This is the mail archive of the
mailing list for the GCC project.
Re: About testing other languages: was RE: [C++ PATCH, committed] Fix libstdc++ regressions due to my access checking changes
Rainer Orth writes:
> Kaveh R. Ghazi writes:
> > I'm aware of that one, however a recent posting indicates that without
> > the closure interface it is impossible to use a large bit of the API.
> > http://gcc.gnu.org/ml/gcc/2003-02/msg00561.html
> > OTOH, one seems to be able to get testsuite results for the library
> > with the line length fix. So I withdraw my comment about testing on
> > irix6.
> Indeed, to the untrained eye testsuite results look somewhat reasonable:
> === libjava Summary for unix ===
> # of expected passes 2316
> # of unexpected failures 20
> # of expected failures 16
> # of untested testcases 111
> Unfortunately, there seems to be no up-to-date documentation on what the
> missing bits are.
This trained eye wants to know what the failures are.