[libc-commits] [libc] [libc][stdfix] Implement sqrtfx for remaining fixed point types (PR #214741)

[via libc-commits]

================
@@ -131,8 +132,25 @@ template <> struct SqrtConfig<unsigned int> {
   using HalfType = unsigned short;
 };
 
-// TODO: unsigned long accum type is 64-bit, and will need 64-bit fract type.
-// Probably we will use DyadicFloat<64> for intermediate computations instead.
+template <typename T> struct FXUnsigned;
----------------
sohail103 wrote:

Thanks, I've removed the signed entrypoints.

https://github.com/llvm/llvm-project/pull/214741