Optimizing grace period detection for preemptible read-copy update on uniprocessor systems

• US 7,953,708 B2
• Filed: 07/28/2008
• Issued: 05/31/2011
• Est. Priority Date: 07/28/2008
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: McKenney, Paul E., Triplett, Joshua A.
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): XUE, BIN
• Application Number:
  US12/181,125
• Publication Number:
  US 20100023559A1
• Time in Patent Office: 1,037 Days
• Field of Search: None
• US Class Current: 707/689
• CPC Class Codes: G06F 9/30087 Synchronisation or serialisation instructions
• CPC Class Codes: G06F 9/52 Program synchronisation; Mutual exclusion, e.g. by means of semaphores

Abstract

A technique for optimizing grace period detection following a data element update operation that affects preemptible data readers. A determination is made whether the data processing system is a uniprocessor system or a multiprocessor system. Grace period detection processing is performed using a first grace period detection technique if the data processing system is a multiprocessor system. Grace period detection processing is performed using a second grace period detection technique if the data processing system is a uniprocessor system. The grace period detection processing according to either technique determines the end of a grace period in which readers that are subject to preemption have passed through a quiescent state and cannot be maintaining references to the pre-update view of the shared data.

Claims

1. A method comprising:

determining if a data processing system that is to perform grace period detection following a data element update operation on shared data is a uniprocessor system or a multiprocessor system;

performing grace period detection processing using a first grace period detection technique if said data processing system is a multiprocessor system;

performing grace period detection processing using a second grace period detection technique if said data processing system is a uniprocessor system; and

said grace period detection processing according to either technique determining an end of a grace period in which readers that are subject to preemption have passed through a quiescent state and cannot be maintaining references to a pre-update view of said shared data.

2. The method of claim 1, wherein said first grace period detection technique comprises multiprocessor coordination processing and said second grace period detection technique does not comprise said multiprocessor coordination processing.

3. The method of claim 2, wherein said multiprocessor coordination processing comprises executing memory barriers, waiting for acknowledgements of new grace periods, or both.

4. The method of claim 1, wherein said determining is performed at compile time so that said grace period detection processing is statically compiled to perform either said first grace period detection technique or said second grace period detection technique.

5. The method of claim 1, wherein said determining is performed at run time so that said grace period detection processing is dynamically conditioned to perform either said first grace period detection technique or said second grace period detection technique.

6. The method of claim 5, wherein said determining is based on system configuration information provided at compile time, at boot time or at run time.

7. A data processing system, comprising:

one or more processors;

a memory coupled to said one or more processors, said memory including a computer useable storage medium tangibly embodying at least one program of instructions executable by said one or more processors to perform operations for optimizing grace period detection following a data element update operation on shared data that affects preemptible data readers, comprising;

determining if said data processing system is a uniprocessor system or a multiprocessor system;

performing grace period detection processing using a first grace period detection technique if said data processing system is a multiprocessor system;

performing grace period detection processing using a second grace period detection technique if said data processing system is a uniprocessor system; and

said grace period detection processing according to either technique determining an end of a grace period in which readers that are subject to preemption have passed through a quiescent state and cannot be maintaining references to a pre-update view of said shared data.

8. The system of claim 7, wherein said first grace period detection technique comprises multiprocessor coordination processing and said second grace period detection technique does not comprise said multiprocessor coordination processing.

9. The system of claim 8, wherein said multiprocessor coordination processing comprises executing memory barriers, waiting for acknowledgements of new grace periods, or both.

10. The system of claim 7, wherein said determining is performed at compile time so that said grace period detection processing is statically compiled to perform either said first grace period detection technique or said second grace period detection technique.

11. The system of claim 7, wherein said determining is performed at run time so that said grace period detection processing is dynamically conditioned to perform either said first grace period detection technique or said second grace period detection technique.

12. The system of claim 11, wherein said determining is based on system configuration information provided at compile time, at boot time or at run time.

13. A computer program product, comprising:

one or more machine-readable storage media;

means provided by said one or more media for programming a data processing system to optimize grace period detection following a data element update operation on shared data that affects preemptible data readers, as by;

determining if said data processing system is a uniprocessor system or a multiprocessor system;

performing grace period detection processing using a first grace period detection technique if said data processing system is a multiprocessor system;

performing grace period detection processing using a second grace period detection technique if said data processing system is a uniprocessor system; and

said grace period detection processing according to either technique determining the end of a grace period in which readers that are subject to preemption have passed through a quiescent state and cannot be maintaining references to a pre-update view of said shared data.

14. The computer program product of claim 13, wherein said first grace period detection technique comprises multiprocessor coordination processing and said second grace period detection technique does not comprise said multiprocessor coordination processing.

15. The computer program product of claim 14, wherein said multiprocessor coordination processing comprises executing memory barriers, waiting for acknowledgements of new grace periods, or both.

16. The computer program product of claim 13, wherein said determining is performed at compile time so that said grace period detection processing is statically compiled to perform either said first grace period detection technique or said second grace period detection technique.

17. The computer program product of claim 13, wherein said determining is performed at run time so that said grace period detection processing is dynamically conditioned to perform either said first grace period detection technique or said second grace period detection technique.

18. The computer program product of claim 17, wherein said determining is based on system configuration information provided at compile time, at boot time or at run time.

19. A method for optimizing grace period detection following a data element update operation in a data processing system having preemptible readers, comprising:

performing a first-phase data update operation on shared data in a manner that temporarily preserves a pre-update view of said shared data for benefit of readers that may be concurrently referencing said shared data during said update operation;

dynamically determining if said data processing system is a uniprocessor system or a multiprocessor system;

performing grace period detection processing using a first grace period detection technique if said data processing system is a multiprocessor system;

performing grace period detection processing using a second grace period detection technique if said data processing system is a uniprocessor system;

said grace period detection processing according to either technique determining an end of a grace period in which readers that are subject to preemption have passed through a quiescent state and cannot be maintaining references to a pre-update view of said shared data; and

performing a second-phase update operation following said grace period.

20. The method of claim 19, wherein said determining is based on system configuration information provided at compile time, at boot time or at run time.