Re: Accuracy of profiling

Andrew Haley <[email protected]>
Newsgroups gmane.linux.oprofile
Message-ID <[email protected]>
On 05/06/15 15:26, Mahmood N wrote:

> Assume without any speculation, ocount reports that the execution
> took 20 seconds. Now we enable some speculations (TLB, PF, ....) and
> now, ocount reports that the execution took 18 seconds. The question
> is, which one reduced the execution time? The caching effect (disk
> and IO caches done by OS) or the speculations?

Hard to say.  I'd not bother with this kind of analysis on a large
application to begin with, but study a small sample and gradually make
it larger.

BTW, you should at least be running this in single user mode.

Andrew.

------------------------------------------------------------------------------
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.