MEMORY FAILURE RECOVERY METHOD, INFORMATION PROCESSING APPARATUS, AND PROGRAM

• US 20100318845A1
• Filed: 02/08/2008
• Published: 12/16/2010
• Est. Priority Date: 02/13/2007
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Kohiga, Akihito
• Granted Patent: US 8,051,333 B2
• Time in Patent Office: Days
• US Class Current: 714/15
• CPC Class Codes: G06F 11/073 in a memory management context, e.g. virtual memory or cache management memory management G06F12/00; testing of static memory units G11C29/00
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40

Abstract

Upon occurrence of a memory error which is difficult to correct, CPU interrupts a process under execution in response to interrupt signal output from error detection device, and transfers the control to the OS. The OS references attributes of a page stored in a page frame which has suffered the memory error. When read-only is indicated, physical memory management table is updated to set the page frame, which has suffered the error, into a disabled state. Then, page table is updated such that a virtual page stored in the page frame which has suffered the error is not assigned to any page frame, followed by resumption of the execution of the process. After execution of the process is resumed, a page fault process is invoked to store the virtual page which has encountered the memory error in another normal page frame.

Claims

1. A memory failure recovery method performed by an information processing apparatus employing a virtual storage scheme for dividing a virtual address space and a physical address space into pages and page frames to manage the virtual address space and physical address space using a page table and a physical memory management table, said method comprising:

interrupting execution of a process by an interrupt when a memory error occurs during execution of the process;

referencing an attribute of a page stored in a page frame, which suffers the memory error, from the page table;

updating the physical memory management table to set the page frame, which suffers the memory error, to a disabled state, when the page is read-only, and updating the page table such that the page will not be allocated to any of the page frames; and

resuming execution of the process.

2. The memory failure recovery method according to claim 1, wherein:

determining whether or not the page has been modified with reference to the physical memory management table when the attribute of the page is not read-only;

setting an address of a swap area into which the contents of the page were swapped the last time into an entry of the page table corresponding to the page when the page was not modified;

updating the physical memory management table to set the page frame which suffers the memory error to a disabled state;

updating the page table such that the page will not be allocated to any of the page frames; and

resuming execution of the process.

3. The memory failure recovery method according to claim 1, wherein:

the occurring memory error is automatically corrected by an error detection device which detects the memory error when the memory error is an automatically correctable memory error, and said information processing apparatus performs the interrupt when the memory error is difficult to correct.

4. An information processing apparatus employing a virtual storage scheme for dividing a virtual address space and a physical address space into pages and page frames to manage the virtual address space and physical address space using a page table and a physical memory management table, said apparatus comprising:

an error detection device configured to detect a memory error in a main storage device to generate an interrupt signal; and

a CPU configured to receive the interrupt signal,

wherein said CPU interrupts execution of a process upon receipt of the interrupt signal during execution of the process, references an attribute of a page stored in a page frame which suffers the memory error, from the page table, updates the physical memory management table to set the page frame, which suffers the memory error, to a disabled state, when the page is read-only, updates the page table such that the page will not be allocated to any of the page frames, and resumes execution of the process.

5. The information processing apparatus according to claim 4, wherein:

said CPU determines whether or not the page has been modified with reference to the physical memory management table when the attribute of the page is not read-only, sets an address of a swap area, into which the contents of the page were swapped the last time, into an entry of the page table corresponding to the page when the page was not modified, updates the physical memory management table to set the page frame which suffers the memory error to a disabled state, updates the page table such that the page will not be allocated to any of the page frames, and resumes execution of the process.

6. The information processing apparatus according to claim 4, wherein:

said error detection device automatically corrects the occurring memory error when the memory error is automatically correctable, and outputs the interrupt signal when the memory error is difficult to correct.

7. A computer-readable medium containing a computer program wherein, in a computer employing a virtual storage scheme for dividing a virtual address space and a physical address space into pages and page frames to manage the virtual address space and physical address space using a page table and a physical memory management table, the program causing said computer to:

interrupt execution of a process by an interrupt when a memory error occurs during execution of the process;

reference an attribute of a page stored in a page frame which suffers the memory error, from the page table;

update the physical memory management table to set the page frame, which suffers the memory error, to a disabled state, when the page is read-only, and update the page table such that the page will not be allocated to any of the page frames; and

resume execution of the process.

8. The computer-readable medium according to claim 7, wherein:

the program further causing said computer to;

determine whether or not the page has been modified with reference to the physical memory management table when the attribute of the page is not read-only;

set an address of a swap area, into which the contents of the page were swapped the last time, into an entry of the page table corresponding to the page when the page was not modified;

update the physical memory management table to set the page frame, which suffers the memory error, to a disabled state;

update the page table such that the page will not be allocated to any of the page frames; and

resume execution of the process.

9. The computer-readable medium according to claim 7, wherein:

the memory error is difficult to be automatically corrected by an error detection device.