Methods, systems, and computer programs are presented for determining the recipe for manufacturing a semiconductor with the use of machine learning (ML) to accelerate the definition of recipes. One general aspect includes a method that includes an operation for performing experiments for processing a component, each experiment controlled by a recipe, from a set of recipes, that identifies parameters for manufacturing equipment. The method further includes an operation for performing virtual simulations for processing the component, each simulation controlled by one recipe from the set of recipes. An ML model is obtained by training an ML algorithm using experiment results and virtual results from the virtual simulations. The method further includes operations for receiving specifications for a desired processing of the component, and creating, by the ML model, a new recipe for processing the component based on the specifications.

Methods, systems, and computer programs are presented for determining the recipe for manufacturing a semiconductor with the use of machine learning (ML) to accelerate the definition of recipes. One general aspect includes a method that includes an operation for performing experiments for processing a component, each experiment controlled by a recipe, from a set of recipes, that identifies parameters for manufacturing equipment. The method further includes an operation for performing virtual simulations for processing the component, each simulation controlled by one recipe from the set of recipes. An ML model is obtained by training an ML algorithm using experiment results and virtual results from the virtual simulations. The method further includes operations for receiving specifications for a desired processing of the component, and creating, by the ML model, a new recipe for processing the component based on the specifications.

An IC is provided. The IC includes an input pin, a controller, a timer, a first memory, a processor, at least one output pin, an output module coupled to the output pin, and a direct memory access (DMA) device coupled between the output module and the first memory. The controller is configured to provide a first control signal in response to a command from the input pin. The timer is configured to periodically provide a trigger signal according to the first control signal. The processor is configured to store first data in the first memory. The DMA device is configured to obtain the first data from the first memory in response to the trigger signal, and transmit the first data to the output module. The output module is configured to provide the first data to the output pin according to a transmission rate.

An IC is provided. The IC includes an input pin, a controller, a timer, a first memory, a processor, at least one output pin, an output module coupled to the output pin, and a direct memory access (DMA) device coupled between the output module and the first memory. The controller is configured to provide a first control signal in response to a command from the input pin. The timer is configured to periodically provide a trigger signal according to the first control signal. The processor is configured to store first data in the first memory. The DMA device is configured to obtain the first data from the first memory in response to the trigger signal, and transmit the first data to the output module. The output module is configured to provide the first data to the output pin according to a transmission rate.

A method includes acquiring a vector data signal associated with a circuit design, performing a timing update to determine timing information for the circuit design, and identifying a glitch in the circuit design based on a shifted vector waveform. The timing information includes a signal delay associated with a cell of the circuit design. The shifted vector waveform is generated by shifting the vector data signal based on the timing information.

A method includes acquiring a vector data signal associated with a circuit design, performing a timing update to determine timing information for the circuit design, and identifying a glitch in the circuit design based on a shifted vector waveform. The timing information includes a signal delay associated with a cell of the circuit design. The shifted vector waveform is generated by shifting the vector data signal based on the timing information.

A soft error-mitigating semiconductor design system and associated methods that tailor circuit design steps to mitigate corruption of data in storage elements (e.g., flip flops) due to Single Events Effects (SEEs). Required storage elements are automatically mapped to triplicated redundant nodes controlled by a voting element that enforces majority-voting logic for fault-free output (i.e., Triple Modular Redundancy (TMR)). Storage elements are also optimally positioned for placement in keeping with SEE-tolerant spacing constraints. Additionally, clock delay insertion (employing either a single global clock or clock triplication) in the TMR specification may introduce useful skew that protects against glitch propagation through the designed device. The resultant layout generated from the TMR configuration may relax constraints imposed on register transfer level (RTL) engineers to make rad-hard designs, as automation introduces TMR storage registers, memory element spacing, and clock delay/triplication with minimal designer input.

In an approach utilizing static analysis, a processor receives a netlist for an integrated circuit. For at least one node of the integrated circuit in the netlist, a processor calculates (i) a total capacitive load of the respective node and (ii) a minimum required driver size. For a driver of the respective node, a processor (i) determines an effective driver size of the driver based on at least a number of fins of the driver and (ii) determines that the effective driver size exceeds the minimum required driver size multiplied by a predefined sizing margin. A processor, responsive to determining that the effective driver size exceeds the minimum required driver size multiplied by the predefined sizing margin, generates a report, where the report includes at least the driver and a suggestion to reduce the effective size of the driver.

In an approach utilizing static analysis, a processor receives a netlist for an integrated circuit. For at least one node of the integrated circuit in the netlist, a processor calculates (i) a total capacitive load of the respective node and (ii) a minimum required driver size. For a driver of the respective node, a processor (i) determines an effective driver size of the driver based on at least a number of fins of the driver and (ii) determines that the effective driver size exceeds the minimum required driver size multiplied by a predefined sizing margin. A processor, responsive to determining that the effective driver size exceeds the minimum required driver size multiplied by the predefined sizing margin, generates a report, where the report includes at least the driver and a suggestion to reduce the effective size of the driver.

Mechanisms are provided for optimizing an integrated circuit device design to obfuscate emissions corresponding to internal logic states of the integrated circuit device design. A first integrated circuit (IC) device design data structure is received and parsed to identify at least one instance of an obfuscation indicator in the data of the IC device design data structure. At least one IC logic element is marked, in the IC device design, which is associated with the at least one instance of the obfuscation indicator. At least one emission obfuscation optimization is applied to the marked at least one IC logic element to obfuscate emissions from the marked at least one IC logic element and generate an emissions obfuscated IC device design data structure. The emissions obfuscated IC device design data structure is output for fabrication of an IC device in accordance with the emissions obfuscated IC device design data structure.