Using transactional execution for reliability and recovery of transient failures

A computer program product for autonomous recovery from a transient hardware failure by executing each portion of a stream of program instructions as a transaction, the computer program product comprising:

• a computer readable storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising:
• creating, by an operating system on a computer system configured to support transactional execution mode processing, a start of a transactional region portion of a stream of program instructions, wherein the portion of the stream of program instructions in the transactional region is to be executed as a transaction in transactional execution mode;
• in response to starting the transactional region in transactional execution mode:
• saving, by the operating system, a snapshot of a system state information,
• executing, by the operating system, one or more instructions in the portion of the stream of program instructions as a transaction;
• and
• creating, by the operating system, an end of the transactional region portion of the stream of program instructions;
• in response to ending execution of the transactional region in transaction execution mode:
• committing, by the operating system, store data of the transaction to memory;
• and
• responsive to a transient hardware failure affecting a program resource required by one or more instructions in the transactional region portion of the stream of program instructions executing as a transaction in transactional execution mode:
• aborting, by the operating system, the transaction without notifying a computer software application that initiated the stream of program instructions;
• and
• re-executing, by the operating system, the transactional region portion of the stream of program instructions as a transaction in transactional execution mode, based on the saved snapshot of the system state information.

A computer system for autonomous recovery from a transient hardware failure by executing each portion of a stream of program instructions as a transaction, the computer system comprising:

• a memory;
• and
• a processor in communication with the memory, wherein the computer system is configured to perform a method, said method comprising:
• creating, by an operating system on a computer system configured to support transactional execution mode processing, a start of a transactional region portion of a stream of program instructions, wherein the portion of the stream of program instructions in the transactional region is to be executed as a transaction in transactional execution mode;
• in response to starting the transactional region in transactional execution mode:
• saving, by the operating system, a snapshot of a system state information,
• executing, by the operating system, one or more instructions in the portion of the stream of program instructions as a transaction;
• and
• creating, by the operating system, an end of the transactional region portion of the stream of program instructions;
• in response to ending execution of the transactional region in transaction execution mode:
• committing, by the operating system, store data of the transaction to memory;
• and
• responsive to a transient hardware failure affecting a program resource required by one or more instructions in the transactional region portion of the stream of program instructions executing as a transaction in transactional execution mode:
• aborting, by the operating system, the transaction without notifying a computer software application that initiated the stream of program instructions;
• and
• re-executing, by the operating system, the transactional region portion of the stream of program instructions as a transaction in transactional execution mode, based on the saved snapshot of the system state information.