Method for making integrated circuits having gated clock trees

Claim 5

The method of claim 2 wherein the plurality of clock sinks of the first portion and the plurality of clock sinks of the second portion are gated by different physical clock gates.

Claim 6

The method of claim 2 wherein the plurality of clock sinks of the first portion are gated by a first clock gate controlled by a first clock gating signal, and the plurality clock sinks of the second portion are gated by a second clock gate controlled by a second clock gating signal.

Claim 7

The method of claim 4 wherein the second clock gating signal is a composite clock gating signal that logically includes the first clock gating signal.

Claim 8

The method of claim 2 wherein the plurality of clock sinks of the first portion are gated by a first clock gate controlled by a first clock gating signal, and the plurality of clock sinks of the second portion are ungated.

Claim 9

The method of claim 2 wherein there are no clock sinks of any other domain between the first region and the second region.

Claim 10

The method of claim 2 wherein there are no clock sinks of any other domain within the first region.

Claim 11

The method of claim 1 wherein at least one clock sink of another clustered domain is within the second region.

Claim 12

An integrated circuit that is driven by a clock signal, comprising: a semiconductor substrate; and a plurality of clock domains each having a plurality of clock sinks; wherein the plurality of clock sinks of one clock domain forms at least one cluster of clock sinks on the semiconductor substrate, wherein a clock sink-density of each cluster of clock sinks is approximately equal to or greater than a clock sink density of the integrated circuit that includes the clock sinks of all of the clock domains, wherein a first portion of the plurality of clock sinks of a clock domain has a higher sink density than a second portion of the plurality of clock sinks of the same clock domain; wherein the first portion has a subregion and the second portion has a subregion, and wherein the subregion of the first portion is adjacent to the region of the second portion.

Claim 13

An integrated circuit that is driven by a clock signal, comprising: a semiconductor substrate; and a plurality of clock domains each having a plurality of clock sinks; wherein the plurality of clock sinks of one clock domain forms at least one cluster of clock sinks on the semiconductor substrate, wherein a clock sink-density of each cluster of clock sinks is approximately equal to or greater than a clock sink density of the integrated circuit that includes the clock sinks of all of the clock domains, wherein a first portion of the plurality of clock sinks of a clock domain has a higher sink density than a second portion of the plurality of clock sinks of the same clock domain; wherein the first portion has a subregion and the second portion has a subregion, and wherein the subregion of the first portion is adjacent to the region of the second portion.

Claim 16

The method of claim 11 wherein the extent to which a domain is in compliance with a target condition is represented by a weighting factor.

Claim 17

The method of claim 12 further comprising changing the weighting factor of the target condition so that the target condition will have no influence upon a subsequent placement refinement.

Claim 18

The method of claim 12 wherein the weighting factor applies to the connections between the clock sinks of the domain.

Claim 19

The method of claim 12 wherein the weighting factor applies to the clock sinks of the domain.

Claim 20

The method of claim 12 wherein the weighting factor is implemented by a numerical value.

Claim 21

The method of claim 12 wherein the weighting factor is implemented by modifying a netlist to change an influence of a particular connection.

Claim 22

The method of claim 11 wherein the step of making a placement refinement results in an increase of the extent of the domain's compliance with the target condition.

Claim 23

The method of claim 11 wherein the placement refinement provides real placement information.

Claim 24

The method of claim 19 wherein the determination of the extent to which a domain is in compliance with the target condition is based upon information provided by a performance of the placement refinement step.

Claim 25

The method of claim 11 wherein the placement refinement is a partitioning min-cut.

Claim 26

The method of claim 21 wherein performance of the placement refinement step places all clock sinks of a given clock gate domain together in one region in compliance with the target condition, wherein the wherein the target condition is a minimum sink density.

Claim 27

The method of claim 11 wherein the netlist includes: a connection between a clock signal and at least one clock sink of a clock domain.

Claim 28

The method of claim 11 wherein the steps of making a determination and of making a placement refinement are repeated.

Claim 29

The method of claim 11 wherein a target condition includes at least one of: a minimum target sink density; a maximum target sink density; a maximum domain switching capacitance; a maximum overlap of sub/domain regions; a minimum overlap of sub/domain regions; a limitation on crossing of clock signal wires; a maximum of heat release per unit area; and a maximum of heat release per unit area per unit time; and a range of heat-release per unit area per unit time.

Claim 30

The method of claim 25 wherein a target condition is a target sink density, and wherein the determination of the extent to which clock domain is in compliance with the target condition includes calculating the density of sinks in a clock domain and comparing the calculated density to the target sink density.

Claim 31

The method of claim 26 wherein the target condition is a minimum sink density, and wherein the minimum sink density is approximately equal to or greater than a clock sink density of the integrated circuit.

Claim 34

A computer program product for use in designing an integrated circuit having a clock signal distributed by a gated clock tree having a plurality of clock domains each having a plurality of clock sinks, the computer program product comprising a medium readable by a computer, the computer readable medium having: program code for creating a model of the integrated circuit, wherein the model includes a netlist, and wherein the model of the integrated circuit includes a netlist of the gated clock tree; program code for changing the gated clock tree's netlist, wherein changing the gated clock tree's netlist includes placing a copy of a domain's gate at a level to implement single-gate gating of each separated subdomain; program code for establishing at least one target condition for a clock domain; program code for making a determination of the extent to which a clock domain is in compliance with the at least one target condition; and program code for making a placement refinement based upon the determination, wherein a performance of the placement refinement divides a clock gate domain into a plurality of separated subdomains.

Claim 35

A computer program product for use in designing an integrated circuit having a clock signal distributed by a gated clock tree having a plurality of clock domains each having a plurality of clock sinks, the computer program product comprising a medium readable by a computer, the computer readable medium having: program code for creating a model of the integrated circuit, wherein the model includes a netlist, and wherein the model of the integrated circuit includes a netlist of the gated clock tree; program code for changing the gated clock tree's netlist, wherein changing the gated clock tree's netlist includes placing a copy of a domain's gate at a level to implement single-gate gating of each separated subdomain; program code for establishing at least one target condition for a clock domain; program code for making a determination of the extent to which a clock domain is in compliance with the at least one target condition; and program code for making a placement refinement based upon the determination, wherein a performance of the placement refinement divides a clock gate domain into a plurality of separated subdomains.