US 11,755,797 B2
System and method for predicting performance, power and area behavior of soft IP components in integrated circuit design
Benny Winefeld, Palo Alto, CA (US)
Assigned to ARTERIS, INC., Campbell, CA (US)
Filed by ARTERIS, INC., Campbell, CA (US)
Filed on Mar. 15, 2021, as Appl. No. 17/202,277.
Application 17/202,277 is a continuation of application No. 16/685,823, filed on Nov. 15, 2019, granted, now 10,949,585, issued on Mar. 16, 2021.
Prior Publication US 2021/0200925 A1, Jul. 1, 2021
Int. Cl. G06F 30/327 (2020.01); G06N 20/00 (2019.01); G06N 3/08 (2023.01); G06F 30/35 (2020.01)
CPC G06F 30/327 (2020.01) [G06F 30/35 (2020.01); G06N 3/08 (2013.01); G06N 20/00 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A method for accumulating data used as a training set for training a machine learning model capable of determining one or more characteristics for any IP block, which has a set of input parameters, in order to optimize an IP block, the method comprising:
determining a plurality of vectors of logical parameters for a plurality of IP arrangements;
determining a plurality of vectors of physical parameters for the plurality of IP arrangements;
generating a plurality of composite vectors based on the plurality of vectors of physical parameters and the plurality of vectors of logical parameters;
synthesizing the plurality of composite vectors to generate a gate level netlist for each combination of the plurality of vectors of physical parameters and the plurality of vectors of logical parameters;
generating a timing characterization for each gate level netlist to produce a plurality of timing characterizations; and
updating the training set, in order to provide optimization characteristic for an IP block that is used to optimize the IP block, using data selected from a set including at least one of:
the plurality of vectors of logical parameters;
the plurality of vectors of physical parameters;
the plurality of composite vectors; and
the timing characterization.