[libc-commits] [compiler-rt] [libc] [compiler-rt][builtins] libc-backed floating-point extend/trunct builtins (PR #209984)

[via libc-commits]

================
@@ -243,6 +243,27 @@ set(GENERIC_TF_SOURCES
   trunctfsf2.c
 )
 
+# Implement extended-precision builtins, assuming long double is 80 bits.
+# long double is not 80 bits on Android or MSVC.
+set(x86_80_BIT_SOURCES
----------------
hulxv wrote:

https://github.com/hulxv/llvm-project/blob/a4c7c2f665ec7ebdf9b5613bc5fc2c40ac31985a/compiler-rt/lib/builtins/CMakeLists.txt#L248

https://github.com/llvm/llvm-project/pull/209984