[llvm] [AMDGPU][True16][CodeGen] use vgpr16 for zext patterns (reopen #153894) (PR #154211)

[Brox Chen via llvm-commits]

================
@@ -0,0 +1,207 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN:  llc -amdgpu-scalarize-global-loads=false  -mtriple=amdgcn -mcpu=gfx1100 -mattr=+real-true16 < %s | FileCheck -check-prefixes=GFX11,GFX11-REAL16 %s
+; RUN:  llc -amdgpu-scalarize-global-loads=false  -mtriple=amdgcn -mcpu=gfx1100 -mattr=-real-true16 < %s | FileCheck -check-prefixes=GFX11,GFX11-FAKE16 %s
+
+define amdgpu_kernel void @zextload_global_i18_to_i16(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
----------------
broxigarchen wrote:

fixed

https://github.com/llvm/llvm-project/pull/154211