Re: [perfmon2] Timestamps and CPU frequency scaling

[Nagy M. <nag...@gm...>]

Yes, but the values displayed (column 6) are obviously not the time in 
ns since epoch. They seem to be nano-seconds passed since processor 
reset. I just need to understand how this field is computed from 
sched_clock().

thanks,
- nagy

stephane eranian wrote:
> On Thu, Jan 22, 2009 at 4:46 PM, Nagy Mostafa <nag...@gm...> wrote:
>   
>> Hi Stephane,
>>
>> I am a little confused now. I thought perfmon uses TSC to timestamp samples.
>> For example, the pfmon command:
>>
>> $ pfmon --smpl-module=compact --with-header
>> -emy_event,unhalted_reference_cycles --long-smpl-periods=140000
>> my_program
>>
>> gives:
>>
>>
>> # description of columns:
>> #       column  1: entry number
>> #       column  2: process id
>> #       column  3: thread id
>> #       column  4: cpu number
>> #       column  5: instruction pointer
>> #       column  6: unique timestamp
>> #       column  7: overflowed PMD index
>> #       column  8: event set
>> #       column  9: initial value of overflowed PMD (sampling period)
>> #       followed by optional sampled PMD values in command line order
>> #
>> #
>> 0            5854     5854  3 0x400621 0x0000002aaadf4476  17     0
>>     -140000 0x3363c
>> 1            5854     5854  3 0x400621 0x0000002aaae0b512  17     0
>>     -140000 0x66c78
>> 2            5854     5854  3 0x400621 0x0000002aaae21813  17     0
>>     -140000 0x9a2bd
>> 3            5854     5854  3 0x400621 0x0000002aaae37a5c  17     0
>>     -140000 0xcd8f9
>> 4            5854     5854  3 0x400621 0x0000002aaae4dc69  17     0
>>     -140000 0x100f3e
>> 5            5854     5854  3 0x400621 0x0000002aaae63e67  17     0
>>     -140000 0x134583
>>
>> So, column #6 is the TSC value divided by the CPU freq (i.e. time from reset
>> in ns).
>> You mentioned a tstamp field in the sample that relies on sched_clock(). Is
>> that another timestamp field ? If yes, how can I display it using pfmon ?
>>
>>     
> We are talking about the same field: tstamp = sched_clock().
> This is the only field. If you try to divide by CPU freq then you will likely be
> affected by scaling. You can lock your CPU at a fixed frequency via /sys.
> I think there exists a command to get/set frequency.
>
>