[llvm] [AMDGPU] SelectionDAG divergence tracking should take into account Target divergency. (PR #144947)

[Jay Foad via llvm-commits]

================
@@ -0,0 +1,32 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+
+declare i32 @llvm.amdgcn.workitem.id.x()
+
+
+define amdgpu_kernel void @test_isel_single_lane(ptr addrspace(1) %in, ptr addrspace(1) %out) #0 {
+; GCN-LABEL: test_isel_single_lane:
+; GCN:       ; %bb.0:
+; GCN-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
+; GCN-NEXT:    v_mov_b32_e32 v0, 0
+; GCN-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN-NEXT:    s_load_dword s0, s[0:1], 0x58
+; GCN-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN-NEXT:    s_lshl_b32 s0, s0, 4
+; GCN-NEXT:    v_mov_b32_e32 v1, s0
+; GCN-NEXT:    global_store_dword v0, v1, s[2:3]
+; GCN-NEXT:    s_endpgm
+	%tid = call i32 @llvm.amdgcn.workitem.id.x()
----------------
jayfoad wrote:

This test passes even without your patch. I guess the workitem call gets simplified to 0?

https://github.com/llvm/llvm-project/pull/144947