Using machine trained network during routing to perform parasitic extraction for an ic design

Abstract

Some embodiments use a machine-trained network during routing to provide the router with sufficient information to improve the quality of routes generated by a router. This machine-trained network in some embodiments is referred to as the “digital twin” of a lengthy design and/or manufacturing process that produces the design of an IC layout and/or manufactures an IC based on a designed IC layout. The digital twin in some embodiments provides information regarding parasitics, regarding redundant vias for insertion or regarding complexity of subsequent manufacturing processes used to manufacture an IC based on the IC design layout.

1 . A method of performing routing to define a plurality of routes for a plurality of nets in an integrated circuit (IC) design layout, the method comprising:
 performing a first routing operation to define a first set of one or more routes for a first set of one or more nets;   supplying the first set of routes to a machine-trained network (MTN) to identify a group of one or more parasitics couplings on a group of one or more routes in the set of routes;   based on the identified parasitic effect, discarding one or more routes in the first set of routes; and   performing a second routing operation to define a new route for any net that had a route discarded from the first set of routes.   
     
     
         2 . The method of  claim 1 , wherein first set of routes comprises a first plurality of routes, and said supplying is performed as part of a rip-up-and-reroute operation that analyzes the first plurality of routes that are defined after the first routing operation to identify any route that does not have desirable characteristics, wherein any route with a parasitic coupling higher than a threshold value is discarded for having an undesirable parasitic coupling. 
     
     
         3 . The method of  claim 1 , wherein first set of routes consists of a first route, and said supplying is performed after the first routing operation identifies the first route, in order to determine whether the first route has parasitic coupling with one or more neighboring components in the IC design that is higher than a threshold parasitic coupling value, said discarding comprising discarding the first route after determining that the first route has parasitic coupling higher than the threshold parasitic coupling value. 
     
     
         4 . The method of  claim 1 , wherein the MTN is trained to output a set of parasitic coupling values for at least one route in the first set of routes. 
     
     
         5 . The method of  claim 4 , wherein the set of parasitic coupling values comprise a set of parasitic capacitance values. 
     
     
         6 . The method of  claim 1 , wherein the MTN is trained to output a set of parasitic parameters, the method further comprising supplying the parasitic parameters to a solver to compute the set of parasitic coupling values for at least one route in the first set of routes. 
     
     
         7 . The method of  claim 6 , wherein the set of parasitic coupling values comprise a set of parasitic capacitance values. 
     
     
         8 . The method of  claim 7 , wherein the set of parasitic coupling values comprises for at least one particular net's route an overall parasitic coupling value representing parasitic coupling on the particular net's route from a plurality of components that neighbor that particular net's route. 
     
     
         9 . The method of  claim 7 , wherein the set of parasitic coupling values comprises for at least one particular net's route a plurality of parasitic coupling value representing a plurality of parasitic coupling on the particular net's route from a plurality of components that neighbor that particular net's route. 
     
     
         10 . The method of  claim 1 , wherein the machine trained network is trained in a training process that uses a first plurality of known input design layout structures with a first plurality of known output parasitic values, said known input design layout structures fed through the MTN during training to produce a second plurality of generated parasitic values that are used in conjunction with the first plurality of known output parasitic values to generate a loss function value, which is used to adjust a set of trainable parameters of the MTN. 
     
     
         11 . A non-transitory machine readable medium storing a program which when executed by at least one processing unit performs routing to define a plurality of routes for a plurality of nets in an integrated circuit (IC) design layout, the program comprising sets of instructions for:
 performing a first routing operation to define a first set of one or more routes for a first set of one or more nets;   supplying the first set of routes to a machine-trained network (MTN) to identify a group of one or more parasitics couplings on a group of one or more routes in the set of routes;   based on the identified parasitic effect, discarding one or more routes in the first set of routes; and   performing a second routing operation to define a new route for any net that had a route discarded from the first set of routes.   
     
     
         12 . The non-transitory machine readable medium of  claim 11 , wherein first set of routes comprises a first plurality of routes, and the set of instructions for said supplying is performed as part of a rip-up-and-reroute operation that analyzes the first plurality of routes that are defined after the first routing operation to identify any route that does not have desirable characteristics, wherein any route with a parasitic coupling higher than a threshold value is discarded for having an undesirable parasitic coupling. 
     
     
         13 . The non-transitory machine readable medium of  claim 11 , wherein first set of routes consists of a first route, and the set of instructions for said supplying is performed after the first routing operation identifies the first route, in order to determine whether the first route has parasitic coupling with one or more neighboring components in the IC design that is higher than a threshold parasitic coupling value, the set of instructions for said discarding comprises a set of instructions for discarding the first route after determining that the first route has parasitic coupling higher than the threshold parasitic coupling value. 
     
     
         14 . The non-transitory machine readable medium of  claim 11 , wherein the MTN is trained to output a set of parasitic coupling values for at least one route in the first set of routes. 
     
     
         15 . The non-transitory machine readable medium of  claim 14 , wherein the set of parasitic coupling values comprise a set of parasitic capacitance values. 
     
     
         16 . The non-transitory machine readable medium of  claim 11 , wherein the MTN is trained to output a set of parasitic parameters, the program further comprises a set of instructions for supplying the parasitic parameters to a solver to compute the set of parasitic coupling values for at least one route in the first set of routes. 
     
     
         17 . The non-transitory machine readable medium of  claim 16 , wherein the set of parasitic coupling values comprise a set of parasitic capacitance values. 
     
     
         18 . The non-transitory machine readable medium of  claim 17 , wherein the set of parasitic coupling values comprises for at least one particular net's route an overall parasitic coupling value representing parasitic coupling on the particular net's route from a plurality of components that neighbor that particular net's route. 
     
     
         19 . The non-transitory machine readable medium of  claim 11 , wherein the set of instructions for supplying the first set of routes to the MTN comprises set of instructions for converting an IC design layout portion that contains the first set of routes from a geometric domain definition to a pixel domain definition, and supplying the pixel-domain definition to the MTN. 
     
     
         20 . The non-transitory machine readable medium of  claim 11 , wherein the MTN is a neural network.