[llvm] [AMDGPU] Allow folding of non-subregs through REG_SEQUENCE (PR #151033)

[Matt Arsenault via llvm-commits]

================
@@ -172,3 +172,19 @@ entry:
   %qm = call i64 @llvm.amdgcn.s.quadmask.i64(i64 %mask)
   ret i64 %qm
 }
+
+define amdgpu_ps void @test_quadmask_half_poison_i64(i32 %in, ptr %out) {
----------------
arsenm wrote:

Reference issue number 

https://github.com/llvm/llvm-project/pull/151033