[llvm] [AMDGPU] Fix unreachable reg bit width (PR #122107)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,12 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5
+# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx90a -run-pass=early-machinelicm -run-pass=postmisched -o - %s | FileCheck %s
----------------
arsenm wrote:

Can do this in one run-pass with , separated pass names. but I'm not sure why you need to run 2 passes, or how either ne stresses this function 

https://github.com/llvm/llvm-project/pull/122107