JP2014191432A - Image processing system and image data processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the load of a host command control unit, and optimally manage a low-speed memory and a high-speed memory regardless of the skill of a programmer regarding the host command control unit.SOLUTION: A CPU 1 issues a display list DL to "read a material n from a low-speed memory 3 and display it on a display unit 5". A prior transfer control unit 21 transfers data for one page relating to the low-speed memory 3, including the material n, when the material n is not stored in a high-speed memory 4 upon receipt of the display list DL, and rewrites the display list DL to "read the material n from the high-speed memory 4 and display it on the display unit 5", to be stored in the high-speed memory 4. The CPU 1 gives a command to a drawing unit 22 to read the rewritten display list DL so as to execute it. The drawing unit 22 reads the rewritten display list DL, executes it, and displays the material n on the display unit 5.

Description

  The present invention relates to an image processing system and an image data processing method, and in particular, a low-speed memory storing data related to a plurality of materials constituting an image displayed on a display unit, and a reading operation faster and smaller than a low-speed memory. The present invention relates to an image processing system having a high-speed memory and a method for processing image data in the system.

  In game machines such as game machines and pachinko machines, the display of images, especially moving images, is an indispensable function, and these images are stored in a read-only memory in advance and read out as necessary. After image processing is performed, the image is displayed on a display unit such as a liquid crystal display. In particular, recently, with the increase in the capacity of such image information, a large-capacity and low-priced memory tends to be used favorably. A typical example of such a memory is a NAND flash memory.

  However, the NAND flash memory is inferior in processing speed as compared with, for example, a NOR flash memory and a DRAM (Dynamic Random Access Memory). In particular, when the memory is randomly accessed locally and image data is repeatedly read out, the NAND flash memory with a slow speed as described above is meaningless and inefficient. . In such a case, it is convenient if there is a mechanism that allows image data read once from a low-speed memory such as a NAND flash memory to be reused instead of reading from the low-speed memory again.

  As a measure to realize such a mechanism, in addition to a read-only low-speed memory in which image data is originally stored, an additional rewritable high-speed memory is provided, and when initially reading from the low-speed memory and displaying it. In addition, there is a measure of transferring the data to the high-speed memory and displaying it again after reading from the high-speed memory. This measure is particularly useful when random access occurs locally. FIG. 6A is a diagram for explaining such a conventional technique. The conventional image processing system shown in FIG. 1 generally includes a CPU 100, an image processor 200, a low speed memory 3, a high speed memory 4, and a display unit 5.

Japanese Patent No. 4748609

  However, as shown in FIG. 6A, in the system to which the high-speed memory 4 is added, whether the image data to be displayed is stored in any of the low-speed memory 3 and the high-speed memory 4. Management had to be performed by the CPU 100 itself. In other words, the programmer of the CPU 100 had to create a program including memory management for the low speed memory 3 and the high speed memory 4.

  FIG. 6B is a diagram showing such a conventional processing procedure of the CPU 100 (a procedure written by a programmer). The figure shows an example in which the material n is to be displayed. That is, the CPU 100 first determines whether or not the material n is currently stored in the high-speed memory 4 with reference to the memory management information related to its creation (step S101). If the material n is not stored in the high-speed memory 4, the CPU 100 transfers the material n from the low-speed memory 3 to the high-speed memory 4 by itself or instructing the image processor 200 (step S102). The memory management information is updated to indicate that the material n is stored in the high speed memory 4 (step S103). Then, the CPU 100 transfers a display list DL that reads out the material n from the high-speed memory 4 and displays it to the image processor 200 (step S104). On the other hand, if the material n is stored in the high speed memory 4 in step S101, the process directly proceeds to step S104. As described above, there is a problem that the burden on the CPU 100 side, that is, the programmer is large. In addition, since it is affected by the skills of individual programmers, memory management is not always optimal.

  A technique for efficiently reading image data by providing a cache memory for a low-speed memory is disclosed in Patent Document 1.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to reduce the load on the host command control unit and to optimally reduce the low-speed memory regardless of the skill of the programmer related to the host command control unit. And an image processing system and an image data processing method capable of managing a high-speed memory.

  In order to achieve the above object, an image processing system according to the present invention includes a low-speed memory in which data relating to a plurality of materials constituting an image displayed on a display unit is stored, a read operation faster than a low-speed memory, and a small capacity A high-speed memory, a high-level command control unit that controls the display so that a desired material is displayed on the display unit, a pre-transfer control unit, and a drawing unit. An image processor for displaying the desired material, wherein the advance transfer control unit reads the desired material from the low-speed memory issued from the higher-order command control unit, and When a display list including at least a command to display on the display unit is received, if it is determined that the desired material is not stored in the high speed memory, the low list is displayed. The desired material stored in the memory is transferred to the high-speed memory, and the display list is rewritten and stored in the high-speed memory so that the desired material is read from the high-speed memory and displayed on the display unit. The drawing unit reads the desired material from the high-speed memory by reading and executing the rewritten display list stored in the high-speed memory based on a command from the host command control unit. The gist is to display on the display unit.

  At that time, it is particularly preferable that the advance transfer control unit reads data in a predetermined area range related to the low-speed memory including the desired material from the low-speed memory and transfers the data to the high-speed memory.

  In the above-described case, it is more effective if the plurality of materials for which data is stored in the low-speed memory are stored closer to each other as the materials that are likely to be displayed continuously.

  Also, typically, the predetermined area range is a page unit in the low-speed memory.

  On the other hand, when determining that the desired material is stored in the high-speed memory, the advance transfer control unit reads the desired material from the high-speed memory and displays it on the display unit. When the display list is not stored in the high-speed memory, the rewritten display list is stored in the high-speed memory.

  Further, the advance transfer control unit obtains information about which material is stored in the high-speed memory and information about which material is rewritten and the display list is stored in the high-speed memory. It is characterized by management.

  The high speed memory may be logically composed of a plurality of memories.

  Also, typically, the low-speed memory is a NAND flash memory, and the high-speed memory is a DRAM.

  In order to achieve the above object, the image data processing method of the present invention has a low-speed memory storing data related to a plurality of materials constituting an image displayed on the display unit, and a reading operation faster than the low-speed memory. A small-capacity high-speed memory, a high-order command control unit that performs command control so that a desired material is displayed on the display unit, a pre-transfer control unit, and a drawing unit. An image data processing method in an image processing system comprising: an image processor for displaying the desired material on a display unit, wherein the upper command control unit displays the desired material on the display unit when the low speed A display list including at least a command to read the desired material from the memory is issued to the image processor, and the display list is received. If the image processor determines that the desired material is not stored in the high-speed memory, the image processor transfers the desired material stored in the low-speed memory to the high-speed memory, and the display list is transferred to the high-speed memory. The desired material is read from the memory and displayed on the display unit and rewritten and stored in the high-speed memory, and the high-order command control unit transmits the rewritten display stored in the high-speed memory to the image processor. The gist is to read and execute a list, and when the image processor receives the command, the image processor reads the desired material from the high-speed memory and displays it on the display unit. .

  According to the present invention, the load on the host command control unit is reduced, and the low-speed memory and the high-speed memory can be optimally managed regardless of the skill of the programmer of the host command control unit. In addition to a desired material, if a predetermined area range, for example, a page unit is read and transferred to a high-speed memory, a high-speed read process can be suitably performed apparently from the viewpoint of the upper command control unit.

1 is a block diagram of a schematic configuration of an embodiment of an image processing system of the present invention. It is a flowchart which shows the process sequence of CPU1. 4 is a flowchart illustrating a processing procedure of a pre-transfer control unit 21 in the image processor 2. 4 is a flowchart illustrating a processing procedure of a drawing unit 22 in the image processor 2. It is a figure for demonstrating transfer of the raw material from the low speed memory 3 to the high speed memory 4. FIG. It is a figure for demonstrating the technique in the past.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a schematic block diagram of an image processing system according to an embodiment of the present invention. The image processing system shown in FIG. 1 is connected to an image processor 2 that controls image display, a CPU (Central Processing Unit) 1 that instructs the image processor 2 to display an image, and the like, and the image processor 2. A display unit 5 for displaying images, a low-speed memory 3 for storing information on a plurality of materials to display images on the display unit 5, and a high-speed memory 4 for temporarily storing information on these materials And. Here, the low-speed memory 3 is a low-cost and relatively large-capacity memory although the operation is low-speed. Typically, it is a NAND flash memory and may be configured as a read-only memory. On the other hand, the high-speed memory 4 is a memory configured with a relatively small capacity because it operates at a high speed but is relatively expensive. Here, it is a rewritable memory, for example, DDR SDRAM (Double-Data-Rate Synchronous
Dynamic Random Access Memory). The high speed memory 4 may be physically mounted inside the image processor 2. An example of such a memory is an embedded DRAM. A plurality of high speed memories 4 may be logically provided. Examples of the material include compression-coded images and 3D vertex stream information. It should be noted that it is preferable to store a plurality of materials stored in the low-speed memory 3 close to each other, which are likely to be read continuously.

  The image processor 2 includes at least a pre-transfer control unit 21, a drawing unit 22, and a decoding unit 23. Typically, these are configured by hardware. Further, the advance transfer control unit 21 includes a transfer processing unit 211, a DL (display list) rewriting unit 212, and a memory management unit 213. Here, the memory management unit 213 includes memory management information MMD that is sequentially updated. The memory management information MMD includes information on each material stored in the high-speed memory 4 and their storage addresses, information on each rewritten display list DL, and their storage addresses.

  The CPU 1 issues a display list DL including each command group “read out the material n from the low-speed memory 3 and display it on the display unit 5”, and after a response from the image processor 2, to the drawing unit 22, A command to read and execute the rewritten display list DL stored in the high-speed memory 4 is given.

  When the advance transfer control unit 21 in the image processor 2 receives the display list DL “Read the material n from the low-speed memory 3 and display it on the display unit 5”, the material n is not stored in the high-speed memory 4. When the determination is made, the data for one page related to the low-speed memory 3 including the material n is read from the low-speed memory 3 and transferred to the high-speed memory 4, and the display list DL is read “Read the material n from the high-speed memory 4. And display it on the display unit 5 ”and store it in the high-speed memory 4. On the other hand, if it is determined that the material n is stored in the high-speed memory 4, it is then determined whether or not the rewritten display list DL related to the material n is stored in the high-speed memory 4. If it is determined that it is not, the display list DL is rewritten as “read out the material n from the high speed memory 4 and displayed on the display unit 5” and stored in the high speed memory 4.

  When the drawing unit 22 in the image processor 2 receives an instruction from the CPU 1 to read and execute the rewritten display list DL stored in the high-speed memory 4, the drawing unit 22 reads and executes the display list DL. The material n is displayed on the display unit 5.

  Next, the processing procedure of an embodiment of the image data processing method of the present invention will be described with reference to FIGS. FIG. 2 is a flowchart showing the processing procedure of the CPU 1. FIG. 3 is a flowchart showing a processing procedure of the advance transfer control unit 21 in the image processor 2. FIG. 4 is a flowchart showing a processing procedure of the drawing unit 22 in the image processor 2. FIG. 5 is a diagram for explaining the transfer of the material from the low-speed memory 3 to the high-speed memory 4.

  First, the display list DL issued by the CPU 1 to the image processor 2 will be described. The display list DL is generally a description in which a plurality of commands (instructions) that can be interpreted by the image processor 2 (particularly, an internal drawing engine) are described in a time series. For example, the display list DL for one frame is displayed. A single screen is displayed by. There are various types of commands, and generally there are a transfer command and a drawing command.

  In particular, in the present invention, regarding the command group for reading out the material from the memory and displaying it, the CPU 1, for example, the command group “Read out the material k from the low-speed memory 3 and display it on the display unit 5”, “ The display list DL including a command group “Read k + 1 from the high-speed memory 4 and display it on the display unit 5” is issued, but the advance transfer control unit 21 of the image processor 2 analyzes the display list DL that is sent. When a command group (display list DL) for reading from the low-speed memory 3 such as “read out the material k from the low-speed memory 3 and display it on the display unit 5” is received for the first time, the material k is read from the low-speed memory 3 at high speed. At the same time, the display list DL is transferred to the memory 4 and “the material k is read from the high speed memory 4 and displayed. Rewrite the display list DL that the case "appears in Part 5, it also is stored in high-speed memory 4. After analyzing and rewriting a series of display lists DL in the advance transfer control unit 21, the CPU 1 reads out the rewritten display list DL stored in the high-speed memory 4 to the drawing unit 22 and executes it. In accordance with the rewritten display list DL, the drawing unit 22 reads the material from the high speed memory 4 instead of the low speed memory 3 and performs processing such as drawing. In short, the CPU 1 issues the display list DL in a state where it does not recognize in which memory the desired material is stored. However, the CPU 1 always uses the function of the advance transfer control unit 21 of the image processor 2. A process of reading out from the high-speed memory 4 and performing drawing or the like is guaranteed.

  As described above, as the processing of the CPU 1, a series of display lists DL including respective command groups “read out the materials 1,..., K,... However, in the following description, for the sake of convenience, as shown in FIG. 2, “read material n (= 1,... K,...) From low-speed memory 3 and display it on display unit 5”. A command group (hereinafter referred to as a display list DL) is transferred to the image processor 2 (step S11).

  Referring to FIG. 3, the advance transfer control unit 21 in the image processor 2 reads the data related to the material n from the CPU 1 from the low-speed memory 3 and displays it on the display unit 5 in a series of analysis processing flows. The CPU 1 determines whether or not the display list DL is received (step S21). As described above, the CPU 1 displays the display list DL "Read the material n from the low-speed memory 3 and display it on the display unit 5". 2, the determination here is affirmative. Next, the advance transfer control unit 21 refers to the memory management information MMD provided in the memory management unit 213, and the data related to the material n is stored in the high-speed memory 4. It is determined whether it is stored (step S22).

  If it is determined in step S22 that the data related to the material n is not stored in the high-speed memory 4 (No determination), the transfer processing unit 211 in the advance transfer control unit 21 is even stored in the low-speed memory 3. Data for one page including the material n (one page in the memory: for example, 16 kbytes) is transferred to the high-speed memory 4 (step S23). Here, the reason for transferring one page at a time is that, when the low-speed memory 3 is a NAND flash memory, an initial operation processing time for reading is required for each reading process. This is because it is efficient because the initial operation processing time for one time is sufficient for pages, and if one page is read at this stage, it becomes more effective as a cache function using the high-speed memory 4.

  Next, the DL rewriting unit 212 in the advance transfer control unit 21 reads the display list DL “Read the material n from the low speed memory 3 and display it on the display unit 5”, “Read the material n from the high speed memory 4. The display list DL is rewritten as “display on the display unit 5”, and the rewritten display list DL is stored in the high-speed memory 4 (step S24). Then, the memory management unit 213 in the advance transfer control unit 21 updates the memory management information MMD (step S25).

  On the other hand, if it is determined in step S22 that the data related to the material n is stored in the high speed memory 4 based on the memory management information MMD (affirmative determination), the advance transfer control unit 21 next selects the memory management information. With reference to the information MMD, it is determined whether or not the rewritten display list DL relating to the material n is stored in the high speed memory 4 (step S26). Here, when it is determined that the rewritten display list DL related to the material n is not stored in the high-speed memory 4, that is, the data related to the material n is stored in the high-speed memory 4, but is read out. If the display list DL is not stored in the high speed memory 4, the process proceeds to step S24, the display list DL is rewritten and stored in the high speed memory 4, and the memory management information MMD is updated in step S25. On the other hand, when it is determined in step S26 that the rewritten display list DL related to the material n is stored in the high-speed memory 4, the pre-transfer control unit 21 ends without processing.

  Here, to explain with an example, the image processor 2 receives the display list DL from the CPU 1 "Read the material k from the low-speed memory 3 and display it on the display unit 5", and the advance transfer control unit 21 performs step S22. If it is determined that the data related to the material k is not stored in the high-speed memory 4, the transfer processing unit 211 in the advance transfer control unit 21 stores the data in the low-speed memory 3 as shown in FIG. One page of data including the material k is transferred to the high-speed memory 4. In the case of the example in FIG. 5, the page including the material k includes a part of the material k−1, the material k + 1, and the material k-2. At the time when step S25 ends, the memory management information MMD includes a part of k-1, k, k + 1, and k-2 for the material, and a display that is rewritten for the material k for the display list DL. The list DL is updated to indicate that it is stored in the high speed memory 4.

  In this state, when the image processor 2 receives a display list DL from the CPU 1 that reads “read the material k + 1 from the low-speed memory 3 and displays it on the display unit 5”, the material k + 1 is already stored in the high-speed memory 4. Therefore, an affirmative determination is made in step S22, and the process proceeds to step S26. Here, for the material k + 1, although the material itself is stored in the high-speed memory 4, the rewritten display list DL for transferring the material k + 1 is not yet stored in the high-speed memory 4, so in step S26 Is negative, and the process proceeds to step S24. In this step, the display list DL “Read the material k + 1 from the low-speed memory 3 and display it on the display unit 5” is changed to the display list DL “Read the material k + 1 from the high-speed memory 4 and display it on the display unit 5”. The data is rewritten and stored in the high-speed memory 4, and an update to that effect is performed in step S25.

  In this state, next, when the image processor 2 receives the display list DL “read the material k + 1 from the low-speed memory 3 and display it on the display unit 5” from the CPU 1, the material k + 1 is already stored in the high-speed memory 4. In addition, since the rewritten display list DL “Read the material k + 1 from the high speed memory 4 and display it on the display unit 5” is also stored in the high speed memory 4, an affirmative determination is made in both step S22 and step S26. And ends without processing.

  Returning to FIG. 2, after the display list DL “Read the material n from the low-speed memory 3 and display it on the display unit 5” is transferred to the processor 2, the fact that the processing is completed (including the storage address) is displayed as an image. When returning from the processor 2, the CPU 1 next gives a command “read and execute the rewritten display list DL relating to the material n from the high-speed memory 4” to the drawing unit 22 in the image processor 2 ( Step S12). This command is a simple command including at least the storage address in the high-speed memory 4.

  Referring to FIG. 4, the drawing unit 22 determines whether or not a command “Read and execute the rewritten display list DL relating to the material n from the high-speed memory 4” has been received (step S31). In this case, the rewritten display list DL related to the material n is read from the high-speed memory 4 (step S32), the read display list DL is executed, and the decoding unit 23 performs decompression processing as necessary, and the material n Is displayed on the display unit 5 (step S33).

  In step S23, when the material n to be read straddles two pages, the two straddling pages may be read.

  As described above, according to the above-described embodiment, the load on the CPU 1 side is reduced, and the low-speed memory 3 and the high-speed memory 4 can be optimally managed regardless of the skill of the programmer of the CPU 1. In addition, since a plurality of materials stored in the low-speed memory 3 are stored in close proximity to each other with a high possibility of being read continuously, for example, one page is read and the high-speed memory is read as described above. If it is transferred to 4, the high-speed reading process can be suitably performed from the viewpoint of the CPU 1.

  The image processing system and the image data processing method of the present invention are useful when a plurality of materials to be continuously displayed on the display unit are stored and used in an inexpensive and large-capacity low-speed memory.

1,100 CPU
2,200 Image processor 21 Advance transfer control unit 211 Transfer processing unit 212 DL rewriting unit 213 Memory management unit 22 Drawing unit 23 Decoding unit 3 Low speed memory 4 High speed memory 5 Display unit MMD Memory management information DL Display list

Claims (9)

A low-speed memory in which data relating to a plurality of materials constituting an image displayed on the display unit is stored;
A high-speed memory with a smaller read capacity and faster read speed than a low-speed memory,
An upper command control unit that performs command control to display a desired material on the display unit;
An image processor having a pre-transfer control unit and a drawing unit, and displaying the desired material on the display unit by command control by the host command control unit;
An image processing system comprising:
When the pre-transfer control unit receives a display list including at least a command issued from the high-order command control unit to read out the desired material from the low-speed memory and display it on the display unit, the desired transfer control unit If it is determined that no material is stored in the high speed memory, the desired material stored in the low speed memory is transferred to the high speed memory, and the display list is read from the high speed memory. Rewrite to display on the display unit, store in the high-speed memory,
The drawing unit reads the desired material from the high-speed memory by reading and executing the rewritten display list stored in the high-speed memory based on a command from the host command control unit. An image processing system characterized by displaying on a display unit.   2. The image according to claim 1, wherein the pre-transfer control unit reads out data in a predetermined area range related to the low-speed memory including the desired material from the low-speed memory and transfers the data to the high-speed memory. Processing system.   The image processing system according to claim 2, wherein the plurality of materials in which data is stored in the low-speed memory are stored closer to each other as the materials that are more likely to be displayed continuously.   The image processing system according to claim 2, wherein the predetermined area range is a page unit in the low-speed memory.   When the pre-transfer control unit determines that the desired material is stored in the high-speed memory, the display list that has been rewritten to read the desired material from the high-speed memory and display it on the display unit 2. The image processing system according to claim 1, wherein when the data is not stored in the high-speed memory, the rewritten display list is stored in the high-speed memory.   The advance transfer control unit manages information on which material is stored in the high-speed memory and information on which material is rewritten and the display list is stored in the high-speed memory. The image processing system according to claim 1, wherein the image processing system is an image processing system.   7. The image processing system according to claim 1, wherein the high-speed memory is logically composed of a plurality of memories.   8. The image processing system according to claim 1, wherein the low-speed memory is a NAND flash memory, and the high-speed memory is a DRAM. A low-speed memory in which data relating to a plurality of materials constituting an image displayed on the display unit is stored, a high-speed memory having a lower reading speed and a smaller capacity than the low-speed memory, and a desired material are displayed on the display unit. An image processor comprising: an upper command control unit that performs command control, and an image processor that has a pre-transfer control unit and a drawing unit, and displays the desired material on the display unit by command control by the upper command control unit An image data processing method in a system,
The upper command control unit, when displaying the desired material on the display unit, issues a display list including at least a command to read the desired material from the low-speed memory to the image processor,
When the image processor that has received the display list determines that the desired material is not stored in the high-speed memory, the image processor transfers the desired material stored in the low-speed memory to the high-speed memory and also displays the display Read the list from the high-speed memory and display the desired material on the display unit, rewrite and store the list in the high-speed memory,
The upper command control unit instructs the image processor to read and execute the rewritten display list stored in the high-speed memory,
When the image processor receives the instruction, the image processor reads the desired material from the high-speed memory and displays it on the display unit.

Images