Re: reliable seg fault

[Jordi Gutiérrez Hermoso]

2011/10/24 Jordi Gutiérrez Hermoso <address@hidden>:
> On 24 October 2011 09:19, Mike Miller <address@hidden> wrote:
>> On Mon, 24 Oct 2011, Francesco Potortì wrote:
>>
>>>> octave:2> n=181; p=39; X=ones(n,p); X'*X;
>>>> panic: Segmentation fault -- stopping myself...
>>>> attempting to save variables to `octave-core'...
>>>> save to `octave-core' complete
>>>> Segmentation fault
>>>>
>>>> That is, when there are 39 columns, a matrix "X" with fewer than 181 rows
>>>> is ok, but with 181 or more rows it seg faults when I attempt X'*X.
>>>
>>> No problem on Debian:
>>>
>>> GNU Octave, version 3.2.4
>>> Octave was configured for "x86_64-pc-linux-gnu".
>>
>>
>> I know that other people won't have this problem.  I've been using Octave
>> for many years and it always worked.  I'm just wondering if anyone here has
>> an idea of what might be causing this problem.
>
> Some old versions of gcc were buggy and producing segfaulty code. Any
> idea what was used to compile this binary?

Oh, I just remembered, there's also an old buggy ATLAS library that
was causing segfaults in cases like this. Are you able to get a stack
trace?

- Jordi G. H.