US20050036614A1 - DMA engine for fetching words in reverse bit order - Google Patents
DMA engine for fetching words in reverse bit order Download PDFInfo
- Publication number
- US20050036614A1 US20050036614A1 US10/736,125 US73612503A US2005036614A1 US 20050036614 A1 US20050036614 A1 US 20050036614A1 US 73612503 A US73612503 A US 73612503A US 2005036614 A1 US2005036614 A1 US 2005036614A1
- Authority
- US
- United States
- Prior art keywords
- word
- data
- sequential
- last
- data words
- 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
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/625—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using discrete cosine transform [DCT]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
- H04N19/423—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
Definitions
- the MPEG-2 standard uses video packets comprising any number of macroblocks to confine bit errors. Frames are represented by a set of macroblocks. The macroblocks are grouped into a data structure known as a video packet. In MPEG-2, all macroblock rows start with a new video packet. During the decoding of a video packet, if an error is encountered, the video decoder can simply drop the remaining macroblocks that followed the bit error. In this manner, only a small amount of information is lost as a result of the bit error.
- MPEG-4 Part 2 also uses video packets to confine bit errors.
- the video data is transmitted as a video elementary stream.
- the portions of the video elementary stream that are at the video packet level and lower are encoded with a variable length code.
- the video packet is defined such that the video packet can be decoded in a forward order, or the reverse order. Accordingly, after encountering an error, the video decoder can go to the end of the video packet, and start decoding in the reverse order, until the same error, or another error is encountered. In this manner, a greater portion of the video packet is recovered and reconstructed, in spite of encountering error(s).
- the video decoder needs to be able to receive and decode the video bitstream in reverse order in real-time.
- the video elementary stream is stored in a memory known as the compressed data buffer in the forward order, along with a table that indicates the starting addresses of each video packet.
- the video decoder receives and decodes the video elementary stream by accessing the compressed data buffer. Upon encountering an error, the video decoder can receive the video packet at the ending address of the video packet and moving in the reverse order.
- Receiving the video packet in reverse order can be made possible by manipulating the memory access.
- the video decoder can sequentially access data words in reverse order. After accessing each data word, the video decoder can use logic to reverse the bit order of the data word.
- the foregoing adds significant operations to the video decoder and makes accessing and decoding in the reverse order difficult to perform in real-time.
- a direct memory access engine for fetching words in reverse order.
- a method for providing a plurality of sequential data words includes receiving a command to provide the plurality of sequential data words, wherein the plurality of sequential data words comprises a first data word and a last data word, and one or more data words between the first data word and the last data word, fetching a sequential portion of the sequential data words, said sequential portion comprising a first intermediate word, the last word, and one or more data words between the intermediate word and the last word, storing the sequential portion, transmitting at least a portion of the last data word, and transmitting at least a portion of the intermediate data words after transmitting at least the portion of the last data word.
- a system for providing a plurality of sequential data words comprises a state logic machine, a memory controller, a local buffer, and a port.
- the state logic machine receives a command to provide the plurality of sequential of sequential data words, wherein the plurality of sequential data words comprises a first data word and a last data word, and one or more data words between the first data word and the last data word.
- the memory controller fetches a sequential portion of the sequential data words, said sequential portion comprising a first intermediate word, the last word, and one or more data words between the intermediate word and the last word.
- the local buffer stores the sequential portion.
- the port transmits at least a portion of the last data word and transmits at least a portion of the intermediate data words after transmitting at least the portion of the last data word.
- a system for decoding a video packet comprises a compressed data buffer, a video decoder, and a direct memory access engine.
- the compressed data buffer comprises a plurality of sequential data words.
- the plurality of sequential data words store a video packet.
- the video decoder decodes the video packet.
- the direct memory access engine provides the video packet to the video decoder and comprises a state logic machine, a memory controller, a local buffer, and a port.
- the state logic machine receives a command to provide the plurality of sequential data words and a control signal indicating reverse order from the video decoder, wherein the plurality of sequential data words comprises a first data word and a last data word, and one or more data words between the first data word and the last data word.
- the memory controller fetches a sequential portion of the sequential data words, said sequential portion comprising a first intermediate word, the last word, and one or more data words between the intermediate word and the last word.
- the local buffer stores the sequential portion.
- the port transmits at least a portion of the last data word and transmits at least a portion of the intermediate data words after transmitting at least the portion of the last data word.
- FIG. 1 is a block diagram describing the encoding of video data
- FIG. 2 is a block diagram describing an exemplary decoder in accordance with an embodiment of the present invention
- FIG. 3 is a block diagram of an exemplary compressed data buffer
- FIG. 4 is a block diagram of an exemplary direct memory access engine in accordance with an embodiment of the present invention.
- FIG. 5 is a flow diagram for accessing data words in reverse order in accordance with an embodiment of the present invention.
- the video data 305 comprises a series of frames 310 .
- Each frame comprises two dimensional grids of luminance Y, chroma red Cr, and chroma blue Cb pixels 315 .
- the two-dimensional grids are divided into 8 ⁇ 8 blocks 335 , where four blocks 335 of luminance pixels Y are associated with a block 335 of chroma red Cr, and a block 335 of chroma blue Cb pixels.
- the four blocks of luminance pixels Y, the block of chroma red Cr, and the chroma blue Cb form a data structure known as a macroblock 337 .
- the macroblock 337 also includes additional parameters, including motion vectors.
- the macroblocks 337 representing a frame are grouped into different video packets 340 .
- the video packet 340 includes the macroblocks 337 in the video packet 340 , as well as additional parameters describing the video packet.
- Each of the video packets 340 forming the frame form the data portion of a picture structure 345 .
- the picture 345 includes the video packets 340 as well as additional parameters.
- the pictures are then grouped together as a group of pictures 350 .
- the group of pictures 350 also includes additional parameters. Groups of pictures 350 are then stored, forming what is known as a video elementary stream 355 .
- the video elementary stream 355 is then packetized to form a packetized elementary sequence 360 .
- Each packet is then associated with a transport header 365 a , forming what are known as transport packets 365 b.
- the transport packets 365 b can be multiplexed with other transport packets 365 b carrying other content, such as another video elementary stream 355 or an audio elementary stream.
- the multiplexed transport packets from what is known as a transport stream.
- the transport stream is transmitted over a communication medium for decoding and presentation.
- a processor that may include a CPU 490 , reads a stream of transport packets 365 b (a transport stream) into a transport stream buffer 432 within an SDRAM 430 .
- the data is output from the transport stream presentation buffer 432 and is then passed to a data transport processor 435 .
- the data transport processor then demultiplexes the MPEG transport stream into its PES constituents and passes the audio transport stream to an audio decoder 460 and the video transport stream to a video transport processor 440 .
- the video transport processor 440 converts the video transport stream into a video elementary stream and provides the video elementary stream to an MPEG video decoder 445 that decodes the video.
- the video elementary stream 355 is stored in a compressed data buffer (CDB) 447 .
- the MPEG video decoder 445 accesses the compressed data buffer (CDB) to receive the video elementary stream 355 .
- the video elementary stream 355 is decoded by the MPEG video decoder 445 resulting in the reconstructed video data 305 .
- the audio data is sent to the output blocks and the video data 305 is sent to a display engine 450 .
- the display engine 450 is responsible for and operable to scale the video picture, render the graphics, and construct the complete display among other functions.
- DAC digital to analog converter
- the digital audio is converted to analog in the audio digital to analog converter (DAC) 465 .
- the compressed data buffer 447 includes any number of data words 505 ( 1 ) . . . 505 ( m ).
- the data words can have any width.
- the data words can comprise 256 bit jumbo words (words).
- the compressed data buffer 447 stores the video elementary stream 355 .
- the video elementary stream 355 comprises any number of video packets 340 .
- the video packets 340 further comprise a video packet header and any number of macroblocks 337 .
- the compressed data buffer 447 also stores a start code table 507 .
- the start code table 507 associates each video packet 340 with its starting address in the compressed data buffer 447 .
- the video packet 340 can be first scanned forward, without evacuating data from memory, then number of bytes/bits can be counted, and then returned using the DMA engine.
- the MPEG video decoder 445 receives the video packets 340 from the video elementary stream 355 and decodes the video packets 340 .
- the video packet 340 is received and decoded by the MPEG video decoder 445 starting from the word 505 ( x ) storing the beginning of the video packet 340 , and proceeding to the word 505 ( n ) storing the end of the video packet 340 .
- a direct memory access (DMA) engine 510 facilitates receipt of the video packets 340 by the MPEG video decoder 445 .
- a processor can facilitate receipt of the video packets 340 .
- DMA engine 510 shall be interpreted to also include a processor that is operable to fetch video packets from memory.
- the MPEG video decoder 445 receives a video packet 340 by looking up the starting address and the ending address of a video packet 340 in the start code table 507 .
- the MPEG video decoder 445 can then command the DMA engine 510 to fetch the words 505 ( x ) . . . 505 ( n ) that store the video packet 340 . Responsive thereto, the DMA engine 510 fetches and provides the words 505 ( x ) . . . 505 ( n ) that store the video packet 340 .
- the DMA engine 510 provides the words 505 ( x ) . . . 505 ( n ) to an extractor 515 within the MPEG video decoder 445 in a serial manner, beginning with word 505 ( x ) and proceeding to the last word 505 ( n ).
- the MPEG decoder 445 decodes the video packet 340 , in a serial manner, beginning decoding with the first word 505 ( x ) and proceeding to the last word 505 ( n ).
- the extractor 515 and the DMA engine 510 operate in conjunction with each other, such that the words 505 are provided to the MPEG video decoder 445 at a dynamic rate that is in substantial relationship to the rate that the MPEG video decoder 445 is decoding the words 505 .
- the video packet 340 is defined such that the video packet 340 can be decoded in a forward order, or the reverse order. Accordingly, if the MPEG video decoder 445 encounters an error, the video decoder can go to the end of the video packet 340 , and start decoding in the reverse order.
- the MPEG video decoder 445 can start decoding the video packet 340 from word 505 ( n ) and decode in the reverse order, e.g., word 505 ( n ⁇ 1), 505 ( n ⁇ 2) . . . , etc.
- the MPEG video decoder 445 Upon detecting an error, the MPEG video decoder 445 transmits a command to the DMA engine 510 to fetch the words storing the video packet 340 , e.g., words 510 ( x ) . . . 510 ( n ), along with a reverse order signal. Responsive thereto, the DMA engine 510 provides the words 510 ( x ) . . . 510 ( n ) in the reverse order to the MPEG video decoder 445 .
- the DMA engine 510 Upon detecting an error, the MPEG video decoder 445 transmits a command to the DMA engine 510 to fetch the words storing the video packet 340 , e.g., words 510 ( x ) . . . 510 ( n ), along with a reverse order signal. Responsive thereto, the DMA engine 510 provides the words 510 ( x ) . . . 510 (
- the DMA engine 510 comprises a state logic machine 605 , a local buffer 610 , and a memory controller 620 .
- the local buffer 610 can comprise any amount of memory with any width of data words.
- the memory can comprise 128 32-bit words 611 ( 0 ) . . . 611 ( 127 ).
- the state logic machine 605 receives a command to fetch data words in an address range, e.g., 505 ( x )- 510 ( n ) from the MPEG video decoder 445 .
- the command can be accompanied by a control signal indicating that the data words in the address range are to be provided to the MPEG video decoder 445 in the reverse order, e.g., 505 ( n ), 505 ( n ⁇ 1) . . . 505 ( x ).
- the state logic machine 605 commands the memory controller 620 to retrieve a batch comprising a the last predetermined number of data words 505 in the provided address range, and store the predetermined number of data words in the local buffer 610 .
- the predetermined number of data words in the batch is less than or equal to the capacity of the local buffer 610 .
- the batch of data words 505 can include the last 16 words in the provided address range, e.g., data words 505 ( n ⁇ 15) . . . 505 ( n ).
- the state logic machine 605 After the batch of data words 505 ( n ⁇ 15) . . . 505 ( n ) is stored in the local buffer 610 , the state logic machine 605 causes the contents of the local buffer 610 to be provided to the MPEG video decoder 445 beginning with word 611 ( 127 ), and proceeding sequentially to word 611 ( 0 ). After the contents of the local buffer 610 , e.g., words 611 ( 127 ) . . . 611 ( 0 ), are provided to the MPEG video decoder 445 , the state logic machine 605 commands the memory controller 620 to fetch another batch comprising the predetermined number of words, e.g., data words 505 ( n ⁇ 31 ) . . .
- the data words 505 ( n ⁇ 15) . . . 505 ( n ) are stored in the local buffer 610 and provided to the MPEG decoder 445 .
- the state logic machine 605 truncates that portion of the predetermined number of data words that precedes the first data word 505 ( x ), and commands the memory controller 620 to fetch the truncated batch comprising the first data word 505 ( x ) and all data words 505 ( x +1), 505 ( x +2) . . . , following the first data word 505 ( x ) that have not been previously transmitted to the MPEG video decoder 445 .
- the foregoing provides the data words 505 ( x ) . . . 505 ( n ) as a set of 32 bit words starting from the last portion of 505 ( n ) and proceeding sequentially to the first portion of 505 ( x ).
- the bits forming the 32-bit words can be reversed with respect to one another, in any number of ways.
- the MPEG video decoder 445 can include logic that reverses the 32 bits of each word.
- the DMA engine 510 can include additional circuitry that causes the 32 bits of each word 611 to be provided to the MPEG video decoder 445 in the reverse order.
- the state logic machine 605 receives a command to fetch data words in an address range, e.g., 505 ( x )- 510 ( n ) from the MPEG video decoder 445 , accompanied by a control signal indicating that the data words in the address range are to be provided to the MPEG video decoder 445 in the reverse order, e.g., 505 ( n ), 505 ( n ⁇ 1) . . . 505 ( x ).
- the state logic machine 605 determines ( 706 ) if a predetermined number of words comprises the first word, 505 ( x ). Where the predetermined number of words comprises the first data word in the address range, e.g., data word 505 ( x ), the state logic machine 605 truncates that portion of the predetermined number of data words that precedes the first data word 505 ( x ) and commands ( 708 ) the memory controller 620 to fetch ( 709 ) the truncated batch comprising the first data word 505 ( x ) and all data words 505 ( x +1), 505 ( x +2) . . .
- the state logic machine 605 determines that the predetermined number of words does not comprise the first word, 505 ( x ), the state logic machine 605 commands ( 710 ) the memory controller 620 to fetch ( 715 ) a batch comprising the last predetermined number of data words 505 in the provided address range.
- the fetched data words 505 are stored ( 720 ) in the local buffer 610 .
- the state logic machine 605 causes the contents of the local buffer 610 to be provided ( 725 ) to the MPEG video decoder 445 beginning with word 611 ( 127 ), and proceeding sequentially to word 611 ( 0 ).
- the contents of the local buffer 610 e.g., words 611 ( 127 ) . . . 611 ( 0 )
- a determination ( 730 ) is made whether the first data word, data word 505 ( x ) has been provided to the MPEG video decoder 445 . If the first data word has not been provided to the MPEG video decoder at 730 , 705 - 730 are repeated. If the first data word has been provided to the MPEG video decoder at 730 , the process is completed.
- One embodiment of the present invention may be implemented as a board level product, as a single chip, application specific integrated circuit (ASIC), or with varying levels integrated on a single chip with other portions of the system as separate components.
- the degree of integration of the system will primarily be determined by speed and cost considerations. Because of the sophisticated nature of modern processors, it is possible to utilize a commercially available processor, which may be implemented external to an ASIC implementation of the present system. Alternatively, if the processor is available as an ASIC core or logic block, then the commercially available processor can be implemented as part of an ASIC device with various functions implemented as firmware.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Discrete Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
Description
- This application claims priority to Provisional Application for U.S. Patent, Ser. No. 60/494,666, (Attorney Docket Number 15137US01) entitled “DMA Engine for Fetching Words in Reverse Order”, filed Aug. 13, 2003, and Provisional Application for U.S. Patent, Ser. No. 60/494,746, (Attorney Docket Number 15138US01) entitled “DMA Engine for Fetching Words in Reverse Order”, filed Aug. 13, 2003, which are incorporated herein by reference for all purposes.
- This application is also related to Application for U.S. Patent, Serial No. ______, entitled “DMA Engine for Fetching Words in Reverse Order”, filed ______.
- [Not Applicable]
- [Not Applicable]
- The MPEG-2 standard uses video packets comprising any number of macroblocks to confine bit errors. Frames are represented by a set of macroblocks. The macroblocks are grouped into a data structure known as a video packet. In MPEG-2, all macroblock rows start with a new video packet. During the decoding of a video packet, if an error is encountered, the video decoder can simply drop the remaining macroblocks that followed the bit error. In this manner, only a small amount of information is lost as a result of the bit error.
- MPEG-4
Part 2 also uses video packets to confine bit errors. The video data is transmitted as a video elementary stream. The portions of the video elementary stream that are at the video packet level and lower are encoded with a variable length code. In MPEG-4Part 2, the video packet is defined such that the video packet can be decoded in a forward order, or the reverse order. Accordingly, after encountering an error, the video decoder can go to the end of the video packet, and start decoding in the reverse order, until the same error, or another error is encountered. In this manner, a greater portion of the video packet is recovered and reconstructed, in spite of encountering error(s). - To take advantage of the foregoing feature, the video decoder needs to be able to receive and decode the video bitstream in reverse order in real-time. During the decoding, the video elementary stream is stored in a memory known as the compressed data buffer in the forward order, along with a table that indicates the starting addresses of each video packet. The video decoder receives and decodes the video elementary stream by accessing the compressed data buffer. Upon encountering an error, the video decoder can receive the video packet at the ending address of the video packet and moving in the reverse order.
- Receiving the video packet in reverse order can be made possible by manipulating the memory access. For example, the video decoder can sequentially access data words in reverse order. After accessing each data word, the video decoder can use logic to reverse the bit order of the data word. However, the foregoing adds significant operations to the video decoder and makes accessing and decoding in the reverse order difficult to perform in real-time.
- Further limitations and disadvantages of conventional and traditional systems will become apparent to one of skill in the art through comparison of such systems with the invention as set forth in the remainder of the present application with reference to the drawings.
- Presented herein is a direct memory access engine for fetching words in reverse order. In one embodiment, there is presented a method for providing a plurality of sequential data words. The method includes receiving a command to provide the plurality of sequential data words, wherein the plurality of sequential data words comprises a first data word and a last data word, and one or more data words between the first data word and the last data word, fetching a sequential portion of the sequential data words, said sequential portion comprising a first intermediate word, the last word, and one or more data words between the intermediate word and the last word, storing the sequential portion, transmitting at least a portion of the last data word, and transmitting at least a portion of the intermediate data words after transmitting at least the portion of the last data word.
- In another embodiment, there is presented a system for providing a plurality of sequential data words. The system comprises a state logic machine, a memory controller, a local buffer, and a port. The state logic machine receives a command to provide the plurality of sequential of sequential data words, wherein the plurality of sequential data words comprises a first data word and a last data word, and one or more data words between the first data word and the last data word. The memory controller fetches a sequential portion of the sequential data words, said sequential portion comprising a first intermediate word, the last word, and one or more data words between the intermediate word and the last word. The local buffer stores the sequential portion. The port transmits at least a portion of the last data word and transmits at least a portion of the intermediate data words after transmitting at least the portion of the last data word.
- In another embodiment, there is presented a system for decoding a video packet. The system comprises a compressed data buffer, a video decoder, and a direct memory access engine. The compressed data buffer comprises a plurality of sequential data words. The plurality of sequential data words store a video packet. The video decoder decodes the video packet. The direct memory access engine provides the video packet to the video decoder and comprises a state logic machine, a memory controller, a local buffer, and a port. The state logic machine receives a command to provide the plurality of sequential data words and a control signal indicating reverse order from the video decoder, wherein the plurality of sequential data words comprises a first data word and a last data word, and one or more data words between the first data word and the last data word. The memory controller fetches a sequential portion of the sequential data words, said sequential portion comprising a first intermediate word, the last word, and one or more data words between the intermediate word and the last word. The local buffer stores the sequential portion. The port transmits at least a portion of the last data word and transmits at least a portion of the intermediate data words after transmitting at least the portion of the last data word.
- These and other advantages and novel features of the present invention, as well as details of illustrated embodiments thereof, will be more fully understood from he following description and drawings.
-
FIG. 1 is a block diagram describing the encoding of video data; -
FIG. 2 is a block diagram describing an exemplary decoder in accordance with an embodiment of the present invention; -
FIG. 3 is a block diagram of an exemplary compressed data buffer; -
FIG. 4 is a block diagram of an exemplary direct memory access engine in accordance with an embodiment of the present invention; and -
FIG. 5 is a flow diagram for accessing data words in reverse order in accordance with an embodiment of the present invention. - Referring now to
FIG. 1 , there is illustrated a block diagram describing MPEG formatting of video data 305. The video data 305 comprises a series offrames 310. Each frame comprises two dimensional grids of luminance Y, chroma red Cr, and chroma blue Cb pixels 315. The two-dimensional grids are divided into 8×8blocks 335, where fourblocks 335 of luminance pixels Y are associated with ablock 335 of chroma red Cr, and ablock 335 of chroma blue Cb pixels. The four blocks of luminance pixels Y, the block of chroma red Cr, and the chroma blue Cb form a data structure known as amacroblock 337. Themacroblock 337 also includes additional parameters, including motion vectors. - The
macroblocks 337 representing a frame are grouped intodifferent video packets 340. Thevideo packet 340 includes themacroblocks 337 in thevideo packet 340, as well as additional parameters describing the video packet. Each of thevideo packets 340 forming the frame form the data portion of apicture structure 345. Thepicture 345 includes thevideo packets 340 as well as additional parameters. The pictures are then grouped together as a group ofpictures 350. The group ofpictures 350 also includes additional parameters. Groups ofpictures 350 are then stored, forming what is known as a videoelementary stream 355. The videoelementary stream 355 is then packetized to form a packetized elementary sequence 360. Each packet is then associated with a transport header 365 a, forming what are known astransport packets 365 b. - The
transport packets 365 b can be multiplexed withother transport packets 365 b carrying other content, such as another videoelementary stream 355 or an audio elementary stream. The multiplexed transport packets from what is known as a transport stream. The transport stream is transmitted over a communication medium for decoding and presentation. - Referring now to
FIG. 2 , there is illustrated a block diagram of an exemplary decoder for decoding compressed video data, configured in accordance with an embodiment of the present invention. A processor, that may include aCPU 490, reads a stream oftransport packets 365 b (a transport stream) into atransport stream buffer 432 within anSDRAM 430. - The data is output from the transport
stream presentation buffer 432 and is then passed to adata transport processor 435. The data transport processor then demultiplexes the MPEG transport stream into its PES constituents and passes the audio transport stream to anaudio decoder 460 and the video transport stream to a video transport processor 440. - The video transport processor 440 converts the video transport stream into a video elementary stream and provides the video elementary stream to an
MPEG video decoder 445 that decodes the video. The videoelementary stream 355 is stored in a compressed data buffer (CDB) 447. TheMPEG video decoder 445 accesses the compressed data buffer (CDB) to receive the videoelementary stream 355. The videoelementary stream 355 is decoded by theMPEG video decoder 445 resulting in the reconstructed video data 305. - The audio data is sent to the output blocks and the video data 305 is sent to a
display engine 450. Thedisplay engine 450 is responsible for and operable to scale the video picture, render the graphics, and construct the complete display among other functions. Once the display is ready to be presented, it is passed to avideo encoder 455 where it is converted to analog video using an internal digital to analog converter (DAC). The digital audio is converted to analog in the audio digital to analog converter (DAC) 465. - Referring now to
FIG. 3 , there is illustrated a block diagram describing an exemplarycompressed data buffer 447. Thecompressed data buffer 447 includes any number of data words 505(1) . . . 505(m). The data words can have any width. In an exemplary case, for example, the data words can comprise 256 bit jumbo words (words). - The
compressed data buffer 447 stores the videoelementary stream 355. The videoelementary stream 355 comprises any number ofvideo packets 340. Thevideo packets 340 further comprise a video packet header and any number ofmacroblocks 337. Thecompressed data buffer 447 also stores a start code table 507. The start code table 507 associates eachvideo packet 340 with its starting address in thecompressed data buffer 447. Alternatively thevideo packet 340 can be first scanned forward, without evacuating data from memory, then number of bytes/bits can be counted, and then returned using the DMA engine. - The
MPEG video decoder 445 receives thevideo packets 340 from the videoelementary stream 355 and decodes thevideo packets 340. Thevideo packet 340 is received and decoded by theMPEG video decoder 445 starting from the word 505(x) storing the beginning of thevideo packet 340, and proceeding to the word 505(n) storing the end of thevideo packet 340. - A direct memory access (DMA)
engine 510 facilitates receipt of thevideo packets 340 by theMPEG video decoder 445. Alternatively, a processor can facilitate receipt of thevideo packets 340. Accordingly,DMA engine 510 shall be interpreted to also include a processor that is operable to fetch video packets from memory. TheMPEG video decoder 445 receives avideo packet 340 by looking up the starting address and the ending address of avideo packet 340 in the start code table 507. TheMPEG video decoder 445 can then command theDMA engine 510 to fetch the words 505(x) . . . 505(n) that store thevideo packet 340. Responsive thereto, theDMA engine 510 fetches and provides the words 505(x) . . . 505(n) that store thevideo packet 340. - The
DMA engine 510 provides the words 505(x) . . . 505(n) to anextractor 515 within theMPEG video decoder 445 in a serial manner, beginning with word 505(x) and proceeding to the last word 505(n). TheMPEG decoder 445 decodes thevideo packet 340, in a serial manner, beginning decoding with the first word 505(x) and proceeding to the last word 505(n). Theextractor 515 and theDMA engine 510 operate in conjunction with each other, such that thewords 505 are provided to theMPEG video decoder 445 at a dynamic rate that is in substantial relationship to the rate that theMPEG video decoder 445 is decoding thewords 505. - In MPEG-4
PART 2, thevideo packet 340 is defined such that thevideo packet 340 can be decoded in a forward order, or the reverse order. Accordingly, if theMPEG video decoder 445 encounters an error, the video decoder can go to the end of thevideo packet 340, and start decoding in the reverse order. For example, if theMPEG video decoder 445 decodes thevideo packet 340 beginning with the first word 505(x), and encounters an error in word 505(x+5), theMPEG video decoder 445 can start decoding thevideo packet 340 from word 505(n) and decode in the reverse order, e.g., word 505(n−1), 505(n−2) . . . , etc. - Upon detecting an error, the
MPEG video decoder 445 transmits a command to theDMA engine 510 to fetch the words storing thevideo packet 340, e.g., words 510(x) . . . 510(n), along with a reverse order signal. Responsive thereto, theDMA engine 510 provides the words 510(x) . . . 510(n) in the reverse order to theMPEG video decoder 445. - Referring now to
FIG. 4 , there is illustrated a block diagram describing anexemplary DMA engine 510 in accordance with an embodiment of the present invention. TheDMA engine 510 comprises astate logic machine 605, alocal buffer 610, and amemory controller 620. Thelocal buffer 610 can comprise any amount of memory with any width of data words. For example, in an exemplary case, the memory can comprise 128 32-bit words 611(0) . . . 611(127). - The
state logic machine 605 receives a command to fetch data words in an address range, e.g., 505(x)-510(n) from theMPEG video decoder 445. The command can be accompanied by a control signal indicating that the data words in the address range are to be provided to theMPEG video decoder 445 in the reverse order, e.g., 505(n), 505(n−1) . . . 505(x). - Responsive to receiving a command to fetch the data words 510(x)-510(n) in the reverse order, the
state logic machine 605 commands thememory controller 620 to retrieve a batch comprising a the last predetermined number ofdata words 505 in the provided address range, and store the predetermined number of data words in thelocal buffer 610. The predetermined number of data words in the batch is less than or equal to the capacity of thelocal buffer 610. For example, in an exemplary embodiment, where thelocal buffer 610 comprises 128 32-bit words, the batch ofdata words 505 can include the last 16 words in the provided address range, e.g., data words 505(n−15) . . . 505(n). - After the batch of data words 505(n−15) . . . 505(n) is stored in the
local buffer 610, thestate logic machine 605 causes the contents of thelocal buffer 610 to be provided to theMPEG video decoder 445 beginning with word 611(127), and proceeding sequentially to word 611(0). After the contents of thelocal buffer 610, e.g., words 611(127) . . . 611(0), are provided to theMPEG video decoder 445, thestate logic machine 605 commands thememory controller 620 to fetch another batch comprising the predetermined number of words, e.g., data words 505(n−31) . . . 505(n−16), that precede the most recently fetched data words, e.g., data words 505(n−15) . . . 505(n). The data words 505(n−15) . . . 505(n) are stored in thelocal buffer 610 and provided to theMPEG decoder 445. - The foregoing is repeated until the next predetermined number of data words comprises the first data word in the address range, e.g., data word 505(x). Where a batch comprises the first data word in the address range, e.g., data word 505(x), the
state logic machine 605 truncates that portion of the predetermined number of data words that precedes the first data word 505(x), and commands thememory controller 620 to fetch the truncated batch comprising the first data word 505(x) and all data words 505(x+1), 505(x+2) . . . , following the first data word 505(x) that have not been previously transmitted to theMPEG video decoder 445. - The foregoing provides the data words 505(x) . . . 505(n) as a set of 32 bit words starting from the last portion of 505(n) and proceeding sequentially to the first portion of 505(x). The bits forming the 32-bit words can be reversed with respect to one another, in any number of ways. For example, the
MPEG video decoder 445 can include logic that reverses the 32 bits of each word. Alternatively, theDMA engine 510 can include additional circuitry that causes the 32 bits of eachword 611 to be provided to theMPEG video decoder 445 in the reverse order. - Referring now to
FIG. 5 , there is illustrated a flow diagram for providing a video packet in a reverse order. At 705, thestate logic machine 605 receives a command to fetch data words in an address range, e.g., 505(x)-510(n) from theMPEG video decoder 445, accompanied by a control signal indicating that the data words in the address range are to be provided to theMPEG video decoder 445 in the reverse order, e.g., 505(n), 505(n−1) . . . 505(x). - Responsive to receiving the command, the
state logic machine 605 determines (706) if a predetermined number of words comprises the first word, 505(x). Where the predetermined number of words comprises the first data word in the address range, e.g., data word 505(x), thestate logic machine 605 truncates that portion of the predetermined number of data words that precedes the first data word 505(x) and commands (708) thememory controller 620 to fetch (709) the truncated batch comprising the first data word 505(x) and all data words 505(x+1), 505(x+2) . . . following the first data word 505(x) that have not been previously transmitted to theMPEG video decoder 445. Where during 706, thestate logic machine 605 determines that the predetermined number of words does not comprise the first word, 505(x), thestate logic machine 605 commands (710) thememory controller 620 to fetch (715) a batch comprising the last predetermined number ofdata words 505 in the provided address range. - The
fetched data words 505 are stored (720) in thelocal buffer 610. After the data words are stored in thelocal buffer 610, thestate logic machine 605 causes the contents of thelocal buffer 610 to be provided (725) to theMPEG video decoder 445 beginning with word 611(127), and proceeding sequentially to word 611(0). After the contents of thelocal buffer 610, e.g., words 611(127) . . . 611(0), are provided to theMPEG video decoder 445, a determination (730) is made whether the first data word, data word 505(x) has been provided to theMPEG video decoder 445. If the first data word has not been provided to the MPEG video decoder at 730, 705-730 are repeated. If the first data word has been provided to the MPEG video decoder at 730, the process is completed. - One embodiment of the present invention may be implemented as a board level product, as a single chip, application specific integrated circuit (ASIC), or with varying levels integrated on a single chip with other portions of the system as separate components. The degree of integration of the system will primarily be determined by speed and cost considerations. Because of the sophisticated nature of modern processors, it is possible to utilize a commercially available processor, which may be implemented external to an ASIC implementation of the present system. Alternatively, if the processor is available as an ASIC core or logic block, then the commercially available processor can be implemented as part of an ASIC device with various functions implemented as firmware.
- While the invention has been described with reference to certain embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the scope of the invention. In addition, many modifications may be made to adapt particular situation or material to the teachings of the invention without departing from its scope. Therefore, it is intended that the invention not be limited to the particular embodiment(s) disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.
Claims (17)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/736,125 US20050036614A1 (en) | 2003-08-13 | 2003-12-15 | DMA engine for fetching words in reverse bit order |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US49466603P | 2003-08-13 | 2003-08-13 | |
US49474603P | 2003-08-13 | 2003-08-13 | |
US10/736,125 US20050036614A1 (en) | 2003-08-13 | 2003-12-15 | DMA engine for fetching words in reverse bit order |
Publications (1)
Publication Number | Publication Date |
---|---|
US20050036614A1 true US20050036614A1 (en) | 2005-02-17 |
Family
ID=34138905
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/735,980 Expired - Fee Related US7284072B2 (en) | 2003-08-13 | 2003-12-15 | DMA engine for fetching words in reverse order |
US10/736,125 Abandoned US20050036614A1 (en) | 2003-08-13 | 2003-12-15 | DMA engine for fetching words in reverse bit order |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/735,980 Expired - Fee Related US7284072B2 (en) | 2003-08-13 | 2003-12-15 | DMA engine for fetching words in reverse order |
Country Status (1)
Country | Link |
---|---|
US (2) | US7284072B2 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070174239A1 (en) * | 2006-01-18 | 2007-07-26 | De Marcken Carl G | Incremental searching in multi-passenger multi-route travel planning |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101034493B1 (en) * | 2004-01-09 | 2011-05-17 | 삼성전자주식회사 | Image transforming apparatus, dma apparatus for image transforming, and camera interface supporting image transforming |
DE102005057568B4 (en) * | 2005-12-02 | 2021-06-17 | Robert Bosch Gmbh | Transmitting device and receiving device |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6271362B1 (en) * | 1994-04-01 | 2001-08-07 | Chugai Seiyaku Kabushiki Kaisha | Gene encoding IGG FC region-binding protein |
US6289053B1 (en) * | 1997-07-31 | 2001-09-11 | Lsi Logic Corporation | Architecture for decoding MPEG compliant video bitstreams meeting 2-frame and letterboxing requirements |
US20020181594A1 (en) * | 2001-03-05 | 2002-12-05 | Ioannis Katsavounidis | Systems and methods for decoding of partially corrupted reversible variable length code (RVLC) intra-coded macroblocks and partial block decoding of corrupted macroblocks in a video decoder |
US6504871B1 (en) * | 1997-07-31 | 2003-01-07 | Lsi Logic Corporation | IDCT processor for use in decoding MPEG compliant video bitstreams meeting 2-frame and letterboxing requirements |
US20030118116A1 (en) * | 2001-09-12 | 2003-06-26 | Gaurav Aggarwal | Command packet system and method supporting improved trick mode performance in video decoding systems |
US20030194008A1 (en) * | 2002-04-11 | 2003-10-16 | Tinku Acharya | Computationally fast and efficient DV to DVD transcoding |
US6842219B2 (en) * | 2001-12-24 | 2005-01-11 | C & S Technology | Moving picture decoding processor for multimedia signal processing |
US7085320B2 (en) * | 2001-07-31 | 2006-08-01 | Wis Technologies, Inc. | Multiple format video compression |
US7333515B1 (en) * | 2002-08-06 | 2008-02-19 | Cisco Technology, Inc. | Methods and apparatus to improve statistical remultiplexer performance by use of predictive techniques |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4608633A (en) * | 1983-04-01 | 1986-08-26 | Honeywell Information Systems Inc. | Method for decreasing execution time of numeric instructions |
US4704641A (en) * | 1985-04-12 | 1987-11-03 | American Telephone And Telegraph Company, At&T Bell Laboratories | Recovery of stored data from mutilated tape data blocks |
CN100502456C (en) * | 1996-07-31 | 2009-06-17 | 富士胶片株式会社 | Printer system and its motion control method |
JP2001113759A (en) * | 1999-10-15 | 2001-04-24 | Ricoh Co Ltd | Image data output device |
US7555042B2 (en) * | 2003-01-08 | 2009-06-30 | Samsung Electronics Co., Ltd. | Method and apparatus for reverse play |
-
2003
- 2003-12-15 US US10/735,980 patent/US7284072B2/en not_active Expired - Fee Related
- 2003-12-15 US US10/736,125 patent/US20050036614A1/en not_active Abandoned
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6271362B1 (en) * | 1994-04-01 | 2001-08-07 | Chugai Seiyaku Kabushiki Kaisha | Gene encoding IGG FC region-binding protein |
US6289053B1 (en) * | 1997-07-31 | 2001-09-11 | Lsi Logic Corporation | Architecture for decoding MPEG compliant video bitstreams meeting 2-frame and letterboxing requirements |
US6504871B1 (en) * | 1997-07-31 | 2003-01-07 | Lsi Logic Corporation | IDCT processor for use in decoding MPEG compliant video bitstreams meeting 2-frame and letterboxing requirements |
US20020181594A1 (en) * | 2001-03-05 | 2002-12-05 | Ioannis Katsavounidis | Systems and methods for decoding of partially corrupted reversible variable length code (RVLC) intra-coded macroblocks and partial block decoding of corrupted macroblocks in a video decoder |
US6990151B2 (en) * | 2001-03-05 | 2006-01-24 | Intervideo, Inc. | Systems and methods for enhanced error concealment in a video decoder |
US7042948B2 (en) * | 2001-03-05 | 2006-05-09 | Intervideo, Inc. | Systems and methods for management of data in a ring buffer for error resilient decoding of a video bitstream |
US7085320B2 (en) * | 2001-07-31 | 2006-08-01 | Wis Technologies, Inc. | Multiple format video compression |
US20030118116A1 (en) * | 2001-09-12 | 2003-06-26 | Gaurav Aggarwal | Command packet system and method supporting improved trick mode performance in video decoding systems |
US6842219B2 (en) * | 2001-12-24 | 2005-01-11 | C & S Technology | Moving picture decoding processor for multimedia signal processing |
US20030194008A1 (en) * | 2002-04-11 | 2003-10-16 | Tinku Acharya | Computationally fast and efficient DV to DVD transcoding |
US7333515B1 (en) * | 2002-08-06 | 2008-02-19 | Cisco Technology, Inc. | Methods and apparatus to improve statistical remultiplexer performance by use of predictive techniques |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070174239A1 (en) * | 2006-01-18 | 2007-07-26 | De Marcken Carl G | Incremental searching in multi-passenger multi-route travel planning |
Also Published As
Publication number | Publication date |
---|---|
US20050038938A1 (en) | 2005-02-17 |
US7284072B2 (en) | 2007-10-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7342967B2 (en) | System and method for enhancing performance of personal video recording (PVR) functions on hits digital video streams | |
CN1121673A (en) | Method and apparatus for efficient addressing of dram in a video decompression processor | |
US20070098072A1 (en) | Command packet system and method supporting improved trick mode performance in video decoding systems | |
US9185407B2 (en) | Displaying audio data and video data | |
US6205181B1 (en) | Interleaved strip data storage system for video processing | |
US6995696B2 (en) | System, method, and apparatus for variable length decoder | |
US7936814B2 (en) | Cascaded output for an encoder system using multiple encoders | |
EP1562383A2 (en) | Multistandard video decoder | |
US7349428B2 (en) | Data alignment of the packetized elementary streams in the coded data buffer for dual decode | |
US7284072B2 (en) | DMA engine for fetching words in reverse order | |
US20050036555A1 (en) | Automatic direct memory access engine | |
US20100007786A1 (en) | System, method, and apparatus for providing massively scaled down video using iconification | |
US20040091159A1 (en) | Image compression device and method for performing a frame skipping process | |
US7702021B2 (en) | Decoding of digital video standard material during variable length decoding | |
US8948263B2 (en) | Read/write separation in video request manager | |
US7889206B2 (en) | Direct memory accessing for fetching macroblocks | |
US7526024B2 (en) | Storing macroblocks for concatenated frames | |
US7386651B2 (en) | System, method, and apparatus for efficiently storing macroblocks | |
US8023564B2 (en) | System and method for providing data starting from start codes aligned with byte boundaries in multiple byte words | |
US7848432B2 (en) | System and method for efficiently storing macroblocks in SD-RAM | |
US20060062388A1 (en) | System and method for command for fast I-picture rewind | |
US7864865B2 (en) | Line address computer for calculating the line addresses of decoded video data | |
US20050169376A1 (en) | Motion vector address computer error detection | |
US8929458B2 (en) | Compressed structure for slice groups in start code table | |
US7610325B2 (en) | System(s), method(s), and apparatus for detecting end of slice groups in a bitstream |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: BROADCOM CORPORATION, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:PAI, RAMADAS LAKSHMIKANTH;KISHORE, CHHAVI;KODIHALLI, HIMAKIRAN;AND OTHERS;REEL/FRAME:014494/0130;SIGNING DATES FROM 20031203 TO 20031209 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- AFTER EXAMINER'S ANSWER OR BOARD OF APPEALS DECISION |
|
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: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD., SINGAPORE Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001 Effective date: 20170120 Owner name: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001 Effective date: 20170120 |
|
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 |