Mapping requests from a processing unit that uses memory-mapped input-output space

• US 6,694,380 B1
• Filed: 12/27/1999
• Issued: 02/17/2004
• Est. Priority Date: 12/27/1999
• Status: Expired due to Term

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Cutter, Daniel, Dolan, Christopher, Bernstein, Debra, Wolrich, Gilbert, Adiletta, Matthew J.
• Primary Examiner(s): Gaffin, Jeffrey
• Assistant Examiner(s): Farooq, Mohammad O.
• Application Number:
  US09/473,271
• Time in Patent Office: 1,513 Days
• Field of Search: 710/5, 710/7, 710/11, 710/20, 710/306
• US Class Current: 710/5
• CPC Class Codes: G06F 12/0223 User address space allocation, e.g. contiguous or non contiguous base addressing
• 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 12/06 Addressing a physical block of locations, e.g. base addressing, module addressing, memory dedication G06F12/08 takes precedence
• CPC Class Codes: G06F 12/0806 Multiuser, multiprocessor or multiprocessing cache systems
• CPC Class Codes: G06F 12/10 Address translation
• CPC Class Codes: G06F 12/109 for multiple virtual address spaces, e.g. segmentation G06F12/1036 takes precedence
• CPC Class Codes: G06F 13/20 for access to input/output bus
• CPC Class Codes: G06F 13/28 using burst mode transfer, e.g. direct memory access DMA, cycle steal G06F13/32 takes precedence
• CPC Class Codes: G06F 13/4027 using bus bridges G06F13/4022 takes precedence
• CPC Class Codes: G06F 15/17318 Parallel communications techniques, e.g. gather, scatter, reduce, roadcast, multicast, all to all
• CPC Class Codes: G06F 15/76 Architectures of general purpose stored program computers with program plugboard G06F15/08; multicomputers G06F15/16
• CPC Class Codes: G06F 2212/1041 Resource optimization
• CPC Class Codes: G06F 2212/206 Memory mapped I/O
• CPC Class Codes: G06F 2212/251 Local memory within processor subsystem
• CPC Class Codes: G06F 3/0622 in relation to access
• CPC Class Codes: G06F 3/0661 Format or protocol conversion arrangements
• CPC Class Codes: G06F 3/0679 Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
• CPC Class Codes: G11C 7/1033 using data registers of which only one stage is addressed for sequentially outputting data from a predetermined number of stages, e.g. nibble read-write mode
• CPC Class Codes: G11C 7/1072 for memories with random access ports synchronised on clock signal pulse trains, e.g. synchronous memories, self timed memories

Abstract

A processor is disclosed that can map a request from a central processing unit that uses memory-mapped input-output space to a second processing domain, such as a multithreaded processing domain. A request addressed to the input-output space of the central processing unit is converted to a corresponding command that simulates an operation between components in the multithreaded processing domain. The command is executed in the multithreaded processing domain. Information is accessed according to the request in response to executing the command.

Claims

1. A method of performing a request from a central processing unit that uses memory-mapped input-output space, the method comprising:

converting a request directed to the input-output space of the central processing unit to a corresponding command that simulates an operation between components in a second processing domain;

executing the command in the second processing domain, wherein the second processing domain comprises a multithreaded processing domain, and wherein the multithreaded domain includes micro-coded processing engines that support multiple threads, the threads are simultaneously active but only one thread is in operation at any one time; and

accessing information according to the request in response to executing the command.

2. The method of claim 1, wherein each micro-coded engine, to support the multiple threads, maintains program counters in hardware and includes states associated with the program counters.

3. A method of writing information from a central processing unit that uses memory-mapped input-output space, the method comprising:

sending a request from the central processing unit to an address in the input-output space, wherein the address is associated with a destination in an bus interface unit;

converting the address and a command type associated with the request to a corresponding command having a format used for communications between an engine in the bus interface unit and one or more micro-coded processing engines, wherein the micro-coded processing engines support multiple threads, the threads are simultaneously active but only one thread is in operation at any one time;

executing the command in the engine in the bus interface unit; and

writing the information to the destination in the bus interface unit in response to executing the command.

4. The method of claim 3 including:

causing a control signal to be sent from the engine in the bus interface unit; and

promoting the information, in response to the control signal, onto a bus used for communications between the bus interface unit and the one or more micro-coded processing engines.

5. The method of claim 4 including giving the command a higher priority than commands from the one or more micro-coded processing engines.

6. A method of writing information from a central processing unit that uses memory-mapped input-output space, the method comprising:

sending a request from the central processing unit to an address in the input-output space, wherein the address is associated with a destination in a micro-coded processing engine;

converting the address and a command type associated with the request to a corresponding command in a format used for communications between an engine in a bus interface unit and the micro-coded processing engine, wherein micro-coded processing engine supports multiple threads, the threads are simultaneously active but only one thread is in operation at any one time;

executing the command in the engine in the bus interface unit; and

writing the information to the destination in the micro-coded processing engine in response to executing the command.

7. The method of claim 6 including:

causing a control signal to be asserted by the engine in the bus interface unit; and

promoting the information, in response to the control signal, onto a bus used for communications between the bus interface unit and the micro-coded processing engine.

8. The method of claim 7 including:

causing the engine in the bus interface unit to assert address signals on a bus used for communications between the bus interface unit and the micro-coded processing engine, wherein the asserted address signals enable the micro-coded processing engine to accept the information.

9. The method of claim 7 including giving the command a higher priority than commands from the micro-coded processing engine.

10. A method of providing information from a bus interface unit to a central processing unit that uses memory-mapped input-output space, the method comprising:

sending a request from the central processing unit to an address in the input-output space, wherein the address is associated with a location in the bus interface unit;

converting the address and a command type associated with the request to a corresponding command having a format used for communications between an engine in the bus interface unit and one or more micro-coded processing engines, wherein the micro-coded processing engines support multiple threads, the threads are simultaneously active but only one thread is in operation at any one time;

executing the command in the engine in the bus interface unit; and

providing the information to the central processing unit in response to executing the command.

11. The method of claim 10 including:

promoting the information from the location in the bus interface unit onto a first bus used for communications between the bus interface unit and the one or more micro-coded processing engines.

12. The method of claim 11 including:

causing a control signal to be asserted by the engine in the bus interface unit; and

promoting the information, in response to the control signal, from the first bus to a second bus coupled to the central processing unit.

13. The method of claim 11 including giving the command a higher priority than commands from the one or more micro-coded processing engines.

14. A method of providing information from a micro-coded processing engine to a central processing unit that uses memory-mapped input-output space, the method comprising:

sending a request from the central processing unit to an address in the input-output space, wherein the address is associated with a location in the micro-coded processing engine;

converting the address and a command type associated with the request to a corresponding command having a format used for communications between an engine in a bus interface unit and the micro-coded processing engine, wherein the micro-coded processing engine supports multiple threads, the threads are simultaneously active but only one thread is in operation at any one time;

executing the command in the engine in the bus interface unit; and

providing the information to the central processing unit in response to executing the command.

15. The method of claim 14 including:

promoting the information from the location in the micro-coded processing engine onto a first bus used for communications between the bus interface unit and the micro-coded processing engine.

16. The method of claim 15 including:

causing a control signal to be asserted by the engine in the bus interface unit; and

promoting the information, in response to the control signal, from the first bus to a second bus coupled to the central processing unit.

17. The method of claim 15 including giving the command a higher priority than commands from the micro-coded processing engine.

18. A parallel hardware-based multithreaded processor comprising:

a central processing unit that coordinates system functions and that uses memory-mapped input-output space;

micro-coded processing engines that support multiple threads, the threads are simultaneously active but only one thread is in operation at any one time;

a bus interface unit;

a first bus coupled to the central processing unit; and

a second bus coupled to the micro-coded processing engines; and

a translator unit coupled between the first bus and the second bus to convert a request that is addressed to the input-output space of the central processing unit to a corresponding command that simulates operations between the bus interface unit and the micro-coded processing units.

19. The processor of claim 18 including a command bus that couples the translator unit to the bus interface unit, wherein the bus interface unit includes a plurality of engines for causing information to be pushed or pulled onto the second bus in response to a command received from the translator unit via the command bus.

20. The processor of claim 19, wherein the bus interface unit includes a plurality of registers, and wherein information stored by a particular one of the registers is passed to the central processing unit via the second bus, the translation unit and the first bus as a result of the command being received and executed by one of the engines in the bus interface unit.

21. The processor of claim 19, wherein the bus interface unit includes a scratchpad, and wherein information stored by the scratchpad is passed to the central processing unit via the second bus, the translation unit and the first bus as a result of the command being received and executed by one of the engines in the bus interface unit.

22. The processor of claim 19, wherein the micro-engines include a plurality of registers, and wherein information stored by a particular one of the registers is passed to the central processing unit via the second bus, the translation unit and the first bus as a result of the command being received and executed by one of the engines in the bus interface unit.

23. The processor of claim 19, wherein the bus interface unit includes a plurality of registers, and wherein information addressed to the input-output space of the central processing unit is passed to a particular one of the registers via the translation unit and the second bus as a result of the command from the translation unit being received and executed by one of the engines in the bus interface unit.

24. The processor of claim 19, wherein the bus interface unit includes a scratchpad, and wherein information addressed to the input-output space of the central processing unit is written to the scratchpad via the translation unit and the second bus as a result of the command being received and executed by one of the engines in the bus interface unit.

25. The processor of claim 19, wherein the micro-engines include a plurality of registers, and wherein information addressed to the input-output space of the central processing unit is written to a particular one of the registers via the translation unit and the second bus as a result of the command being received and executed by one of the engines in the bus interface unit.

26. The processor of claim 19 including a bus for sending control signals from the bus interface unit to the translation unit, wherein the translation unit includes a plurality of registers that respond to control signals from the engines in the bus interface unit to supply targeted information onto the second bus or to receive targeted information from the second bus.

27. An article comprising a computer-readable medium which stores computer-executable instructions for causing a computer system to:

convert a request addressed to input-output space of a central processing unit to a corresponding command that simulates an operation between components in a multithreaded processing domain, wherein the multithreaded processing domain includes micro-coded processing engines that support multiple threads, the threads are simultaneously active but only one thread is in operation at any one time;

execute the command in the multithreaded processing domain; and

cause information to be accessed according to the request in response to executing the command.