Machine-learning based datapath extraction
View Patent ↗A datapath extraction tool uses machine-learning models to selectively classify clusters of cells in an integrated circuit design as either datapath logic or non-datapath logic based on cluster features. A support vector machine and a neural network can be used to build compact and run-time efficient models. A cluster is classified as datapath if both the support vector machine and the neural network indicate that it is datapath-like. The cluster features may include automorphism generators for the cell clusters, or physical information based on the cell locations from a previous (e.g., global) placement, such as a ratio of a total cell area for a given cluster to a half-perimeter of a bounding box for the given cluster.
1. A computer-implemented method of extracting datapath logic from an integrated circuit design, comprising:
receiving a circuit description for the integrated circuit design which includes a plurality of cells interconnected to form a plurality of nets, by executing first instructions in a computer system;
generating cell clusters from the circuit description, by executing second instructions in the computer system;
evaluating the cell clusters to identify one or more cluster features in the cell clusters, by executing third instructions in the computer system; and
selectively classifying the cell clusters as either datapath logic or non-datapath logic using one or more machine-learning models based on the one or more cluster features, by executing fourth instructions in the computer system wherein said classifying uses at least two machine-learning models each providing an indication of whether a given one of the cell clusters is datapath logic, and the given cell cluster is classified as datapath only when both of the two machine-learning models indicate that the given cell cluster is datapath logic wherein a first one of the machine-learning models is a support vector machine, and a second one of the machine-learning models is a neural network.
2. The method of claim 1 wherein the cluster features include automorphism generators for the cell clusters.
3. The method of claim 1 wherein the circuit description further includes locations for the cells from a previous placement, and the cluster features include physical information based on the cell locations.
4. The method of claim 3 wherein the physical information includes a ratio of a total cell area for a given cluster to a half-perimeter of a bounding box for the given cluster.
5. A computer system comprising:
one or more processors which process program instructions;
a memory device connected to said one or more processors; and
program instructions residing in said memory device for extracting datapath logic from an integrated circuit design by receiving a circuit description for the integrated circuit design which includes a plurality of cells interconnected to form a plurality of nets, generating cell clusters from the circuit description, evaluating the cell clusters to identify one or more cluster features in the cell clusters, and selectively classifying the cell clusters as either datapath logic or non-datapath logic using one or more machine-learning models based on the one or more cluster features wherein said program instructions classify the cell clusters using at least two machine-learning models each providing an indication of whether a given one of the cell clusters is datapath logic, and the given cell cluster is classified as datapath only when both of the two machine-learning models indicate that the given cell cluster is datapath logic wherein a first one of the machine-learning models is a support vector machine, and a second one of the machine-learning models is a neural network.
6. The computer system of claim 5 wherein the cluster features include automorphism generators for the cell clusters.
7. The computer system of claim 5 wherein the circuit description further includes locations for the cells from a previous placement, and the cluster features include physical information based on the cell locations.
8. The computer system of claim 7 wherein the physical information includes a ratio of a total cell area for a given cluster to a half-perimeter of a bounding box for the given cluster.
9. A computer program product comprising:
a computer-readable storage medium; and
program instructions residing in said storage medium for extracting datapath logic from an integrated circuit design by receiving a circuit description for the integrated circuit design which includes a plurality of cells interconnected to form a plurality of nets, generating cell clusters from the circuit description, evaluating the cell clusters to identify one or more cluster features in the cell clusters, and selectively classifying the cell clusters as either datapath logic or non-datapath logic using one or more machine-learning models based on the one or more cluster features wherein said program instructions classify the cell clusters using at least two machine-learning models each providing an indication of whether a given one of the cell clusters is datapath logic, and the given cell cluster is classified as datapath only when both of the two machine-learning models indicate that the given cell cluster is datapath logic wherein a first one of the machine-learning models is a support vector machine, and a second one of the machine-learning models is a neural network.
10. The computer program product of claim 9 wherein the cluster features include automorphism generators for the cell clusters.
11. The computer program product of claim 9 wherein the circuit description further includes locations for the cells from a previous placement, and the cluster features include physical information based on the cell locations.
12. The computer program product of claim 11 wherein the physical information includes a ratio of a total cell area for a given cluster to a half-perimeter of a bounding box for the given cluster.