[llvm] [APFloat] Add APFloat support for E8M0 type (PR #107127)

[Durgadoss R via llvm-commits]

================
@@ -396,7 +412,8 @@ static inline Error createError(const Twine &Err) {
 }
 
 static constexpr inline unsigned int partCountForBits(unsigned int bits) {
-  return ((bits) + APFloatBase::integerPartWidth - 1) / APFloatBase::integerPartWidth;
+  return std::max(1u, ((bits) + APFloatBase::integerPartWidth - 1) /
+                          APFloatBase::integerPartWidth);
----------------
durga4github wrote:

Accepted this change in the latest revision,
Resolving.

https://github.com/llvm/llvm-project/pull/107127