Question about simple profiling implementation

[Andrew de Quincey]

On Tuesday 02 December 2003 18:21, Shachar Shemesh wrote:
> Dimitrie O. Paun wrote:
> >On December 2, 2003 08:52 am, Andrew de Quincey wrote:
> >>As you say, relay debugging adds a huge overhead... however, would you
> >> say that this overhead would be fairly constant for each particular
> >> function?
> >
> >And herein lies the problem: we're adding a _big_ constant overhead (O)
> >to a variable cost (c).
>
> When you are trying to profile an application, you don't really care how
> much it takes to run. A profiling that makes the program run slower is
> not really a problem. Profiling that changes the relative weights of the
> program's parts are.

Exactly. The percentage thing (which as Dimitrie points out is useless) was an 
off-the-top-of-the-head method to try and express the time used by each 
function without actually using the raw time values as they're very 
misleading.

I can see this is going to be contentious: I think I will just keep the 
patches to myself for my own uses.