Assertion `result && "Zero-sized vectors are not allowed"'

[pravg-amd]

For the below IR,

module {
  func.func @fcos_resnet50(%arg0: !torch.vtensor<[],i1>, %arg1: !torch.vtensor<[?,3],si64>) -> !torch.vtensor<[?],si64> attributes {torch.onnx_meta.ir_version = 8 : si64, torch.onnx_meta.opset_version = 21 : si64, torch.onnx_meta.opset_versions = {ai.onnx.ml = 5 : si64, ai.onnx.preview.training = 1 : si64, ai.onnx.training = 1 : si64, com.microsoft = 1 : si64, com.microsoft.experimental = 1 : si64, com.microsoft.nchwc = 1 : si64, org.pytorch.aten = 1 : si64}, torch.onnx_meta.producer_name = "pytorch", torch.onnx_meta.producer_version = "2.1.0"} {
    %1254 = torch.operator "onnx.If"(%arg0) : (!torch.vtensor<[],i1>) -> !torch.vtensor<[?],si64> {
      %1274 = torch.operator "onnx.Constant"() {torch.onnx.value = dense_resource<_3819> : tensor<1xsi64>} : () -> !torch.vtensor<[1],si64>
      %1277 = torch.operator "onnx.Constant"() {torch.onnx.value = dense_resource<_3817> : tensor<1xsi64>} : () -> !torch.vtensor<[1],si64>
      %1292 = torch.operator "onnx.Gather"(%arg1, %1277) {torch.onnx.axis = 1 : si64} : (!torch.vtensor<[?,3],si64>, !torch.vtensor<[1],si64>) -> !torch.vtensor<[?,1],si64>
      %1293 = torch.operator "onnx.Squeeze"(%1292, %1274) : (!torch.vtensor<[?,1],si64>, !torch.vtensor<[1],si64>) -> !torch.vtensor<[?],si64>
      torch.operator_terminator %1293 : !torch.vtensor<[?],si64>
    }, {
      %1273 = torch.operator "onnx.Constant"() {torch.onnx.value = dense_resource<_3790> : tensor<0xsi64>} : () -> !torch.vtensor<[0],si64>
      torch.operator_terminator %1273 : !torch.vtensor<[0],si64>
    }
    return %1254 : !torch.vtensor<[?],si64>
  }
}

{-#
  dialect_resources: {
    builtin: {
      _3819: "0x080000000100000000000000",
      _3817: "0x080000000200000000000000",
      _3790: "0x08000000"
    }
  }
#-}

Getting error as

iree-compile: ~/iree/third_party/llvm-project/mlir/lib/Dialect/Arith/IR/InferIntRangeInterfaceImpls.cpp:57: void mlir::arith::ConstantOp::inferResultRanges(ArrayRef<ConstantIntRanges>, SetIntRangeFn): Assertion `result && "Zero-sized vectors are not allowed"' failed.

Command:

 iree-compile --iree-hal-target-backends=llvm-cpu --iree-llvmcpu-target-cpu=host -o test.vmfb model_torch_onnx.mlir 

Models Impacted:

fcos_resnet50_fpn_Opset16_torchvision
fcos_resnet50_fpn_Opset17_torchvision
retinanet_resnet50_fpn_v2_Opset17_torchvision
retinanet_resnet50_fpn_v2_Opset18_torchvision
retinanet_resnet50_fpn_v2_Opset16_torchvision

[pravg-amd]

The constant in the imported IR doesn't seem to be a valid number

%1273 = torch.operator "onnx.Constant"() {torch.onnx.value = dense_resource<_3790> : tensor<0xsi64>} : () -> !torch.vtensor<[0],si64>

{-#
  dialect_resources: {
    builtin: {
      _3790: "0x08000000"
    }
  }
#-}