WO2012174807A1 - Method and system for compressing and decompressing configuration data stream of programmable logic device - Google Patents

Abstract

Disclosed are a method and a system for compressing and decompressing a configuration data stream of a programmable logic device. The compression method comprises: acquiring a data frame to be compressed in a configuration data stream, and compressing a data unit in the data frame to be compressed. Through the foregoing technical solution, the present invention provides a more efficient method and system for compressing and decompressing a configuration data stream of a programmable logic device.

Description

 Programmable logic device configuration data stream compression and decompression processing method and system Technical field

The present invention relates to the field of data processing, and in particular, to a method and system for compressing and decompressing a configuration data stream of a programmable logic device.

Background technique

Traditional programmable logic device , PLD), such as Field Program Gate FPGA (Field Program Gate) Array) is configured by the corresponding configuration data stream, the configuration data stream is downloaded to the programmable logic device, and then stored in the configuration storage unit, and finally the programmable logic device is configured according to the content of the configuration storage unit, thereby A programmable logic device is constructed as a device of a particular function preset by a user.

Usually, a design file contains millions or more of configuration data bits, which requires a large external memory to store the configuration data stream, and the download operation of the configuration data stream will also cost a lot. time. Therefore, in order to save storage space and save download time, compressing the configuration data stream is a very effective technical measure.

Compression is generally divided into lossy compression and lossless compression. One of the famous most commonly used lossless compression algorithms is a dictionary-based LZ (according to the original author Lempel-Ziv named) series. The traditional FPGA developers basically use LZ-based compression algorithms. The LZ compression algorithm is mainly for the compression of strings or data strings (hereinafter referred to as strings), such as saving the latest compressed data in a buffer buffer. Stream, when there is data input, the LZ algorithm looks for the matching of the input data with the data of the buffer, finds the longest matching data, and then outputs a piece of code instead of the matched data. The encoding consists of three parts, including matching. The position, length, and character or data of the last digit of the data string. A part of the conventional FPGA is configured in units of frame frames, and is compressed in units of frames, that is, compression outside the frame. 1 is a schematic diagram of an FPGA of the prior art, including an FPGA 11, a programmable logic unit 12, and a frame 13. The FPGA 11 represents a structure of an FPGA, and each dotted line bar represents a frame (the number of dashed lines does not mean any meaning) , here is just a schematic diagram), the FPGA completes the configuration by frame by frame. A frame of data will contain a number of bits or fields. Depending on the type and capacity of the FPGA, the number of bits or fields it contains will be different. Suppose that the number of fields in a frame of data is m, and the corresponding number of bits is 32*m; use uppercase letter W The lower footer is used to indicate the internal field of the frame and its order. As shown in FIG. 2, FIG. 2 is a schematic diagram of a frame structure in an FPGA in the prior art. The first word is represented by W0, and the second word is represented by W1. In this way, the last word is represented by Wm-1. Usually, when the number of fields included in one frame of data is m, the length of this frame is said to be m.

The traditional method of external compression of the frame considers the structural characteristics and programming features of the FPGA to some extent, but does not take into account the data characteristics inside the frame. Therefore, the traditional method of compressing the frame based on the frame unit has certain limitations. , low efficiency.

technical problem

The present invention provides a more efficient method and system for compressing and decompressing a configuration data stream of a programmable logic device.

Technical solution

In order to solve the above technical problems, the present invention adopts the following technical solutions:

A method for compressing a configuration data stream of a programmable logic device, comprising:

Obtaining a data frame to be compressed in the configuration data stream;

Compressing the data unit inside the data frame to be compressed.

The data unit includes one or more bit values, or the data unit includes one or more field values.

Performing compression processing on the data unit inside the data frame to be compressed to obtain a reference value and a command unit.

The process of obtaining the reference value and the command unit after performing compression processing on the data unit inside the data frame to be compressed is specifically:

Determining whether the number of consecutive and identical data units is greater than or equal to the preset value by comparing the number of consecutive and identical data units in the frame obtained by the statistics with a preset value;

If yes, the command unit obtained by processing the consecutive and identical plurality of data units includes command type information for indicating a compression type, and an address for indicating that the consecutive and identical plurality of data units are located inside the frame Information and number information for indicating the plurality of consecutive and identical data units;

If not, the command unit obtained by processing the consecutive and identical plurality of data units or 1 data unit includes command type information for indicating an uncompressed type, and is used to represent the continuous and identical plurality of data. The unit or one data unit is located in the address information inside the frame and is used to indicate the number information of the consecutive and identical plurality of data units or one data unit.

The preset value is an integer greater than or equal to 3.

Also includes:

Compressing the frame units in the configuration data stream.

A method for decompressing a configuration data stream of a programmable logic device, comprising:

Obtaining a data frame to be decompressed in the configuration data stream;

Decompressing the data unit inside the data frame to be decompressed.

The process of decompressing the data unit inside the data frame to be decompressed is specifically:

Obtaining a command unit and a reference value inside the data frame to be decompressed;

And decompressing the data unit inside the data frame to be decompressed according to the command unit and the reference value.

The process of decompressing the data unit inside the data frame to be decompressed according to the command unit and the reference value is specifically:

Obtaining command type information in the command unit;

If the command type information in the command unit is a compression type, according to the address information in the command unit for indicating that the consecutive and identical plurality of data units are located inside the frame and for indicating the continuous and the same Number information of the plurality of data units, the reference value is copied to a corresponding address in the frame;

If the command type information in the command unit is an uncompressed type, according to the address information in the command unit for indicating that the consecutive and identical plurality of data units or one data unit are located inside the frame and for indicating The number information of the consecutive and identical plurality of data units or 1 data unit copies the reference value to a corresponding address within the frame.

Also includes:

Decompressing the frame units in the configuration data stream.

A compression processing system for configuring a data stream of a programmable logic device, comprising: a data acquisition module to be compressed and a first compression processing module, wherein

The data processing module to be compressed is used to acquire a data frame to be compressed in the configuration data stream;

The first compression processing module is configured to perform compression processing on the data unit inside the data frame to be compressed.

The first compression processing module is configured to perform compression processing on the data unit inside the data frame to be compressed to obtain a reference value and a command unit.

The first compression processing module is specifically configured to: compare the number of consecutive and identical data units in the statistically derived frame with a preset value, and determine whether the number of the consecutive and identical data units is greater than or equal to the a preset value; if yes, the command unit obtained by processing the consecutive and identical plurality of data units includes command type information for indicating a compression type, and is used to represent the continuous and identical plurality of data units Address information located inside the frame and number information for indicating the continuous and identical plurality of data units; if not, the command unit obtained by processing the consecutive and identical plurality of data units or 1 data unit includes Command type information for indicating an uncompressed type, address information for indicating that the continuous and identical plurality of data units or one data unit is located inside the frame, and for indicating the continuous and identical plurality of data units or The number information of 1 data unit.

A second compression processing module is further included, the second compression processing module is configured to perform compression processing on the frame unit in the configuration data stream.

A decompression processing system for configuring a data stream of a programmable logic device, comprising: a data acquisition module to be decompressed and a first decompression module, wherein:

The data to be decompressed processing module is configured to acquire a data frame to be decompressed in the configuration data stream;

The first decompression module is configured to perform decompression processing on the data unit inside the data frame to be decompressed.

The first decompression module is specifically configured to acquire a command unit and a reference value inside the data frame to be decompressed; and to use a data unit in the data frame to be decompressed according to the command unit and the reference value. Perform decompression processing.

The first decompression module is specifically configured to acquire command type information in the command unit; if the command type information in the command unit is a compression type, according to the command unit, the continuous and identical Address data of the plurality of data units located inside the frame and number information for indicating the continuous and identical plurality of data units, copying the reference value to a corresponding address within the frame; if in the command unit The command type information is an uncompressed type, and according to the address information in the command unit for indicating that the consecutive and identical multiple data units or one data unit are located inside the frame, and for indicating the continuous and the same The number of data units or the number of data units is copied to the corresponding address within the frame.

A second decompression module is further included, and the second decompression module is configured to perform decompression processing on the frame unit in the configuration data stream.

The present invention provides a compression and decompression processing method and system for a configuration data stream of a programmable logic device, wherein the compression method includes performing compression processing on data units inside a data frame to be compressed, and compressing data units inside the frame. In order to improve the compression efficiency, the present invention can also select an optimal compression ratio for the characteristics of the data frame or the internal data unit of the frame by setting a preset value, and therefore, has a strong flexibility.

DRAWINGS

1 is a schematic diagram of an FPGA of the prior art;

2 is a schematic diagram of a frame structure in an FPGA in the prior art;

3 is a flowchart of a configuration data stream compression processing method of a programmable logic device according to an embodiment of the present invention;

4 is a flowchart of a method for compressing a frame in a configuration data stream compression processing method of a programmable logic device according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a command unit obtained by processing a plurality of consecutive and identical data units whose number is greater than or equal to the preset value according to an embodiment of the present invention;

FIG. 6 is a schematic structural diagram of a command unit obtained by processing a plurality of consecutive or identical data units having a number smaller than the preset value according to an embodiment of the present invention;

7 is a schematic diagram of internal compressed data of a frame in a configuration data stream compression processing method of a programmable logic device according to an embodiment of the present invention;

8 is a flowchart of a configuration data stream decompression processing method of a programmable logic device according to an embodiment of the present invention;

FIG. 9 is a schematic diagram of decompressing data in a frame in a configuration data stream decompression processing method of a programmable logic device according to an embodiment of the present invention; FIG.

10 is a block diagram of a configuration data stream compression processing system of a programmable logic device according to an embodiment of the present invention;

11 is a block diagram of a configuration data stream compression processing system of a programmable logic device according to another embodiment of the present invention;

12 is a block diagram of a configuration data stream decompression processing system of a programmable logic device according to an embodiment of the present invention;

FIG. 13 is a block diagram of a configuration data stream decompression processing system of a programmable logic device according to another embodiment of the present invention.

Embodiments of the invention

The present invention will be further described in detail below with reference to the accompanying drawings.

A method for compressing a configuration data stream of a programmable logic device, comprising:

Obtaining a data frame to be compressed in the configuration data stream;

Compressing the data unit inside the data frame to be compressed.

FIG. 3 is a flowchart of a configuration data stream compression processing method of a programmable logic device according to an embodiment of the present invention. Referring to FIG. 3, this embodiment uses a compression method of an FPGA configuration data stream as an example:

S31. Generate a data stream by using a design file.

S32, it is determined whether the frame outside frame unit based compression processing can be performed, and if so, step S33 is performed, otherwise step S34 is performed;

S33, performing frame-based frame unit-based compression processing on the data stream, after the compression of the frame external frame unit is completed, executing S34;

S34: Perform a compression process on the data stream based on the internal data unit of the frame;

S35. Store the compression result in the compressed data stream memory.

In this embodiment, it is preferable to perform external compression of the frame and then perform internal compression of the frame. The present invention is also applicable to a method of performing internal compression of the frame and then external compression of the frame.

The specific process of performing intra-frame compression on the data stream in step S34 includes:

Comparing the number of consecutive and identical data units in the frame obtained by the statistics with a preset value, determining whether the number of consecutive and identical data units is greater than or equal to the preset value; if yes, the determination result is satisfied The compression condition performs compression processing on a plurality of consecutive and identical data units that satisfy the compression condition.

Depending on the type of data in the data stream, the data unit can be a bit value or a field value or the like, and the compression and decompression methods are similar. The present invention is not limited to compression with a bit value or a field value as a data unit, and may also be compressed with a plurality of bit values or a plurality of field values as one data unit, and the preset value is an integer greater than or equal to 3.

4 is a flowchart of a method for compressing a frame in a configuration data stream compression processing method of a programmable logic device according to an embodiment of the present invention. Referring to FIG. 4, in this embodiment, a data unit is a field value in a frame:

S341, loading the data stream into the buffer A. Preferably, the size of the buffer A is one bit longer than the configuration data length of one frame, and is used as a flag bit. The preferred flag bit is the first bit of the buffer A;

S342: Determine whether the data stream is subjected to frame-based external unit frame-based compression processing. In this embodiment, the frame compression command or other flag attached to the frame after the frame unit-based compression processing is used to determine, if yes, execute S343. Otherwise, execute S344;

S343, in the step S341, the flag bit reserved by the buffer A, that is, the first bit of the buffer A is written with a "1" flag, and S345 is executed;

S344, in step S341, the flag bit reserved by the buffer A, that is, the first bit of the buffer A is written with a "0" flag, and S345 is executed;

S345, copying the data stream in the buffer A to the buffer B of the same size, after the copying, the buffer A continues to receive the new data stream, and the buffer B is addressed except the first bit. The basic addressing unit is a field, that is, 32 bits is an addressing unit. For example, the buffer B size is 106×32 bits, and there are 106 fields, and each field is addressed, and the address is addressed. Is 1-106, expressed in binary is 00000001-01101010, expressed in hexadecimal is 00-6A;

S346, wherein one field value is one data unit, and the processing sequence is performed according to the address of the buffer B from low to high, and each field is processed in turn, and the number of consecutive and identical field values is counted, and a preset is made. Value comparison, determining whether the number of consecutive and identical field values is greater than or equal to the preset value, and if so, executing S347, otherwise, executing S348;

S347. Process the consecutive and identical multiple field values of the number greater than or equal to the preset value to obtain the following form: command unit + reference value. Wherein the command unit includes command type information for indicating a compression type, address information for indicating that the continuous and identical plurality of field values are located inside the frame, and number information for indicating the continuous and identical plurality of field values , execute S349;

FIG. 5 is a schematic structural diagram of a command unit obtained by processing a plurality of consecutive and identical data units whose number is greater than or equal to the preset value according to an embodiment of the present invention. Referring to FIG. 5, the command unit in this embodiment is a command field. The width of the command field may be 32 bits, wherein the first part of the content is used to indicate the command type information of the compression type, and may be identified by one binary bit “0”, indicating that the command type is a compression type, and the second part is used. The address information indicating that the consecutive and identical multiple field values are located inside the frame, the address information refers to the order or number of each field in a frame of data, or may be the compressed consecutive and identical multiple field values. Address information of the first field value in the field, since these fields are continuous, after knowing the address information of the first field, the address information of other fields can be estimated; the third part is used to indicate the continuous and identical The number information of multiple field values.

S348. Process the consecutive and identical multiple field values or one field value that is less than the preset value to obtain the following form: command unit + reference value, where the command unit includes command type information for indicating an uncompressed type, S349 is performed to indicate that the consecutive and identical plurality of data units or one data unit is located inside the frame and the number information indicating the continuous and identical plurality of data units or one data unit, S349;

FIG. 6 is a schematic structural diagram of a command unit obtained by processing a plurality of consecutive or identical data units having a number smaller than the preset value according to an embodiment of the present invention. Referring to FIG. 6, the command unit in this embodiment is an instruction. Field, the width of the command word may be 32 bits, and its structure, format and meaning are similar to the above-mentioned command unit obtained by compressing a plurality of consecutive and identical data units whose number is greater than or equal to a preset value, wherein the first part of the content Used to indicate the uncompressed type, the highest bit, that is, the 31st bit is identified by 1 binary bit "1", indicating that the command type is uncompressed, and the second part is used to indicate that the number is less than the preset value and The same multiple field value or one field value is located in the address information inside the frame, the address information refers to the order or number of each field in a frame of data, or the address information of the first field in these fields, due to these The fields are contiguous, so after knowing the address information of the first field, the address information of other fields can be extrapolated; the third part is used for the table. The number information of the continuous and identical plurality of field values or 1 field value is shown.

FIG. 7 is a schematic diagram of internal compressed data of a frame in a configuration data stream compression processing method of a programmable logic device according to an embodiment of the present invention. Referring to FIG. 7, assuming a frame data length m=106, according to the definition of the command field, its address is encoded from 0000 to 0069, pointing to the corresponding field, and the content of each field is also filled with data, where the address and field values are expressed in hexadecimal, the omitted part of the field value represents the same number of field values, The omitted part of the address indicates an increment of several address values, the 0th field value is: 0000_0000; the first field value to the 15th field value are the same, being 1111_1111; the 16th field value to the 20th field value are respectively :2222_2222,3333_3333, 4444_4444, 4444_4444, 5555_5555; the 21st field value is the same as the 102nd field value, which is 6666_6666; the 103th field value to 105 field values are: 7777_7777, 6666_6666, 6666_6666, assuming the default value is 10, assuming this An embodiment obtains a command unit obtained by compressing a plurality of consecutive and identical plurality of field values that are greater than or equal to a preset value, and is a CMD1, The command unit obtained by processing the consecutive and identical multiple or one field values whose number is less than the preset value is CMD2, and the process is: since the number of the 0th field value is 1, it is processed as CMD2, 0000_0000, that is, the command unit and the reference value, the reference value is the field value itself, the first field value to the 15th field value are consecutive and identical, and the number of consecutive and identical field values is 15, which is greater than the preset value, so Its compression is: CMD1, 1111_1111, that is, the command unit and the reference value, the reference value is the value of the field itself. Similarly, the remaining field values are respectively processed as: CMD2, 2222_2222, 3333_3333, 4444_4444, 4444_4444, 5555_5555; CMD1, 6666_6666; CMD2, 7777_7777, 6666_6666, 6666_6666.

By specifying CMD1 and CMD2, you can get: 8000_0001, 0000_0000, 0002_000E, 1111_1111, 8010_0005, 2222_2222, 3333_3333, 4444_4444, 4444_4444, 5555_5555, 0015_0056, 6666_6666, 8067_0003, 7777_7777, 6666_6666, 6666_6666.

In this embodiment, 106 fields of one frame can be compressed into 16 fields, which improves the efficiency of data compression.

a plurality of consecutive and identical data units having a number less than a preset value, that is, when the number of consecutive and identical data units is an integer greater than or equal to 2, and the number is less than a preset value, in which case the command unit is included for Command type information indicating an uncompressed type, address information indicating that the continuous and identical plurality of data units are located inside the frame, and number information indicating the continuous and identical plurality of data units, wherein Address information indicating that the consecutive and identical plurality of data units are located inside the frame includes address information in which each data unit is located inside the frame, or address information in which the first data unit of the same data unit is located inside the frame, for The number information indicating the continuous and identical plurality of data units is the number of the same data units, such as the 18th field value and the 19th field value in FIG. 7: 4444_4444, 4444_4444; the number is less than the preset value The data unit, that is, the number of consecutive and identical data units is 1, and the number is less than a preset value, in which case the command unit is included for Command information indicating an uncompressed type, address information indicating that the data unit is located inside the frame, and number information indicating the data unit, wherein the number information of the data unit is 1, as shown in FIG. Field 2222_2222, field 17 3333_3333.

S349. The processing results of steps S347 and S348 are placed in the buffer C until the data of the buffer B is completely processed, and the content of the buffer C is output.

In this embodiment, the command type information in the command unit, that is, the command type information for indicating the compression type and the command type information for indicating the uncompressed type, can be facilitated by identifying the command type information in the command unit during decompression. The region is divided into internal compression of the frame and data that has not been internally compressed by the frame. First, the existing technology is used in decompression, and only two additional decoding lines can be added to distinguish between intra-frame compression and intra-frame compression. The data.

The embodiment further includes a data decompression processing method. FIG. 8 is a flowchart of a configuration data stream decompression processing method of the programmable logic device according to an embodiment of the present invention.

S81. Acquire a data stream to be decompressed and put it into the buffer D.

S82, parsing the data frame, determining whether the compression processing of the internal data unit of the frame is passed, and if so, executing S83, otherwise executing step S84;

S83, performing decompression processing on the data unit inside the frame, and ending the internal decompression processing of the frame, executing S84;

S84. Put the data frame that has undergone intra-frame decompression processing and the intra-frame decompression processing into the buffer D, and detect whether the value of the first bit in the buffer D is “1”, and if yes, execute S85, otherwise Execute S86;

S85, performing frame-based frame unit-based decompression processing, executing S87;

S86, without frame external frame unit based decompression processing, executing S87;

S87. Download the data stream obtained in steps S85 and S86 to the FPGA.

In this embodiment, it is preferable to perform intra-frame decompression and then external decompression of the frame. The present invention is also applicable to a method of performing external decompression of the frame and then decompressing the intra-frame.

FIG. 9 is a schematic diagram of decompressing data in a frame in a configuration data stream decompression processing method of a programmable logic device according to an embodiment of the present invention. Referring to FIG. 9, for example, a received frame data compressed internally by a frame is as follows:

8000_0001, 0000_0000, 0002_000E, 1111_1111, 8010_0005, 2222_2222, 3333_3333, 4444_4444, 4444_4444, 5555_5555, 0015_0056, 6666_6666, 8067_0003, 7777_7777, 6666_6666, 6666_6666.

Among them, 8000_0001, 8010_0005, and 8067_0003 are CMD1 for CMD20002_000E and 0015_0056.

When receiving 8000_0001, it informs the decompression circuit to directly put 0000_0000 on the corresponding address of the buffer D, and receives 0002_000E. When notified, the decompression circuit next decompresses a plurality of consecutive identical fields, followed by 0002_000E The address information indicating that the plurality of identical fields are located inside the frame and the number information for indicating the plurality of identical fields, and copying the corresponding number of 1111_1111 to the corresponding address in the frame, and then decompressing the operation , get a frame of data: the 0th field value is 0000_0000; the first field value to the 15th field value data is the same, is 1111_1111; the 16th field value to the 20th field value are: 2222_2222, 3333_3333, 4444_4444, 4444_4444, 5555_5555; the 21st field value is the same as the 102nd field value, which is 6666_6666; the 103th field value to the 105th field value are 7777_7777, 6666_6666, 6666_6666, respectively.

At present, a typical FPGA configuration data stream file compression method in the industry is a frame external frame-based unit compression processing method adopted by Xilinx XC2V40 device. This embodiment uses ISE10.1 development tool and uses Verilog. As the input file of the design, HDL generates a standard configuration stream file through ISE10.1, and uses the generated standard configuration stream file with ISE10.1. The external frame is compressed based on the frame unit compression algorithm to obtain the compressed first compressed data stream file. In addition, the XC2V40 configuration data stream file generated by ISE10.1 is used in ISE10.1. The external frame is compressed based on the frame unit-based compression algorithm, and then compressed by the compression method of the frame internal data unit of the present invention to obtain a compressed second compressed data stream file obtained by the above two compression methods. The file size of the compressed data stream file is shown in Table 1:

 Table 1

Input design file	Standard configuration stream file size (byte)	The size of the first compressed data stream file (byte)	The size of the second compressed data stream file (byte)
Cuonter20.v	57.4K	21.6K	11.0K
Div5.v	57.4K	12.1K	8.6K
Mult2.v	57.4K	13.1K	9.0K
Mult4.v	57.4K	15.9K	9.7K
Mult8.v	57.4K	19.7K	10.7K
Mult16.v	57.4K	25.5K	12.2K
Add4.v	57.4K	15.9K	9.5K
Add8.v	57.4K	19.3K	10.1K
Add16.v	57.4K	21.4K	11.3K
Add24.v	57.4K	26.3K	13.3K

The compression of the compressed data stream file obtained by the above two compression methods is respectively calculated as shown in Table 2. The compression ratio refers to the percentage of the result obtained by dividing the compressed file size by the size of the original file.

 Table 2

Input design file	The compression ratio of the first compressed data stream file	The compression ratio of the second compressed data stream file	Compression ratio increase
Cuonter20.v	37.6%	19.2%	49.0%
Div5.v	21.1%	15.0%	28.9%
Mult2.v	22.9%	15.7%	31.4%
Mult4.v	26.8%	16.9%	36.9%
Mult8.v	34.3%	18.6%	45.8%
Mult16.v	44.4%	21.3%	52.0%
Add4.v	27.7%	16.6%	40.0%
Add8.v	33.6%	17.6%	47.6%
Add16.v	37.3%	19.7%	47.2%
Add24.v	45.8%	23.2%	49.3%
average value	33.15%	18.38%	42.81%

The increase ratio of the compression ratio refers to: the size of the first compressed data stream file is subtracted from the size of the second compressed data stream file, and the obtained result is divided by the size of the first compressed data stream file, and the obtained percentage is obtained. .

As can be seen from the comparison data of Table 2, the present invention combines the compression ratio obtained by combining the conventional frame-based frame unit-based compression method with the frame internal compression method, and the compression ratio obtained by using the frame unit-based compression method alone. Increased by 42.81%. Therefore, the compression method proposed by the present invention can greatly improve the compression performance of the FPGA configuration stream file.

The embodiment further includes a data compression processing system. FIG. 10 is a block diagram of a configuration data stream compression processing system of a programmable logic device according to an embodiment of the present invention. Please refer to FIG. 10:

A data compression processing system includes a data processing module 101 to be compressed and a first compression processing module 102. The data processing module 101 to be compressed is used to acquire a data frame to be compressed in the configuration data stream. The first compression processing module 102 uses And compressing the data unit inside the data frame to be compressed.

Further, the first compression processing module 102 is configured to perform compression processing on the data unit inside the data frame to be compressed to obtain a reference value and a command unit.

Further, the first compression processing module 102 is specifically configured to compare the number of consecutive and identical data units in the statistically derived frame with a preset value, and determine whether the number of consecutive and identical data units is greater than or equal to the number. a preset value; if yes, the command unit obtained by processing the consecutive and identical plurality of data units includes command type information for indicating a compression type, and is used to represent the continuous and identical plurality of data units Address information located inside the frame and number information for indicating the continuous and identical plurality of data units; if not, the command unit obtained by processing the consecutive and identical plurality of data units or 1 data unit includes Command type information for indicating an uncompressed type, address information for indicating that the continuous and identical plurality of data units or one data unit is located inside the frame, and for indicating the continuous and identical plurality of data units or The number information of 1 data unit.

FIG. 11 is a block diagram of a configuration data stream compression processing system of a programmable logic device according to another embodiment of the present invention. Referring to FIG. 11, a second compression processing module 103 is further included, and the second compression processing module 103 is used for Compressing the frame units in the configuration data stream.

FIG. 12 is a block diagram of a configuration data stream decompression processing system of a programmable logic device according to an embodiment of the present invention. Please refer to FIG. 12:

A data decompression processing system includes a data processing module 121 to be decompressed and a first decompression module 122. The data to be decompressed is used to obtain a data frame to be decompressed in the configuration data stream. The decompression module 122 is configured to perform decompression processing on the data unit inside the data frame to be decompressed.

Further, the first decompression module 122 is specifically configured to acquire a command unit and a reference value inside the data frame to be decompressed; and to perform internal data of the data frame to be decompressed according to the command unit and the reference value. The unit performs decompression processing.

Further, the first decompression module 122 is specifically configured to acquire command type information in the command unit; if the command type information in the command unit is a compression type, according to the command unit, the continuous Address information of the same plurality of data units located inside the frame and number information for indicating the continuous and identical plurality of data units, copying the reference value to a corresponding address within the frame; if the command unit is Command type information is an uncompressed type, according to address information in the command unit for indicating that the consecutive and identical plurality of data units or one data unit is located inside the frame and for indicating the continuous and identical The number information of a plurality of data units or 1 data unit is copied to a corresponding address within the frame.

FIG. 13 is a block diagram of a configuration data stream decompression processing system of a programmable logic device according to another embodiment of the present invention. Referring to FIG. 3, a second decompression module 123 is further included. The frame unit in the configuration data stream is decompressed.

The above is a further detailed description of the present invention in connection with the specific embodiments, and the specific embodiments of the present invention are not limited to the description. It will be apparent to those skilled in the art that the present invention may be made without departing from the spirit and scope of the invention.

Claims (25)

1.  A method for compressing a configuration data stream of a programmable logic device, comprising:

   Obtaining a data frame in the configuration data stream;

   Compressing each data unit inside the data frame.
2. The method of claim 1 wherein said data unit comprises one or more bits or said data unit comprises one or more fields.
3. The method of claim 1 wherein compressing processing of consecutive data units within said data frame results in a reference value and a command unit; said reference value comprising data content of the data unit; said command unit including command Type information, address information of a data unit, and quantity information of a data unit.
4. The method according to claim 3, wherein the process of performing compression processing on each data unit inside the data frame is specifically: determining whether a continuous data unit satisfies a compression condition, and if the compression condition is satisfied, satisfying the compression The continuous data unit of the condition is subjected to compression processing to obtain a reference value and a command unit; the reference value is data content of one of the consecutive data units satisfying the compression condition, and the command unit includes Type of command type information, address information of consecutive data units satisfying the compression condition, and number information of consecutive data units satisfying the compression condition.
5. The method according to claim 4, wherein if the compression condition is not satisfied, the continuous data unit that does not satisfy the compression condition is processed to obtain a reference value and a command unit; and the reference value is the unsatisfied compression All data contents of a conditional continuous data unit, the command unit including command type information indicating an uncompressed type, address information of consecutive data units not satisfying the compression condition, and the unsatisfied compression condition The number of consecutive data units.
6. The method according to claim 4, wherein the process of determining whether the continuous data unit satisfies the compression condition is specifically: comparing the number of consecutive and identical data units obtained by statistics with a preset value; if greater than or If it is equal to the preset value, the judgment result is that the compression condition is satisfied; otherwise, the judgment result is that the compression condition is not satisfied.
7.  The method of claim 6 wherein said predetermined value is an integer greater than or equal to three.
8. The method according to any one of claims 1 to 7, wherein before performing compression processing on each data unit inside the data frame, the method further comprises: performing data processing on the frameless internal compression processing Frame external compression processing.
9. The method according to any one of claims 1 to 7, wherein after performing compression processing on each data unit inside the data frame, the method further comprises: performing compression processing on the frame after internal compression processing Frame external compression processing.
10. A method for decompressing a configuration data stream of a programmable logic device, comprising:

    Obtaining a compressed frame in the configuration data stream;

    Performing intra-frame decompression processing on the compressed frame to obtain each data unit after the intra-frame decompression process.
11.  The method according to claim 10, wherein the process of performing intra-frame decompression processing on the compressed frame to obtain each data unit after decompressing the intra-frame is specifically:

    Obtaining a command unit and a reference value inside the compressed frame; the reference value includes data content of the data unit; the command unit includes command type information, address information of the data unit, and quantity information of the data unit;

    Performing intra-frame decompression processing on the compressed frame according to the command unit and the reference value to obtain each data unit after the intra-frame decompression process.
12. The method according to claim 11, wherein the intra-frame decompression process is performed on the compressed frame according to the command unit and the reference value, and the process of obtaining each data unit after the intra-frame decompression process is specifically :

    Obtaining command type information in the command unit;

    If the command type information in the command unit is a compression type, copying the reference value to a corresponding address according to address information and number information of consecutive data units satisfying the compression condition in the command unit, the reference The value is the data content of one of the consecutive data units that satisfy the compression condition.
13. The method according to claim 12, wherein if the command type information in the command unit is of an uncompressed type, address information and number information of consecutive data units that do not satisfy the compression condition in the command unit are And writing the reference value to a corresponding address, where the reference value is the entire data content of the consecutive data units that do not satisfy the compression condition.
14. The method according to any one of claims 10 to 13, further comprising performing frame external decompression processing on the compressed frame before performing intraframe decompression processing on the compressed frame.
15. The method according to any one of claims 10 to 13, wherein after performing intraframe decompression processing on the compressed frame, the method further comprises performing frame external decompression processing on the data frame after intraframe decompression processing. .
16. A compression processing system for configuring a data stream of a programmable logic device, comprising: a data acquisition module to be compressed and a first compression processing module, wherein

    The to-be-compressed processing data acquiring module is configured to acquire one data frame in the configuration data stream;

    The first compression processing module is configured to perform compression processing on each data unit inside the data frame.
17. The system according to claim 16, wherein said first compression processing module is configured to perform compression processing on consecutive data units in said data frame to obtain reference values and command units; said reference value comprising data units Data content; the command unit includes command type information, address information of a data unit, and quantity information of a data unit.
18. The system according to claim 17, wherein the first compression processing module is specifically configured to: determine whether a continuous data unit satisfies a compression condition, and if the compression condition is satisfied, perform continuous data units that satisfy the compression condition. Deriving a reference value and a command unit; the reference value is data content of one of the consecutive data units satisfying the compression condition, the command unit including command type information for indicating a compression type, Address information of consecutive data units satisfying the compression condition, and information of the number of consecutive data units satisfying the compression condition.
19. The system according to claim 18, wherein the first compression processing module is further configured to: if the compression condition is not satisfied, process the consecutive data units that do not satisfy the compression condition to obtain a reference value and a command unit; The reference value is the entire data content of the consecutive data units that do not satisfy the compression condition, and the command unit includes command type information indicating an uncompressed type, and an address of the consecutive data units that do not satisfy the compression condition. Information, the number information of consecutive data units that do not satisfy the compression condition.
20. The system of claim 18, wherein the first compression processing module is specifically configured to compare the number of consecutive and identical data units obtained by statistics with a preset value; if greater than or equal to the predetermined If the value is set, the judgment result is that the compression condition is satisfied; otherwise, the judgment result is that the compression condition is not satisfied.
21. The system according to any one of claims 16 to 20, further comprising a second compression processing module, wherein the second compression processing module is configured to be internal to the data frame by the first compression processing module Before each data unit is compressed, Performing frame external compression processing on the data frame that is not subjected to intra-frame compression processing; or performing internal compression processing on the frame after the first compression processing module performs compression processing on each data unit inside the data frame The compressed frame is subjected to frame external compression processing.
22. A decompression processing system for configuring a data stream of a programmable logic device, comprising: a data acquisition module to be decompressed and a first decompression module, wherein:

    The data to be decompressed processing module is configured to acquire a compressed frame in the configuration data stream;

    The first decompression module is configured to perform intra-frame decompression processing on the compressed frame to obtain each data unit after the intra-frame decompression process.
23. The system according to claim 22, wherein the first decompression module is specifically configured to acquire a command unit and a reference value inside the compressed frame; the reference value includes data content of the data unit; The unit includes command type information, address information of the data unit, and quantity information of the data unit; and performs internal frame decompression processing on the compressed frame according to the command unit and the reference value, and obtains each of the internal decompression processing of the frame. Data unit.
24. The system according to claim 23, wherein the first decompression module is specifically configured to acquire command type information in the command unit; if the command type information in the command unit is a compression type, The first decompression module is configured to copy the reference value to a corresponding address according to address information and number information of consecutive data units satisfying a compression condition in the command unit, where the reference value satisfies the compression condition The data content of one data unit in a continuous data unit.
25. The system according to claim 24, wherein if the command type information in the command unit is of an uncompressed type, the first decompression module is configured to perform continuous according to the command unit that does not satisfy the compression condition The address information and the number information of the data unit are restored to the corresponding address, and the reference value is the entire data content of the consecutive data units that do not satisfy the compression condition.

    26. The system according to any one of claims 22 to 25, further comprising a second decompression module, wherein the second decompression module performs a frame internal solution on the compressed frame in the first decompression module Before the compression process, the frame is externally decompressed for the compressed frame; or the second decompression module is used after the first decompression module performs intraframe decompression processing on the compressed frame. The frame external decompression process is performed on the data frame after the frame internal decompression process.

Images