Software Only Intra-Compute Unit Redundant Multithreading for GPUs

摘要

A system, method and computer program product to execute a first and a second work-item, and compare the signature variable of the first work-item to the signature variable of the second work-item. The first and the second work-items are mapped to an identifier via software. This mapping ensures that the first and second work-items execute exactly the same data for exactly the same code without changes to the underlying hardware. By executing the first and second work-items independently, the underlying computation of the first and second work-item can be verified. Moreover, system performance is not substantially affected because the execution results of the first and second work-items are compared only at specified comparison points.

主权项

1. A system comprising:
a processor configured to perform operations comprising:
executing a first and a second work-item, wherein the first work-item performs first write operations to a first signature variable and the second work-item performs second write operations to a second signature variable; andcomparing the first signature variable to the second signature variable at a comparison point.