[Mlir-commits] [mlir] [mlir][spirv] Add 8-bit float type emulation (PR #148811)

[Krzysztof Drewniak]

================
@@ -39,6 +39,10 @@ struct SPIRVConversionOptions {
   /// The number of bits to store a boolean value.
   unsigned boolNumBits{8};
 
+  /// Whether to emulate unsupported floats with integer types of same bit
+  /// width.
+  bool emulateUnsupportedFloatTypes{true};
----------------
krzysz00 wrote:

I think the theory is that you'll generally want to do software emulation of the small floats earlier?

https://github.com/llvm/llvm-project/pull/148811