Tracking down superlinear reductions
Ulf Wiger (AL/EAB)
Fri Jun 16 12:52:49 CEST 2006
One could imagine using QuickCheck and specifying as a property that the
number of reductions (returned from statistics(reductions) should be
roughly proportional to the size of the input.
QuickCheck will then generate random input, according to a
specification, and report if it finds a case where the number of
reductions is out of bounds for the given input.
I haven't actually tried it, though.
> -----Original Message-----
> [mailto:] On Behalf Of
> Andrew Lentvorski
> Sent: den 16 juni 2006 12:13
> To: erlang-questions
> Subject: Tracking down superlinear reductions
> pman seems to be reporting that I have a gen_server that
> appears to be doing a superlinear (probably closer to
> quadratic, but I haven't done the full analysis, yet) number
> of reductions (200,000,000+) for what should only be a linear
> amount of work (14,000 packets).
> Dialyzer doesn't report anything obviously stupid, so I'm
> going to need to dig into this further.
> Which tool should I use to analyze this? pman seems to be
> choking on the trace. I see a couple of different programs
> listed in the "Tool Applications" section of the manual, but
> nothing jumps out at me as being the obvious choice.
> What I'd like to do is the Erlang equivalent of getting the
> number of times a function is called. This would avoid the
> problem of Erlang trying to print out the arguments and
> flooding the I/O system.
> What's the best way to do this?
More information about the erlang-questions