Two stage memory allocation using a cache

• US 9,658,794 B2
• Filed: 05/08/2015
• Issued: 05/23/2017
• Est. Priority Date: 05/08/2015
• Status: Active Grant

• Current Assignee: Quest Software, Inc.
• Original Assignee: SonicWALL, Inc. (SonicWall Holdings Ltd.)
• Inventors: Zhang, Xiangyang
• Primary Examiner(s): Rossiter, Sean D
• Application Number:
  US14/708,110
• Publication Number:
  US 20160328175A1
• Time in Patent Office: 746 Days
• Field of Search: 711163
• CPC Class Codes: G06F 3/0604 Improving or facilitating administration, e.g. storage management
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/0644 Management of space entities, e.g. partitions, extents, pools
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0673 Single storage device
• CPC Class Codes: G06F 9/5016 the resource being the memory
• CPC Class Codes: G06F 9/52 Program synchronisation; Mutual exclusion, e.g. by means of semaphores

Abstract

The presently claimed invention manages memory in a multi-processor system. The presently claimed invention may use a combination of global and local locks when allocating memory and de-allocating memory in a multi-processor system. A method consistent with the presently claimed invention may first receive an allocation of a first memory space in the system memory of a multi-core processing system. The allocation of the first memory space may globally locks the first memory space where the memory space may administered by a software module using one or more local locks.

Claims

1. A method for allocating memory on a multi-core processing system, the method comprising:

receiving a first allocation of a first memory space of system memory of the multi-core processing system at a first software module, wherein the first allocation of the first memory space globally locks the first memory space;

receiving a first request for locally locking a first portion of memory, wherein the first portion of memory is of a size that corresponds to a memory size that spans a range of memory addresses from a first starting memory address;

locking the first portion of memory with a first local lock, wherein the first portion of memory includes memory addresses in the first memory space; and

storing the first starting memory address in a first entry of a data structure.

2. The method of claim 1, further comprising:

receiving a second request for locally locking a second portion of memory, wherein the second portion of memory corresponds to the size of the first portion of memory that spans a range of memory addresses from a second starting memory address;

locking the second portion of memory with a second local lock, wherein the second portion of memory includes memory addresses in the second memory space;

storing the second starting memory address in a second entry in the data structure;

receiving a third request for locally locking a third portion of memory, wherein the third portion of memory corresponds to the size of the first portion of memory that spans a range of memory addresses from a third starting memory address,

locking the third portion of memory with a third local lock, wherein the third portion of memory includes memory addresses in the third memory space; and

storing the third starting memory address in a third entry in the data structure.

3. The method of claim 2, wherein:

the first entry in the data structure includes a pointer to the first starting memory address;

the second entry in the data structure includes a pointer to the second starting memory address; and

the third entry in the data structure includes a pointer to the third starting memory address.

4. The method of claim 2, wherein:

the first portion of memory is exclusively accessed by a first thread executed by one or more processors at the multi-core processor;

the second portion of memory is exclusively accessed by a second thread executed by one or more processors at the multi-core processor; and

the third portion of memory is exclusively accessed by a third thread executed by one or more processors at the multi-core processor.

5. The method of claim 3, further comprising:

receiving an indication that identifies the first local lock, the second local lock, or the third local lock is no longer required;

unlocking the local lock identified by the indication; and

deleting an entry in the data structure identified by the indication.

6. The method of claim 3, wherein:

the data structure is a linked list,

the first entry in the data structure includes a pointer to the second entry in the data structure; and

the second entry in the data structure includes a pointer to the third entry in the data structure.

7. The method of claim 3, further comprising:

receiving an allocation of a second memory space of the system memory of the computer system at the first software module, wherein the allocation of the second memory space globally locks the second memory space; and

receiving a fourth request for locally locking a portion of memory, wherein the size of the fourth portion of memory corresponds to the size of the first portion of memory that spans a range of memory addresses from a fourth starting memory address.

8. The method of claim 7, further comprising:

receiving an indication that the fourth local lock is no longer required;

unlocking the local lock associated with the fourth local lock; and

unlocking the global lock locking the second memory space of the system memory.

9. The method of claim 2, wherein the size of the first, the second, and the third portions of memory are equal and correspond to a size of memory relating to a process performed by an application program executed by one or more cores of the multi-core processor.

10. The method of claim 9, wherein the application program process corresponds to at least one of storing data packets, a secure socket layer session, managing an endpoint of an Internet protocol security tunnel, network address translation object management, dynamic firewall policy management, and managing a session in a firewall.

11. A non-transitory computer readable storage medium having embodied thereon a program executable by a processor for performing a method for allocating memory on a multi-core processing system, the method comprising:

receiving a first allocation of a first memory space of system memory of the multi-core processing system, wherein the first allocation of the first memory space globally locks the first memory space;

receiving a first request for locally locking a first portion of memory, wherein the first portion of memory is of a size that corresponds to a memory size that spans a range of memory addresses from a first starting memory address;

locking the first portion of memory with a first local lock, wherein the first portion of memory includes memory addresses in the first memory space; and

storing the first starting memory address in a first entry in a data structure.

12. The non-transitory computer readable storage medium of claim 11, the program further executable to:

receive a second request for locally locking a second portion of memory, wherein the second portion of memory corresponds to the size of the first portion of memory that spans a range of memory addresses from a second starting memory address;

lock the second portion of memory with a second local lock, wherein the second portion of memory includes memory addresses in the second memory space;

store the second starting memory address in a second entry in the data structure;

receive a third request for locally locking a third portion of memory, wherein the third portion of memory corresponds to the size of the first portion of memory that spans a range of memory addresses from a third starting memory address,

lock the third portion of memory with a third local lock, wherein the third portion of memory includes memory addresses in the third memory space; and

store the third starting memory address in a third entry in the data structure.

13. The non-transitory computer readable storage medium of claim 12, wherein:

the first entry in the data structure includes a pointer to the first starting memory address;

the second entry in the data structure includes a pointer to the second starting memory address; and

the third entry in the data structure includes a pointer to the third starting memory address.

14. The non-transitory computer readable storage medium of claim 12, wherein:

the first portion of memory is exclusively accessed by a first thread executed by one or more processors at the multi-core processor;

the second portion of memory is exclusively accessed by a second thread executed by one or more processors at the multi-core processor; and

the third portion of memory is exclusively accessed by a third thread executed by one or more processors at the multi-core processor.

15. The non-transitory computer readable storage medium of claim 13, the program further executable to:

receive an indication that identifies the first local lock, the second local lock, or the third local lock is no longer required;

unlock the local lock identified by the indication; and

delete an entry in the data structure identified by the indication.

16. The non-transitory computer readable storage medium of claim 13, wherein:

the data structure is a linked list,

the first entry in the data structure includes a pointer to the second entry in the data structure; and

the second entry in the data structure includes a pointer to the third entry in the data structure.

17. The non-transitory computer readable storage medium of claim 13, the program further executable to:

receive an allocation of a second memory space of the system memory of the computer system, wherein the allocation of the second memory space globally locks the second memory space; and

receive a fourth request for locally locking a portion of memory, wherein the size of the fourth portion of memory corresponds to the size of the first portion of memory that spans a range of memory addresses from a fourth starting memory address.

18. The non-transitory computer readable storage medium of claim 17, the program further executable to:

receive an indication that a fourth local lock is no longer required;

unlock the local lock associated with the fourth local lock; and

unlock the global lock locking the second memory space of the system memory.

19. The non-transitory computer readable storage medium of claim 12, wherein the size of the first, the second, and the third portions of memory are equal and correspond to a size of memory relating to a process performed by an application program executed by one or more cores of the multi-core processor.

20. The non-transitory computer readable storage medium of claim 19, wherein the application program process corresponds to at least one of storing data packets, a secure socket layer session, managing an endpoint of an Internet protocol security tunnel, network address translation object management, dynamic firewall policy management, and managing a session in a firewall.

21. An apparatus for allocating memory, the apparatus comprising:

a memory; and

multiple processors, wherein at least one of the one or more multiple processors executes instructions out of the memory, wherein the at least one or more processors;

receives a first allocation of a first memory space of system memory of the multi-core processing system at a first software module, the first allocation of the first memory space globally locks the first memory space;

receives a first request for locally locking a first portion of memory, the first portion of memory is of a size that corresponds to a memory size that spans a range of memory addresses from a first starting memory address;

locks the first portion of memory with a first local lock, wherein the first portion of memory includes memory addresses in the first memory space; and

stores the first starting memory address in a first entry of a data structure.