[compiler-rt] [builtins] Support building the 128-bit float functions on i386 (PR #122658)

[Saleem Abdulrasool via llvm-commits]

================
@@ -12,7 +12,8 @@
 #define QUAD_PRECISION
 #include "fp_lib.h"
 
-#if defined(CRT_HAS_TF_MODE) && __LDBL_MANT_DIG__ == 64 && defined(__x86_64__)
+#if defined(CRT_HAS_TF_MODE) && __LDBL_MANT_DIG__ == 64 &&                     \
----------------
compnerd wrote:

This change is concerning. Not all 32-bit environments support FP80. Concretely, Windows, FreeBSD, and Darwin definitely do support FP80 even though the hardware does. This can make it easy to get ABI breakages.

https://github.com/llvm/llvm-project/pull/122658