</pre>

<h2 id="Expressions">Expressions</h2>

<p>
An expression specifies the computation of a value by applying
operators and functions to operands.
</p>

<h3 id="Operands">Operands</h3>

<p>
Operands denote the elementary values in an expression. An operand may be a
literal, a (possibly <a href="#Qualified_identifiers">qualified</a>)
non-<a href="#Blank_identifier">blank</a> identifier denoting a

//
// All HLO infeed ops expect a HLO token type operand and produce a tuple
// containing a token. This HLO token type is used to order multiple infeed
// operations within a computation. The token type can come from other
// infeed/outfeed/send/recv ops or can be generated using create_token op with
// no operands. Here we emit a create_token op to generate the token type

  func.return %0: tensor<3x5xf32>
}

// SparseMatMul where one operand needs to be transposed and the other one not.
//
// CHECK-LABEL:   @test_sparse_mat_mul_with_transpose
// CHECK-SAME:      %[[ARG0:.*]]: tensor<3x4xf32>
// CHECK-SAME:      %[[ARG1:.*]]: tensor<5x4xf32>

// CHECK-LABEL:   func @convert_gather_to_slice_dynamic_error
func.func @convert_gather_to_slice_dynamic_error(%arg0: tensor<3x?xi32>, %arg1: tensor<3x2xi32>) -> tensor<3x1504xi32> {
  // expected-error @+1 {{Dynamic shaped operand is not supported.}}
  %0 = "mhlo.gather"(%arg0, %arg1) {
    dimension_numbers = #mhlo.gather<
      offset_dims = [1],
      collapsed_slice_dims = [0],
      start_index_map = [0, 1],