Scalable Spin Locks for Multiprogrammed Systems

Synchronization primitives for large scale multiprocessors need to provide low latency and low contention to achieve good performance. Queue-based locks (implemented in software with fetch and instructions) can greatly reduce contention and improve overall performance by arranging for processors to spin only on local locations. Unfortunately, queued locks exhibit poor behavior in the presence of multiprogramming: a process near the end of the queue, in addition to waiting for any process that is preempted during its critical section, must also wait for any preempted processes ahead of it in the queue. To solve this problem, we present two queue-based locks that recover from in-queue preemption. The rst lock employs the kernel interface of the NYU Symunix project. The second employs an extended interface that shares information in both directions across the user-kernel boundary, resulting in simpler code and better performance. We describe experiments with these locks in both real and synthetic applications on Silicon Graphics and Kendall Square multiprocessors. Results demonstrate the feasibility of high performance software locks with multiprogramming on scalable systems, and show competitive behavior on smaller, bus based machines. This work was supported in part by NSF Institutional Infrastructure award number CDA-8822724, NSF grant number CCR-9005633, and ONR research contract number N00014-92-J-1801 (in conjunction with the ARPA Research in Information Science and Technology|High Performance Computing, Software Science and Technology program, ARPA Order No. 8930).