Re: Code analysis, call graphs for Common Lisp
- From: Juho Snellman <jsnell@xxxxxx>
- Date: 29 Apr 2005 16:51:28 GMT
<wpower@xxxxxxxxxxxxxxxxxxxxxxxx> wrote:
> I've been having a bit of a look around for some code to do call
> graphs given Common Lisp source code input. One of my purposes is to
> find functions that don't get utilised at all (due to code
> refactoring, for instance).
[...]
> I did notice a reference to something in the SBCL profiler that
> outputs a run-time call graph or something. Is that the best deal out
> there?
The profiler in question works by taking periodic samples of the
instruction pointer and the stack frames during program execution, so
the results are only approximate. This doesn't matter for profiling,
but makes the call-graphs just about useless for your intended
purpose.
--
Juho Snellman
"Premature profiling is the root of all evil."
.
- References:
- Code analysis, call graphs for Common Lisp
- From: David Trudgett
- Code analysis, call graphs for Common Lisp
- Prev by Date: Re: Comparing Lisp conditions to Java Exceptions
- Next by Date: Re: Lisp/Unix impedance [a programming challenge]
- Previous by thread: Re: Code analysis, call graphs for Common Lisp
- Next by thread: Re: Code analysis, call graphs for Common Lisp
- Index(es):