Barrier instruction

The barrier instruction is a single-qubit control instruction that is used to constrain the optimization of a scheduler. It tells a scheduler that instructions on the specified qubit(s) cannot be scheduled across the barrier. See the wait instruction, to impose a time delay following a barrier.

The general form of the barrier instruction is as follows:

 barrier qubit-argument

Grammar for barrier instruction

barrier-instruction:
  barrier qubit-argument

qubit-argument:
  qubit-variable
  qubit-index

qubit-variable:
  identifier

qubit-index:
  index

Note

The barrier instruction accepts SGMQ notation. SGMQ notation allows you to express the application of an instruction to multiple qubits. However, SGMQ notation does not guarantee simultaneity. In general, a compiler will unpack this notation to separate consecutive single-qubit instructions on each respective qubit. This will depend on the scheduling optimization that is applied during the compilation process.

For certain backends, groups of consecutive barrier instructions are linked together to form a uniform barrier, across which no instructions on the specified qubits can be scheduled. Note that one can create a group of consecutive barrier instructions, i.e. a uniform barrier, using SGMQ notation.

Example

1
2
3
4
qubit q
X q
barrier q
X q

1
2
3
4
5
6
qubit[3] q
X q[0]
barrier q[0, 1]
H q[0]
X q[2]
H q[1]

In the examples above it is shown how the barrier instruction can be used with a single qubit and multiple qubits. In the case of the single qubit, an optimizer would generally fuse the two X gates into a single identity gate, I. However, because of the presence of the barrier instruction, the user explicitly states that the instructions on q are not permitted to be optimized across the barrier. The second example shows that a barrier can be placed for multiple qubits. Note that the barrier instruction is applied to qubits q[0] and q[1]. Qubit q[2] can be optimized freely across this circuit, i.e., be scheduled before or after the barrier.

The following code snippet illustrates how the barrier instruction might be used in context.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
version 3.0

qubit[2] q
init q

// Phi+ state
H q[0]
CNOT q[0], q[1]
barrier q
b[0,1] = measure q

barrier q
reset q

// Psi+ state
H q[0]
X q[1]
CNOT q[0], q[1]
barrier q
b[2, 3] = measure q

In the above code snippet, the \(\Phi_{+}\) and \(\Psi_{+}\) Bell states are generated and measured, successively. The barrier instruction is used to guarantee that the measure instruction is scheduled after the respective states have been generated. Moreover, an extra barrier instruction is used to separate the generation and measurement of the two Bell states.