Memory write protection for memory corruption detection architectures

• US 10,521,361 B2
• Filed: 02/26/2018
• Issued: 12/31/2019
• Est. Priority Date: 06/19/2015
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Stark, Tomer, Gabor, Ron, Tal, Ady, Nuzman, Joseph
• Primary Examiner(s): Ehne, Charles
• Application Number:
  US15/904,717
• Publication Number:
  US 20180181501A1
• Time in Patent Office: 673 Days
• 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 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/0766 Error or fault reporting or storing
• CPC Class Codes: G06F 11/0772 Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers
• CPC Class Codes: G06F 11/0787 Storage of error reports, e.g. persistent data storage, storage using memory protection
• CPC Class Codes: G06F 11/1666 where the redundant component is memory or memory area
• CPC Class Codes: G06F 12/023 Free address space management
• CPC Class Codes: G06F 12/0238 Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
• CPC Class Codes: G06F 12/0284 Multiple user address space allocation, e.g. using different base addresses interprocessor communication G06F15/163
• CPC Class Codes: G06F 12/1425 the protection being physical, e.g. cell, word, block
• CPC Class Codes: G06F 12/1441 for a range
• CPC Class Codes: G06F 12/1466 Key-lock mechanism
• CPC Class Codes: G06F 2201/80 Database-specific techniques
• CPC Class Codes: G06F 2212/1032 Reliability improvement, data loss prevention, degraded operation etc
• CPC Class Codes: G06F 2212/1052 Security improvement

Abstract

Memory corruption detection technologies are described. A method may store in a register an address of a memory corruption detection (MCD) table. The method receives, from an application, a memory store request to store data in a first portion of a contiguous memory block of a memory and sends, to the application, a fault message when a fault event associated with the first portion occurs in view of a protection mode of the first portion, wherein the protection mode indicates that the first portion is write protected.

Claims

1. A method comprising:

storing in a register an address of a memory corruption detection (MCD) table;

receiving, from an application, an allocation request for allocation of a memory object with a contiguous memory block in memory;

allocating the contiguous memory block for the memory object in view of a size of the memory object requested;

writing, into the MCD table, a MCD metadata word, wherein the MCD metadata word comprises;

a first MCD unique identifier (ID) associated with the contiguous memory block; and

a write protection indicator that indicates a first protection mode of a first portion of the contiguous memory block;

receiving, from the application, a memory store request to store data in the first portion of the contiguous memory block of the memory object of the memory, wherein the memory store request comprises a first pointer indicating a first location of the first portion in the contiguous memory block to store the data;

retrieving, from the MCD table, the write protection indicator that indicates the first protection mode of the first portion; and

sending, to the application, a fault message when a fault event associated with the first portion occurs based on the first protection mode of the first portion.

2. The method of claim 1, wherein the contiguous memory block comprises the first portion of the contiguous memory block and a second portion of the contiguous memory block, wherein the first protection mode is a write protection mode and a second protection mode of the second portion is a non-write protection mode.

3. The method of claim 2, further comprising receiving, from the application, the memory store request to write data to the first portion.

4. The method of claim 1, further comprising storing at least one of allocation metadata, a linked-list pointer, or a return address of a function for a stack in the first portion.

5. The method of claim 1, further comprising:

comparing the first MCD unique ID to a second MCD unique ID in the memory store request;

writing the data to the first portion when the first MCD unique ID matches the second MCD unique ID; and

sending the fault message when the first MCD unique identifier does not match the second MCD unique ID.

6. The method of claim 1, wherein the contiguous memory block is 64 bytes in size.

7. The method of claim 1, wherein the MCD metadata word comprises a first byte and a second byte, and wherein the first byte comprises the first MCD unique ID, a mode indicator of the second byte, and a reserved bit, and wherein the second byte comprises byte level granularity detection information and the write protection indicator.

8. A system comprising:

a memory device; and

a processor coupled to the memory device, wherein the processor comprises;

a register to store an address of a memory corruption detection (MCD) table; and

a memory controller coupled to the memory device, wherein the processor is to;

receive, from an application, a memory store request to store data in a first portion of a contiguous memory block of a memory; and

send, to the application, a fault message when a fault event associated with the first portion occurs in view of a protection mode of the first portion, wherein the protection mode indicates that the first portion is write protected.

9. The system of claim 8, wherein the processor is further to:

receive, from the application, an allocation request for an allocation of a memory object with the contiguous memory block in the memory;

allocate the contiguous memory block for the memory object in view of a size of the memory object requested; and

write, into the MCD table, a MCD metadata word.

10. The system of claim 8, wherein the memory store request comprises:

a first pointer indicating a location of the first portion in the contiguous memory block to store the data, wherein the first pointer includes a first MCD unique ID; and

a second pointer indicating a location of a memory object with the contiguous memory block, wherein the second pointer includes a second MCD unique ID, and wherein the processor is further to write the data to the first portion when the first MCD unique ID matches the second MCD unique ID.

11. The system of claim 10, wherein the processor is further to send the fault message when the first MCD unique ID does not match the second MCD unique ID.

12. The system of claim 8, wherein the contiguous memory block comprises the first portion of the contiguous memory block and a second portion of the contiguous memory block, wherein the protection mode is a write protection mode and a second protection mode of the second portion is a non-write protection mode.

13. The system of claim 8, wherein the first portion stores at least one of allocation metadata, a linked-list pointer, or a return address of a function for a stack.

14. The system of claim 8, wherein the contiguous memory block is 64 bytes in size.

15. The system of claim 8, wherein a MCD metadata word in the MCD table comprises a first byte and a second byte, and wherein the first byte comprises a first MCD unique ID, a mode indicator of the second byte, and a reserved bit, and wherein the second byte comprises byte level granularity detection information and a write protection indicator.

16. A method comprising:

storing in a register an address of a memory corruption detection (MCD) table;

receiving, from an application, a memory store request to store data in a first portion of a contiguous memory block of a memory; and

sending, to the application, a fault message when a fault event associated with the first portion occurs in view of a protection mode of the first portion, wherein the protection mode indicates that the first portion is write protected.

17. The method of claim 16, further comprising retrieving, from the MCD table, a write protection indicator that indicates the protection mode of the first portion, wherein the write protection indicator is a color value.

18. The method of claim 17, wherein the color value is a 6-bit metadata value.

19. The method of claim 16, wherein the contiguous memory block comprises the first portion of the contiguous memory block and a second portion of the contiguous memory block, wherein the protection mode of the first portion is a write protection mode and a second protection mode of the second portion is a non-write protection mode.

20. The method of claim 16, further comprising storing at least one of allocation metadata, a linked-list pointer, or a return address of a function for a stack in the first portion.