[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Problem with FFT and Intel compiler in "test fft2 verbose" or make c
From: |
Andy Buckle |
Subject: |
Re: Problem with FFT and Intel compiler in "test fft2 verbose" or make check |
Date: |
Mon, 6 Sep 2010 16:08:50 +0100 |
On Mon, Sep 6, 2010 at 12:56 PM, RUSS BRENNAN <address@hidden> wrote:
> I have tried using fftw 3.2.2 as well as the MKL fft routines but I get the
> same error with both when compiling with the intel compiler. The fft2 test
> fails with a segmentation fault. I am thinking that it probably has
> something to do with intel providing it's own fft routines, and the compiler
> getting confused about which one to call... but that wouldn't explain why I
> get the seg fault when trying to use the MKL libraries.
>
> Has anyone seen this before, and if not how should I go about debugging
> this? Is there an easy way to run octave in valgrind?
>
> TIA,
>
> Russell
I can't help with most of your question. I do frequently use valgrind
to test my oct files. write a short m-file script that will use the
code you want to test. then
valgrind octave -q testscript.m
(possibly -qf, if required)
Andy