US 4,231,089 A · Data processing system with apparatus for correcting microinstruction errors

Abstract

In a data processing system, a method and apparatus which enable an erroneous microinstruction word to be rewritten before it is executed. After the microinstruction word is written from memory into a control register, a parity network coupled to the control register determines whether a correct microinstruction is being executed. Upon a parity error being detected, the clock pulses to the data paths coupled to the control register are inhibited. The original microinstruction word is then fetched from secondary storage and rewritten into the microinstruction memory with the aid of a separate register providing the address to the microinstruction memory. Upon rewriting the microinstruction memory, the signal inhibiting the clock pulses is removed thereby allowing a new microinstruction to be executed. An additional feature includes apparatus for determining whether an error has previously occurred for the same microinstruction word, in which case a critical fault occurs and remedial maintenance is necessary.

Claims10

Dependent claims

Apparatus for correcting instruction errors in a data processing system comprising, in combination:

first and second instruction memory means for storing instructions in memory locations defined by addresses;
register means for receiving and holding instructions from said first and second instruction memory means;
instruction error detection means operable in association with the register means for evaluating an instruction held in said register means to determine whether the instruction is valid or erroneous;
means for freezing the state of at least a portion of the data processing system in response to the detection of an erroneous instruction by the instruction error detection means;
and
means operable in response to the detection of an erroneous instruction for replacing the erroneous instruction in said register means with a corresponding valid instruction retrieved from the second instruction memory means.

3 more elements

A method of correcting instruction errors in a data processing system comprising the steps of:

storing instruction information words in first and second sets of memory locations, each memory location being defined by an address and each address in the first instruction set corresponding to an address in the second set, such that under faultless operating conditions the same instruction information is written into corresponding addresses in the first and second set of memory locations;
for executing instructions, reading instruction information from an address in the first set of memory locations;
checking such instruction information to detect the presence of errors therein;
in response to detecting an error in such instruction information, inhibiting the operation of at least a portion of the data processing system whose operation is dependent upon the instruction information;
obtaining the instruction information from the corresponding address in the second set of memory locations and substituting it for the instruction information read from the address in the first set of memory locations;
and
re-establishing operation of the inhibited portion of the data processing system,
whereby such portion of the data processing system operates responsively to the instruction information obtained from the address in the second set of memory locations rather than the instruction information obtained from the address in the first set of memory locations.

5 more elements

Dependencies are inferred from the claim text.

Data processing system with apparatus for correcting microinstruction errors

Life of this patent

Data processing system with apparatus for correcting microinstruction errors

Life of this patent

Abstract

Claims10