On Fri, 12 Feb 2010, heechul Yun wrote:
> I'm experimenting performance counter to find deterministic events.
>
> I originally thought that the number of retired instructions will be the
> same for every run of the program with the same input. However, it was not.
>
> ///
>
> What is the reason of this unexpected non-determinism?
You can find an entire paper about the non-determinism of the retired
instruction performance counter here:
http://www.csl.cornell.edu/~vince/papers/iiswc08/
There's a lot of reasons for this, though the one that causes the most
trouble is address space randomization, which can be turned off under
Linux.
Vince
------------------------------------------------------------------------------
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________
perfmon2-devel mailing list
perfmon2-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
