Hello~

Thanks for replying.

I have modified my source code accordingly and I could successfully
retrieve the UNHALTED_CORE_CYCLE.
However, the L2_RQSTS:SELF is still zero.
Could anyone give me some hints about this problem?

The self_smpl_multi.c has a structure named fd2ov to match the fd to
the corresponding ov.
I am wondering that this is necessary if fd and ov are threads'
private variables.

Any help would be grateful!

Dennis

2009/4/20 stephane eranian <eran...@googlemail.com>:
> Hi,
>
> On Sun, Apr 19, 2009 at 11:23 AM, Dionysus <shinrai...@gmail.com> wrote:
>>    if(pfm_self_start(fd))
>>        fprintf(stderr,"pfm_self_start error\n");
>>        t_next = t->execute(); // This is the function I want to monitor.
>>    if(pfm_self_stop(fd))
>>        fprintf(stderr,"pfm_self_stop error\n");
>
>>    fprintf(stderr,"This task's L2 references is %d\n", ov.pd[0].reg_num);
>
> You are not reading the results out. You need to call pfm_read_pmds().
> And then the results are in the reg_value field, not reg_num (which is the
> register index).
>

------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
perfmon2-devel mailing list
perfmon2-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/perfmon2-devel

Reply via email to