Thanks Jim and Adriano both of those look like they will do what I wanted.

I will try them both out, I must admit "Valgrind's one, and kcachegrind"
on the surface look very appealing.

On 27/12/08 21:27, paulruizendaal wrote:
> Whilst all measurement is good, it may have little reward in this
> case. I found Dmitry's arguments convincing: on reflection my
> characterization of looper as a tree walker is only superficially
> correct.

Now was that the Socratic method at work there :-)

My need was a bit broader than this particular question, and I did want
something that could help, in part, quantify the results of a change.

But often the devil is in the detail, as Roman (Simakov) recently
reminded us, with a find in fb-devel, picking up a
Jrd::RuntimeStatistics::operator- taking 20% of the time.

Cheers - Mark