Re: May nearly have simple statistical profiler working (need help).

[Rob Browning]

Rob Browning <address@hidden> writes:

> Wait a minute.  I think maybe I was misinterpreting the stack
> behaviors.  Don't waste too much time on my response until I play
> around a bit.

OK, detected minor brain inversion that was causing the trouble.
Righting cerebrum now.

Note that in the sampling handler I was careful to not accumulate time
if the profiling time went off when we were already inside the
apply-frame handler, but I still sampled the entire stack.  Since the
apply-frame handler doesn't count any of the calls generated by it's
own code, all of these functions might be sampled, but never
apply-frame counted.  Hence the anomolies.

I'll need to fix this...

-- 
Rob Browning
rlb @defaultvalue.org, @linuxdevel.com, and @debian.org
Previously @cs.utexas.edu
GPG=1C58 8B2C FB5E 3F64 EA5C  64AE 78FE E5FE F0CB A0AD