
> scheduled process since the measurement time is based on jiffies
> incrementation.

if you mean "scheduled" as "marked as runnable", then yes, many 
processes might be scheduled in one tick interval. But, only one 
process is selected to actually run on a single CPU core. So in my 
opinion, the kernel can still track the CPU time pretty accurately (by 
1/HZ precision)

>  Most output got from relayfs shows that CPU time is zero as:
>
>  PID   : Name: Jiffies     : CPU Time
>  4353  : Xorg: 152680   : 0
>  4521  : artsd: 152680    : 0
>  4353  : Xorg: 152680   : 0

Maybe because those processes are mostly in sleeping state? BTW, can you 
paste your code so we can analyze how do you do the time accounting?

regards

Mulyadi


--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/


[Index of Archives]     [Newbies FAQ]     [Linux Kernel Mentors]     [Linux Kernel Development]     [IETF Annouce]     [Git]     [Networking]     [Security]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux SCSI]     [Linux ACPI]
  Powered by Linux