An interactive graphical editor for parameterized layout generators, based on symbolic layout, is discussed. One basic idea is to have built-in rules, corresponding to basic knowledge about elementary things in VLSI design, in order to make the tool behave as an experienced partner to the designer. It was found that the fundamental problem was not to implement but to select a proper set of rules to define the semantics of the tool. The problem was attacked by actually implementing various experimental versions and using them to create layout. These experiments and the conclusions are summarized. For example, the fundamental importance of proper mechanisms to explicitly express geometrical constraints, such as pitch-matching, is shown. It is also shown how parameterization can be introduced in a user-friendly and natural way through the graphical, symbolic layout view.>

Movie is a special-purpose environment for silicon compilation tools, providing services in the same way that an operating system does to application problems. Movie also contains the functionality of an advanced module generator tool. An internal design representation was developed that allows a combination of hierarchically parameterized module generators, symbolic layout, and interactive graphical editing. This design representation has been formalized into a representation language with well-defined semantics. Movie provides a set of operations, formally defined using the representation language, for use by the tools. A set of instantiation mechanisms ensure certain correctness of the result independently of the tools and closes the design-verification loop. Movie was designed to take advantage of fast color graphics, special-purpose data structures and incremental updating to achieve interactivity with immediate visual feedback by real-time animation. By integration high-level tools, a hierarchy of levels of abstraction is maintained. Silicon compilation systems based on Movie can be expected to be open-ended, transparent, and highly interactive.>

Design principles for MIMD multiprocessor computers with virtual memory based on a common, global and uniform logical address space, supporting parallel, procedural languages such as Ada (Ada is a registered trademark of the US Government, AJPO), are discussed. The major design issues are identified and suggested solutions given, the most important of which are distributed, associative address translation, and local mechanisms supporting efficient resource allocation policies to reduce over-all communication costs. Arguments are given for using shared memory and bus-based global communication. Some preliminary studies of bus-based intercommunication schemes, parallel language implementation, capacity simulation and VLSI implementation are reviewed, as well as a number of existing experimental and commercial multiprocessors. Finally an experimental system for evaluation of different mechanisms and policies in systems of the suggested type is outlined.

Abstract This paper reports the work on implementing a parallel version of Pascal on a small scale multiprocessor computer. A few simple primitives were included to support parallel programming. The code generation, linking and loading procedures are described. An overview of the hardware is included. Finally some programming experience with the system is reported.

An experimental multiprocessor computer was designed and built in order to explore the feasibility of certain internal communication mechanisms. The system consisted of seven processing elements, each containing a part of the global memory connected to a local bus. For each processor the global memory is seen as one single, linearly addressable structure. The processing elements were all connected to a common, global bus, consisting of three separate busses in order to increase the capacity. A bus selection unit was designed, capable of making a unique bus selection for each request, within a fraction of a memory cycle. The experiments have shown that communication structures based on distributed global memory and global bus systems can be used efficiently for medium scale systems.