> From: Ian Dall <ian@beware.dropbear.id.au>
>:::
> This is clumsy though. What would it take to make a more capable profiling
> method? If you could call profil(2) or something like it, for each of
> the dynamically loaded modules/libraries. The main catch I see is you
> need an efficient way of sticking the sampled PC in what is effectively
> a sparse array. Also you would want some corresponding user land changes.

If one is going to invent a new profiling mechanism, I think that extending
the existing method of compression-into-bins into compression-into-sparse-
bins is too much work for a historically important but suboptimal sampling
algorithm.

I would propose instead that a new profiler simply accumulate and
write out a linear series of PC samples.

	ross.harvey@computer.org