US20090282172A1 - Memory access engine having multi-level command structure - Google Patents

Memory access engine having multi-level command structure Download PDF

Info

Publication number
US20090282172A1
US20090282172A1 US12/506,778 US50677809A US2009282172A1 US 20090282172 A1 US20090282172 A1 US 20090282172A1 US 50677809 A US50677809 A US 50677809A US 2009282172 A1 US2009282172 A1 US 2009282172A1
Authority
US
United States
Prior art keywords
dma
command
accessed
entry
commands
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US12/506,778
Inventor
Alexander G. MacInnis
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to US12/506,778 priority Critical patent/US20090282172A1/en
Publication of US20090282172A1 publication Critical patent/US20090282172A1/en
Assigned to BANK OF AMERICA, N.A., AS COLLATERAL AGENT reassignment BANK OF AMERICA, N.A., AS COLLATERAL AGENT PATENT SECURITY AGREEMENT Assignors: BROADCOM CORPORATION
Assigned to BROADCOM CORPORATION reassignment BROADCOM CORPORATION TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENTS Assignors: BANK OF AMERICA, N.A., AS COLLATERAL AGENT
Abandoned legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/20Handling requests for interconnection or transfer for access to input/output bus
    • G06F13/28Handling requests for interconnection or transfer for access to input/output bus using burst mode transfer, e.g. direct memory access DMA, cycle steal
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00General purpose image data processing
    • G06T1/60Memory management

Definitions

  • the present invention relates generally to direct memory access (DMA), and, more particularly, to a DMA engine having a multi-level command structure.
  • DMA direct memory access
  • One aspect of the present invention is directed to a direct memory access system utilizing a local memory that stores a plurality of DMA command lists, each comprising at least one DMA command.
  • a command queue can hold a plurality of entries, each entry comprising a pointer field and a sequence field. The pointer field pointing to one of the DMA command lists. The sequence field holds a sequence value.
  • a DMA engine accesses an entry in the command queue and then accesses the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry. The DMA engine performs the DMA operations specified by the accessed DMA commands. The DMA engine makes available the sequence value held in the sequence field of the accessed entry when all of the DMA commands in the accessed command list have been performed.
  • the command queue is part of the DMA engine.
  • Another embodiment of the present invention is directed to a digital media processing system including a memory element, a command queue, a media processor and a DMA engine.
  • the memory element stores a plurality of DMA command lists, each comprising at least one DMA command.
  • the command queue is adapted to hold a plurality of entries, each entry comprising a pointer field and a sequence field.
  • the pointer field points to one of the DMA command lists.
  • the sequence field holds a sequence value.
  • the media processor is adapted to process digital media data elements.
  • the media processor provides entries to the command queue in order to effect the performance of corresponding DMA operations. Each entry corresponds to a specified media data element.
  • FIG. 2 is a functional block diagram of a decoding system according to an illustrative embodiment of the present invention.
  • FIG. 4 is a functional block diagram depicting a command structure of DMA engine according to an illustrative embodiment of the present invention.
  • the analog video decoder 126 digitizes and decodes an analog video signal (NTSC or PAL) and provides the decoded data to the display engine 120 .
  • the graphics engine 122 processes graphics data in the data stream and provides the processed graphics data to the display engine 120 .
  • the display engine 120 prepares decoded video and graphics data for display and provides the data to display encoders 124 , which provide an encoded video signal to a display device.
  • aspects of an illustrative embodiment of the present invention relate to the architecture of digital video decoder 116 .
  • aspects of the present invention can also be employed in decoders of other types of media, for example, audio decoder 104 .
  • the acceleration modules 208 and 210 are hardware accelerators that accelerate special decoding tasks that would otherwise be bottlenecks for real-time media decoding if these tasks were handled by the core processor 202 alone. This helps the core processor 202 achieve the required performance.
  • the co-processor 206 is also a hardware accelerator that communicates with the core processor 202 via a co-processor interface of the core processor 202 .
  • the decoding system 200 is a video decoding system
  • the co-processor 206 is a variable-length decoder
  • the acceleration modules perform one or more video decoding tasks, such as inverse quantization, inverse discrete cosine transformation, pixel filtering, motion compensation and deblocking.
  • the system of FIGS. 2 and 3 are illustrative only. In accordance with the present invention, the decoding system 200 can have any number of hardware accelerators.
  • the core processor 202 also orchestrates a data unit processing pipeline (such as a macroblock processing pipeline) for the acceleration modules 206 , 208 and 210 and fetches the required data from main memory 110 via the bridge module 204 .
  • the core processor 202 also handles some data processing tasks.
  • picture level processing including sequence headers, GOP headers, picture headers, time stamps, macroblock-level information, except the block coefficients, and buffer management, are performed directly and sequentially by the core processor 202 , without using the accelerators 206 , 208 , 210 , except for using a variable-length decoder 206 to accelerate general bitstream parsing.
  • the core processor 202 is a MIPS processor, such as a MIPS32 implementation, for example.
  • FIG. 4 is a functional block diagram depicting a command structure of DMA engine 204 according to an illustrative embodiment of the present invention.
  • Each command list 470 , 480 , 490 has at least one command, and the commands are all sequential in memory, with an indication as to which command is the last command in a given list.
  • the command lists 470 , 480 , 490 are not linked lists, they are just lists.
  • the DMA goes to the list 470 , 480 , 490 pointed to by the next entry 410 , 420 , 430 , 440 in the queue 400 .
  • the sequence number of that entry is sent to a DMA Status output port, which is available to the core processor 202 as part of the central status register.
  • Firmware on the core processor 202 creates the sequence numbers. By policy, the sequence numbers should be sequential, i.e., increasing modulo 16 (0, 1, 2, . .
  • the firmware can identify, at any time, which entry in the queue 400 , i.e., which list ( 470 , 480 , 490 ) of DMA commands, has been completed. This is very beneficial if the list of commands spans more than one MBlock time.

Abstract

A direct memory access system utilizing a local memory that stores a plurality of DMA command lists, each comprising at least one DMA command. A command queue can hold a plurality of entries, each entry comprising a pointer field and a sequence field. The pointer field points to one of the DMA command lists. The sequence field holds a sequence value. A DMA engine accesses an entry in the command queue and then accesses the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry. The DMA engine performs the DMA operations specified by the accessed DMA commands. The DMA engine makes available the sequence value held in the sequence field of the accessed entry when all of the DMA commands in the accessed command list have been performed. In one embodiment, the command queue is part of the DMA engine.

Description

    PRIORITY CLAIM TO RELATED APPLICATION
  • This application is a continuation of U.S. patent application Ser. No. 11/945,702 which is a continuation of U.S. patent application Ser. No. 10/404,074 (now U.S. Pat. No. 7,302,503), filed Apr. 1, 2003, which claims priority to and claims benefit from U.S. Provisional Patent Application Ser. No. 60/369,210, entitled “MEMORY ACCESS ENGINE HAVING MULTI-LEVEL COMMAND STRUCTURE” filed on Apr. 1, 2002, all of which are hereby expressly incorporated herein by reference.
  • INCORPORATION BY REFERENCE OF RELATED APPLICATIONS
  • The following U.S. patent applications are related to the present application and are hereby specifically incorporated by reference: patent application Ser. No. 10/114,679, entitled “METHOD OF OPERATING A VIDEO DECODING SYSTEM” (Attorney Ref. No. 13305US01); patent application Ser. No. 10/114,797, entitled “METHOD OF COMMUNICATING BETWEEN MODULES IN A DECODING SYSTEM” (Attorney Ref. No. 13304US01); patent application Ser. No. 10/114,798, entitled “VIDEO DECODING SYSTEM SUPPORTING MULTIPLE STANDARDS” (Attorney Ref. No. 13301US01); patent application Ser. No. 10/114,619, entitled “INVERSE DISCRETE COSINE TRANSFORM SUPPORTING MULTIPLE DECODING PROCESSES” (Attorney Ref No. 13303US01); patent application Ser. No. 10/114,886, entitled “MEMORY SYSTEM FOR VIDEO DECODING SYSTEM” (Attorney Ref. No. 13388US01); and patent application Ser. No. 10/113,094, entitled “RISC PROCESSOR SUPPORTING ONE OR MORE UNINTERRUPTIBLE CO-PROCESSORS” (Attorney Ref. No. 13306US01); all filed on Apr. 1, 2002; patent application Ser. No. 10/293,633, entitled “PROGRAMMABLE VARIABLE LENGTH DECODER” (Attorney Ref No. 13391US02); filed on Nov. 12, 2002; and patent application Ser. No. 10/404,387, entitled “VIDEO DECODING SYSTEM HAVING A PROGRAMMABLE VARIABLE-LENGTH DECODER” (Attorney Ref No. 13300US02); and patent application Ser. No. 10/404,389, entitled “INVERSE QUANTIZER SUPPORTING MULTIPLE DECODING PROCESSES” (Attorney Ref. No. 13387US02); both filed on Apr. 1, 2003.
  • FIELD OF THE INVENTION
  • The present invention relates generally to direct memory access (DMA), and, more particularly, to a DMA engine having a multi-level command structure.
  • BACKGROUND OF THE INVENTION
  • Direct memory access, or DMA, is a method for direct communication from a peripheral device to memory with no programming involved. The data is moved to memory via the bus without program intervention. The only effect on the executing program is some slowing down of execution time because the DMA activity “steals” bus cycles that would otherwise be used to access memory for program execution.
  • DMA is a prime example of a hardware function in the chip that benefits from queued commands. There are two main aspects to this: (1) there are multiple DMA operations that need to be performed in one macroblock (MB) time interval, and (2) there are performance advantages in giving the DMA two macroblock times to complete each macroblock set of DMA operations, i.e., longer latency deadlines can help to tolerate DRAM latency. Because the DMA engine receives multiple commands in a given macroblock time interval, some form of command list is necessary for acceptable performance. There are up to (approximately) twelve DMA operations per macroblock for prediction fetches, and up to (approximately) four DMA operations per macroblock for write-backs, leading to a need for at least 16 DMA operations per macroblock. With the pipeline structured to allow two macroblock times of latency for each one macroblock set of operations, the complete list of commands is not expected to be empty at any time during normal operation. For example, commands for a first macroblock are sent to the DMA engine. Then commands for a second macroblock are sent to the DMA engine, while DMA operations for the first macroblock need not be complete yet. The first macroblock operations should be done prior to the time when the DMA commands for a third macroblock are sent to the DMA engine, while the second macroblock operations need not be done by that time. This means that it is not possible for firmware (or anything else) to determine that the DMA engine is “done” with a particular macroblock by checking to see if the entire command queue (assuming there is one) has been completed.
  • Further limitations and disadvantages of conventional and traditional approaches will become apparent to one of skill in the art through comparison of such systems with the present invention as set forth in the remainder of the present application with reference to the drawings.
  • SUMMARY OF THE INVENTION
  • One aspect of the present invention is directed to a direct memory access system utilizing a local memory that stores a plurality of DMA command lists, each comprising at least one DMA command. A command queue can hold a plurality of entries, each entry comprising a pointer field and a sequence field. The pointer field pointing to one of the DMA command lists. The sequence field holds a sequence value. A DMA engine accesses an entry in the command queue and then accesses the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry. The DMA engine performs the DMA operations specified by the accessed DMA commands. The DMA engine makes available the sequence value held in the sequence field of the accessed entry when all of the DMA commands in the accessed command list have been performed. In one embodiment, the command queue is part of the DMA engine.
  • Another embodiment of the present invention is directed to a method of implementing direct memory access. According to the method, a plurality of DMA command lists are stored, each comprising at least one DMA command. A command queue is maintained. The command queue is designed to hold a plurality of entries. Each entry includes a pointer field and a sequence field. The pointer field points to one of the DMA command lists. The sequence field holding a sequence value. An entry in the command queue is accessed. The DMA commands of the DMA command list pointed to by the pointer field of the accessed entry are then accessed. The DMA operations specified by the accessed DMA commands are performed. The sequence value held in the sequence field of the accessed entry is made available when all of the DMA commands in the accessed command list have been performed.
  • Another embodiment of the present invention is directed to a digital media processing system including a memory element, a command queue, a media processor and a DMA engine. The memory element stores a plurality of DMA command lists, each comprising at least one DMA command. The command queue is adapted to hold a plurality of entries, each entry comprising a pointer field and a sequence field. The pointer field points to one of the DMA command lists. The sequence field holds a sequence value. The media processor is adapted to process digital media data elements. The media processor provides entries to the command queue in order to effect the performance of corresponding DMA operations. Each entry corresponds to a specified media data element. The DMA engine accesses an entry in the command queue and then accesses the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry. The DMA engine performs DMA operations specified by the accessed DMA commands. The DMA engine provides the sequence value held in the sequence field of the accessed entry to the media processor when all of the DMA commands in the accessed command list have been performed.
  • It is understood that other embodiments of the present invention will become readily apparent to those skilled in the art from the following detailed description, wherein embodiments of the invention are shown and described only by way of illustration of the best modes contemplated for carrying out the invention. As will be realized, the invention is capable of other and different embodiments, and its several details are capable of modification in various other respects, all without departing from the spirit and scope of the present invention. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not as restrictive.
  • DESCRIPTION OF THE DRAWINGS
  • These and other features, aspects, and advantages of the present invention will become better understood with regard to the following description, appended claims, and accompanying drawings where:
  • FIG. 1 is a functional block diagram of a digital media system in which the present invention may be illustratively employed.
  • FIG. 2 is a functional block diagram of a decoding system according to an illustrative embodiment of the present invention.
  • FIG. 3 is a functional block diagram of a decoding system according to an illustrative embodiment of the present invention.
  • FIG. 4 is a functional block diagram depicting a command structure of DMA engine according to an illustrative embodiment of the present invention.
  • DETAILED DESCRIPTION
  • FIG. 1 is a functional block diagram of a digital media system in which the present invention may be illustratively employed. It will be noted, however, that the present invention can be employed in systems of widely varying architectures and widely varying designs.
  • The digital media system of FIG. 1 includes transport processor 102, audio decoder 104, direct memory access (DMA) controller 106, system memory controller 108, system memory 110, host CPU interface 112, host CPU 114, digital video decoder 116, display feeder 118, display engine 120, graphics engine 122, display encoders 124 and analog video decoder 126. The transport processor 102 receives and processes a digital media data stream. The transport processor 102 provides the audio portion of the data stream to the audio decoder 104 and provides the video portion of the data stream to the digital video decoder 116. In one embodiment, the audio and video data is stored in main memory 110 prior to being provided to the audio decoder 104 and the digital video decoder 116. The audio decoder 104 receives the audio data stream and produces a decoded audio signal. DMA controller 106 controls data transfer amongst main memory 110 and memory units contained in elements such as the audio decoder 104 and the digital video decoder 116. The system memory controller 108 controls data transfer to and from system memory 110. In an illustrative embodiment, system memory 110 is a dynamic random access memory (DRAM) unit. The digital video decoder 116 receives the video data stream, decodes the video data and provides the decoded data to the display engine 120 via the display feeder 118. The analog video decoder 126 digitizes and decodes an analog video signal (NTSC or PAL) and provides the decoded data to the display engine 120. The graphics engine 122 processes graphics data in the data stream and provides the processed graphics data to the display engine 120. The display engine 120 prepares decoded video and graphics data for display and provides the data to display encoders 124, which provide an encoded video signal to a display device.
  • Aspects of an illustrative embodiment of the present invention relate to the architecture of digital video decoder 116. However aspects of the present invention can also be employed in decoders of other types of media, for example, audio decoder 104.
  • FIG. 2 is a functional block diagram of a media decoding system 200, according to an illustrative embodiment of the present invention. The digital media decoding system 200 of FIG. 2 can illustratively be employed to implement the digital video decoder 116 of FIG. 1, or, alternatively, to implement audio decoder 104. Decoding system 200 includes a core decoder microprocessor 202, bridge module 204, co-processor 206, two hardware accelerators 208 and 210, decoder memory module 212, register bus 214 and system bus 216. Register bus 214 and system bus 216 communicate with the external host 114 and main memory 110. The bridge module 204 includes direct memory access (DMA) functionality. In an illustrative embodiment, the bridge module 204 is a “switch center” to arbitrate and interface between different modules. In an alternative embodiment, the bridge module 204 operates such that buses connect different modules directly, as shown in FIG. 3.
  • The acceleration modules 208 and 210 are hardware accelerators that accelerate special decoding tasks that would otherwise be bottlenecks for real-time media decoding if these tasks were handled by the core processor 202 alone. This helps the core processor 202 achieve the required performance. In an illustrative embodiment, the co-processor 206 is also a hardware accelerator that communicates with the core processor 202 via a co-processor interface of the core processor 202. In an illustrative embodiment wherein the decoding system 200 is a video decoding system, the co-processor 206 is a variable-length decoder, and the acceleration modules perform one or more video decoding tasks, such as inverse quantization, inverse discrete cosine transformation, pixel filtering, motion compensation and deblocking. The system of FIGS. 2 and 3 are illustrative only. In accordance with the present invention, the decoding system 200 can have any number of hardware accelerators.
  • The core processor 202 is the central control unit of the decoding system 200. In an illustrative embodiment of the present invention, the core processor 202 receives the data units from the bitstream to be decoded. The core processor 202 prepares the data for decoding. In an embodiment wherein the data being decoded is video data, the data unit comprises macroblock coefficient data. The core processor extracts the data for each data unit. After extracting the data for each data unit, the core processor 202 illustratively deposits the data in decoder memory 212. In an alternative embodiment, the core processor 202 provides the data directly to the co-processor 206 for processing by the co-processor 206. The core processor 202 also orchestrates a data unit processing pipeline (such as a macroblock processing pipeline) for the acceleration modules 206, 208 and 210 and fetches the required data from main memory 110 via the bridge module 204. The core processor 202 also handles some data processing tasks. Where decoding system 200 is a video decoding system, picture level processing, including sequence headers, GOP headers, picture headers, time stamps, macroblock-level information, except the block coefficients, and buffer management, are performed directly and sequentially by the core processor 202, without using the accelerators 206, 208, 210, except for using a variable-length decoder 206 to accelerate general bitstream parsing. In an illustrative embodiment of the present invention, the core processor 202 is a MIPS processor, such as a MIPS32 implementation, for example.
  • The co-processor 206 retrieves data that was placed in decoder memory 212 by the core processor 202 and performs one or more decoding functions on the retrieved data. In an alternative embodiment, the co-processor 206 receives the data to be processed directly from the core processor 202. After processing the received data, the co-processor 206 deposits the processed data in decoder memory 212. The data (such as DCT coefficients) deposited in decoder memory 212 by the co-processor 206 are processed by hardware accelerator module 208. After processing the data, hardware accelerator 208 deposits the processed data in decoder memory 212. The data deposited in decoder memory 212 by hardware accelerator 208 are processed by hardware accelerator module 210. After processing the data, hardware accelerator 210 deposits the processed data in decoder memory 212.
  • The bridge module 204 arbitrates and moves data between decoder memory 212 and main memory 110. The bridge interface 204 includes a direct memory access (DMA) engine. The bridge interface 204 illustratively includes an internal bus network that includes arbiters and the DMA engine. The bridge module 204 serves as an asynchronous interface to the system buses.
  • Decoder memory 212 is used to store data unit data and other time-critical data used during the decoding process. Each hardware block 206, 208, 210 accesses decoder memory 212 to either read the data to be processed or write processed data back. In an illustrative embodiment of the present invention, all currently used data is stored in decoder memory 212 to minimize access to main memory. The co-processor 206 and hardware accelerators 208 and 210 use the decoder memory 212 as the source and destination memory for their normal operation. Each module accesses the data in decoder memory 212 as the data units (such as macroblock data units) are processed through the system. In an illustrative embodiment, decoder memory 202 is a static random access memory (SRAM) unit. The CPU 114 has access to decoder memory 212, and the bridge module 204 can transfer data between decoder memory 212 and the main system memory (DRAM) 110. The arbiter for decoder memory 212 is in the bridge module 204.
  • FIG. 4 is a functional block diagram depicting a command structure of DMA engine 204 according to an illustrative embodiment of the present invention. There can be any number of lists 470, 480, 490 of DMA commands placed in decoder memory 212. Each command list 470, 480, 490 has at least one command, and the commands are all sequential in memory, with an indication as to which command is the last command in a given list. In an illustrative embodiment, the command lists 470, 480, 490 are not linked lists, they are just lists. There is also a command queue 400 in the DMA (in the bridge 204) itself. Each entry 410, 420, 430, 440 in the command queue 400 has two fields: a pointer 460 to a command list 470, 480, 490 and a sequence number 450. Entries are pushed onto the queue by the core processor 202 writing to a single queue input address. The DMA engine 204 takes the commands off the queue 400 in FIFO order and performs the operations specified by the commands in the command list (such as list 470, 480 or 490) that is pointed to by the command queue entry (such as 410, 420, 430, 440). Once one list 470, 480, 490 has been completed, the DMA goes to the list 470, 480, 490 pointed to by the next entry 410, 420, 430, 440 in the queue 400. After all of the commands (in a list such as list 470, 480 or 490) corresponding to an entry (such as 410, 420, 430 or 440) in the queue 400 have been completed, the sequence number of that entry is sent to a DMA Status output port, which is available to the core processor 202 as part of the central status register. Firmware on the core processor 202 creates the sequence numbers. By policy, the sequence numbers should be sequential, i.e., increasing modulo 16 (0, 1, 2, . . . , 15, 0, . . . ). With sequence numbers, the firmware can identify, at any time, which entry in the queue 400, i.e., which list (470, 480, 490) of DMA commands, has been completed. This is very beneficial if the list of commands spans more than one MBlock time.
  • Each DMA command, in the lists, should be as compact as possible, while providing the required functionality. According to an illustrative embodiment of the present invention, each command fits into two 32-bit words, with one word (e.g., the second) containing the DRAM address for the transfer, and the other word (first) containing bits for: direction of transfer, start address in SRAM/module; two bits to select which SRAM or module; special transaction type (for prediction reads); length of transfer; and whether this is the last command in the list.
  • Although a preferred embodiment of the present invention has been described, it should not be construed to limit the scope of the appended claims. For example, the present invention is applicable to any type of coded data, in addition to the media data illustratively described herein. Those skilled in the art will understand that various modifications may be made to the described embodiment. Moreover, to those skilled in the various arts, the invention itself herein will suggest solutions to other tasks and adaptations for other applications. It is therefore desired that the present embodiments be considered in all respects as illustrative and not restrictive, reference being made to the appended claims rather than the foregoing description to indicate the scope of the invention.

Claims (20)

1. A direct memory access (DMA) system comprising:
a memory element storing a plurality of DMA command lists, each comprising at least one DMA command;
a command queue adapted to hold a plurality of entries, each entry comprising a pointer field and a sequence field, the pointer field pointing to one of the DMA command lists, the sequence field holding a sequence value; and
a DMA engine adapted to access an entry in the command queue and to access the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry, the DMA engine further adapted to perform DMA operations specified by the accessed DMA commands, the DMA engine further adapted to make available the sequence value held in the sequence field of the accessed entry when all of the DMA commands in the accessed command list have been performed.
2. The DMA system of claim 1 wherein the command queue is included in the DMA engine.
3. The DMA system of claim 1 wherein a last DMA command in each command list includes an indication that it is the last command in the list.
4. The DMA system of claim 1 further comprising a processor adapted to provide the entries to the command queue in order to effect the performance of corresponding DMA operations.
5. The DMA system of claim 4 wherein the DMA engine is adapted to provide the sequence value to the processor when all of the DMA commands in the accessed command list have been performed.
6. The DMA system of claim 1 wherein when all of the DMA commands in the accessed command list have been performed, the DMA engine is adapted to access a next entry in the command queue and to access the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry, the DMA engine further adapted to perform DMA operations specified by the accessed DMA commands, the DMA engine further adapted to make available the sequence value held in the sequence field of the accessed entry when all of the DMA commands in the accessed command list have been performed.
7. The DMA system of claim 1 wherein the memory element is a static RAM memory element.
8. A method of implementing direct memory access (DMA) comprising:
storing a plurality of DMA command lists, each comprising at least one DMA command;
maintaining a command queue adapted to hold a plurality of entries, each entry comprising a pointer field and a sequence field, the pointer field pointing to one of the DMA command lists, the sequence field holding a sequence value;
accessing an entry in the command queue;
accessing the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry;
performing DMA operations specified by the accessed DMA commands; and
making available the sequence value held in the sequence field of the accessed entry when all of the DMA commands in the accessed command list have been performed.
9. The method of claim 8 wherein a last DMA command in each command list includes an indication that it is the last command in the list.
10. The method of claim 8 wherein a processor provides the entries to the command queue in order to effect the performance of corresponding DMA operations.
11. The method of claim 10 wherein making available the sequence value comprises providing the sequence value to the processor when all of the DMA commands in the accessed command list have been performed.
12. The method of claim 8 wherein the plurality of DMA command lists are stored in a static RAM memory element.
13. The method of claim 8 further comprising:
when all of the DMA commands in the accessed command list have been performed, accessing a next entry in the command queue;
accessing the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry;
performing DMA operations specified by the accessed DMA commands; and
making available the sequence value held in the sequence field of the accessed entry when all of the DMA commands in the accessed command list have been performed.
14. A digital media processing system comprising:
a memory element storing a plurality of direct memory access (DMA) command lists, each comprising at least one DMA command;
a command queue adapted to hold a plurality of entries, each entry comprising a pointer field and a sequence field, the pointer field pointing to one of the DMA command lists, the sequence field holding a sequence value;
a media processor adapted to process digital media data elements and adapted to provide entries to the command queue in order to effect the performance of corresponding DMA operations, each entry corresponding to a specified media data element; and
a DMA engine adapted to access an entry in the command queue and to access the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry, the DMA engine further adapted to perform DMA operations specified by the accessed DMA commands, the DMA engine further adapted to provide the sequence value held in the sequence field of the accessed entry to the media processor when all of the DMA commands in the accessed command list have been performed.
15. The digital media processing system of claim 14 wherein the command queue is included in the DMA engine.
16. The digital media processing system of claim 14 wherein a last DMA command in each command list includes an indication that it is the last command in the list.
17. The digital media processing system of claim 14 wherein when all of the DMA commands in the accessed command list have been performed, the DMA engine is adapted to access a next entry in the command queue and to access the DMA commands of the DMA command list pointed to by the pointer field of the accessed entry, the DMA engine further adapted to perform DMA operations specified by the accessed DMA commands, the DMA engine further adapted to provide the sequence value held in the sequence field of the accessed entry to the media processor when all of the DMA commands in the accessed command list have been performed.
18. The digital media processing system of claim 14 wherein the memory element is a static RAM memory element.
19. The digital media processing system of claim 14 wherein the media processor is a video processor adapted to process digital video data elements and adapted to provide entries to the command queue in order to effect the performance of corresponding DMA operations, each entry corresponding to a specified video data element.
20. The digital media processing system of claim 19 wherein the video processor is adapted to process macroblock data elements and adapted to provide entries to the command queue in order to effect the performance of corresponding DMA operations, each entry corresponding to a specified macroblock data element.
US12/506,778 2002-04-01 2009-07-21 Memory access engine having multi-level command structure Abandoned US20090282172A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US12/506,778 US20090282172A1 (en) 2002-04-01 2009-07-21 Memory access engine having multi-level command structure

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US36921002P 2002-04-01 2002-04-01
US10/404,074 US7302503B2 (en) 2002-04-01 2003-04-01 Memory access engine having multi-level command structure
US11/945,702 US7565462B2 (en) 2002-04-01 2007-11-27 Memory access engine having multi-level command structure
US12/506,778 US20090282172A1 (en) 2002-04-01 2009-07-21 Memory access engine having multi-level command structure

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US11/945,702 Continuation US7565462B2 (en) 2002-04-01 2007-11-27 Memory access engine having multi-level command structure

Publications (1)

Publication Number Publication Date
US20090282172A1 true US20090282172A1 (en) 2009-11-12

Family

ID=31997102

Family Applications (3)

Application Number Title Priority Date Filing Date
US10/404,074 Expired - Fee Related US7302503B2 (en) 2002-04-01 2003-04-01 Memory access engine having multi-level command structure
US11/945,702 Expired - Lifetime US7565462B2 (en) 2002-04-01 2007-11-27 Memory access engine having multi-level command structure
US12/506,778 Abandoned US20090282172A1 (en) 2002-04-01 2009-07-21 Memory access engine having multi-level command structure

Family Applications Before (2)

Application Number Title Priority Date Filing Date
US10/404,074 Expired - Fee Related US7302503B2 (en) 2002-04-01 2003-04-01 Memory access engine having multi-level command structure
US11/945,702 Expired - Lifetime US7565462B2 (en) 2002-04-01 2007-11-27 Memory access engine having multi-level command structure

Country Status (1)

Country Link
US (3) US7302503B2 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110276775A1 (en) * 2010-05-07 2011-11-10 Mosaid Technologies Incorporated Method and apparatus for concurrently reading a plurality of memory devices using a single buffer
US20150331638A1 (en) * 2012-08-23 2015-11-19 Apple Inc. Efficient enforcement of command execution order in solid state drives
US10241946B2 (en) 2017-01-18 2019-03-26 Nxp Usa, Inc. Multi-channel DMA system with command queue structure supporting three DMA modes
US11445227B2 (en) 2018-06-12 2022-09-13 Ela KLIOTS SHAPIRA Method and system for automatic real-time frame segmentation of high resolution video streams into constituent features and modifications of features in each frame to simultaneously create multiple different linear views from same video source
US11943489B2 (en) 2022-08-03 2024-03-26 Snakeview Data Science, Ltd. Method and system for automatic real-time frame segmentation of high resolution video streams into constituent features and modifications of features in each frame to simultaneously create multiple different linear views from same video source

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7757061B2 (en) 2005-05-03 2010-07-13 Micron Technology, Inc. System and method for decoding commands based on command signals and operating state
US7831801B1 (en) * 2006-08-30 2010-11-09 Xilinx, Inc. Direct memory access-based multi-processor array
US20080126779A1 (en) * 2006-09-19 2008-05-29 Ned Smith Methods and apparatus to perform secure boot
US7822885B2 (en) * 2007-10-16 2010-10-26 Applied Micro Circuits Corporation Channel-less multithreaded DMA controller
US20090133097A1 (en) * 2007-11-15 2009-05-21 Ned Smith Device, system, and method for provisioning trusted platform module policies to a virtual machine monitor
KR20110013868A (en) * 2009-08-04 2011-02-10 삼성전자주식회사 Multi processor system having multi command set operation and priority operation function
US8458377B2 (en) * 2010-03-05 2013-06-04 Lsi Corporation DMA engine capable of concurrent data manipulation
KR20140129607A (en) * 2013-04-30 2014-11-07 주식회사 칩스앤미디어 Method and apparatus for processing moving image
US10013385B2 (en) 2014-11-13 2018-07-03 Cavium, Inc. Programmable validation of transaction requests
US20160139806A1 (en) * 2014-11-13 2016-05-19 Cavium, Inc. Independent Ordering Of Independent Transactions

Citations (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4783730A (en) * 1986-09-19 1988-11-08 Datapoint Corporation Input/output control technique utilizing multilevel memory structure for processor and I/O communication
US4807111A (en) * 1987-06-19 1989-02-21 International Business Machines Corporation Dynamic queueing method
US5237661A (en) * 1989-05-29 1993-08-17 Hitachi, Ltd. Buffer management method and system therefor using an I/O buffer on main memory and utilizing virtual memory and page fixing
US5255367A (en) * 1987-09-04 1993-10-19 Digital Equipment Corporation Fault tolerant, synchronized twin computer system with error checking of I/O communication
US5404522A (en) * 1991-09-18 1995-04-04 International Business Machines Corporation System for constructing a partitioned queue of DMA data transfer requests for movements of data between a host processor and a digital signal processor
US5555396A (en) * 1994-12-22 1996-09-10 Unisys Corporation Hierarchical queuing in a system architecture for improved message passing and process synchronization
US5577211A (en) * 1994-05-11 1996-11-19 Ibm Corporation System and method using chained structure queues for ordering of message delivery between connected nodes wherein unsuccessful message portion is skipped and retried
US5600646A (en) * 1995-01-27 1997-02-04 Videoserver, Inc. Video teleconferencing system with digital transcoding
US5634099A (en) * 1994-12-09 1997-05-27 International Business Machines Corporation Direct memory access unit for transferring data between processor memories in multiprocessing systems
US5678062A (en) * 1993-02-22 1997-10-14 Hitachi, Ltd. Input/output control method and data processor
US5708849A (en) * 1994-01-26 1998-01-13 Intel Corporation Implementing scatter/gather operations in a direct memory access device on a personal computer
US5797041A (en) * 1992-06-30 1998-08-18 Hitachi, Ltd. Communication control system utilizing a shared buffer composed of first and second descriptors which are managed by high and low level protocols
US5809334A (en) * 1996-09-24 1998-09-15 Allen-Bradley Company, Llc Receive packet pre-parsing by a DMA controller
US5923852A (en) * 1996-09-04 1999-07-13 Advanced Micro Devices, Inc. Method and system for fast data transmissions in a processing system utilizing interrupts
US5933654A (en) * 1996-09-24 1999-08-03 Allen-Bradley Company, Llc Dynamic buffer fracturing by a DMA controller
US5968153A (en) * 1996-06-21 1999-10-19 Digital Equipment Corporation Mechanism for high bandwidth DMA transfers in a PCI environment
US5970069A (en) * 1997-04-21 1999-10-19 Lsi Logic Corporation Single chip remote access processor
US6052387A (en) * 1997-10-31 2000-04-18 Ncr Corporation Enhanced interface for an asynchronous transfer mode segmentation controller
US6070194A (en) * 1997-12-17 2000-05-30 Advanced Micro Devices, Inc. Using an index and count mechanism to coordinate access to a shared resource by interactive devices
US6145016A (en) * 1998-09-03 2000-11-07 Advanced Micro Devices, Inc. System for transferring frame data by transferring the descriptor index data to identify a specified amount of data to be transferred stored in the host computer
US6167465A (en) * 1998-05-20 2000-12-26 Aureal Semiconductor, Inc. System for managing multiple DMA connections between a peripheral device and a memory and performing real-time operations on data carried by a selected DMA connection
US6189064B1 (en) * 1998-11-09 2001-02-13 Broadcom Corporation Graphics display system with unified memory architecture
US6205494B1 (en) * 1998-12-18 2001-03-20 Western Digital Corporation Controller with automatic generation of linked list of data transfer descriptors for sequential commands, with linked list being used for execution of sequential data transfers
US6212593B1 (en) * 1998-06-01 2001-04-03 Advanced Micro Devices, Inc. Method and apparatus for generating interrupts on a buffer by buffer basis in buffer descriptor ring direct memory access system
US20010021949A1 (en) * 1997-10-14 2001-09-13 Alacritech, Inc. Network interface device employing a DMA command queue
US6314477B1 (en) * 1998-10-30 2001-11-06 Agilent Technologies, Inc. Performance of fibre channel protocol sequence reassembly using expected frame information and buffer list calculations
US20010049755A1 (en) * 2000-06-02 2001-12-06 Michael Kagan DMA doorbell
US6538656B1 (en) * 1999-11-09 2003-03-25 Broadcom Corporation Video and graphics system with a data transport processor
US6681296B2 (en) * 2000-04-07 2004-01-20 Nintendo Co., Ltd. Method and apparatus for software management of on-chip cache
US6714553B1 (en) * 1998-04-15 2004-03-30 Top Layer Networks, Inc. System and process for flexible queuing of data packets in network switching
US6721320B1 (en) * 1998-12-18 2004-04-13 Lsi Logic Corporation Method and apparatus for fibre channel identification and retrieval
US6738358B2 (en) * 2000-09-09 2004-05-18 Intel Corporation Network echo canceller for integrated telecommunications processing

Patent Citations (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4783730A (en) * 1986-09-19 1988-11-08 Datapoint Corporation Input/output control technique utilizing multilevel memory structure for processor and I/O communication
US4807111A (en) * 1987-06-19 1989-02-21 International Business Machines Corporation Dynamic queueing method
US5255367A (en) * 1987-09-04 1993-10-19 Digital Equipment Corporation Fault tolerant, synchronized twin computer system with error checking of I/O communication
US5237661A (en) * 1989-05-29 1993-08-17 Hitachi, Ltd. Buffer management method and system therefor using an I/O buffer on main memory and utilizing virtual memory and page fixing
US5404522A (en) * 1991-09-18 1995-04-04 International Business Machines Corporation System for constructing a partitioned queue of DMA data transfer requests for movements of data between a host processor and a digital signal processor
US5797041A (en) * 1992-06-30 1998-08-18 Hitachi, Ltd. Communication control system utilizing a shared buffer composed of first and second descriptors which are managed by high and low level protocols
US5678062A (en) * 1993-02-22 1997-10-14 Hitachi, Ltd. Input/output control method and data processor
US5708849A (en) * 1994-01-26 1998-01-13 Intel Corporation Implementing scatter/gather operations in a direct memory access device on a personal computer
US5577211A (en) * 1994-05-11 1996-11-19 Ibm Corporation System and method using chained structure queues for ordering of message delivery between connected nodes wherein unsuccessful message portion is skipped and retried
US5634099A (en) * 1994-12-09 1997-05-27 International Business Machines Corporation Direct memory access unit for transferring data between processor memories in multiprocessing systems
US5555396A (en) * 1994-12-22 1996-09-10 Unisys Corporation Hierarchical queuing in a system architecture for improved message passing and process synchronization
US5600646A (en) * 1995-01-27 1997-02-04 Videoserver, Inc. Video teleconferencing system with digital transcoding
US5968153A (en) * 1996-06-21 1999-10-19 Digital Equipment Corporation Mechanism for high bandwidth DMA transfers in a PCI environment
US5923852A (en) * 1996-09-04 1999-07-13 Advanced Micro Devices, Inc. Method and system for fast data transmissions in a processing system utilizing interrupts
US5809334A (en) * 1996-09-24 1998-09-15 Allen-Bradley Company, Llc Receive packet pre-parsing by a DMA controller
US5933654A (en) * 1996-09-24 1999-08-03 Allen-Bradley Company, Llc Dynamic buffer fracturing by a DMA controller
US5970069A (en) * 1997-04-21 1999-10-19 Lsi Logic Corporation Single chip remote access processor
US20010021949A1 (en) * 1997-10-14 2001-09-13 Alacritech, Inc. Network interface device employing a DMA command queue
US6052387A (en) * 1997-10-31 2000-04-18 Ncr Corporation Enhanced interface for an asynchronous transfer mode segmentation controller
US6070194A (en) * 1997-12-17 2000-05-30 Advanced Micro Devices, Inc. Using an index and count mechanism to coordinate access to a shared resource by interactive devices
US6714553B1 (en) * 1998-04-15 2004-03-30 Top Layer Networks, Inc. System and process for flexible queuing of data packets in network switching
US6167465A (en) * 1998-05-20 2000-12-26 Aureal Semiconductor, Inc. System for managing multiple DMA connections between a peripheral device and a memory and performing real-time operations on data carried by a selected DMA connection
US6212593B1 (en) * 1998-06-01 2001-04-03 Advanced Micro Devices, Inc. Method and apparatus for generating interrupts on a buffer by buffer basis in buffer descriptor ring direct memory access system
US6145016A (en) * 1998-09-03 2000-11-07 Advanced Micro Devices, Inc. System for transferring frame data by transferring the descriptor index data to identify a specified amount of data to be transferred stored in the host computer
US6314477B1 (en) * 1998-10-30 2001-11-06 Agilent Technologies, Inc. Performance of fibre channel protocol sequence reassembly using expected frame information and buffer list calculations
US6189064B1 (en) * 1998-11-09 2001-02-13 Broadcom Corporation Graphics display system with unified memory architecture
US6205494B1 (en) * 1998-12-18 2001-03-20 Western Digital Corporation Controller with automatic generation of linked list of data transfer descriptors for sequential commands, with linked list being used for execution of sequential data transfers
US6721320B1 (en) * 1998-12-18 2004-04-13 Lsi Logic Corporation Method and apparatus for fibre channel identification and retrieval
US6538656B1 (en) * 1999-11-09 2003-03-25 Broadcom Corporation Video and graphics system with a data transport processor
US6681296B2 (en) * 2000-04-07 2004-01-20 Nintendo Co., Ltd. Method and apparatus for software management of on-chip cache
US20010049755A1 (en) * 2000-06-02 2001-12-06 Michael Kagan DMA doorbell
US6738358B2 (en) * 2000-09-09 2004-05-18 Intel Corporation Network echo canceller for integrated telecommunications processing

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110276775A1 (en) * 2010-05-07 2011-11-10 Mosaid Technologies Incorporated Method and apparatus for concurrently reading a plurality of memory devices using a single buffer
US20150331638A1 (en) * 2012-08-23 2015-11-19 Apple Inc. Efficient enforcement of command execution order in solid state drives
US10019196B2 (en) * 2012-08-23 2018-07-10 Apple Inc. Efficient enforcement of command execution order in solid state drives
US10241946B2 (en) 2017-01-18 2019-03-26 Nxp Usa, Inc. Multi-channel DMA system with command queue structure supporting three DMA modes
US11445227B2 (en) 2018-06-12 2022-09-13 Ela KLIOTS SHAPIRA Method and system for automatic real-time frame segmentation of high resolution video streams into constituent features and modifications of features in each frame to simultaneously create multiple different linear views from same video source
US11943489B2 (en) 2022-08-03 2024-03-26 Snakeview Data Science, Ltd. Method and system for automatic real-time frame segmentation of high resolution video streams into constituent features and modifications of features in each frame to simultaneously create multiple different linear views from same video source

Also Published As

Publication number Publication date
US20040047194A1 (en) 2004-03-11
US7302503B2 (en) 2007-11-27
US7565462B2 (en) 2009-07-21
US20080133786A1 (en) 2008-06-05

Similar Documents

Publication Publication Date Title
US7565462B2 (en) Memory access engine having multi-level command structure
US9032104B2 (en) Method and system for performing DMA in a multi-core system-on-chip using deadline-based scheduling
JP4426099B2 (en) Multiprocessor device having shared memory
US6944746B2 (en) RISC processor supporting one or more uninterruptible co-processors
US6167475A (en) Data transfer method/engine for pipelining shared memory bus accesses
US5884050A (en) Mechanism for high bandwidth DMA transfers in a PCI environment
US5522080A (en) Centralized control SIMD processor having different priority levels set for each data transfer request type and successively repeating the servicing of data transfer request in a predetermined order
KR100227278B1 (en) Cache control unit
US7630388B2 (en) Software defined FIFO memory for storing a set of data from a stream of source data
US20030185305A1 (en) Method of communicating between modules in a decoding system
KR20010110651A (en) A System for Transferring Data of Reusing a Plurality of Data Transferring Area Cyclically
KR20120070591A (en) Address translation unit with multiple virtual queues
WO2009130871A1 (en) Decoding device
US20050253858A1 (en) Memory control system and method in which prefetch buffers are assigned uniquely to multiple burst streams
JP2001216194A (en) Arithmetic processor
KR100218619B1 (en) Video processing with non-stalling interrupt service
JPH08241288A (en) Multitask processing system and image-processing system
US6700582B2 (en) Method and system for buffer management
US7007031B2 (en) Memory system for video decoding system
JP3260456B2 (en) Computer system, integrated circuit suitable for it, and requirement selection circuit
EP1351514A2 (en) Memory acces engine having multi-level command structure
US20090235010A1 (en) Data processing circuit, cache system, and data transfer apparatus
US20020184450A1 (en) Multifunctional I/O organizer unit for multiprocessor multimedia chips
JP5379223B2 (en) Information processing device
JPH11175464A (en) Arbitrating device and method therefor

Legal Events

Date Code Title Description
STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION

AS Assignment

Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH CAROLINA

Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001

Effective date: 20160201

Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH

Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001

Effective date: 20160201

AS Assignment

Owner name: BROADCOM CORPORATION, CALIFORNIA

Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENTS;ASSIGNOR:BANK OF AMERICA, N.A., AS COLLATERAL AGENT;REEL/FRAME:041712/0001

Effective date: 20170119