A “Dogleg” channel router

Abstract

This paper presents an algorithm for interconnecting two sets of terminals across an intervening channel. It is assumed that the routing is-done on two distinct levels with all horizontal paths being assigned to one level and all vertical paths to the other. Connections between the levels are made through contact windows. A single net may result in many horizontal and vertical segments. Experimental results indicate that this algorithm is very successful in routing-channels that contain severe constraints. Usually, the routing is accomplished within one track of the mathematical lower bound. The routing algorithm presented here Was developed as part of LTX, a computer-aided design system for integrated circuit layout and was implemented on an HP-2100 minicomputer. A typical channel (300 terminals, 100 nets) can be routed in less than 5 seconds. Routing results are presented both for polycell chips under development at Bell. Laboratories and for examples that exist in the published literature. For the latter, reductions of 10% in the wiring area were. typical.

Switchbox routing by stepwise reshaping

“…PACKER can solve small and medium-size channel routing problems, that have the mentioned property. Big problems, such as the Deutsch dificult channel [28] are out of the scope of PACKER because of the nonlinear time complexity (see Section XI).…”

A Minimum-Impact Routing Algorithm

“…This scheme may be viewed as a generalization of the "left-edge packing" ordering schemes often used for the ~wo-sided channel routing problem (e.g. see [4]). …”

Minimizing channel density in standard cell layout

“…Much experience has shown this bound to be fairly tight. That is, using any of several channel routing algorithms in the literature (e.g., [2], [4], [15], and [18]), the number of tracks used is typically only slightly greater than the density. Hence, some placement algorithms have aimed to minimize density (e.g., [11]).…”

“…In the consistency subslice both a-row groups have a net endpoint pair from position 1 to position 8. Also, two overlapping nets are added to the a-row groups: nets (2,5) and (4,7). 8 Finally, we note that all of the nets shown in Figure 6 belong to a particular consistency subslice.…”

“…Without loss of generality, assume that the a-false net stopping at position 3 is assigned to the channel below row 2. Then the only net at position 5 that can share the track with the a-false net is the a-row3 net (4,7). In addition, the a-false net at position 8 is the only row 2 net that can share this track.…”