1 / 18

Mask-Cost-Aware ECO Routing

Mask-Cost-Aware ECO Routing. Hsi -An Chieny , Zhen-Yu Pengy , Yun-Ru Wuz , Ting- Hsiung Wangz , Hsin -Chang Linz, Chi-Feng Wuz , and Ting-Chi Wangy. DATE’14. Outline. I. INTRODUCTION II. PROBLEM FORMULATION III. OUR APPROACH ECO Routing Phase Connectivity-Breaking Phase

maine
Download Presentation

Mask-Cost-Aware ECO Routing

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Mask-Cost-Aware ECO Routing Hsi-An Chieny, Zhen-Yu Pengy, Yun-Ru Wuz, Ting-HsiungWangz, Hsin-Chang Linz, Chi-Feng Wuz, and Ting-Chi Wangy DATE’14

  2. Outline • I. INTRODUCTION • II. PROBLEM FORMULATION • III. OUR APPROACH • ECO Routing Phase • Connectivity-Breaking Phase • IV. EXPERIMENTAL RESULTS • V. CONCLUSION

  3. INTRODUCTION • engineering change order (ECO) plays an important role during circuit design for fixing functional/timing errors.

  4. INTRODUCTION

  5. INTRODUCTION

  6. PROBLEM FORMULATION

  7. PROBLEM FORMULATION • ML the set of n metal layers • CL the set of n-1 via layers • C the set of 2n -1 mask re-spin costs for these metal and via layers • EN the set of ECO nets to be routed • FP the set of pins to become floating • RW the set of reusable routes

  8. PROBLEM FORMULATION • Problem (Mask-Cost-Aware ECO Routing) • given EN, FP, RW, and the routing grid graph of a design • make all pins in FP become floating, route as many nets in EN as possible, and minimize the total mask re-spin cost induced by the set of changed metal and via layers

  9. ECO Routing Phase • ECO net into a set of 2-pin sub-nets according to the minimum spanning tree • iteratively add an adjacent layer of a lower mask cost to L to enlarge the layer range and reroute all ECO nets • Dijkstra’s algorithm to get a target shortest path algorithm for routing each 2-pin sub-net

  10. ECO Routing Phase • each sub-net that is ripped up or failed will be re-routed in the next iteration • expands the routing region • terminate when there is no failed or ripped-up sub-net, or it reaches a given iteration limit

  11. Connectivity-Breaking Phase • Problem (Mask-Cost-Aware Connectivity-Breaking) • Given a set of unused routes each of which is associated with a disconnection set • remove routing objects from each unused route for disconnecting all vertices in the associated disconnection set such that the total mask re-spin cost induced by the changed layers is minimized.

  12. Connectivity-Breaking Phase

  13. Connectivity-Breaking Phase • reduced to the problem of selecting a collection of layer combinations

  14. Connectivity-Breaking Phase

  15. EXPERIMENTAL RESULTS • implemented our ECO routing approach in C++ language • all experiments were conducted on a Linux workstation with a 2.9 GHz CPU and 48G memory • three industrial ECO routing cases are used and has 6 metal layers in 55nm node

  16. EXPERIMENTAL RESULTS

  17. EXPERIMENTAL RESULTS

  18. CONCLUSION • an ECO routing approach that considers the reuse of old routes to save the mask re-spin cost

More Related