Firehose dump of SRAM write cache data to non-volatile memory using a supercap

• US 8,195,901 B2
• Filed: 02/05/2009
• Issued: 06/05/2012
• Est. Priority Date: 02/05/2009
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Harper, Michael L., Klein, Craig A., Lucas, Gregg S., Marquez, Mary A. J., Medlin, Robert E.
• Primary Examiner(s): LO, KENNETH M
• Application Number:
  US12/365,923
• Publication Number:
  US 20100199021A1
• Time in Patent Office: 1,216 Days
• Field of Search: 711/103, 711/162, 711/112, 711/166
• US Class Current: 711/162
• CPC Class Codes: G06F 11/1441 Resetting or repowering
• CPC Class Codes: G06F 12/0804 with main memory updating G06F12/0806 takes precedence
• CPC Class Codes: G06F 2212/2022 Flash memory
• CPC Class Codes: G11B 2220/455 said record carriers being in one device and being used as primary and secondary/backup media, e.g. HDD-DVD combo device, or as source and target media, e.g. PC and portable player

Abstract

A mechanism is provided for firehose dumping modified data in a static random access memory of a hard disk drive to non-volatile memory of the hard disk drive during a power event. Responsive an indication of a power event in the hard disk drive, hard disk drive command processing is suspended. A token is set in the non-volatile storage indicating that flash memory in the non-volatile memory contains modified data. A portion of a static random access memory cache table containing information on the modified data in the static random access memory is copied to the flash memory. The modified data from the static random access memory is then copied to the flash memory. Responsive to a determination that the power event that initiated the copy of the modified data in the static random access memory to the flash memory is still present, the hard disk drive is shut down.

Claims

1. A method, in a data processing system, for firehose dumping modified data in a static random access memory of a hard disk drive to non-volatile memory of the hard disk drive during a power event, the method comprising:

receiving an indication of the power event in the hard disk drive;

suspending command processing in the hard disk drive;

setting a token in the non-volatile storage indicating that a flash memory in the non-volatile memory contains the modified data;

copying a portion of a static random access memory cache table containing information on the modified data in the static random access memory to the flash memory;

copying the modified data from the static random access memory to the flash memory, wherein copying the modified data from the static random access memory to the flash memory further comprises;

initializing a cache block pointer to point to a first logical block address of the modified data in the static random access memory;

copying the modified data stored in the static random access memory that is associated with the first logical block address pointed to by the cache block pointer to the flash memory at an address that corresponds to an address where the modified data is stored in the static random access memory;

determining whether all cached logical block addresses have been copied from the static random access memory to corresponding locations in the flash memory;

responsive to all of the cached logical block addresses failing to have been copied from the static random access memory to the corresponding locations in the flash memory, incrementing the cache block pointer to the next logical block address in the static random access memory;

copying the modified data stored in the static random access memory that is associated with the next logical block address pointed to by the cache block pointer to the flash memory at an address that corresponds to an address where the modified data is stored in the static random access memory; and

repeating the incrementing and copying steps until all of the cached logical block addresses have been copied from the static random access memory to corresponding locations in the flash memory;

determining whether the power event that initiated the copy of the modified data in the static random access memory to the flash memory is still present; and

responsive to the power event that initiated the copy still being present, shutting down the hard disk drive.

2. The method of claim 1, wherein the indication of the power event is initiated by at least one of a +12 volt power connection falling below a marginal level associated with the +12 volt power connection, a +5 volt power connection falling below a marginal level associated with the +5 volt power connection, or an Emergency Power discrete signal being asserted, wherein the Emergency Power discrete signal is initiated based the removal of the hard disk drive from a backplane due to an Emergency Power pin in a set of pins on the hard disk drive being a first pin to be disconnected because the Emergency Power pin is the shortest pin in the set of pins.

3. The method of claim 1, wherein the hard disk drive is powered by a supercap during the power event so that the modified data is copied from the static random access memory to the flash memory, wherein the supercap capacitively stores a higher voltage than is required by the hard disk drive to copy the modified data stored in the static random access memory to the flash memory, and wherein a voltage regulator in the hard disk drive reduces the higher voltage supplied by the supercap to a voltage that is usable by the hard disk drive to copy the modified data stored in the static random access memory to the flash memory.

4. The method of claim 1, wherein shutting down the hard disk drive further comprises:

parking read/write heads of the hard disk drive;

spinning down a rotating magnetic disk of the hard disk drive; and

presenting permanent error sense data that indicates that a non-recoverable error has occurred to all hosts with outstanding commands.

5. The method of claim 1, further comprising:

responsive to the power event that initiated the copy failing to be present, synchronizing the modified data in the static random access memory with a rotating magnetic disk of the hard disk drive;

erasing all data stored in the flash memory;

setting the token in the non-volatile memory to indicate that the flash memory has been erased;

determining whether a new power event has occurred;

responsive to an absence of the new power event, determining whether a number of power events exceeds a predetermined threshold for power events;

responsive to the number of power events failing to exceed the predetermined threshold for power events, creating sense data that indicates a temporary power event was detected;

presenting the sense data to each host that interacts with the hard disk drive on a next command sent by the hard disk drive to each host;

presenting the sense data to each host that has an outstanding command awaiting execution with the hard disk drive; and

resuming normal command processing.

6. The method of claim 5, further comprising:

responsive to an existence of the new power event or responsive to the number of power events exceeding the predetermined threshold for power events, shutting down the hard disk drive, wherein shutting down the hard disk drive further comprises;

parking read/write heads of the hard disk drive;

spinning down a rotating magnetic disk of the hard disk drive; and

presenting permanent error sense data that indicates that a non-recoverable error has occurred to all hosts with outstanding commands.

7. The method of claim 1, further comprising:

responsive to the hard disk drive being initialized, booting up a processor in the hard disk drive;

spinning-up a rotating magnetic disk of the hard disk drive;

determining whether a token in the non-volatile storage indicates the modified data in the flash memory;

responsive to the modified data existing in the flash memory, copying the portion of the static random access memory cache table in the flash memory to the static random access memory;

copying the modified data from the flash memory to the static random access memory;

erasing all data stored in the flash memory;

setting the token in the non-volatile memory to indicate that the flash memory has been erased; and

resuming normal command processing.

8. The method of claim 7, further comprising:

responsive to the modified data failing to exist in the flash memory, determining whether the token in the non-volatile storage indicates that the flash memory has been erased;

responsive to the token indicating that the flash memory has not been erased, erasing all data stored in the flash memory;

setting the token in the non-volatile memory to indicate that the flash memory has been erased; and

resuming normal command processing.

9. The method of claim 8, further comprising:

responsive to the token indicating that the flash memory has been erased, resuming normal command processing.

10. An apparatus, comprising:

a processor; and

a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to;

receive an indication of the power event in the hard disk drive;

suspend command processing in the hard disk drive;

set a token in the non-volatile storage indicating that a flash memory in the non-volatile memory contains the modified data;

copy a portion of a static random access memory cache table containing information on the modified data in the static random access memory to the flash memory;

copy the modified data from the static random access memory to the flash memory, wherein the instructions to copy the modified data from the static random access memory to the flash memory further cause the processor to;

initialize a cache block pointer to point to a first logical block address of the modified data in the static random access memory;

copy the modified data stored in the static random access memory that is associated with the first logical block address pointed to by the cache block pointer to the flash memory at an address that corresponds to an address where the modified data is stored in the static random access memory;

determine whether all cached logical block addresses have been copied from the static random access memory to corresponding locations in the flash memory;

responsive to all of the cached logical block addresses failing to have been copied from the static random access memory to the corresponding locations in the flash memory, increment the cache block pointer to the next logical block address in the static random access memory;

copy the modified data stored in the static random access memory that is associated with the next logical block address pointed to by the cache block pointer to the flash memory at an address that corresponds to an address where the modified data is stored in the static random access memory; and

repeat the incrementing and copying steps until all of the cached logical block addresses have been copied from the static random access memory to corresponding locations in the flash memory;

determine whether the power event that initiated the copy of the modified data in the static random access memory to the flash memory is still present; and

responsive to the power event that initiated the copy still being present, shut down the hard disk drive.

11. The apparatus of claim 10, wherein the indication of the power event is initiated by at least one of a +12 volt power connection falling below a marginal level associated with the +12 volt power connection, a +5 volt power connection falling below a marginal level associated with the +5 volt power connection, or an Emergency Power discrete signal being asserted, wherein the Emergency Power discrete signal is initiated based the removal of the hard disk drive from a backplane due to an Emergency Power pin in a set of pins on the hard disk drive being a first pin to be disconnected because the Emergency Power pin is the shortest pin in the set of pins.

12. The apparatus of claim 10, wherein the hard disk drive is powered by a supercap during the power event so that the modified data is copied from the static random access memory to the flash memory, wherein the supercap capacitively stores a higher voltage than is required by the hard disk drive to copy the modified data stored in the static random access memory to the flash memory, and wherein a voltage regulator in the hard disk drive reduces the higher voltage supplied by the supercap to a voltage that is usable by the hard disk drive to copy the modified data stored in the static random access memory to the flash memory.

13. The apparatus of claim 10, wherein the instructions to shut down the hard disk drive further cause the processor to:

park read/write heads of the hard disk drive;

spin down a rotating magnetic disk of the hard disk drive; and

present permanent error sense data that indicates that a non-recoverable error has occurred to all hosts with outstanding commands.

14. The apparatus of claim 10, wherein the instructions further cause the processor to:

responsive to the power event that initiated the copy failing to be present, synchronize the modified data in the static random access memory with a rotating magnetic disk of the hard disk drive;

erase all data stored in the flash memory;

set the token in the non-volatile memory to indicate that the flash memory has been erased;

determine whether a new power event has occurred;

responsive to an absence of the new power event, determine whether a number of power events exceeds a predetermined threshold for power events;

responsive to the number of power events failing to exceed the predetermined threshold for power events, create sense data that indicates a temporary power event was detected;

present the sense data to each host that interacts with the hard disk drive on a next command sent by the hard disk drive to each host;

present the sense data to each host that has an outstanding command awaiting execution with the hard disk drive; and

resume normal command processing.

15. The apparatus of claim 14, wherein the instructions further cause the processor to:

responsive to an existence of the new power event or responsive to the number of power events exceeding the predetermined threshold for power events, shut down the hard disk drive, wherein the instructions to shut down the hard disk drive further cause the processor to;

park read/write heads of the hard disk drive;

spin down a rotating magnetic disk of the hard disk drive; and

present permanent error sense data that indicates that a non-recoverable error has occurred to all hosts with outstanding commands.

16. The apparatus of claim 10, wherein the instructions further cause the processor to:

responsive to the hard disk drive being initialized, boot up a processor in the hard disk drive;

spin-up a rotating magnetic disk of the hard disk drive;

determine whether a token in the non-volatile storage indicates the modified data in the flash memory;

responsive to the modified data existing in the flash memory, copy the portion of the static random access memory cache table in the flash memory to the static random access memory;

copy the modified data from the flash memory to the static random access memory;

erase all data stored in the flash memory;

set the token in the non-volatile memory to indicate that the flash memory has been erased; and

resume normal command processing.

17. The apparatus of claim 16, wherein the instructions further cause the processor to:

responsive to the modified data failing to exist in the flash memory, determine whether the token in the non-volatile storage indicates that the flash memory has been erased;

responsive to the token indicating that the flash memory has been erased, resume normal command processing;

responsive to the token indicating that the flash memory has not been erased, erase all data stored in the flash memory;

set the token in the non-volatile memory to indicate that the flash memory has been erased; and

resume normal command processing.

18. A computer program product comprising a non-transitory computer readable storage medium having a computer readable program recorded thereon, wherein the computer readable program, when executed on a computing device, causes the computing device to:

receive an indication of the power event in the hard disk drive;

suspend command processing in the hard disk drive;

set a token in the non-volatile storage indicating that a flash memory in the non-volatile memory contains the modified data;

copy a portion of a static random access memory cache table containing information on the modified data in the static random access memory to the flash memory;

copy the modified data from the static random access memory to the flash memory, wherein the computer readable program to copy the modified data from the static random access memory to the flash memory further causes the computing device to;

initialize a cache block pointer to point to a first logical block address of the modified data in the static random access memory;

copy the modified data stored in the static random access memory that is associated with the first logical block address pointed to by the cache block pointer to the flash memory at an address that corresponds to an address where the modified data is stored in the static random access memory;

determine whether all cached logical block addresses have been copied from the static random access memory to corresponding locations in the flash memory;

responsive to all of the cached logical block addresses failing to have been copied from the static random access memory to the corresponding locations in the flash memory, increment the cache block pointer to the next logical block address in the static random access memory;

copy the modified data stored in the static random access memory that is associated with the next logical block address pointed to by the cache block pointer to the flash memory at an address that corresponds to an address where the modified data is stored in the static random access memory; and

repeat the incrementing and copying steps until all of the cached logical block addresses have been copied from the static random access memory to corresponding locations in the flash memory;

determine whether the power event that initiated the copy of the modified data in the static random access memory to the flash memory is still present; and

responsive to the power event that initiated the copy still being present, shut down the hard disk drive.

19. The computer program product of claim 18, wherein the computer readable program further causes the computing device to:

responsive to the power event that initiated the copy failing to be present, synchronize the modified data in the static random access memory with a rotating magnetic disk of the hard disk drive;

erase all data stored in the flash memory;

set the token in the non-volatile memory to indicate that the flash memory has been erased;

determine whether a new power event has occurred;

responsive to an absence of the new power event, determine whether a number of power events exceeds a predetermined threshold for power events;

responsive to the number of power events failing to exceed the predetermined threshold for power events, create sense data that indicates a temporary power event was detected;

present the sense data to each host that interacts with the hard disk drive on a next command sent by the hard disk drive to each host;

present the sense data to each host that has an outstanding command awaiting execution with the hard disk drive; and

resume normal command processing.

20. The computer program product of claim 18, wherein the computer readable program further causes the computing device to:

responsive to an existence of the new power event or responsive to the number of power events exceeding the predetermined threshold for power events, shut down the hard disk drive, wherein the instructions to shut down the hard disk drive further cause the processor to;

park read/write heads of the hard disk drive;

spin down a rotating magnetic disk of the hard disk drive; and

present permanent error sense data that indicates that a non-recoverable error has occurred to all hosts with outstanding commands.