IF the trig funs work THEN this happens

[Branko]

Phil Edwards wrote:
> 
> Well, if I add the cosf/sinf/sqrtf.lo files to LIBMATHOBJS and
> libmath_la_LIBADD targets manually, and then also apply the following patch
> (fabs needs to be declared; and "iff" LD_LIBRARY_PATH is set then we need
> to prepend LIB_PATH during checking), then everything builds and only two
> tests fail:
> 
> - 3     97.657  227097  3952            21_strings/ctor_copy_dtor.cc
> - 6     99.337  23973   184             21_strings/ctor_copy_dtor.cc
> - 7     0.000   548887  8000            27_io/istream_extractor_arith.cc
> - 6     0.000   51187   1192            27_io/istream_extractor_arith.cc
> 
> Both static and dynamic fail with runtime errors, so we get 134/4 results.
> 
> I'm going to go ahead and check in the complex_inserters_extractors patch,
> but I won't do the mkcheck.in yet -- opinions?

On Solaris 2.6, I get:

- 10    0.000   text    415054          27_io/istream_extractor_arith.cc
- 9     0.000   text    46331           27_io/istream_extractor_arith.cc
- 11    0.000   text    422094          27_io/istream_unformatted.cc
- 9     0.000   text    48436           27_io/istream_unformatted.cc

I think the mkcheck.in patch can go in, too.

    Brane

-- 
Branko Čibej                 <branko.cibej@hermes.si>
HERMES SoftLab, Litijska 51, 1000 Ljubljana, Slovenia
voice: (+386 1) 586 53 49     fax: (+386 1) 586 52 70