AMDGPU: Add a test to show how later optimization works

llvm/test/CodeGen/AMDGPU/while-break.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -march=amdgcn -mcpu=gfx1010 -verify-machineinstrs < %s | FileCheck %s -check-prefix=GCN
+
+define amdgpu_ps float @while_break(i32 %z, float %v, i32 %x, i32 %y) #0 {
+; GCN-LABEL: while_break:
+; GCN:       ; %bb.0: ; %entry
+; GCN-NEXT:    s_mov_b32 s1, -1
+; GCN-NEXT:    s_mov_b32 s0, 0
+; GCN-NEXT:    s_branch .LBB0_2
+; GCN-NEXT:  .LBB0_1: ; %Flow2
+; GCN-NEXT:    ; in Loop: Header=BB0_2 Depth=1
+; GCN-NEXT:    s_or_b32 exec_lo, exec_lo, s4
+; GCN-NEXT:    v_mov_b32_e32 v4, v1
+; GCN-NEXT:    v_mov_b32_e32 v1, v5
+; GCN-NEXT:    s_and_b32 s2, exec_lo, s3
+; GCN-NEXT:    s_or_b32 s0, s2, s0
+; GCN-NEXT:    s_andn2_b32 exec_lo, exec_lo, s0
+; GCN-NEXT:    s_cbranch_execz .LBB0_8
+; GCN-NEXT:  .LBB0_2: ; %header
+; GCN-NEXT:    ; =>This Inner Loop Header: Depth=1
+; GCN-NEXT:    s_add_i32 s1, s1, 1
+; GCN-NEXT:    s_mov_b32 s2, 0
+; GCN-NEXT:    v_cmp_ge_i32_e32 vcc_lo, s1, v2
+; GCN-NEXT:    s_and_saveexec_b32 s3, vcc_lo
+; GCN-NEXT:    s_xor_b32 s3, exec_lo, s3
+; GCN-NEXT:  ; %bb.3: ; %else
+; GCN-NEXT:    ; in Loop: Header=BB0_2 Depth=1
+; GCN-NEXT:    v_cmp_lt_i32_e32 vcc_lo, s1, v3
+; GCN-NEXT:    s_and_b32 s2, vcc_lo, exec_lo
+; GCN-NEXT:  ; %bb.4: ; %Flow
+; GCN-NEXT:    ; in Loop: Header=BB0_2 Depth=1
+; GCN-NEXT:    s_or_saveexec_b32 s3, s3
+; GCN-NEXT:    v_mov_b32_e32 v5, v1
+; GCN-NEXT:    s_xor_b32 exec_lo, exec_lo, s3
+; GCN-NEXT:  ; %bb.5: ; %if
+; GCN-NEXT:    ; in Loop: Header=BB0_2 Depth=1
+; GCN-NEXT:    v_add_f32_e32 v5, 1.0, v1
+; GCN-NEXT:    s_or_b32 s2, s2, exec_lo
+; GCN-NEXT:  ; %bb.6: ; %Flow1
+; GCN-NEXT:    ; in Loop: Header=BB0_2 Depth=1
+; GCN-NEXT:    s_or_b32 exec_lo, exec_lo, s3
+; GCN-NEXT:    s_mov_b32 s3, -1
+; GCN-NEXT:    s_and_saveexec_b32 s4, s2
+; GCN-NEXT:    s_cbranch_execz .LBB0_1
+; GCN-NEXT:  ; %bb.7: ; %latch
+; GCN-NEXT:    ; in Loop: Header=BB0_2 Depth=1
+; GCN-NEXT:    v_cmp_lt_i32_e32 vcc_lo, s1, v0
+; GCN-NEXT:    v_mov_b32_e32 v1, v5
+; GCN-NEXT:    s_orn2_b32 s3, vcc_lo, exec_lo
+; GCN-NEXT:    s_branch .LBB0_1
+; GCN-NEXT:  .LBB0_8: ; %end
+; GCN-NEXT:    s_or_b32 exec_lo, exec_lo, s0
+; GCN-NEXT:    v_mov_b32_e32 v0, v4
+; GCN-NEXT:    ; return to shader part epilog
+entry:
+  br label %header
+
+header:
+  %v.1 = phi float [ %v, %entry ], [ %v.2, %latch ]
+  %ind = phi i32 [ 0, %entry], [ %ind.inc, %latch ]
+  %cc = icmp slt i32 %ind, %x
+  br i1 %cc, label %if, label %else
+
+if:
+  %v.if = fadd float %v.1, 1.0
+  br label %latch
+
+else:
+  %cc2 = icmp slt i32 %ind, %y
+  br i1 %cc2, label %latch, label %end
+
+latch:
+  %v.2 = phi float [ %v.if, %if ], [ %v.1, %else ]
+  %ind.inc = add i32 %ind, 1
+  %cc3 = icmp slt i32 %ind, %z
+  br i1 %cc3, label %end, label %header
+
+end:
+  %r = phi float [ %v.2, %latch ], [ %v.1, %else ]
+  ret float %r
+}
+
+; Just different dfs order from while_break.
+define amdgpu_ps float @while_break2(i32 %z, float %v, i32 %x, i32 %y) #0 {
+; GCN-LABEL: while_break2:
+; GCN:       ; %bb.0: ; %entry
+; GCN-NEXT:    s_mov_b32 s1, -1
+; GCN-NEXT:    s_mov_b32 s0, 0
+; GCN-NEXT:    s_branch .LBB1_2
+; GCN-NEXT:  .LBB1_1: ; %Flow2
+; GCN-NEXT:    ; in Loop: Header=BB1_2 Depth=1
+; GCN-NEXT:    s_or_b32 exec_lo, exec_lo, s4
+; GCN-NEXT:    v_mov_b32_e32 v4, v1
+; GCN-NEXT:    v_mov_b32_e32 v1, v5
+; GCN-NEXT:    s_and_b32 s2, exec_lo, s3
+; GCN-NEXT:    s_or_b32 s0, s2, s0
+; GCN-NEXT:    s_andn2_b32 exec_lo, exec_lo, s0
+; GCN-NEXT:    s_cbranch_execz .LBB1_8
+; GCN-NEXT:  .LBB1_2: ; %header
+; GCN-NEXT:    ; =>This Inner Loop Header: Depth=1
+; GCN-NEXT:    s_add_i32 s1, s1, 1
+; GCN-NEXT:    s_mov_b32 s2, 0
+; GCN-NEXT:    v_cmp_ge_i32_e32 vcc_lo, s1, v2
+; GCN-NEXT:    ; implicit-def: $vgpr5
+; GCN-NEXT:    s_and_saveexec_b32 s3, vcc_lo
+; GCN-NEXT:    s_xor_b32 s3, exec_lo, s3
+; GCN-NEXT:  ; %bb.3: ; %if
+; GCN-NEXT:    ; in Loop: Header=BB1_2 Depth=1
+; GCN-NEXT:    v_add_f32_e32 v5, 1.0, v1
+; GCN-NEXT:    s_mov_b32 s2, exec_lo
+; GCN-NEXT:  ; %bb.4: ; %Flow
+; GCN-NEXT:    ; in Loop: Header=BB1_2 Depth=1
+; GCN-NEXT:    s_andn2_saveexec_b32 s3, s3
+; GCN-NEXT:  ; %bb.5: ; %else
+; GCN-NEXT:    ; in Loop: Header=BB1_2 Depth=1
+; GCN-NEXT:    v_cmp_lt_i32_e32 vcc_lo, s1, v3
+; GCN-NEXT:    v_mov_b32_e32 v5, v1
+; GCN-NEXT:    s_andn2_b32 s2, s2, exec_lo
+; GCN-NEXT:    s_and_b32 s4, vcc_lo, exec_lo
+; GCN-NEXT:    s_or_b32 s2, s2, s4
+; GCN-NEXT:  ; %bb.6: ; %Flow1
+; GCN-NEXT:    ; in Loop: Header=BB1_2 Depth=1
+; GCN-NEXT:    s_or_b32 exec_lo, exec_lo, s3
+; GCN-NEXT:    s_mov_b32 s3, -1
+; GCN-NEXT:    s_and_saveexec_b32 s4, s2
+; GCN-NEXT:    s_cbranch_execz .LBB1_1
+; GCN-NEXT:  ; %bb.7: ; %latch
+; GCN-NEXT:    ; in Loop: Header=BB1_2 Depth=1
+; GCN-NEXT:    v_cmp_lt_i32_e32 vcc_lo, s1, v0
+; GCN-NEXT:    v_mov_b32_e32 v1, v5
+; GCN-NEXT:    s_orn2_b32 s3, vcc_lo, exec_lo
+; GCN-NEXT:    s_branch .LBB1_1
+; GCN-NEXT:  .LBB1_8: ; %end
+; GCN-NEXT:    s_or_b32 exec_lo, exec_lo, s0
+; GCN-NEXT:    v_mov_b32_e32 v0, v4
+; GCN-NEXT:    ; return to shader part epilog
+entry:
+  br label %header
+
+header:
+  %v.1 = phi float [ %v, %entry ], [ %v.2, %latch ]
+  %ind = phi i32 [ 0, %entry], [ %ind.inc, %latch ]
+  %cc = icmp slt i32 %ind, %x
+  br i1 %cc, label %else, label %if
+
+if:
+  %v.if = fadd float %v.1, 1.0
+  br label %latch
+
+else:
+  %cc2 = icmp slt i32 %ind, %y
+  br i1 %cc2, label %latch, label %end
+
+latch:
+  %v.2 = phi float [ %v.if, %if ], [ %v.1, %else ]
+  %ind.inc = add i32 %ind, 1
+  %cc3 = icmp slt i32 %ind, %z
+  br i1 %cc3, label %end, label %header
+
+end:
+  %r = phi float [ %v.2, %latch ], [ %v.1, %else ]
+  ret float %r
+}
+
+attributes #0 = { nounwind }