Method for managing the threads of execution in a computer unit, and computer unit configured to implement said method

• US 9,158,582 B2
• Filed: 07/17/2013
• Issued: 10/13/2015
• Est. Priority Date: 07/20/2012
• Status: Active Grant

• Current Assignee: IDEMIA Identity and Security France SAS (Advent International Corporation)
• Original Assignee: Morpho (Advent International Corporation)
• Inventors: Darbois, Matthieu, Bronsart, Sebastien, Thuillier, Cedric
• Primary Examiner(s): Thammavong, Prasith
• Application Number:
  US14/415,986
• Publication Number:
  US 20150242241A1
• Time in Patent Office: 818 Days
• Field of Search: 711/151
• US Class Current: 1/1
• CPC Class Codes: G06F 13/1663 Access to shared memory
• CPC Class Codes: G06F 13/18 based on priority control G06F13/1605 takes precedence
• CPC Class Codes: G06F 13/37 using a physical-position-dependent priority, e.g. daisy chain, round robin or token passing
• CPC Class Codes: G06F 9/3009 Thread control instructions
• CPC Class Codes: G06F 9/4887 involving deadlines, e.g. rate based, periodic
• CPC Class Codes: G06F 9/5016 the resource being the memory
• CPC Class Codes: G06F 9/526 Mutual exclusion algorithms

Abstract

A method of managing execution threads launched by processes being executed in a computer unit having at least one calculation core connected to a shared memory. The method includes the steps of:

• • using a zone of the shared memory that is accessible to all of the processes and execution threads for the purpose of managing calculation tokens;
  • when a thread seeks to execute, that thread verifies that a calculation token is available;
  • if a calculation token is available, the thread allocates the calculation token to itself by updating the shared memory and it continues its execution, and then releases the calculation token at the end of its execution; and
  • each execution thread has a priority index allocated thereto, and each thread having a task that is being executed is caused periodically to verify that a thread has not been put on standby that has a priority index higher than its own, and where appropriate, causing the thread that is executing to stop executing and release the corresponding calculation token.

A computer unit for performing the method.

Claims

1. A method of managing execution threads launched by processes being executed in a computer unit having at least one calculation core connected to a shared memory, the method comprising the following steps:

using a zone of the shared memory that is accessible to all of the processes and execution threads for the purpose of managing calculation tokens;

when a thread associated with a task seeks to execute, causing the thread to verify whether a calculation token is available;

if a calculation token is available, causing the thread to allocate itself the calculation token by updating the shared memory, followed by the thread executing, and then by the thread releasing the calculation token at the end of its execution;

if no token is available, causing the thread for execution to go on standby until a calculation token is released by a thread that has terminated its execution;

when a calculation token is released by a thread, causing that thread to identify the next thread to which the token is to be transferred, and to transfer the token thereto, where appropriate; and

each execution thread has a priority index allocated thereto, and each thread having a task that is being executed is caused periodically to verify that a thread has not been put on standby that has a priority index higher than its own, and where appropriate, causing the thread that is executing to stop executing and release the corresponding calculation token.

2. A method according to claim 1, wherein each process having a thread that is executing periodically indicates that it is active by means of the shared memory zone which maintains an up-to-date list of processes having a thread that is being executed, and one of these processes releases the calculation token of a registered process that has failed to signal that it is active.

3. A method according to claim 1, wherein each process is capable of loading into memory the data entity enabling calculation tokens to be managed and the calculation token data entity is loaded into memory by the first process seeking to execute in the context of a session and is unloaded at the end of execution of the last process to be executing.

4. A method according to claim 1, including a step of providing an allocation of a limited number calculation tokens to at least one process.

5. A method according to claim 1, including a step of reserving at least one calculation token for at least one process.

6. A method according to claim 1, including a step of authorizing a program to access the shared zone.

7. A method according to claim 6, wherein the program is a program for collecting information suitable for use for statistical purposes.

8. A method according to claim 6, wherein the program is a debugging program.

9. A computer unit including at least one calculation core connected to a memory and arranged to execute processes, the computer unit wherein it is arranged to perform the method as described in claim 1.