# RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-unconditionally-use-set-output-shapes-on-import -tf-enable-shape-inference-on-import=false -tf-graph-as-function %s -o -  | FileCheck %s
# RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-unconditionally-use-set-output-shapes-on-import -tf-enable-shape-inference-on-import=true -tf-graph-as-function %s -o - | FileCheck %s

# Verify importing with _output_shapes enabled works as expected.

node {
  name: "_Arg"

        //
        //   shape = RandomUniformInt();
        //   reshape = Reshape(input, shape)
        //
        // Both RandomUniformInt and Reshape are compilable by XLA so, absent
        // any other reason, we will try to put both shape and reshape in the
        // same cluster.  However, since XLA only supports statically shaped

  op: "InfeedDequeueTuple"
  attr {
    key: "shapes"
    value {
      list {
        shape {
          dim {
            size: -1
          }
          dim {
            size: -1
          }
          dim {
            size: -1
          }
        }
        shape {
          unknown_rank: true
        }
        shape {
          dim {
            size: 7
          }

  {
    GraphDefBuilder shape2(GraphDefBuilder::kFailImmediately);
    Node* key_constant = KeyPlaceholder("F1", shape2.opts());
    Node* recv1 = RecvAtHost(
        ops::NodeOut(key_constant, 0), "F1", "F1", "O1", {DT_FLOAT, DT_FLOAT},
        shape2.opts().WithAttr(kXlaHasHostTransferAttrName, true));
    Node* e = Binary(ops::NodeOut(recv1, 0), ops::NodeOut(recv1, 1),
                     shape2.opts()

// RUN: tf-mlir-translate -mlir-tf-to-hlo-text %s -tf-input-shapes=10,17:17,19 -tf-xla-emit-use-tuple-args -tf-xla-emit-return-tuple | FileCheck %s
// RUN: tf-mlir-translate -mlir-tf-to-hlo-text %s -tf-input-shapes=10,17:17,19 | FileCheck -check-prefix=NO_TUPLES %s
// RUN: tf-mlir-translate -mlir-tf-to-hlo-text-via-builder %s -tf-input-shapes=10,17:17,19 | FileCheck -check-prefix=NO_TUPLES %s

module attributes {tf.versions = {producer = 179 : i32}} {

// RUN: tf-opt -split-input-file -verify-diagnostics -tf-tpu-annotate-dynamic-shape-inputs %s | FileCheck %s

// Test that annotate the inputs of the cluster func to be dynamic shaped.

module attributes {tf.devices = ["/job:worker/replica:0/task:0/device:CPU:0", "/job:worker/replica:0/task:0/device:TPU_SYSTEM:0", "/job:worker/replica:0/task:0/device:TPU:0"]} {
   func.func @main(
      %arg0: tensor<2048xi64> {tf.device = "/job:localhost/replica:0/task:0/device:CPU:0"},

# RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-shape-inference-on-import=false %s -o - | FileCheck %s

node {
  name: "input0"
  op: "Placeholder"
  attr {
    key: "dtype"
    value {
      type: DT_INT32
    }
  }
}
versions {
  producer: 29
  min_consumer: 12
}

# Verify that functions from the library are properly imported.

# CHECK-LABEL:  func @main() {
# CHECK:    "tf.Placeholder"

// CHECK-NEXT:     value {
// CHECK-NEXT:       type: DT_INT32
// CHECK-NEXT:     }
// CHECK:          key: "_output_shapes"
// CHECK-NEXT:     value {
// CHECK-NEXT:       list {
// CHECK-NEXT:         shape {
// CHECK-NEXT:           dim {
// CHECK-NEXT:             size: 10
// CHECK:        name: "output0"
// CHECK-NEXT:   op: "_Retval"
// CHECK-NEXT:   input: "input0"
// CHECK-NEXT:   attr {
// CHECK-NEXT:     key: "T"

Matthew Dempsky <******@****.***> 1646087539 -0800

// RUN: tf-mlir-translate -mlir-tf-to-hlo-text %s -tf-input-shapes=8,16,16,64:64 -tf-xla-emit-use-tuple-args -tf-xla-emit-return-tuple | FileCheck %s
// RUN: tf-mlir-translate -mlir-tf-to-hlo-text-via-builder %s -tf-input-shapes=8,16,16,64:64 | FileCheck %s

module attributes {tf.versions = {producer = 179 : i32}} {