Incorporating timing constraints explicitly imposed by the data and control paths during clock network synthesis can enhance the robustness of the synthesized clock networks. With these constraints, a clock scheduler can be used to guide the synthesis of a clock network by specifying a set of feasible arrival times at the respective sequential elements. Clock scheduling can be either static or dynamic. In static clock scheduling, a clock schedule is first specified; next, a clock network is constructed realizing the prescribed schedule. Clock trees constructed using this approach may consume significant routing resources. In dynamic clock scheduling, the clock tree and clock schedule are both simultaneously constructed and determined, respectively. In earlier studies, the scalability of dynamic clock scheduling, which is essentially a shortest path problem, has been limited. The bottleneck is in finding the shortest paths between different vertices in an incrementally changing weighted graph. In this work, we present two clock schedulers that address the scalability issues by exploiting the sparsity of this weighted graph. Experimental results show that the proposed clock schedulers are one to two orders of magnitude faster compared to a published scheduler in an earlier work. The proposed clock schedulers are scalable, and are tested on a synthesized circuit with 348 710 cells, 57 491 sequential elements, and 496 727 explicit timing constraints.

The clock networks of modern circuits must be able to operate in multiple corners and multiple modes (MCMM). Earlier studies on clock network synthesis for MCMM designs focus on the legalization of an initial clock network that has timing violations in different corners or modes. We propose a mode reconfigurable clock tree (MRCT) that is based on a correct-by-construction approach. An MRCT consists of multiple clock trees. Depending on the active mode, the MRCT is reconfigured such that one of the clock trees is activated to deliver the clock signal. To limit the overhead, the bottom part of the network (closer to the clock sinks) is shared among all of the clock trees, and only the top part of the network (closer to the clock source) is mode reconfigurable. The reconfiguration is realized using or-gates and a single one-input-multiple-output demultiplexer. The MRCT is constructed in a bottom-up fashion by iteratively merging subtrees to form larger subtrees. When two subtrees cannot be merged because of mode-incompatible constraints, an or-gate is inserted to separate the incompatible modes. Corner-incompatible constraints are resolved by reducing safety margins of appropriate skew constraints. The experimental results show that for a set of synthesized MCMM circuits with 715 to 13; 216 sequential elements, the proposed approach can achieve high yield.