MAXIMIZING ENCODINGS OF VERSION CONTROL BITS FOR MEMORY CORRUPTION DETECTION

• US 20130036332A1
• Filed: 08/05/2011
• Published: 02/07/2013
• Est. Priority Date: 08/05/2011
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Gove, Darryl J., Radovic, Zoran, Adams, Jonathan
• Granted Patent: US 8,572,441 B2
• Time in Patent Office: Days
• US Class Current: 714/54
• CPC Class Codes: G06F 11/0724 in a multiprocessor or a multi-core unit multiprocessors per se G06F15/80
• CPC Class Codes: G06F 11/0763 by bit configuration check, e.g. of formats or tags
• CPC Class Codes: G06F 21/52 during program execution, e.g. stack integrity ; Preventing unwanted data erasure; Buffer overflow

Abstract

Systems and methods for maximizing a number of available states for a version number used for memory corruption detection. A physical memory may be a DRAM comprising a plurality of regions. Version numbers associated with data structures allocated in the physical memory may be generated so that version numbers of adjacent data structures in a virtual address space are different. A reserved set and an available set of version numbers are associated with each one of the plurality of regions. A version number in a reserved set of a given region may be in an available set of another region. The processor detects no memory corruption error in response to at least determining a version number stored in a memory location in a first region identified by a memory access operation is also in a reserved set associated with the first region.

Claims

1. A computing system comprising:

a processor;

a physical memory coupled to the processor; and

an operating system (OS), wherein the OS is configured to;

partition the memory into a plurality of regions;

determine a reserved set of one or more version numbers for each one of the plurality of regions; and

determine an available set of one or more version numbers for each one of the plurality of regions, wherein an available set for a given region comprises version numbers not included in a reserved set for the given region;

wherein the processor is configured to;

generate a first address to identify a memory location in a first region of the plurality of regions, wherein the first address comprises a first version number; and

detect an error responsive to determining the version number is in a reserved set of version numbers for the first region.

2. The computing system as recited in claim 1, wherein a second version number in a reserved set associated with a second region of the plurality of regions is in the available set associated with the first region.

3. The computing system as recited in claim 1, wherein the first region corresponds to a lower half of an address space and the second region corresponds to an upper half of an address space.

4. The computing system as recited in claim 1, wherein a version number in a reserved set for a particular region has all bits equal to uppermost bits of addresses corresponding to the particular region.

5. The computing system as recited in claim 1, wherein the processor is configured to indicate no memory corruption error in response to detecting the first version number is in an available set of version numbers for the first region.

6. The computing system as recited in claim 5, wherein the processor is further configured to indicate a memory corruption error in response to determining the first version number is not equal to a version number previously stored in association with the first address.

7. The computing system as recited in claim 6, wherein the version number previously stored in association with the first address is stored in a spare memory bank that corresponds to a memory bank storing data associated with the first address.

8. The computing system as recited in claim 1, wherein the processor is further configured to:

complete a memory access operation using the identified memory location in response to determining there is no memory corruption error; and

send an indication to a trap handler in response to determining there is a memory corruption error.

9. A method for a computing system, the method comprising:

partitioning a memory into a plurality of regions;

determining a reserved set of one or more version numbers for each one of the plurality of regions; and

determining an available set of one or more version numbers for each one of the plurality of regions, wherein an available set for a given region comprises version numbers not included in a reserved set for the given region;

generating a first address to identify a memory location in a first region of the plurality of regions, wherein the first address comprises a first version number; and

detecting an error responsive to determining the version number is in a reserved set of version numbers for the first region.

10. The method as recited in claim 9, wherein a second version number in a reserved set associated with a second region of the plurality of regions is in the available set associated with the first region.

11. The method as recited in claim 9, wherein the first region corresponds to a lower half of an address space and the second region corresponds to an upper half of an address space.

12. The method as recited in claim 9, wherein a version number in a reserved set for a particular region has all bits equal to uppermost bits of addresses corresponding to the particular region.

13. The method as recited in claim 9, further comprising indicating a memory corruption error in response to determining the first version number is not equal to a version number previously stored in association with the first address.

14. The method as recited in claim 13, wherein the version number previously stored in association with the first address is stored in a spare memory bank that corresponds to a memory bank storing data associated with the first address.

15. The method as recited in claim 9, further comprising indicating no memory corruption error in response to determining the first version number is equal to a version number previously stored in association with the first address.

16. The method as recited in claim 15, further comprising:

completing a memory access operation using the identified memory location in response to determining there is no memory corruption error; and

sending an indication to a trap handler in response to determining there is a memory corruption error.

17. A processor comprising:

an instruction fetch unit configured to fetch instructions from a physical memory comprising a plurality of regions;

a decoder coupled to the instruction fetch unit;

an execution unit; and

a load-store unit;

wherein the decoder is configured to determine a received instruction is a memory access instruction corresponding to a first region of the plurality of regions;

wherein the execution unit is configure to generate a first address to identify a memory location in the first region, wherein the first address comprises a first version number; and

wherein the load-store unit is configured to;

generate a first address to identify a memory location in a first region of the plurality of regions, wherein the first address comprises a first version number; and

detect an error responsive to determining the version number is in a reserved set of version numbers for the first region.

18. The processor as recited in claim 17, wherein a second version number in a reserved set associated with a second region of the plurality of regions is in the available set associated with the first region.

19. The processor as recited in claim 17, wherein the first region corresponds to a lower half of an address space and the second region corresponds to an upper half of an address space.

20. The processor as recited in claim 17, wherein a version number in a reserved set for a particular region has all bits equal to uppermost bits of addresses corresponding to the particular region.