[clang] [llvm] [HLSL][SPIRV][DXIL] Implement `dot4add_i8packed` intrinsic (PR #113623)

[Finn Plummer via llvm-commits]

================
@@ -1694,6 +1698,84 @@ bool SPIRVInstructionSelector::selectIntegerDot(Register ResVReg,
   return Result;
 }
 
+// Since pre-1.6 SPIRV has no DotProductInput4x8BitPacked implementation,
+// extract the elements of the packed inputs, multiply them and add the result
+// to the accumulator.
----------------
inbelic wrote:

Okay sounds good thanks. In terms of when the operation is not available, do you agree with the current expansion implementation?

https://github.com/llvm/llvm-project/pull/113623