On 9/13/21 11:07 AM, Tobias Burnus wrote:
> On 13.09.21 18:59, Sandra Loosemore wrote:
>> On 9/13/21 10:51 AM, Jakub Jelinek wrote: >>> Wouldn't it be better to use the __LDBL_* macros anyway and not rely on
>>> float.h?  The header doesn't want to test what float.h tells about the
>>> long double type, but what the compiler knows about it.
>> I originally wrote the code to use the internal GCC __LDBL_* macros as 
>> you suggest, but Tobias complained that then the gfortran-provided .h 
>> file could not be used to compile the C parts of the program with some 
>> other C compiler. 
> For instance, clang does not seem to provide those - and in some cases, 
> it can be useful to mix gfortran code with code complied by other 
> compilers (icc, clang, ...).
>> Maybe it needs to first check the internal macros and then look for 
>> the float.h versions if it can't find them?
> 
> I think that makes sense. (Adding a comment that #include <float.h> is 
> for non-GCC compilers, only.)

Here's a patch.  Gerald, can you check that this fixes your bootstrap 
problem on i586-unknown-freebsd11?

-Sandra