Selected stage: loom-lower-scf-to-dfg. Test exactly output obligation mlir-obligation:f9b33846c2d82708b7bf1867 with its governing context; report any stage attr

<stdin>:15:7: error: loom-lower-graph-memory: parallel addresses have no comparable lane projection; address rank is 1
      scf.parallel (%iv0) = (%zero) to (%two) step (%one) {
      ^
<stdin>:15:7: note: see current operation: 
"scf.parallel"(%0, %2, %1) <{operandSegmentSizes = array<i32: 1, 1, 1, 0>}> ({
^bb0(%arg3: index):
  "scf.forall"() <{operandSegmentSizes = array<i32: 0, 0, 0, 0>, staticLowerBound = array<i64: 0>, staticStep = array<i64: 1>, staticUpperBound = array<i64: 2>}> ({
  ^bb0(%arg4: index):
    %5 = "arith.muli"(%arg3, %2) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %6 = "arith.addi"(%5, %arg4) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    "memref.store"(%1, %arg2, %6) : (index, memref<?xindex>, index) -> ()
    "scf.forall.in_parallel"() ({
    ^bb0:
    }) : () -> ()
  }) : () -> ()
  "scf.reduce"() : () -> ()
}) : (index, index, index) -> ()

<stdin>:15:7: error: loom-lower-graph-memory: parallel addresses have no comparable lane projection; address rank is 1
      scf.parallel (%iv0) = (%zero) to (%two) step (%one) {
      ^
<stdin>:15:7: note: see current operation: 
"scf.parallel"(%0, %2, %1) <{operandSegmentSizes = array<i32: 1, 1, 1, 0>}> ({
^bb0(%arg3: index):
  "scf.forall"() <{operandSegmentSizes = array<i32: 0, 0, 0, 0>, staticLowerBound = array<i64: 0>, staticStep = array<i64: 1>, staticUpperBound = array<i64: 2>}> ({
  ^bb0(%arg4: index):
    %3 = "arith.muli"(%arg3, %2) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %4 = "arith.addi"(%3, %arg4) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    "memref.store"(%1, %arg2, %4) : (index, memref<?xindex>, index) -> ()
    "scf.forall.in_parallel"() ({
    ^bb0:
    }) : () -> ()
  }) : () -> ()
  "memref.store"(%1, %arg2, %arg3) : (index, memref<?xindex>, index) -> ()
  "scf.reduce"() : () -> ()
}) : (index, index, index) -> ()

<stdin>:11:7: error: loom-lower-graph-memory: parallel addresses have no comparable lane projection; address rank is 1
      scf.forall (%iv0) in (2) {
      ^
<stdin>:11:7: note: see current operation: 
"scf.forall"() <{operandSegmentSizes = array<i32: 0, 0, 0, 0>, staticLowerBound = array<i64: 0>, staticStep = array<i64: 1>, staticUpperBound = array<i64: 2>}> ({
^bb0(%arg7: index):
  "scf.parallel"(%0, %2, %1) <{operandSegmentSizes = array<i32: 1, 1, 1, 0>}> ({
  ^bb0(%arg8: index):
    %8 = "arith.muli"(%arg7, %2) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %9 = "arith.addi"(%8, %arg8) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %10 = "memref.load"(%arg2, %9) : (memref<?xindex>, index) -> index
    %11 = "arith.addi"(%10, %1) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    "memref.store"(%11, %arg2, %9) : (index, memref<?xindex>, index) -> ()
    "scf.reduce"() : () -> ()
  }) : (index, index, index) -> ()
  "scf.forall.in_parallel"() ({
  ^bb0:
  }) : () -> ()
}) : () -> ()

<stdin>:0:0: error: canonical Dataflow publication failed: graph @g_thread_0 completion witness #0 is not statically one-shot
<stdin>:0:0: note: see current operation: 
"builtin.module"() ({
  "llvm.func"() <{CConv = #llvm.cconv<ccc>, function_type = !llvm.func<i64 (i64)>, linkage = #llvm.linkage<external>, sym_name = "imported_callable", unnamed_addr = 0 : i64, visibility_ = 0 : i64}> ({
  }) : () -> ()
  "func.func"() <{function_type = (i32) -> i32, sym_name = "native_callable", sym_visibility = "private"}> ({
  ^bb0(%arg6: i32):
    "func.return"(%arg6) : (i32) -> ()
  }) : () -> ()
  "dataflow.thread"() <{domain = #dataflow.thread_domain<dense>, function_type = (index, memref<?xindex>) -> (), sym_name = "thread_0", sym_visibility = "private"}> ({
  ^bb0(%arg3: index, %arg4: memref<?xindex>, %arg5: none):
    %80 = "dataflow.graph.launch"(%arg5, %arg3, %arg4) <{callee = @g_thread_0, operandSegmentSizes = array<i32: 1, 1, 0, 1, 0>, resultSegmentSizes = array<i32: 0, 0, 1>, source_maps = []}> : (none, index, memref<?xindex>) -> none
    "dataflow.thread.yield"(%80) : (none) -> ()
  }) : () -> ()
  "dataflow.graph"() <{arg_attrs = [{}, {llvm.noalias}], function_type = (index, memref<?xindex>) -> (), input_segments = array<i32: 1, 0, 1>, result_segments = array<i32: 0, 0, 0>, sym_name = "g_thread_0", sym_visibility = "private"}> ({
  ^bb0(%arg0: none, %arg1: index, %arg2: memref<?xindex>):
    %0 = "dataflow.constant"(%arg0) <{const_value = 0 : index}> : (none) -> index
    %1 = "dataflow.constant"(%arg0) <{const_value = 1 : index}> : (none) -> index
    %2 = "dataflow.constant"(%arg0) <{const_value = 2 : index}> : (none) -> index
    %3 = "arith.index_cast"(%0) : (index) -> i32
    %4 = "arith.index_cast"(%2) : (index) -> i32
    %5 = "arith.index_cast"(%1) : (index) -> i32
    %6:2 = "dataflow.stream"(%3, %4, %5) <{predicate = 2 : i64, step_kind = 0 : i32}> : (i32, i32, i32) -> (i32, i1)
    %7 = "dataflow.carry"(%6#1, %arg0, %8#1) : (i1, none, none) -> none
    %8:2 = "dataflow.demux"(%6#1, %7) : (i1, none) -> (none, none)
    %9 = "dataflow.invariant"(%6#1, %0) : (i1, index) -> index
    %10:2 = "dataflow.gate"(%6#1, %9) : (i1, index) -> (i1, index)
    %11:2 = "dataflow.demux"(%10#0, %10#1) : (i1, index) -> (index, index)
    %12 = "dataflow.invariant"(%6#1, %1) : (i1, index) -> index
    %13:2 = "dataflow.gate"(%6#1, %12) : (i1, index) -> (i1, index)
    %14:2 = "dataflow.demux"(%13#0, %13#1) : (i1, index) -> (index, index)
    %15 = "dataflow.carry"(%6#1, %arg0, %23) : (i1, none, none) -> none
    %16 = "dataflow.carry"(%6#1, %arg0, %23) : (i1, none, none) -> none
    %17:2 = "dataflow.demux"(%6#1, %15) : (i1, none) -> (none, none)
    %18:2 = "dataflow.demux"(%6#1, %16) : (i1, none) -> (none, none)
    %19:2 = "dataflow.sync"(%8#1, %17#1) : (none, none) -> (none, none)
    %20:2 = "dataflow.load"(%arg2, %10#1, %19#0) : (memref<?xindex>, index, none) -> (index, none)
    %21:2 = "dataflow.sync"(%18#1, %20#1) : (none, none) -> (none, none)
    %22 = "arith.addi"(%20#0, %13#1) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %23 = "dataflow.store"(%arg2, %10#1, %22, %21#0) : (memref<?xindex>, index, index, none) -> none
    %24 = "arith.cmpi"(%3, %4) <{predicate = 2 : i64}> : (i32, i32) -> i1
    %25:2 = "dataflow.demux"(%24, %8#0) : (i1, none) -> (none, none)
    %26:3 = "dataflow.sync"(%25#1, %11#0, %14#0) : (none, index, index) -> (none, index, index)
    %27 = "dataflow.mux"(%24, %25#0, %26#0) : (i1, none, none) -> none
    %28 = "arith.index_cast"(%0) : (index) -> i32
    %29 = "arith.index_cast"(%2) : (index) -> i32
    %30 = "arith.index_cast"(%1) : (index) -> i32
    %31:2 = "dataflow.stream"(%28, %29, %30) <{predicate = 2 : i64, step_kind = 0 : i32}> : (i32, i32, i32) -> (i32, i1)
    %32 = "dataflow.carry"(%31#1, %27, %33#1) : (i1, none, none) -> none
    %33:2 = "dataflow.demux"(%31#1, %32) : (i1, none) -> (none, none)
    %34 = "dataflow.invariant"(%31#1, %0) : (i1, index) -> index
    %35:2 = "dataflow.gate"(%31#1, %34) : (i1, index) -> (i1, index)
    %36:2 = "dataflow.demux"(%35#0, %35#1) : (i1, index) -> (index, index)
    %37 = "dataflow.invariant"(%31#1, %1) : (i1, index) -> index
    %38:2 = "dataflow.gate"(%31#1, %37) : (i1, index) -> (i1, index)
    %39:2 = "dataflow.demux"(%38#0, %38#1) : (i1, index) -> (index, index)
    %40 = "dataflow.carry"(%31#1, %17#0, %48) : (i1, none, none) -> none
    %41 = "dataflow.carry"(%31#1, %18#0, %48) : (i1, none, none) -> none
    %42:2 = "dataflow.demux"(%31#1, %40) : (i1, none) -> (none, none)
    %43:2 = "dataflow.demux"(%31#1, %41) : (i1, none) -> (none, none)
    %44:2 = "dataflow.sync"(%33#1, %42#1) : (none, none) -> (none, none)
    %45:2 = "dataflow.load"(%arg2, %35#1, %44#0) : (memref<?xindex>, index, none) -> (index, none)
    %46:2 = "dataflow.sync"(%43#1, %45#1) : (none, none) -> (none, none)
    %47 = "arith.addi"(%45#0, %38#1) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %48 = "dataflow.store"(%arg2, %35#1, %47, %46#0) : (memref<?xindex>, index, index, none) -> none
    %49 = "arith.cmpi"(%28, %29) <{predicate = 2 : i64}> : (i32, i32) -> i1
    %50:2 = "dataflow.demux"(%49, %33#0) : (i1, none) -> (none, none)
    %51:3 = "dataflow.sync"(%50#1, %36#0, %39#0) : (none, index, index) -> (none, index, index)
    %52 = "dataflow.mux"(%49, %50#0, %51#0) : (i1, none, none) -> none
    %53 = "arith.index_cast"(%0) : (index) -> i32
    %54 = "arith.index_cast"(%2) : (index) -> i32
    %55 = "arith.index_cast"(%1) : (index) -> i32
    %56:2 = "dataflow.stream"(%53, %54, %55) <{predicate = 2 : i64, step_kind = 0 : i32}> : (i32, i32, i32) -> (i32, i1)
    %57 = "dataflow.carry"(%56#1, %27, %58#1) : (i1, none, none) -> none
    %58:2 = "dataflow.demux"(%56#1, %57) : (i1, none) -> (none, none)
    %59 = "dataflow.invariant"(%56#1, %1) : (i1, index) -> index
    %60:2 = "dataflow.gate"(%56#1, %59) : (i1, index) -> (i1, index)
    %61:2 = "dataflow.demux"(%60#0, %60#1) : (i1, index) -> (index, index)
    %62 = "dataflow.invariant"(%56#1, %1) : (i1, index) -> index
    %63:2 = "dataflow.gate"(%56#1, %62) : (i1, index) -> (i1, index)
    %64:2 = "dataflow.demux"(%63#0, %63#1) : (i1, index) -> (index, index)
    %65 = "dataflow.carry"(%56#1, %17#0, %73) : (i1, none, none) -> none
    %66 = "dataflow.carry"(%56#1, %18#0, %73) : (i1, none, none) -> none
    %67:2 = "dataflow.demux"(%56#1, %65) : (i1, none) -> (none, none)
    %68:2 = "dataflow.demux"(%56#1, %66) : (i1, none) -> (none, none)
    %69:2 = "dataflow.sync"(%58#1, %67#1) : (none, none) -> (none, none)
    %70:2 = "dataflow.load"(%arg2, %60#1, %69#0) : (memref<?xindex>, index, none) -> (index, none)
    %71:2 = "dataflow.sync"(%68#1, %70#1) : (none, none) -> (none, none)
    %72 = "arith.addi"(%70#0, %63#1) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %73 = "dataflow.store"(%arg2, %60#1, %72, %71#0) : (memref<?xindex>, index, index, none) -> none
    %74 = "arith.cmpi"(%53, %54) <{predicate = 2 : i64}> : (i32, i32) -> i1
    %75:2 = "dataflow.demux"(%74, %58#0) : (i1, none) -> (none, none)
    %76:3 = "dataflow.sync"(%75#1, %61#0, %64#0) : (none, index, index) -> (none, index, index)
    %77 = "dataflow.mux"(%74, %75#0, %76#0) : (i1, none, none) -> none
    %78:2 = "dataflow.sync"(%52, %77) : (none, none) -> (none, none)
    %79:2 = "dataflow.sync"(%43#0, %68#0) : (none, none) -> (none, none)
    "dataflow.graph.return"(%78#0, %79#0) <{operandSegmentSizes = array<i32: 0, 0, 0, 2>}> : (none, none) -> ()
  }) : () -> ()
}) : () -> ()

<stdin>:11:7: error: loom-lower-graph-memory: parallel addresses have no comparable lane projection; address rank is 1
      scf.forall (%iv0) in (2) {
      ^
<stdin>:11:7: note: see current operation: 
"scf.forall"() <{operandSegmentSizes = array<i32: 0, 0, 0, 0>, staticLowerBound = array<i64: 0>, staticStep = array<i64: 1>, staticUpperBound = array<i64: 2>}> ({
^bb0(%arg4: index):
  "scf.forall"() <{operandSegmentSizes = array<i32: 0, 0, 0, 0>, staticLowerBound = array<i64: 0>, staticStep = array<i64: 1>, staticUpperBound = array<i64: 2>}> ({
  ^bb0(%arg5: index):
    %4 = "arith.muli"(%arg4, %1) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %5 = "arith.addi"(%4, %arg5) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    %6 = "memref.load"(%arg2, %5) : (memref<?xindex>, index) -> index
    %7 = "arith.addi"(%6, %0) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
    "memref.store"(%7, %arg2, %5) : (index, memref<?xindex>, index) -> ()
    "scf.forall.in_parallel"() ({
    ^bb0:
    }) : () -> ()
  }) : () -> ()
  %2 = "memref.load"(%arg2, %arg4) : (memref<?xindex>, index) -> index
  %3 = "arith.addi"(%2, %0) <{overflowFlags = #arith.overflow<none>}> : (index, index) -> index
  "memref.store"(%3, %arg2, %arg4) : (index, memref<?xindex>, index) -> ()
  "scf.forall.in_parallel"() ({
  ^bb0:
  }) : () -> ()
}) : () -> ()