[AMDGPU] Autogenerate R600 packetizer checks (#166570)

Author: jayfoad

llvm/test/CodeGen/AMDGPU/packetizer.ll

@@ -1,13 +1,49 @@
-; RUN: llc < %s -mtriple=r600 -mcpu=redwood | FileCheck %s
-; RUN: llc < %s -mtriple=r600 -mcpu=cayman | FileCheck %s
-
-; CHECK: {{^}}test:
-; CHECK: BIT_ALIGN_INT T{{[0-9]}}.X
-; CHECK: BIT_ALIGN_INT T{{[0-9]}}.Y
-; CHECK: BIT_ALIGN_INT T{{[0-9]}}.Z
-; CHECK: BIT_ALIGN_INT * T{{[0-9]}}.W
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
+; RUN: llc < %s -mtriple=r600 -mcpu=redwood | FileCheck %s -check-prefix=R600
+; RUN: llc < %s -mtriple=r600 -mcpu=cayman | FileCheck %s -check-prefix=CM
 
 define amdgpu_kernel void @test(ptr addrspace(1) %out, i32 %x_arg, i32 %y_arg, i32 %z_arg, i32 %w_arg, i32 %e) {
+; R600-LABEL: test:
+; R600:       ; %bb.0: ; %entry
+; R600-NEXT:    ALU 12, @4, KC0[CB0:0-32], KC1[]
+; R600-NEXT:    MEM_RAT_CACHELESS STORE_RAW T0.X, T1.X, 1
+; R600-NEXT:    CF_END
+; R600-NEXT:    PAD
+; R600-NEXT:    ALU clause starting at 4:
+; R600-NEXT:     ADD_INT T0.Y, KC0[3].X, 1,
+; R600-NEXT:     ADD_INT T0.Z, KC0[3].Y, 1,
+; R600-NEXT:     ADD_INT T0.W, KC0[2].Z, 1,
+; R600-NEXT:     ADD_INT * T1.W, KC0[2].W, 1,
+; R600-NEXT:     BIT_ALIGN_INT T0.X, PS, PS, KC0[3].Z,
+; R600-NEXT:     BIT_ALIGN_INT T1.Y, PV.W, PV.W, KC0[3].Z,
+; R600-NEXT:     BIT_ALIGN_INT T0.Z, PV.Z, PV.Z, KC0[3].Z,
+; R600-NEXT:     BIT_ALIGN_INT * T0.W, PV.Y, PV.Y, KC0[3].Z,
+; R600-NEXT:     OR_INT T0.W, PV.W, PV.Z,
+; R600-NEXT:     OR_INT * T1.W, PV.Y, PV.X,
+; R600-NEXT:     OR_INT T0.X, PS, PV.W,
+; R600-NEXT:     LSHR * T1.X, KC0[2].Y, literal.x,
+; R600-NEXT:    2(2.802597e-45), 0(0.000000e+00)
+;
+; CM-LABEL: test:
+; CM:       ; %bb.0: ; %entry
+; CM-NEXT:    ALU 12, @4, KC0[CB0:0-32], KC1[]
+; CM-NEXT:    MEM_RAT_CACHELESS STORE_DWORD T0.X, T1.X
+; CM-NEXT:    CF_END
+; CM-NEXT:    PAD
+; CM-NEXT:    ALU clause starting at 4:
+; CM-NEXT:     ADD_INT T0.X, KC0[3].X, 1,
+; CM-NEXT:     ADD_INT T0.Y, KC0[3].Y, 1,
+; CM-NEXT:     ADD_INT T0.Z, KC0[2].Z, 1,
+; CM-NEXT:     ADD_INT * T0.W, KC0[2].W, 1,
+; CM-NEXT:     BIT_ALIGN_INT T1.X, PV.W, PV.W, KC0[3].Z,
+; CM-NEXT:     BIT_ALIGN_INT T1.Y, PV.Z, PV.Z, KC0[3].Z,
+; CM-NEXT:     BIT_ALIGN_INT T0.Z, PV.Y, PV.Y, KC0[3].Z,
+; CM-NEXT:     BIT_ALIGN_INT * T0.W, PV.X, PV.X, KC0[3].Z,
+; CM-NEXT:     OR_INT T0.Z, PV.W, PV.Z,
+; CM-NEXT:     OR_INT * T0.W, PV.Y, PV.X,
+; CM-NEXT:     OR_INT * T0.X, PV.W, PV.Z,
+; CM-NEXT:     LSHR * T1.X, KC0[2].Y, literal.x,
+; CM-NEXT:    2(2.802597e-45), 0(0.000000e+00)
 entry:
   %shl = sub i32 32, %e
   %x = add i32 %x_arg, 1