In this article, we will discuss about few important UPF command syntax that is used here to write an UPF for a given power intent.
create_power_switch
It is used to define a switch required in a gated power domain.
Syntax:
create_power_switch switch_name
– domain domain_name
– input_supply_port port_name supply_net_name
– output_supply_port port_name supply_net_name
– control_port port_name net_name
– on_state state_name input_supply_port {Boolean_function}
| Argument | Brief Description |
| switch_name | It is just the switch name, thus the user can give any name. |
| – domain domain_name | The domain containing the switch. |
| – input_supply_port port_name supply_net_name | The input supply port of the switch and the net where this port is connected. |
| – output_supply_port port_name supply_net_name | The output supply port of the switch and the net where this port is connected. |
| – control_port port_name net_name | The control port of the switch and the net where this port is connected. |
| – on_state state_name input_supply_port {Boolean_function} | A user defined state, the input supply port for which this is defined and the corresponding Boolean function. |
set_isolation
It is used to specify the elements in the domain to isolate using the specified strategy. You can check here, why we need isolation.
Syntax:
set_isolation isolation_name
– domain domain_name
– isolation power_net net_name | isolation_ground_net net_name | isolation_power_net net_name -isolation_ground_net net_name | no_isolation
– elements list
– clamp_value < 0 | 1 | latch | Z >
– applies_to < inputs | outputs | both >
| Argument | Brief Description |
| isolation_name | It is just the isolation strategy name, thus the user can give any name. |
| – domain domain_name | The domain where the strategy is applied. |
| – isolation power_net net_name | isolation_ground_net net_name | isolation_power_net net_name -isolation_ground_net net_name | no_isolation | The supply net(s) used to supply the isolation logic inferred by this isolation strategy. |
| – elements list | A list of ports/nets to which strategy is applied. |
| – clamp_value < 0 | 1 | latch | Z > | The value to which the ports mentioned in the list will be clamped. Default is 0. |
| – applies_to < inputs | outputs | both > | Whether the domain’s input ports or output ports or both are isolated. Default is outputs. |
set_isolation_control
It is used to specify the control signals for a previously defined isolation strategy.
Syntax:
set_isolation_control isolation_name
– domain domain_name
– isolation_signal signal_name
– isolation_sense < high | low >
– isolation < self | parent | sibling | fanout | automatic >
| Argument | Brief Description |
| isolation_name | Correponding isolation strategy name. |
| – domain domain_name | The domain where the strategy is applied. |
| – isolation_signal signal_name | The control signal that causes the specified element to drive its clamp value. |
| – isolation_sense < high | low > | It specified whether the isolation is enabled when the isolation control signal is high or low. |
| – isolation < self | parent | sibling | fanout | automatic > | Where the isolation cells are placed in the logic hierarchy. Default is automatic. |
set_level_shifter
It is used to specify a level shifter strategy for signals crossing from one voltage domain to another. You can check here why we need level shifters.
Syntax:
set_level_shifter level_shifter_name
– domain domain_name
– elements list
– applies_to < inputs | outputs | both >
– threshold value
– rule < low_to_high | high_to_low | both >
– location < self | parent | sibling | fanout | automatic >
| Argument | Brief Description |
| level_shifter_name | It is just the level shifter strategy name, thus the user can give any name. |
| – domain domain_name | The domain where the strategy is applied. |
| – elements list | A list of ports/nets to which strategy is applied. |
| – applies_to < inputs | outputs | both > | Whether the domain’s input ports, output ports, or both are level shifted. Default is both. |
| – threshold value | The voltage (V) threshold for determining when level shifters are required. Default is 0. |
| – rule < low_to_high | high_to_low | both > | Which type of level shifters are required. Default is both. |
| – isolation < self | parent | sibling | fanout | automatic > | Where the level shifter is placed in the logic hierarchy. Default is automatic. |
set_retention
It is used to specify the elements in the domain to retain its value when power is gated.
Syntax:
set_retention retention_name
– domain domain_name
– retention_power_net net_name
– retention_ground_net net_name
– elements list
| Argument | Brief Description |
| retention_name | It is just the retention strategy name, thus the user can give any name. |
| – domain domain_name | The domain where the strategy is applied. |
| – retention_power_net net_name | The supply net used to supply the retention logic inferred by this retention strategy. |
| – retention_ground_net net_name | The ground net used to supply the retention logic inferred by this retention strategy. |
| – elements list | A list of registers to which strategy is applied. |
set_retention_control
It is used to specify the control signals for a previously defined retention strategy.
Syntax:
set_retention_control retention_name
– domain domain_name
– save_signal signal < high | low >
– restore_signal signal < high | low >
| Argument | Brief Description |
| retention_name | Corresponding retention strategy name. |
| – domain domain_name | The domain where the strategy is applied. |
| – save_signal signal < high | low > | The control signal that causes the register to save its state before power is gated. And if it is an active low or active high signal. |
| – restore_signal signal < high | low > | The control signal that causes the register to restore its saved state after power is back. And if it is an active low or active high signal. |
NOTE: Normally the save_signal and restore_signal is same but with different active levels. This same signal is connected to the SLEEP port of the retention cell, discussed here.
add_port_state
It is used specify the value the supply nets or ports can take on. These values are referred to as port states.
Syntax:
add_port_state port_name
– state(s) {name < voltage | off >}
| Argument | Brief Description |
| port_name | The name of the already defined port, whose state(s) we are specifying. |
| – state(s) {name < voltage | off >} | User defined state name(s) and its value. |
create_pst
It is used to create a power state table with a specific ordering of supply nets. Power state table basically defines the combination of states that can exist at the same time during operation of the design.
Syntax:
create_pst table_name
– supplies list
| Argument | Brief Description |
| table_name | User defined power state table name. |
| – supplies list | The list of supply nets or ports to include in each power state of the design. |
add_pst_state
It is used to define the states of each of the supply nets for one possible state of the design.
Syntax:
add_pst_state state_name
– pst table_name
– state supply_states
| Argument | Brief Description |
| state_name | User defined power state name. |
| – pst table_name | The power state table to which this state applies. |
| – state supply_states | The list of supply nets state names [defined in add_port_state command], listed in same order as that of -supplies listed in the create_pst command. |
VLSI Tutorials 