Network processing pipeline chipset for routing and host packet processing

• US 7,362,772 B1
• Filed: 12/13/2002
• Issued: 04/22/2008
• Est. Priority Date: 12/13/2002
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: NVIDIA Corporation
• Inventors: Sidenblad, Paul J., Hicok, Gary D., Alfieri, Robert A., Parris, Mark A.
• Primary Examiner(s): To; Doris H.
• Assistant Examiner(s): Moore; Ian N.
• Application Number:
  US10/319,798
• Time in Patent Office: 1,957 Days
• Field of Search: 370/461, 370/412, 370/429, 370/235, 370/465, 370/462, 370/463, 709/225, 709/227, 709/230, 709/246, 709/238, 709/250, 711/118, 711/144, 711/170, 711/119, 712/228, 712/218
• US Class Current: 370/429
• CPC Class Codes: H04L 45/00 Routing or path finding of packets in data switching networks routing or path finding in wireless networks H04W40/00
• CPC Class Codes: H04L 49/3009 Header conversion, routing tables or routing tags

Abstract

A novel network architecture that integrates the functions of an internet protocol (IP) router into a network processing unit (NPU) that resides in a host computer'"'"'s chipset such that the host computer'"'"'s resources are perceived as separate network appliances. The NPU appears logically separate from the host computer even though, in one embodiment, it is sharing the same chip.

Claims

1. A network processing system, said system comprising:

a host central processing unit (CPU) configured to virtualize host resources to provide multiple virtual host computers through a host media access controller (HM);

a network processing unit attached to each one of the virtual host computers and having a plurality of processing subunits for performing both router packet processing and host packet processing, wherein one of said plurality of processing subunits is the HM that is configured to communicate with a first virtual host computer of the multiple virtual host computers using standard networking protocols to;

indicate that a packet processing event has occurred and said network processing unit is unable to process a packet corresponding to said packet processing event, and

forward said packet to host CPU for resolution;

a first software driver configured to manage communications between said host CPU and said network processing unit for host packet processing; and

a second software driver configured to manage communications between said network processing unit and other processing units in said network processing system for router packet processing using standard networking protocols without involvement by the CPU.

2. The system of claim 1, wherein said packet processing event relates to a decapsulation of said packet.

3. The system of claim 1, wherein said packet processing event relates to an authentication of said packet.

4. The system of claim 1, wherein said packet processing event relates to a decryption of said packet.

5. The system of claim 1, wherein said packet processing event relates to a decompression of said packet.

6. The system of claim 1, wherein said packet processing event relates to a checksum of said packet.

7. The system of claim 1, wherein said packet processing event relates to a routing of said packet.

8. The system of claim 1, wherein said packet processing event relates to a bridging of said packet.

9. The system of claim 1, wherein said packet processing event relates to a multicasting of said packet.

10. The system of claim 1, wherein said packet processing event relates to a firewalling of said packet.

11. The system of claim 1, wherein said packet processing event relates to a compression of said packet.

12. The system of claim 1, wherein said packet processing event relates to a encryption of said packet.

13. The system of claim 1, wherein said packet processing event relates to a fragmentation of said packet.

14. The system of claim 1, wherein said first software driver communicates with said first virtual host computer using a first circular push buffer that queues commands for output from said first software driver to said first virtual host computer and a second circular push buffer that queues commands for output from said first virtual host computer to said first software driver.

15. The system of claim 14, further comprising a system memory coupled to the host CPU and configured to store said first circular push buffer and said second circular push buffer.

16. The system of claim 14, wherein said first circular push buffer and said second circular push buffer also queue pointers to frame data for processing.

17. The system of claim 1, wherein one of said plurality of processing subunits is a sequence processor (SP) that is configured to generate random numbers.

18. The system of claim 1, further comprising a graphics processing unit (GPU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the GPU is configured to render graphics primitives.

19. The system of claim 1, further comprising an audio processing unit (APU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the APU is configured to perform encapsulation and decapsulation of audio packets.

20. The system of claim 1, further comprising a video processing unit (VPU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the VPU is configured to compress and decompress video packets.

21. The system of claim 1, further comprising a physics processing unit (PPU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the PPU is configured to process object positions, current velocity vectors, and force equations to produce new positions, velocity vectors, and collision information.

22. The system of claim 1, wherein the first virtual host computer of the multiple virtual host computers is configured to run a first operating system and a second virtual host computer of the multiple virtual host computers is configured to run a second operating system that is different than the first operating system.

23. A network processing system, said system comprising:

a host central processing unit (CPU);

a system memory of said host CPU;

a network processing unit having a plurality of processing subunits for performing both router packet processing and host packet processing, wherein one of said plurality of processing subunits is a media access controller interface (MI) subunit for communicating with at least one media access controller (MAC) to serve as an input and output point for traversal of packets, and to provide an asynchronous boundary interface between a first clock domain of said MAC and a second clock domain of said MI subunit; and

a host media access controller (HM) for communicating directly with said MI subunit and said host CPU, wherein said HM accesses a plurality of push buffers that are stored in said system memory to transfer commands or data between said HM and said host CPU; and

a first software driver configured to manage communications between said host CPU and said network processing unit for host packet processing, wherein the first software driver is configured to load host originated frames into the plurality of push buffers in order for the HM to select one push buffer from said plurality of push buffers by detecting how many of said plurality of push buffers that are allocated to store commands or data are not empty to produce non-empty push buffers for consumption by said MI and determine a priority of each non-empty push buffer.

24. The system of claim 23, wherein said network processing unit has a high priority interface to said system memory of said host CPU.

25. The system of claim 24, wherein said high priority interface to said system memory of said host CPU provides a higher priority than said host CPU.

26. The system of claim 24, wherein said high priority interface to said system memory of said host CPU provides a guaranteed latency to said system memory of said host CPU.

27. The system of claim 23, wherein said at least one media access controller (MAC) is a media independent interface MAC (miiMAC).

28. The system of claim 23, wherein said at least one media access controller (MAC) is a digital signal processor media access controller (dspMAC) that is connected to said NPU.

29. The system of claim 23, wherein said at least one media access controller (MAC) is a wireless media access controller (wiMAC) that is configured to operate according to a standard wireless communication protocol and connect to an external baseband.

30. The system of claim 23, wherein said first software driver communicates with the HM using a first circular push buffer that queues commands for output from said first software driver to said HM and a second circular push buffer that queues commands for output from said HM to said first software driver.

31. The system of claim 30, wherein the system memory is configured to store said first circular push buffer and said second circular push buffer.

32. The system of claim 30, wherein said first circular push buffer and said second circular push buffer also queue pointers to frame data for processing.

33. The system of claim 23, wherein one of said plurality of processing subunits is a sequence processor (SP) that is configured to generate random numbers.

34. The system of claim 23, further comprising a graphics processing unit (GPU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the GPU is configured to render graphics primitives.

35. The system of claim 23, further comprising an audio processing unit (APU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the APU is configured to perform encapsulation and decapsulation of audio packets.

36. The system of claim 23, further comprising a video processing unit (VPU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the VPU is configured to compress and decompress video packets.

37. The system of claim 23, further comprising a physics processing unit (PPU) that is coupled to said network processing unit that is allowed to communicate with another processing unit or with any external computer using standard internet protocols without involvement of said host CPU, wherein the PPU is configured to process object positions, current velocity vectors, and force equations to produce new positions, velocity vectors, and collision information.