Xilinx, Inc.

Grigor S Gasparyan

Xiao Dong

Abhishek Joshi

Guenter Stenz

Mehrdad Eslami Dehkordi

Paul D Kundarewich

Xiaojian Yang

Garik Mkrtchyan

Yau-Tsun Steven Li

Sridhar Krishnamurthy

Aaron Ng

Dinesh D Gaitonde

Satish B Sivaswamy

Marvin Tom

Nitin Deshmukh

Aditya Chaubal

Raoul Badaoui

Zhaoxuan Shen

Amish Pandya

Sridhar Kirshnamurthy

Maogang Wang

Performing partition wire assignment for routing a multi-partition circuit design can include performing, using computer hardware, a global assignment phase by clustering a plurality of super-long lines (SLLs) into a plurality of SLL bins, clustering loads of nets of a circuit design into a plurality of load clusters, and assigning the plurality of SLL bins to the plurality of load clusters. For each SLL bin, a detailed assignment phase can be performed wherein each net having a load cluster assigned to the SLL bin is assigned one or more particular SLLs of the SLL bin using the computer hardware.

Partition wire assignment for routing multi-partition circuit designs

Disclosed approaches for processing a circuit design include interrupting processing of a circuit design by an electronic design automation (EDA) tool at a selected phase of processing. The tool serializes EDA state data into serialized state data while processing is interrupted and writes the serialized state data for subsequent restoration of tool state. To resume processing at the point of interruption, the EDA tool can read the serialized state data and deserialize the serialized state data. The EDA tool bypasses one or more phases of processing after reading the serialized state data and thereafter resumes processing of the circuit design.

Serialization in electronic design automation flows

Examples described herein provide a non-transitory computer-readable medium storing instructions, which when executed by one or more processors, cause the one or more processors to perform operations. The operations include: generating, using the one or more processors, a plurality of child processes according to a number of programmable dies of the multi-die device, each of the plurality of child processes corresponding to a respective programmable die of the multi-die device, wherein the plurality of child processes execute on different processors; partitioning a design for the multi-die device into a plurality of portions, each of the portions to be used to configure one of the programmable dies of the multi-die device; transmitting the plurality of portions of the design to the plurality of child processes for placement; and receiving placements from the plurality of child processes.

Multiprocessing flow and massively multi-threaded flow for multi-die devices

Some examples described herein relate to routing in routing elements. In an example, a design system includes a processor and a memory, storing instruction code, coupled to the processor. The processor is configured to execute the instruction code to model a communication network comprising switches interconnected in an array of data processing engines (DPEs), generate global routes of nets in the modeled communication network, generate detailed routes of the nets using the global routes, and translate the detailed routes to a file. Each of the switches has multiple input or output channels connected to another switch that are modeled as a single input or output edge, respectively, connected to the other switch. Each global route is generated through edge(s) of the switches. Each detailed route is generated comprising identifying one of the multiple input or output channels modeled by each edge through which the respective global route is generated.

Data processing engine (DPE) array routing

Disclosed approaches for processing a circuit design include providing access to checkpoint data of a design checkpoint of a circuit design and starting child processes by a parent process. An initial intermediate representation is generated by the parent process, and concurrent with the generating of the initial intermediate representation, the child processes load the checkpoint data into respective memory spaces. The parent process produces incremental updates to the design checkpoint. The parent process signals availability of the incremental updates to the child processes, which apply the incremental updates to the checkpoint data in the respective memory spaces. The child processes process the circuit design in response to completion of producing incremental updates by the parent placer process.

Incremental initialization by parent and child placer processes in processing a circuit design

Some examples described herein relate to global mapping of program nodes of a netlist of an application. In an example, a design system includes a processor and a memory coupled to the processor. The memory stores instruction code. The processor is configured to execute the instruction code to obtain a netlist of an application. The netlist contains program nodes and respective edges between the program nodes. The application is to be implemented on a device comprising an array of data processing engines. The processor is also configured to execute the instruction code to generate a global mapping of the program nodes based on a representation of the array of data processing engines and using an integer linear programming (ILP) algorithm; generate a detailed mapping of the program nodes based on the global mapping; and translate the detailed mapping to a file.

Data processing engine (DPE) array global mapping

An example method for compiling by a processor-based system includes obtaining a netlist of an application, the netlist containing program nodes and respective edges between the program nodes, the application to be implemented on a device comprising an array of data processing engines; generating a global mapping of the program nodes based on a representation of the array of data processing engines; generating a detailed mapping of the program nodes based on the global mapping, the detailed mapping assigning input/outputs of programmable logic (PLIOs) of the device to channels in an interface of the array of data processing engines, the detailed mapping further assigning buffers of the application to individual memory banks in the array of data processing engines; and translating the detailed mapping to a file.

Data processing engine (DPE) array detailed mapping

An example method for compiling includes, by a processor-based system: obtaining a netlist of an application, the netlist containing program nodes and respective edges between the program nodes, the application to be implemented on a device comprising an array of data processing engines; partitioning the netlist into a plurality of partitions; for each of the plurality of partitions: generating a global mapping of the program nodes based on a representation of the array of data processing engines and using an integer linear programming (ILP) algorithm; generating a detailed mapping of the program nodes based on the global mapping; and translating the detailed mapping for each of the plurality of partitions to a file.

Netlist partitioning for designs targeting a data processing engine array

Closing timing for a circuit design can include displaying, using a display device, a first region having a plurality of controls corresponding to a plurality of data sets generated at different times during a phase of a design flow for a circuit design, wherein each control selects a data set associated with the control, and displaying, using the display device, a second region configured to display a list of critical paths for data sets selected from the first region using one of the plurality of controls. Closing timing further can include displaying, using the display device, a third region configured to display a representation of a target integrated circuit including layouts for the critical paths of the list for implementations of the circuit design specified by the selected data sets.

Timing closure of circuit designs for integrated circuits

Processing a circuit design that specifies application logic and debugging logic includes partitioning the circuit design. Each partition includes a part of the application logic and a part of the debugging logic, each partition is specified for implementation on a respective IC die, and the circuit design specifies connections between a part of the application logic in one partition and a part of the debugging logic in another partition. The connections between the part of the application logic in the one partition and the part of the debugging logic in the other partition are changed to connections from the part of the application logic in the one partition to a part of the debugging logic in the one partition. The part of the application logic and the part of the debugging logic of each partition are placed and routed on the respective IC die.

Processing of a circuit design for debugging

Designing circuits can include, within a circuit design, detecting, using a processor, a high fan-out net having loads with a same timing requirement, wherein the circuit design is technology specific for a target integrated circuit (IC), determining, using the processor, a region having a predetermined shape and an area sized to fit loads of the high fan-out net within the region on the target IC, and determining, using the processor, a delay of the high fan-out net based upon a distance from a center of the region to an edge of the region. Designing circuits can also include assigning, using the processor, the delay to the high fan-out net.

Delay modeling for high fan-out nets within circuit designs

Partitioning a circuit design can include determining, using a processor, a target area utilization and a target cut utilization by iterating over a range of timing violations and determining, using the processor, a worst allowed timing violation based upon the target area utilization and the target cut utilization. Circuit elements of the circuit design can be assigned to partitions, using the processor, for implementation of the circuit design in a multi-die integrated circuit based upon a partition cost calculated using the target area utilization, the target cut utilization, and the worst allowed timing violation.

Partitioning circuit designs for implementation within multi-die integrated circuits

Placing a circuit design may include partitioning circuit elements of the circuit design into circuit element sets and grouping bins of an integrated circuit into bin sets. The bins include circuit elements of the circuit design from an initial placement. Placing a circuit design also may include determining a dependency connectivity metric for the circuit elements and, using a processor, selectively relocating circuit elements concurrently, for a plurality of iterations, using a cost metric for relocating the circuit elements and using an order of processing the circuit elements determined from the bin sets, the circuit element sets, and the dependency connectivity metrics.

Multithreaded scheduling for placement of circuit designs using connectivity and utilization dependencies

Re-budgeting connections includes detecting a budget event for a circuit design and, responsive to detecting the budget event, calculating, using a processor, a delta for a selected combinatorial circuit element of the circuit design using an incoming slack and an outgoing slack of the selected combinatorial circuit element. Using the processor, a delay budget for a connection of the selected combinatorial circuit element is adjusted using the delta responsive to detecting the budget event.

Re-budgeting connections of a circuit design

Routing a multi-fanout net includes selecting a driver component of the multi-fanout net of a circuit design, wherein the circuit design is specified programmatically, and determining a plurality of targets of the driver component. A source wave is created at each of a plurality of nodes of the driver component. One target is assigned to each source wave. Each source wave is expanded.

Growing community of inventors

San Jose, CA, United States of America

Grigor S Gasparyan