Queuing spin lock algorithms to support timing predictability

Craig, T.S.

doi:10.1109/real.1993.393505

Cited by 52 publications

(49 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To avoid long non-preemptable sections, prior work has explored the design and implementation of preemptable spin locks [3,12,19], where busy-waiting jobs can be preempted at any time and only the actual critical sections are executed non-preemptably. The primary benefit of preemptable spinning is that the maximum non-preemptable section length is independent of the number of processors (which, given rising core counts, is highly desirable), albeit at the cost of increased worst-case blocking, which arises because preempted jobs must "reissue" their lock requests after a preemption (e.g., in FIFO-ordered spin locks, preempted jobs must re-queue again at the end of the queue [3,12]). While preemptable spinning may be unavoidable in demanding latencysensitive applications, no analysis of the associated increase in blocking has been proposed to date, which renders such locks unsafe in the context of predictable hard real-time systems.…”

Section: A Motivation: Spin Locks In Theory and Practicementioning

confidence: 99%

“…The seminal work on efficient, list-based spin locks with FIFO semantics is due to Mellor-Crummey and Scott [22]. Based on their technique, priority-ordered variants for use in real-time systems were later developed [12,18,21], as were extensions that allow spinning jobs to be preempted [3,12,19]. Most preemptable spin locks, and especially list-based spin locks, require OS support to reliably detect preempted jobs [19].…”

Section: Related Workmentioning

confidence: 99%

“…First of all, it is worth emphasizing that even in the case of preemptable spinning, the actual critical section must still be executed non-preemptably (e.g., see [3,12,27]). This is required to avoid the preemption of a lock-holding job, which could result in excessive delays (i.e., if lock holders could be preempted, bounds on worst-case blocking would have to reflect entire job execution costs, which is undesirable since job execution costs are typically much larger than critical sections).…”

Section: B Unordered Preemptable Spin Locksmentioning

confidence: 99%

See 2 more Smart Citations

On Spin Locks in AUTOSAR: Blocking Analysis of FIFO, Unordered, and Priority-Ordered Spin Locks

Wieder

Brandenburg

2013

2013 IEEE 34th Real-Time Systems Symposium

View full text Add to dashboard Cite

Abstract-Motivated by the widespread use of spin locks in embedded multiprocessor real-time systems, the worst-case blocking in spin locks is analyzed using mixed-integer linear programming. Four queue orders and two preemption models are studied: (i) FIFO-ordered spin locks, (ii) unordered spin locks, (iii) priorityordered spin locks with unordered tie-breaking, and (iv) priorityordered spin locks with FIFO-ordered tie-breaking, each analyzed assuming both preemptable and non-preemptable spinning. Of the eight lock types, seven have not been analyzed in prior work. Concerning the sole exception (non-preemptable FIFO spin locks), the new analysis is asymptotically less pessimistic and typically much more accurate since no critical section is accounted for more than once. The eight lock types are empirically compared in schedulability experiments. While the presented analysis is generic in nature and applicable to real-time systems in general, it is specifically motivated by the recent inclusion of spin locks into the AUTOSAR standard, and four concrete suggestions for an improved AUTOSAR spin lock API are derived from the results.

show abstract

Section: A Motivation: Spin Locks In Theory and Practicementioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: B Unordered Preemptable Spin Locksmentioning

confidence: 99%

See 1 more Smart Citation

On Spin Locks in AUTOSAR: Blocking Analysis of FIFO, Unordered, and Priority-Ordered Spin Locks

Wieder

Brandenburg

2013

2013 IEEE 34th Real-Time Systems Symposium

View full text Add to dashboard Cite

show abstract

“…In this case, its overall RMR time complexity for entering and exiting its critical section is O(k). By defining the renaming tree's height to be Θ(log N ), the 6 In the conference version of this paper [3], we presented a DSM algorithm as a direct transformation of Algorithm G-CC. We later discovered that an additional "if " clause is necessary, as explained here.…”

Section: Theorem 1 Using Any Fetch-and-φ Primitive Of Rank R ≥ 2 Stamentioning

confidence: 99%

A generic local-spin fetch-and--based mutual exclusion algorithm

Anderson

Kim

2007

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

We present a generic fetch-and-φ-based local-spin mutual exclusion algorithm, with O(1) time complexity under the remote-memory-references time measure. This algorithm is "generic" in the sense that it can be implemented using any fetch-and-φ primitive of rank 2N , where N is the number of processes. The rank of a fetch-and-φ primitive is a notion introduced herein; informally, it expresses the extent to which processes may "order themselves" using that primitive. This algorithm breaks new ground because it shows that O(1) time complexity is possible using a wide range of primitives. In addition, by applying our generic algorithm within an arbitration tree, one can easily construct a Θ(max(1, log r N )) algorithm using any primitive of rank r, where 2 ≤ r < N.

show abstract

“…Unfortunately, non-preemptive spinlocks can cause unacceptable delays for latency-sensitive workloads because tasks may remain non-preemptive for prolonged times. While it is possible to reduce the negative effects by allowing spinning tasks to remain preemptive (e.g., see [2,20,35]), critical sections are still executed non-preemptively in such spinlocks and thus have a major impact on latency, not unlike the priority-boosting of critical sections under the MPCP.…”

Section: Introductionmentioning

confidence: 99%

A Fully Preemptive Multiprocessor Semaphore Protocol for Latency-Sensitive Real-Time Applications

Brandenburg¹

2013

2013 25th Euromicro Conference on Real-Time Systems

View full text Add to dashboard Cite

Independence preservation, a property in real-time locking protocols that isolates latency-sensitive tasks from delays due to unrelated critical sections, is identified, formalized, and studied in detail. The key to independence preservation is to ensure that tasks remain fully preemptive at all times. For example, on uniprocessors, the classic priority inheritance protocol is independence-preserving. It is shown that, on multiprocessors, independence preservation is impossible if job migrations are disallowed. The O(m) independencepreserving protocol (OMIP), a new, asymptotically optimal binary sempahore protocol based on migratory priority inheritance, is proposed and analyzed. The OMIP is the first independence-preserving, real-time, suspension-based locking protocol for clustered job-level fixed-priority scheduling. It is shown to benefit latency-sensitive workloads, both analytically by means of schedulability experiments, and empirically using response-time measurements in LITMUS RT .

show abstract

Queuing spin lock algorithms to support timing predictability

Cited by 52 publications

References 7 publications

On Spin Locks in AUTOSAR: Blocking Analysis of FIFO, Unordered, and Priority-Ordered Spin Locks

On Spin Locks in AUTOSAR: Blocking Analysis of FIFO, Unordered, and Priority-Ordered Spin Locks

A generic local-spin fetch-and--based mutual exclusion algorithm

A Fully Preemptive Multiprocessor Semaphore Protocol for Latency-Sensitive Real-Time Applications

Contact Info

Product

Resources

About