[gfortran] PR libfortran/21376

[Tobias Schlüter]

FX Coudert wrote:
>>you
>>should add a testcase for it, though.  The testcase I posted is not good
>>enough given that I got the exponents of the corner cases wrong.
> 
> 
> There can is no easy testcase for this PR. In fact, there are already 
> cases in the testsuite where we compute the format of 1.0, but the PR 
> only fails if some external C code modifies the FPU flags to issue FPE 
> in such a case.
> 
> So, I don't think this can be done in a portable way (and using the 
> dejagnu framework).

Well, we can't test for the same error, but we can verify that the output is
indeed correct.  This is what I meantt the testcase should be for.

- Tobi