As 'processing' a signalling NaN may convert it to a quiet NaN. Apparently
happening on 32-bit x86 using the x87 FPU, but not on x86_64 with SSE.
This fixes potential issues with different constants being used when
mixing natively and cross-compiled objects/libs and failing bitwise `is`
comparisons etc.
See https://forum.dlang.org/post/nsp1ql$ivu$1@digitalmars.com for Martin
Nowak's summarized findings.
While parsing of floating-point literals and CTFE still operate with the
host LDC's real type, compile-time reals can in principle be emitted in
arbitrary precision via LLVM software conversion, therefore paving the way
for cross-compilation to all targets.
The representable constants are still limited by the compile-time real_t
precision. E.g., LDC on Windows with its 64-bit reals can't hold and emit
an 80-bit `real.max` when cross-compiling to a non-Windows x86(_64)
target; the compile-time value will silently overflow to infinity and
later be emitted as 80-bit infinity.
LDC on AArch64 with its 128-bit quad-precision reals on the other hand can
hold and emit reals for all targets, making it a universal cross-compiler
with quad-precision compile-time reals in hardware.
We don't use the strange 2x64-bit PPC double-double format (see
`getRealType()` in `ir/irtype.cpp`), but would more or less support it
(the type properties (max, min_normal...) still need to be determined;
LLVM isn't sure about those either...).
