USING SHARED MEMORY WITH AN EXECUTE-IN-PLACE PROCESSOR AND A CO-PROCESSOR

• US 20080126749A1
• Filed: 11/07/2006
• Published: 05/29/2008
• Est. Priority Date: 11/07/2006
• Status: Active Grant

• Current Assignee: Monterey Research, LLC (Vector Capital Corporation)
• Original Assignee: Spansion LLC (Infineon Technologies AG)
• Inventors: Natarajan, Venkat, Tom, Joe Y.
• Granted Patent: US 7,512,743 B2
• Time in Patent Office: Days
• US Class Current: 712/30
• CPC Class Codes: G06F 12/1408 by using cryptography for digital transmission H04L9/00
• CPC Class Codes: G06F 12/1416 by checking the object accessibility, e.g. type of access defined by the memory independently of subject rights G06F12/1458 takes precedence
• CPC Class Codes: G06F 21/85 interconnection devices, e.g. bus-connected or in-line devices
• CPC Class Codes: G06F 9/44573 Execute-in-place [XIP]

Abstract

The claimed subject matter provides systems and/or methods that facilitate sharing of a memory, having a single channel of access, between two or more processors. A host processor can be operatively connected to a co-processor and the memory in series. The host processor can execute in place to enable it to execute code directly from the memory, and can arbitrate access to the memory bus and thus the memory, so that the host processor can perform all memory fetches to the memory without interruption by the co-processor. The co-processor can be implemented as a finite state machine, and only accesses the memory during read or write cycles issued by the host processor. Various types of co-processors can be employed to perform various functions, such as cryptography and digital signal processing, for example. The memory can be volatile or non-volatile memory.

Claims

1. A system that facilitates sharing of a memory, comprising:

a host processor that performs at least one of a data read or data write by accessing the memory via an access channel and generates at least one of a read cycle or write cycle associated with at least one other processor when the host processor wants the at least one other processor to perform a function; and

the at least one other processor that is implemented as a finite state machine and cannot access the memory via the access channel unless the at least one other processor receives the at least one of a read cycle or write cycle from the host processor.

2. The system of claim 1, the host processor executes in place and executes code directly from the memory.

3. The system of claim 1, the memory is at least one of volatile memory or non-volatile memory.

4. The system of claim 3, the non-volatile memory is flash memory.

5. The system of claim 1, the memory is partitioned so that a subset of memory addresses is accessed by the host processor and another subset of memory addresses is accessed by the other processor, wherein the partitioning of the memory is dynamic.

6. The system of claim 1, the at least one other processor performs at least one process associated with at least one of a digital signal process, a Fast Fourier Transform process, a Finite Impulse Response filter process, a cryptography process, an image process, or a data compression process.

7. The system of claim 1, the access channel is a bus, comprising control lines, memory address lines, and data lines.

8. A device that facilitates sharing of memory, comprising:

a host processor that retrieves data by accessing a memory via an access channel and generates a read cycle or write cycle associated with at least one other processor at a time when the host processor wants the at least one other processor to perform an operation; and

the at least one other processor that is implemented as a finite state machine and cannot access the memory via the access channel unless the at least one other processor receives the read cycle or write cycle from the host processor.

9. The device of claim 8, the host processor executes in place and executes code directly from the memory.

10. The device of claim 8, the memory is at least one of volatile or non-volatile memory.

11. The device of claim 10, the non-volatile memory is flash memory.

12. The device of claim 8, the memory is partitioned so that a subset of one or more memory addresses in the memory is associated with the host processor and at least one other subset of one or more memory addresses is associated with the at least one other processor.

13. The device of claim 8, the at least one other processor is at least one of a digital signal processor, a Fast Fourier Transform processor, a Finite Impulse Response filter processor, a cryptography processor, an image processor, or a data compression processor.

14. A method for facilitating access to a memory, having a single access channel, shared by a host processor and at least one other processor, comprising:

generating at least one of a read or write associated with the host processor;

selecting a bypass mode to provide the host processor access to the memory via the single access channel; and

retrieving data, associated with the at least one of a read, or write stored in the memory.

15. The method of claim 14, further comprising:

generating at least one of a read cycle or write cycle associated with the at least one other processor;

de-selecting the bypass mode to the provide the at least one other processor access to the memory via the single access channel;

writing data to the memory via the single access channel when a write cycle associated with the at least one other processor is generated; and

reading data from the memory via the single access channel when a read cycle associated with the at least one other processor is generated.

16. The method of claim 15, further comprising:

transferring data from the memory to the at least one other processor when a read cycle associated with the at least one other processor is generated, or transferring data from the host processor to the at least one other processor when a write cycle associated with the at least one other processor is generated; and

performing at least one process on the data transferred to the at least one other processor.

17. The method of claim 14, the host processor generates at least one of an instruction fetch cycle, a read cycle, or a write cycle.

18. The method of claim 14, the at least one other processor is at least one of a digital signal processor, a Fast Fourier Transform processor, a Finite Impulse Response filter processor, a cryptography processor, an image processor, or a data compression processor.

19. The method of claim 14, the at least one other processor is implemented as a finite state machine.

20. The method of claim 14, the memory is at least one of volatile or non-volatile memory.