[llvm] [PHIElim][AMDGPU]: shrink source subranges to lane-specific uses (PR #228073)

[Matt Arsenault via llvm-commits]

================
@@ -306,6 +311,19 @@ bool PHIEliminationImpl::run(MachineFunction &MF) {
   }
 
   LoweredPHIs.clear();
+
+  // Different lanes may be used by different PHI source copies, or may already
+  // be dead in a predecessor. The main range's last use is therefore not a
+  // valid endpoint for every subrange. Wait until all PHIs have been removed
+  // before shrinking subranges to their remaining lane-specific uses.
+  for (Register Reg : PHISrcRegsToShrink) {
----------------
arsenm wrote:

This looks like the brute force version of #226209

https://github.com/llvm/llvm-project/pull/228073