This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: g77 -ieee crash in 1.0.3a


>>>>>> "Craig" == Craig Burley <burley@gnu.org> writes:
>
> Craig> Dave, could you add this test case to g77.f-torture/compile
> Craig> as, say, 980729-0.f?
>
>Afraid I still don't know how to operate the system-specifc/
>option-specifc testcases.  I'll look into it as soon as I can and try
>to act Tcl-ish.

I'd say don't worry about it for now, stuff it in anyway (not the 1.1
branch for sure, which I forgot to mention; 1.1's test suite is frozen),
and we'll worry later about it.

In particular, -mieee should be one of the options that gets exercised
for *all* tests on Alphas, at least for the g77 tests (but why stop
there?!).  The commentary already mentions it, so that's taken care
of.  And if we have to simply stuff a .x file in for it someday to
just take care of that case, we can do that, but I'd rather get the
test in to the mainline (just so I can stop worrying about it) and
deal with -mieee later.

>BTW, I don't understand the reason for -mieee with LAPACK (was it?) in
>the first place.

Maybe LAPACK needs the larger range, smaller numbers, and such; even
if it doesn't, -mieee surely should do nothing worse to any code
than to slow it down.  Crashing isn't nice; especially on a three-line
test case.  :)

        tq vm, (burley)


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]