SELECTIVELY KILLING TRAPPED MULTI-PROCESS SERVICE CLIENTS SHARING THE SAME HARDWARE CONTEXT

摘要

A method for handling parallel processing clients associated with a server in a GPU, the method comprising: receiving a failure indication for at least client running a thread in the GPU; determining threads in the GPU associated with the failing client; exiting threads in the GPU associated with the failing client; and continuing to execute remaining threads in the GPU for other clients running threads in the GPU.

主权项

1. A method for handling parallel processing clients associated with a graphics processor, the method comprising:
receiving a failure indication for at least one client of a plurality of clients running threads in the graphics processor; determining threads in the graphics processor associated with the failing client; exiting threads in the graphics processor associated with the failing client; and continuing to execute remaining threads in the graphics processor for other clients of the plurality of clients running threads in the graphics processor.