Abstract: A computer bus includes a first original signal line, a second redundant signal line, circuitry connected to the first original signal line and the second redundant signal line for driving the first original signal line and the second redundant signal line so as to convey on each identical information, circuitry for receiving signals on the first original signal line and the second redundant signal line, and error checking circuitry for comparing the signals on the first original signal line and the second redundant signal line and for indicating an error if the signals differ. By providing redundant signals for each signal that cannot be check with parity (for example wired-OR signals), the potential for single undetected points of failure is eliminated. In accordance with another embodiment of the invention, a computer having multiple modules connected by a backplane bus.
Type:
Grant
Filed:
March 25, 1997
Date of Patent:
July 28, 1998
Assignee:
Pyramid Technology Corporation
Inventors:
Mark Myers, Stacey Lloyd, Richard Stout, Robert Takasumi, John Lynch
Abstract: A user-initiated diagnostic test of a computer system is performed using a hardware abstraction layer including a diagnostic subsystem file and a diagnostic definition file, and also using diagnostic control routines and diagnostic status response routines. A selection is accepted from a user of a diagnostic subsystem to be tested from among a list of diagnostic subsystems stored in the diagnostic subsystem file. In accordance with information stored in the diagnostic definition file, the user is prompted for parameters of the selected diagnostic test, which are accepted from a user. These parameters are passed to the diagnostic control routine, which in response starts the selected diagnostic test. Diagnostic status response information from the diagnostic status response routine is received in the diagnostic control routine. The diagnostic status response information is presented to the user through a hardware-independent mechanism.
Abstract: A computer having multiple modules connected by a backplane bus includes multiple competition signal lines and multiple class signal lines. Access to the backplane bus to engage in one or more of multiple types of bus transactions is arbitrated between the modules by classifying the bus transactions into different classes and, during each of a succession of competition cycles, when a module wants access to the backplane bus to engage in a particular type of bus transaction, asserting a class signal line corresponding to a class in which the particular type of bus transaction has been classified. Based on information presented on the class signal lines, it is determined which modules are or are not eligible to compete for access to the backplane bus. When a module is eligible to compete for access to the backplane bus, it drives an identification code associated with the module on the competition signal lines.
Type:
Grant
Filed:
November 15, 1995
Date of Patent:
December 3, 1996
Assignee:
Pyramid Technology Corporation
Inventors:
Mark Myers, Stacey Lloyd, Richard Stout, Robert Takasumi, John Lynch
Abstract: A cache coherency test exercises cache coherency logic exhaustively such that any cache coherency failures liable to occur will occur. The CPU(s) which caused the failure is automatically identified by performing an automatic CPU sort. In particular, cache coherency is tested by causing each processor in the system to perform a sequence of read and write accesses to main memory and to its own cache memory so as to cause substantially every possible sequence of cache coherency bus operations. Each processor tests consistency of data read by it with data written by it. As long as no processor detects an error, read and write accesses are continued for a predetermined period of time. When any processor detects an error, each CPU is disabled, one at a time, to see if the remaining CPUs can run the test successfully. If they do not, then every combination of two CPUs are disabled, then every combination of three, etc. In this manner, a maximum running set of CPUs is identified.