Method and apparatus for performing clock allocation for a system implemented on a programmable device
A method for designing a system on a target device includes placing the system on the target device. A netlist retiming is performed on the placed system. A clock allocation and a clock region optimization are performed utilizing information from the placing and the netlist retiming.
1. A method for designing a system on a target device, the method comprising:
placing the system on the target device;
performing a netlist retiming on the placed system; and
performing a clock allocation that includes identifying a size and a location on the target device for a clock region that is occupied by a clock network by utilizing information from the placing and the netlist retiming.
2. The method of claim 1 , wherein the size and the location on the target device for the clock region is not limited to a predefined size and a predetermined area on the target device.
3. The method of claim 1 , wherein the identifying the size and the location for the clock region comprises:
generating an initial clock region in response to a placement of all registers in the clock network; and
generating a final clock region by moving bounding registers towards a center of the initial clock region while respecting one or more of worst negative slack, density metrics, wiring usage, wiring congestion, and legality constraints in the clock network.
4. The method of claim 1 , wherein the clock region comprises a plurality of disjointed areas.
5. The method of claim 1 , wherein the placing the system comprises reducing a size of a clock region projected for the clock network.
6. The method of claim 1 , wherein the placing the system comprises reducing a number of sectors on the target device that are crossed by a clock region projected for the clock network.
7. The method of claim 1 , wherein the placing the system comprises reducing a number of sectors on the target device that are crossed by a clock region projected for the clock network by re-shaping the clock region projected for the clock network.
8. The method of claim 1 further comprising performing physical clustering and detailed placement refinements after the performing clock allocation.
9. The method of claim 1 further comprising routing the system.
10. The method of claim 1 further comprising incrementally refining the size and the location for the clock region during one or more subsequent stages of compilation.
11. The method of claim 1 , wherein clock networks are optimized during a placement taking into consideration one or more of overall wire usage, routing congestion, timing, a total number of clustered block elements created in a clock region, and legality constraints.
12. A non-transitory computer readable medium including a sequence of instructions stored thereon for causing a computer to execute a method for designing a system on a target device, the method comprising:
performing a synthesis where a clock network is defined; and
performing a placement in response to a clock region projected for the clock network.
13. The non-transitory computer readable medium of claim 12 , wherein the performing a placement in response to a clock region projected for the clock network comprises reducing a size of the clock region projected for the clock network.
14. The non-transitory computer readable medium of claim 12 , wherein the performing a placement in response to a clock region projected for the clock network comprises reducing a number of sectors on the target device that are occupied by the clock region projected for the clock network.
15. The non-transitory computer readable medium of claim 14 , wherein the number of sectors that are occupied are reduced by changing a location of the clock region projected for the clock network.
16. The non-transitory computer readable medium of claim 14 , wherein the number of sectors that are occupied are reduced by changing a shape of the clock region projected for the clock network.
17. The non-transitory computer readable medium of claim 14 , wherein the clock network is optimized during a placement taking into consideration one or more of overall wire usage, routing congestion, timing, and legality constraints.
18. The non-transitory computer readable medium of claim 12 , wherein the method further comprises performing a physical retiming to optimize a timing and an area of a design of the system, wherein the physical retiming also determines a placement for new flip-flops added to the design.
19. A system designer, comprising:
a placement unit that places a system on a target device;
a retiming unit that performs a netlist retiming on the placed system; and
a clock allocation unit that performs a clock allocation by identifying a size and a location on the target device for a clock region that is occupied by a clock network by utilizing information from the placing and the netlist retiming.
20. The system designer of claim 19 , wherein the placement unit reduces a size of a clock region projected for a clock network.
21. The system designer of claim 19 , wherein the placement unit reduces a number of sectors on the target device that are crossed by a clock region projected for a clock network.
22. The system designer of claim 19 , wherein the placement unit optimizes clock networks taking into consideration one or more of overall wire usage, routing congestion, timing, and legality constraints.