make check of linalg . Solaris 7/Ultra 10/gcc 2.7.2.3

Brian Gough bjg@network-theory.co.uk
Sun Feb 13 13:34:00 GMT 2000


Thanks -- that is useful information. We'll improve those tests.

More, Richard writes:
 > FYI,
 > 
 >   I had these problems with the 'make check' of gsl 0.5
 > Here's the output. I don't think all these FAILs are in the KNOWN-PROBLEMS
 > file.
 > 
 > Rich More
 > 
 > =============================
 > Making check in linalg
 > make  test_la
 > `test_la' is up to date.
 > make  check-TESTS
 > PASS: Matrix Multiply
 > PASS: Matrix Multiply with Modification
 > PASS:   solve_LU hilbert(2)
 > PASS:   solve_LU hilbert(3)
 > PASS:   solve_LU hilbert(4)
 >  12[0]:   -1523.77518385648727                    -1728
 >  12[1]:     219511.16757080768                   245388
 >  12[2]:    -7715210.2688227715                 -8528520
 >  12[3]:    115954444.966090292                127026900
 >  12[4]:   -927911537.858231306              -1009008000
 >  12[5]:    4412597738.54375744               4768571808
 >  12[6]:   -13211944461.5272522             -14202796608
 >  12[7]:    25544704655.7053375              27336497760
 >  12[8]:   -31822600256.5581818             -33921201600
 >  12[9]:    24653651482.3935051              26189163000
 >  12[10]:   -10800041470.9100952             -11437874448
 >  12[11]:    2043096934.37019181               2157916488
 > FAIL:   solve_LU hilbert(12)
 >  ....


More information about the Gsl-discuss mailing list