Bull Hn Information Systems, Inc.

Other

Bull Hn

David W Selway

David A Bowman

Donald R Kesner

Clinton B Eckard

Russell Wayne Guenthner

Stefan R Bohult

James H Phillips

Leonard Rabins

Boubaker Shaiek

Russell W Guenther

Clark D McCaslin

As manufacturers of very fast and powerful commodity processors continue to improve the capabilities of their products, it has become practical to emulate the proprietary hardware and operating systems of powerful older computers on platforms built using commodity processors such that the manufacturers of the older computers can provide new systems which allow their customers to continue to use their highly-regarded proprietary legacy software on state-of-the-art new computer systems by emulating the older computer in software that runs on the new systems. In an example of the subject invention, a 64-bit Cobol Virtual Machine instruction provides the capability of adding to or improving the performance of legacy 36-bit Cobol code. Legacy Cobol instructions can be selectively diverted, in the host CPU, to a 64 bit Virtual Machine Implementation. The output legacy and new Cobol code is compiled in a dedicated implementation of the Cobol compiler, and the output of the special purpose compiler is emulated in a special purpose software emulator, separate from the main software emulator that handles the normal 36-bit stream of legacy code.

Host computer system emulating target system legacy software and providing for incorporating more powerful application program elements into the flow of the legacy software

As fast and powerful commodity processors have been developed, it has become practical to emulate on platforms built using commodity processors the proprietary hardware systems of powerful older computers. High performance is typically a key requirement for a system even when built using emulation software. In a hardware design many special cases and conditions which may cause exceptions are detected by logic operating in parallel with the instruction execution. In software these checks can cost extra cycles of processor time during emulation of each instruction and be a significant detriment to performance. Avoiding some of these checks by relying upon the underlying hardware checks of the host system and then using a signal handler and special software to recover from these signals is a way to improve the performance and simplify the coding of the software emulation system.

Performance improvement for software emulation of central processor unit utilizing signal handler

As fast and powerful commodity processors have been developed, it has become practical to emulate the proprietary hardware systems of powerful older computers on platforms built using commodity processors. The systems being emulated are often large mainframe computers with large numbers of disks, communications systems and other attached hardware. Because of the size and expense, and also because databases involved must reside in only one location, it is difficult to replicate these systems for testing, development, debug or for providing alternative options to customers. A method for providing a single emulated computer system which provides for multiple views or options in control of the emulator is disclosed in which the options are dependent and selected based on job or user basis. The mechanism continues to provide for high performance and a single copy of the operating system with multiple processes, jobs and threads being emulated under user controlled parameters.

Runtime selection of code variants in a multiprogram computer hardware emulation system

A process for determining the optimum load driving capacity for each driving node in a complex logic circuit is disclosed. First, the logic equations of the logic circuit are extracted from a circuit description. Then, the fan-out of each driving node is analyzed to determine if the total number of pass transistor loads of the analyzed node is excessive compared to a predetermined driving capacity. For each flagged node, logic equations are added which represent the sum of the node's pass transistor loads, and further logic equations are added to compare the number of pass transistors turned on from one to the absolute maximum for the node. Then, a formal proof program is used to analyze the logic circuit and determine which of the comparators have a true output. For each flagged node, the comparator for the largest number which has a possible true output is identified to determine the highest possible actual load for the node; and, if necessary, the driving capacity of the node is adjusted to handle the determined highest possible actual load.

Formal proof methods for analyzing circuit loading problems under operating conditions

A redundant computer system including two systems capable of independent operation. The two systems correspondingly employ two independent clock generation and distribution (CGD) units which each issue clock and clock definer signals. The clock and definer signals of each system are used internally and are also sent to the other system. When the two systems are split, phase locked loops in each system are disabled, and each system is controlled by a precision oscillator in its own CGD unit When the two systems are merged, one CGD is designated as master and remains under control of its internal oscillator. The clock and definer signals of the master system are employed in the slave system to derive a signal which is used as the reference input to the slave system's phase locked loop from which the slave system's clock and definer signals are developed. Preferably, dual flip-flop phase detector type phase locked loops are employed. For higher frequency operation, it is desirable to incorporate certain correction circuitry which minimizes phase offset at apparent phase lock which is an inherent characteristic of this type of phase locked loop.

Apparatus for phase synchronizing clock signals in a fully redundant

A redundant computer system including two systems capable of independent operation. The two systems correspondingly employ two independent clock generation and distribution (CGD) units which each issue clock and clock definer signals. When the two systems are split, each system is controlled by the oscillator signal generated by its own CGD unit. When the two systems are merged, one oscillator is designated as master, and its output is employed to derive the clock and definer signals on both sides of the redundant system. Special logic included in each CGD unit ensures that the change from master to slave (or slave to master) operation is performed without error. This special logic includes circuitry which places a temporary hold at a predetermined logic level on the local oscillator signal, which is in use when the switch is made, when the local clock and definer signals are both at the predetermined logic level. The hold continues until the oscillator which is to 'take over' is also at the predetermined logic level.

Apparatus for coordinating clock oscillators in a fully redundant

A redundant computer system including two systems capable of independent operation. The two systems correspondingly employ two independent clock generation and distribution (CGD) units which each issue clock and clock definer signals. When the two systems are split, each system is controlled by the clock and definer signals generated by its own CGD unit. When the two systems are merged, one CGD unit is designated as master, and its clock and definer signals drives both sides of the redundant system. Special logic included in each CGD unit ensures that the change from master to slave (or slave to master) operation is performed without error. This special logic includes circuitry which places a temporary hold at a predetermined logic level on the local clock and definer signals, which are in use when the switch is made, when the local clock and definer signals are both at the predetermined logic level. The hold continues until the clock and definer signals which are to 'take over' are also at the predetermined logic level.

Apparatus for coordinating clock distribution in a fully redundant

A reconfigurable computer system which includes two computer subsystems, corresponding lines of the system busses of the two computer subsystems being interconnected by solid state switches. Each of the computer subsystems includes a control component, a service processor, which when an error is detected that would render the subsystem inoperative, causes the solid state switches to open to sever the connection between the system busses of the two computer subsystems so that the computer subsystem that has not suffered such a failure can continue to operate. A communication link is also established between the two service processors. Either, or both, service processors can sever the link between them.

Growing community of inventors

Phoenix, AZ, United States of America

David W Selway