Compilation errors on IRIX64 with optimizations

[Daniel Konkle <daniel dot konkle at dynetics dot com>]

I am in the process of recompiling the GSL libraries for a few different
platforms.  My need is to get the GSL libraries compiled with '-O3'
optimizations.

While working on multi-processor SGI, I had a lot of successor.  After
compilation, I ran make check.  If 'make check' failed in a directory, I
commented that directory out and continued with the checks.

There were only 3 directories that failed:  sys specfunc & multiroot

I'm showing my compilation options below as well as the various errors
that were reported by 'make check' in hopes that this information might
help.

Compilation/Configure:
 CC=cc CFLAGS="-OPT:IEEE_comparisons=ON -woff 728 -O3 -mips4 -64"
LDFLAGS="-64" ./configure


Directory 'sys' errors:
PASS: gsl_isinf(0) (0 observed vs 0 expected)
PASS: gsl_isinf(1) (0 observed vs 0 expected)

FAIL: gsl_isinf(inf) (0 observed vs 1 expected)
FAIL: gsl_isinf(-inf) (0 observed vs -1 expected)

PASS: gsl_isinf(nan) (0 observed vs 0 expected)
PASS: gsl_isnan(0) (0 observed vs 0 expected)
PASS: gsl_isnan(1) (0 observed vs 0 expected)
PASS: gsl_isnan(inf) (0 observed vs 0 expected)
PASS: gsl_isnan(nan) (1 observed vs 1 expected)
PASS: gsl_finite(0) (1 observed vs 1 expected)
PASS: gsl_finite(1) (1 observed vs 1 expected)

FAIL: gsl_finite(inf) (1 observed vs 0 expected)
FAIL: gsl_finite(nan) (1 observed vs 0 expected)

PASS: gsl_fdiv(2,3) (0.666667 observed vs 0.666667 expected)
FAIL: test

Directory 'spec' errors: 

see attached file

Directory 'multiroots' errors:

PASS: dnewton on Discrete BVP (1), 2 iterations, residual = 8.2e-08
FAIL: dnewton on Trig (1), 1000 iterations, residual = nan0x7fffffff
PASS: broyden on Rosenbrock (1), 4 iterations, residual = 2.2e-14



Hope this helps.
Danny

Attachment: specfunc_errors.txt
Description: Text document