ECO-system: Embracing the Change in Placement

Abstract

In a realistic design flow, circuit and system optimizations must interact with physical aspects of the design. For example, improvements in timing and power may require replacing large modules with variants that have different power/delay trade-off, shape and connectivity. New logic may be added late in the design flow, subject to interconnect optimization. To support such flexibility in design flows we develop a robust system for performing Engineering Change Orders (ECOs). In contrast with existing stand-alone tools that offer poor interfaces to the design flow and cannot handle a full range of modern VLSI layouts, our ECO-system reliably handles fixed objects and movable macros in instances with widely varying amounts of whitespace. It detects geometric regions and sections of the netlist that require modification and applies an adequate amount of change in each case. Given a reasonable initial placement, it applies minimal changes, but is capable of re-placing large regions to handle pathological cases. ECO-system can be used in the range from high-level synthesis, to physical synthesis and detail placement.

A statistical framework for post-silicon tuning through body bias clustering

“…To overcome this problem, we ran Capo [1,8] in an extension of the Engineering Change Order (ECO) placement algorithm described in [8]. In this mode, Capo makes incremental changes to a given placement (which in this case is the initial placement used to form the correlation grid in Figure 3) and can build contiguous regions of similarly clustered cells.…”

“…By limiting the number of clusters to just a few, the overhead is already drastically reduced compared to approaches that use individual gate-level ABB control. We show that modern placers [1,8] can be used to incrementally perturb an initial placement leading to only small increases in area and wirelength and that the gains of the method far outweigh these small penalties.…”

Design time body bias selection for parametric yield improvement

“…That is, most gates are intrinsically clustered within the physically continuous regions, which helps reduce the well spacing overheads. Moreover, we employ the incremental placer CAPO [25] to minimize the gate displacement and area overhead, following a similar flow as in [12]. CAPO works in its Engineering Change Order (ECO) mode to make limited changes to the initial placement based on certain constraints [25].…”

“…Moreover, we employ the incremental placer CAPO [25] to minimize the gate displacement and area overhead, following a similar flow as in [12]. CAPO works in its Engineering Change Order (ECO) mode to make limited changes to the initial placement based on certain constraints [25]. Figure 10 demonstrates the resulting layout for the Vit1 circuit with both two and three clusters after applying CAPO to the initial placement.…”