Programmable integrated circuit underlay
A method for implementing a programmable device is provided. The method may include extracting an underlay from an existing routing network on the programmable device and then mapping a user design to the extracted underlay. The underlay may represent a subset of fast routing wires satisfying predetermined constraints. The underlay may be composed of multiple repeating adjacent logic blocks, each implementing some datapath reduction operation. Implementing circuit designs in this way can dramatically improve circuit performance while cutting down compile times by more than half.
1. A method of using design tools to implement a logic circuit on a programmable device, comprising:
extracting an underlay from a routing network on the programmable device, wherein the extracted underlay comprises a subset of routing wires in the routing network satisfying target routing constraints; and
mapping a first portion of the logic circuit to the extracted underlay and a second portion of the logic circuit to the routing network at least partially outside of the underlay, wherein the first portion corresponds to a first operation type and the second portion corresponds to a second operation type.
2. The method of claim 1 , wherein extracting the underlay comprises:
accessing a database to obtain information on the routing network.
3. The method of claim 2 , wherein extracting the underlay further comprises:
receiving the target routing constraints, wherein the target routing constraints comprise constraints selected from the group consisting of: source coordinates, timing requirements, speed requirements, types of routing resources, routing direction, and crosstalk properties.
4. The method of claim 1 , further comprising:
using the extracted underlay on at least one other region on the programmable device.
5. The method of claim 1 , wherein the extracted underlay comprises a plurality of adjacent programmable logic blocks.
6. The method of claim 1 , wherein the extracted underlay comprises a plurality of 2:1 datapath reduction operators.
7. The method of claim 6 , wherein the plurality of 2:1 datapath reduction operators comprises a plurality of 2:1 multiplexers.
8. The method of claim 6 , wherein the plurality of 2:1 datapath reduction operators comprises a plurality of adders.
9. The method of claim 6 , wherein the plurality of 2:1 datapath reduction operators comprises a plurality of logic gates.
10. The method of claim 6 , wherein the plurality of 2:1 datapath reduction operators have different ingress and egress patterns.
11. The method of claim 1 , wherein the first operation type comprises operations of an accelerator, and the second operation type comprises operations not of the accelerator.
12. The method of claim 1 , wherein the first operation type uses a higher clock rate than the second operation type.
13. An integrated circuit, comprising:
a programmable routing network; and
a logic circuit comprising:
a first portion implementing a first operation type using an underlay extracted from the programmable routing network; and
a second portion implementing a second operation type at least partially outside of the extracted underlay, wherein the extracted underlay comprises a wiring pattern within the programmable routing network satisfying target routing constraints.
14. The integrated circuit of claim 13 , wherein the extracted underlay comprises a plurality of programmable logic blocks.
15. The integrated circuit of claim 14 , wherein at least one of the plurality of programmable logic blocks in the extracted underlay is used to implement a 2:1 datapath reduction operator.
16. The integrated circuit of claim 15 , wherein the 2:1 datapath reduction operator comprises a 2:1 multiplexer.
17. The integrated circuit of claim 15 , wherein the 2:1 datapath reduction operator comprises an adder.
18. The integrated circuit of claim 15 , wherein the 2:1 datapath reduction operator comprises a logic gate.
19. The integrated circuit of claim 13 , wherein the extracted underlay comprises a plurality of adjacent programmable logic blocks.
20. The integrated circuit of claim 13 , wherein the target routing constraints comprise a timing constraint.
21. A non-transitory computer-readable storage medium comprising instructions to:
extract a subset of routing paths in a programmable interconnect fabric, where the extracted subset of routing paths satisfies predetermined performance criteria; and
map a first portion of a logic circuit to the extracted subset of routing paths and a second portion of the logic circuit to the routing paths at least partially outside of the subset, wherein the first portion corresponds to a first operation type and the second portion corresponds to a second operation type.