System and method for recovering from memory failures in computer systems

A method for recoverable programming, comprising the steps of:

• identifying a predetermined instruction sequence;
• checkpointing a predetermined set of system data;
• masking a machine check abort handle;
• monitoring for memory access errors;
• logging a memory access error in an error logging register;
• polling the register for any logged memory access error during execution of the instruction sequence;
• raising exceptions, if the memory access error is logged;
• updating pointers, if the memory access error is logged;
• recovering from the memory access error using the checkpointed system data, if the memory access error is logged during execution of the instruction sequence.;
• re-executing the memory access request, if software so commands;
• and
• enabling the machine check abort handle.

A computer-usable medium embodying computer program code for commanding a computer to perform recoverable programming, comprising the steps of:

• identifying a predetermined instruction sequence;
• monitoring for memory access errors;
• logging a memory access error in an error logging register;
• polling the register for any logged memory access error during execution of the instruction sequence;
• and
• raising exceptions, if the memory access error is logged.

A system for recoverable programming, comprising:

• means for identifying a predetermined instruction sequence;
• means for monitoring for memory access errors;
• means for logging a memory access error in an error logging register;
• means for polling the register for any logged memory access error during execution of the instruction sequence;
• and
• means for raising exceptions, if the memory access error is logged.