Re: [Octave-bug-tracker] Octave with Clang

[Michael D. Godfrey]

On 03/06/2014 09:57 AM, PrasannaKumar Muralidharan wrote:

A few days back I compiled octave with clang. Did not make any changes
to Makefile. Make check reported 2 failures. Please find them below.

Is this something to look at?

I guess gcc is following gnu89 standard while clang follows c99 by
default. Is this the cause for the problem?

Failure 1:

>>>>> processing /home/i21337/Software/octave/octave/libinterp/corefcn/data.cc-tst

  ***** assert (log2 (complex (0,Inf)), Inf + log2 (i))
!!!!! test failed
ASSERT errors for:  assert (log2 (complex (0, Inf)),Inf + log2 (i))

  Location  |  Observed  |  Expected  |  Reason
     ()        Inf+NaNi   Inf+2.2662i    'NaN' mismatch


Failure 2:

>>>>> processing /home/i21337/Software/octave/octave/scripts/polynomial/roots.m

  ***** assert (roots ([1e-200, -1e200 * 1i, 1]), -1e-200 * 1i)
!!!!! test failed
ASSERT errors for:  assert (roots ([1e-200, -1e200 * 1i, 1]),-1e-200 * 1i)

  Location  |  Observed  |  Expected  |  Reason
     ()           O            E         real != complex


Seeing a huge number of tests passing without any problem I feel
Octave code seems to be portable across compilers. I also found some
warnings which I am planning to report soon.

Thanks and Regards,
PrasannaKumar

This is very useful.  I think that there was a previous attempt at using clang, and the
Mac OS X 10.9 system also includes clang, as does Fedora. 

It may make sense to consider use of clang, where available, as the preferred compiler.

The above problem may not lead to incorrect results, but any use of uninitialized
variables should be eliminated as a matter of principle.

Michael