[llvm] [AMDGPU][True16][MC] VOP2 update instructions with fake16 format (PR #114436)

[Ivan Kosarev via llvm-commits]

================
@@ -988,16 +998,17 @@ let FPDPRounding = 1, isReMaterializable = 1, FixedSize = 1 in {
 let SubtargetPredicate = isGFX10Plus, True16Predicate = NotHasTrue16BitInsts in {
 def V_FMAMK_F16 : VOP2_Pseudo <"v_fmamk_f16", VOP_MADMK_F16, [], "">;
 }
-let SubtargetPredicate = HasTrue16BitInsts in {
-def V_FMAMK_F16_t16 : VOP2_Pseudo <"v_fmamk_f16_t16", VOP_MADMK_F16_t16, [], "">;
+let True16Predicate = UseFakeTrue16Insts in {
+ def V_FMAMK_F16_fake16 : VOP2_Pseudo <"v_fmamk_f16_fake16", VOP_MADMK_F16_fake16, [], "">;
----------------
kosarev wrote:

Is this supposed to be a two-space-character indentation? We don't seem to indent the defs just before and after this snippet, so maybe stick to how it is downstream and remove it?

+Same below.

https://github.com/llvm/llvm-project/pull/114436