[llvm] [AMDGPU] Use subtarget feature for flat offset bit width instead of arch checks (PR #183742)

[Diana Picus via llvm-commits]

================
@@ -59,6 +59,7 @@ class AMDGPUSubtarget {
   unsigned LocalMemorySize = 0;
   unsigned AddressableLocalMemorySize = 0;
   char WavefrontSizeLog2 = 0;
+  unsigned FlatOffsetBitWidth = 13;
----------------
rovka wrote:

Yeah, I think it would be ok to set the default to 0 and update it to 13 in initializeSubtargetDependencies. That would be consistent with a bunch of other features (e.g. AddressableLocalMemorySize).

https://github.com/llvm/llvm-project/pull/183742