[Bug target/92083] -mlong-double-64 strange results

[joseph at codesourcery dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92083

--- Comment #2 from joseph at codesourcery dot com <joseph at codesourcery dot com> ---
Note also that glibc does not support being built with a different long 
double ABI from the default one.  On architectures where more than one 
long double format is supported by glibc, either a format is paired with 
other ABI choices (mips, where long double is binary64 for the o32 ABI and 
binary128 for the other two ABIs), or more than one long double ABI is 
supported in the same glibc build and the headers automatically select 
different functions accordingly (alpha, powerpc, s390, 32-bit sparc; in 
the powerpc64le case, work is underway to support binary128 as a third 
long double format in a single glibc build, but much of that is not yet in 
glibc).  x86 is not one of those architectures, so alternative long double 
formats are not supported for it in glibc at all.  (Before NaCl died, I 
think Roland planned to add support for different formats for use by the 
x86 NaCl port, but that never got merged and all the Arm NaCl support was 
removed.)