JP4828006B2 - Image processing device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus.
[0002]
[Prior art]
Computer graphics are often used in various CAD (Computer Aided Design) systems and amusement machines. In particular, with the recent development of image processing technology, systems using three-dimensional computer graphics are rapidly spreading.
In such 3D computer graphics, when determining the color corresponding to each pixel (pixel), the color value of each pixel is calculated, and the calculated color value is used as the display buffer corresponding to the pixel. Rendering processing to write to the (frame buffer) address.
[0003]
One of the rendering processing methods is polygon rendering. In this method, a three-dimensional model is expressed as a combination of triangular unit graphics (polygons), and the color of the display screen is determined by drawing with the polygon as a unit.
[0004]
In the polygon rendering, the coordinates (x, y, z), color data (R, G, B), and the coordinate coordinates (s) of the texture data indicating the image pattern of pasting for each vertex of the triangle in the physical coordinate system. , T) and the value of the homogeneous term q are input, and a process of interpolating these values inside the triangle is performed.
Here, the homogeneous term q is simply an enlargement / reduction ratio, and the coordinates of the actual texture buffer in the UV coordinate system, that is, the texture coordinate data (u, v) are represented by the homogeneous coordinates (s , T) divided by the homogeneous term q is multiplied by the texture sizes USIZE and VSIZE, respectively, and “t / q”.
[0005]
FIG. 28 is a system configuration diagram showing a basic concept of a three-dimensional computer graphics system.
[0006]
In this three-dimensional computer graphics system, graphics rendering data is rendered via a main bus 4 from the main memory 2 of the main processor 1 or from an I / O interface circuit 3 that receives graphics data from the outside. 5a and a rendering circuit 5 having a frame buffer memory 5b.
[0007]
The rendering processor 5a is coupled with a frame buffer 5b intended to hold data for display and a texture memory 6 holding texture data to be attached to the surface of a graphic element (for example, a triangle) to be drawn. ing.
Then, the rendering processor 5a performs a process of drawing a graphic element with a texture on the surface of each graphic element in the frame buffer 5b.
[0008]
The frame buffer 5b and the texture memory 6 are generally constituted by a DRAM (Dynamic Random Access Memory).
In the system of FIG. 28, the frame buffer 5b and the texture memory 6 are configured as physically separate memory systems.
[0009]
[Problems to be solved by the invention]
By the way, in the above-described three-dimensional computer graphic system or the like, in order to efficiently perform the read operation and write operation of the DRAM, for example, the address designation at the time of accessing the DRAM is limited in consideration of the characteristics of graphic processing. It is possible.
However, when such a restriction is provided, in the system shown in FIG. 28, when data is transmitted / received between the rendering processor 5a and the main processor 1, and the frame buffer memory 5b and the texture memory 6, the DRAM is not connected. There is a need to solve various problems caused by addressing restrictions during access.
[0010]
The present invention has been made in view of such circumstances, and it is an object of the present invention to provide an image processing apparatus capable of accurately accessing a memory even when there is a restriction on address designation when accessing the memory. Objective.
[0011]
[Means for Solving the Problems]
  According to the present invention,
  Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic An image processing circuit that performs rendering processing using data and generates image data including a plurality of pixels;
  Writing of image data composed of the plurality of pixels generated by the rendering process in units of a block having a predetermined sizeThe writing is performed in units of the first rectangular area with the representative point as a reference, and the first and second directions orthogonal to the first rectangular area are the first. There is a restriction that only addresses that are a multiple of the number can be specified.A plurality of memory circuits;
  When the address of the representative point of the block including the image data composed of the plurality of pixels generated by the rendering process input from the image processing circuit is specified based on the restriction,Input from the image processing circuitThe image data isBased on the constraintsIn the plurality of memory circuitsThe corresponding first rectangular area of the corresponding memory circuitTo be written on,in frontA data rearrangement circuit for rearranging the image data in units of pixel data;
  bookWith a built-in circuit
  Comprising
  The rearrangement circuit is connected to the image processing circuit.Image data composed of the plurality of pixels generated by the input rendering processThe plurality of memory circuitsSwitch toIn the first transfer mode for sending and storing,
    The rendered first image data having the first bit width input from the image processing circuit is converted into second image data having the second bit width.Placed in the
    When the address of the representative point of the block including the second image data is designated based on the restriction, the first image data is stored in the plurality of storage circuits.The corresponding first rectangular storage area of the corresponding storage circuitThe second image data is written toBased on the control signal, the shift process, selection process, and hold process are repeated,Sorting,
  The writing circuit distributes and writes the blocks including the rearranged second image data to corresponding storage circuits of the plurality of storage circuits;
  An image processing apparatus is provided.
[0012]
  Also according to the invention,
  Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic Writing and reading of image data composed of a plurality of pixel data generated by a rendering process in which data is stored in a distributed manner and stored in a storage area in units of blocks of a predetermined size are performed.A plurality of storage circuits, wherein the writing is performed in units of a first rectangular area with a representative point as a reference, and a multiple of a first number in the first and second directions orthogonal to the first rectangular area There is a first restriction that only the address of the second rectangular area can be specified, and the reading is performed in units of the second rectangular area with the representative point as a reference, and an arbitrary address is set in the first direction perpendicular to the second rectangular area. There is a second restriction that the address can be specified only in multiples of the second number in the second direction.A plurality of memory circuits;
  An image processing circuit for inputting first image data having a first bit width and performing a rendering process on the first image data;
  SaidSecond restrictionA readout circuit for designating a representative point based on the block and reading out the image data from the plurality of storage circuits in units of the block;
  Data rearrangement for outputting the first image data of the first bit width generated by rearranging the second image data of the second bit width obtained by the reading in units of pixel data to the image processing circuit Circuit and
  Comprising
  In the second transfer mode in which the rearrangement circuit transfers image data from the storage circuit to the image processing circuit,
    An address corresponding to the address input from the image processing circuit is generated and output to the reading circuit to read out corresponding image data from the plurality of storage circuits,
    Converting to image data of the first bit width defined by the image processing circuit and outputting to the image processing circuit;
  An image processing apparatus is provided.
[0013]
  According to the present invention,
  Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic The data is distributed, the image data generated by the rendering process is written in units of a predetermined size block, and the image data generated by the distributed writing process is read. Is doneA plurality of memory circuits, wherein the writing is performed in units of a first rectangular area with a representative point as a reference, and a first number of first and second directions orthogonal to the first rectangular area There is a first restriction that only multiple addresses can be specified, and the reading is performed in units of the second rectangular area with the representative point as a reference, and the first direction orthogonal to the second rectangular area is arbitrary. There is a second restriction that the address can be specified but the address can only be specified in multiples of the second number in the second direction.A plurality of memory circuits;
  SaidSecondA readout circuit that reads out the image data from the plurality of storage circuits in units of the blocks by designating representative points based on restrictions;
  Processing is performed according to either a first transfer mode in which image data is transferred from the image processing circuit to the storage circuit, or a second transfer mode in which image data is transferred from the storage circuit to the image processing circuit. A rearrangement circuit, wherein the address of the representative point of the block including the read image data isFirst limitWhen the image data is specified based onApplicableIn memory circuitThe corresponding first rectangular areaA data rearrangement circuit for rearranging the read image data in units of pixel data so as to be written to
  SaidFirstA writing circuit that designates a representative point based on a restriction and writes the block including the rearranged image data to the storage circuit;
  Comprising
  In the first transfer mode, the rearrangement circuit transfers and stores the image data from the image processing circuit to the plurality of storage circuits.
    The rendered first image data having the first bit width input from the image processing circuit is converted into second image data having the second bit width.Placed inside,
    The address of the representative point of the block including the second image data isFirst limitThe first image data is stored in the plurality of storage circuits when designated based onThe corresponding rectangular area of the corresponding memory circuitThe second image data is written toBased on the control signal, the shift process, selection process, and hold process are repeated,Sorting,
  The writing circuit distributes and writes the blocks including the rearranged second image data to corresponding storage circuits of the plurality of storage circuits,
  In the second transfer mode in which the rearrangement circuit transfers image data from the storage circuit to the image processing circuit,
    An address corresponding to the address input from the image processing circuit is generated and output to the reading circuit to read out corresponding image data from the plurality of storage circuits,
    Image data of the first bit width defined by the image processing circuitPlaced insideOutput to the image processing circuit;
  An image processing apparatus is provided.
[0014]
  According to the present invention,
  Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic An image processing apparatus that performs rendering processing using data,
  A plurality of storage circuits, wherein display data consisting of a plurality of pixel bits and texture data required by at least one graphic element are distributed and stored in the plurality of storage circuits, and a block of a predetermined size is unit The image data generated by the rendering process is written asA plurality of storage circuits, wherein the writing is performed in units of a first rectangular area with a representative point as a reference, and a multiple of a first number in the first and second directions orthogonal to the first rectangular area There is a first restriction that only the address of the second rectangular area can be specified, and the reading is performed in units of the second rectangular area with the representative point as a reference, and an arbitrary address is set in the first direction perpendicular to the second rectangular area. There is a second restriction that the address can be specified only in multiples of the second number in the second direction.A plurality of memory circuits;
  An interpolation data generation circuit that interpolates polygon rendering data at the vertices of the unit graphic and generates interpolation data of pixels located in the unit graphic;
  The texture homogeneous coordinates (s, t) included in the interpolation data are divided by the homogeneous term q to generate “s / q” and “t / q”, and the “s / q” and “t / q”. Texture processing that reads texture data from the plurality of storage circuits using a texture address according to the image, and pastes the read texture data on the surface of the graphic element to generate display data composed of a plurality of pixel bits Circuit,
  A data rearrangement circuit for performing data rearrangement processing according to a first transfer mode for transferring image data from the texture processing circuit to the storage circuit, and comprising a plurality of pixel bits input from the texture processing circuit The address of the representative point of the block containing display data isFirst limitThe display data is stored in the plurality of storage circuits when specified based onApplicable memory circuitA data rearrangement circuit for rearranging the display data in units of image data so as to be written in a distributed manner;
  A writing circuit that writes the blocks including the rearranged display data in a distributed manner to the plurality of storage circuits;
  Comprising
  In the first transfer mode, the rearrangement circuit transfers and stores the image data from the image processing circuit to the plurality of storage circuits.
    The rendered first image data having the first bit width input from the image processing circuit is converted into second image data having the second bit width.Placed inside,
    The address of the representative point of the block including the second image data isFirst limitThe second image data so that the first image data is written into a predetermined storage area of the plurality of storage circuits when designated based onBased on the control signal, the shift process, selection process, and hold process are repeated,Sorting,
  The writing circuit distributes and writes the blocks including the rearranged second image data to corresponding storage circuits of the plurality of storage circuits;
  An image processing apparatus is provided.
[0015]
  According to the present invention,
  Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic An image processing apparatus that performs rendering processing using data,
  A plurality of storage circuits, wherein display data and texture data required by at least one graphic element are distributed and stored in the plurality of storage circuits, and generated by the above rendering process in units of a predetermined size block. Written image data is writtenA plurality of memory circuits, wherein the writing is performed in units of a first rectangular area with a representative point as a reference, and a first number of first and second directions orthogonal to the first rectangular area There is a first restriction that only multiple addresses can be specified, and the reading is performed in units of the second rectangular area with the representative point as a reference, and the first direction orthogonal to the second rectangular area is arbitrary. There is a second restriction that the address can be specified but the address can only be specified in multiples of the second number in the second direction.A plurality of memory circuits;
  An interpolation data generation circuit that interpolates polygon rendering data at the vertices of the unit graphic and generates interpolation data of pixels located in the unit graphic;
  The texture homogeneous coordinates (s, t) included in the interpolation data are divided by the homogeneous term q to generate “s / q” and “t / q”, and the “s / q” and “t / q”. A texture processing circuit that generates display data by applying a texture data read from the plurality of storage circuits to a surface of a graphic element using a texture address according to
  SaidSecond restrictionA readout circuit that designates a representative point based on the block and reads the texture data from the plurality of storage circuits in units of the block;
  An interface circuit;
  A data rearrangement circuit for rearranging data according to a second transfer mode for transferring display data read from the storage circuit to the texture processing circuit, wherein the read texture data is transferred to the interface. A data rearrangement circuit for rearranging according to the output format of the circuit and outputting to the interface circuit;
  Comprising
  The rearrangement circuit includes:SaidIn the second transfer mode, an address corresponding to an address input from the image processing circuit from the texture circuit via the interface is generated and output to the readout circuit, and corresponding image data is output from the plurality of storage circuits. Read and image data of the first bit width defined by the image processing circuitPlaced insideOutput to the image processing circuit;
  An image processing apparatus is provided.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, in the present embodiment, a 3D computer graphics system that is applied to a personal computer or the like and displays a desired 3D image of an arbitrary 3D object model on a display such as a CRT (Cathode Ray Tube) at high speed. Will be described.
[0017]
FIG. 1 is a system configuration diagram of a three-dimensional computer graphics system 10 as an image processing apparatus according to the present invention.
[0018]
The three-dimensional computer graphics system 10 represents a three-dimensional model as a combination of triangles (polygons) that are unit figures, determines the color of each pixel on the display screen by drawing the polygon, and renders the polygon on the display. It is a system that performs processing.
The three-dimensional computer graphics system 10 represents a three-dimensional object using z coordinates representing depth in addition to (x, y) coordinates representing a position on a plane, and this (x, y, z). Any one point in the three-dimensional space is specified by the three coordinates.
[0019]
As shown in FIG. 1, a three-dimensional computer graphics system 10 includes a main processor 11, a main memory 12, an I / O interface circuit 13, and a rendering circuit 14 connected via a main bus 15.
Hereinafter, the function of each component will be described.
[0020]
For example, the main processor 11 reads out necessary graphic data from the main memory 12 in accordance with the progress of the application and the like, and geometry (Geometry) such as clipping processing and lighting processing for the graphic data. Processing is performed to generate polygon rendering data. The main processor 11 outputs the polygon rendering data S11 to the rendering circuit 14 via the main bus 15.
[0021]
The I / O interface circuit 13 receives movement control information, polygon rendering data, or the like from the outside as necessary, and outputs this to the rendering circuit 14 via the main bus 15.
[0022]
Here, the polygon rendering data includes data of (x, y, z, R, G, B, s, t, q) of each of the three vertices of the polygon.
Here, the (x, y, z) data indicates the three-dimensional coordinates of the top of the apple, and the (R, G, B) data indicates the red, green, and blue luminance values in the three-dimensional coordinates, respectively. Yes.
Of the (s, t, q) data, (s, t) indicates the homogeneous coordinates of the corresponding texture, and q indicates the homogeneous term. Here, “s / q” and “t / q” are multiplied by the texture sizes USIZE and VSIZE, respectively, to obtain texture coordinate data (u, v). Access to the texture data stored in the texture buffer 147a is performed using the texture coordinate data (u, v).
That is, the polygon rendering data is a physical coordinate value of each vertex of the triangle, a color of each vertex, and texture data.
[0023]
Hereinafter, the rendering circuit 14 will be described in detail.
As shown in FIG. 1, the rendering circuit 14 includes a host interface circuit 149, a DDA (Digital Differential Anarizer) setup circuit 141, a triangle DDA circuit 142, a texture engine circuit 143, a memory interface (I / F) circuit 144, and a CRT control circuit. 145, a RAMDAC circuit 146, a DRAM 147, and an SRAM (Static RAM) 148.
In the rendering circuit 14 according to this embodiment, a logic circuit and a DRAM 147 that stores at least display data and texture data are mixedly mounted in one semiconductor chip.
[0024]
DRAM147
The DRAM 147 functions as a texture buffer 147a, a display buffer 147b, a z buffer 147c, and a texture CLUT (Color Look Up Table) buffer 147d.
The DRAM 147 is divided into a plurality (four in this embodiment) of modules having the same function, as will be described later.
[0025]
In addition, in order to store more texture data, the DRAM 147 stores an index in the index color and a color lookup table value for the index color in the texture CLUT buffer 147d.
The index and color lookup table values are used for texture processing. That is, the texture element is usually expressed by 24 bits in total of 8 bits for each of R, G, and B. However, since the data amount is expanded, one color is selected from, for example, 256 colors selected in advance, The data is used for texture processing. Thus, if there are 256 colors, each texture element can be expressed by 8 bits. Although a conversion table from an index to an actual color is necessary, the texture data becomes more compact as the texture resolution increases.
Thereby, the texture data can be compressed, and the built-in DRAM can be used efficiently.
[0026]
The DRAM 147 stores depth information of an object to be drawn in order to perform hidden surface processing in parallel with drawing.
Note that the display data, depth data, and texture data are stored as follows: display data is stored continuously from the beginning of the memory block, then depth data is stored, and the remaining space is stored for each texture type. Texture data is stored in a continuous address space. Thereby, texture data can be stored efficiently.
[0027]
FIG. 2 is a block diagram illustrating a specific configuration example of the DRAM 147, the SRAM 148, and the memory I / F circuit 144 that accesses the DRAM 147 and the SRAM 148.
[0028]
As shown in FIG. 2, the DRAM 147 and the SRAM 148 shown in FIG. 1 are divided into four memory modules 200, 210, 220, and 230 as described above.
[0029]
The memory module 200 includes memories 201 and 202.
The memory 201 includes banks 201A and 201B that constitute a part of the DRAM 147, and banks 201C and 201D that constitute a part of the SRAM 148.
The memory 202 includes banks 202A and 202B that constitute a part of the DRAM 147, and banks 202C and 202D that constitute a part of the SRAM 148.
It should be noted that the banks 201C, 201D, 202C, 202D constituting the SRAM 148 can be accessed simultaneously.
[0030]
The memory module 210 includes memories 211 and 212.
The memory 211 includes banks 211A and 211B that constitute a part of the DRAM 147, and banks 211C and 211D that constitute a part of the SRAM 148.
The memory 212 includes banks 212A and 212B that constitute part of the DRAM 147, and banks 212C and 212D that constitute part of the SRAM 148.
Note that the banks 211C, 211D, 212C and 212D constituting the SRAM 148 can be accessed simultaneously.
[0031]
The memory module 220 includes memories 221 and 222.
The memory 221 includes banks 221A and 221B that constitute a part of the DRAM 147, and banks 221C and 221D that constitute a part of the SRAM 148.
The memory 222 includes banks 222A and 222B that constitute a part of the DRAM 147, and banks 222C and 222D that constitute a part of the SRAM 148.
Note that the banks 221C, 221D, 222C, and 222D constituting the SRAM 148 can be accessed simultaneously.
[0032]
The memory module 230 includes memories 231 and 232.
The memory 231 includes banks 231A and 231B that constitute a part of the DRAM 147, and banks 231C and 231D that constitute a part of the SRAM 148.
The memory 232 includes banks 232A and 232B that constitute part of the DRAM 147, and banks 232C and 232D that constitute part of the SRAM 148.
Note that the banks 231C, 231D, 232C, and 232D constituting the SRAM 148 can be accessed simultaneously.
[0033]
Here, each of the memory modules 200, 210, 220, and 230 has all the functions of the texture buffer 147a, the display buffer 147b, the Z buffer 147c, and the texture CLUT buffer 147d shown in FIG.
That is, each of the memory modules 200, 210, 220, and 230 stores all of texture data, drawing data ((R, G, B) data), z data, and texture color lookup table data of the corresponding pixel.
However, the memory modules 200, 210, 220, and 230 store data about different pixels.
Here, texture data, drawing data, z data, and texture color look-up table data for 16 pixels processed simultaneously are different from each other in banks 201A, 201B, 202A, 202B, 211A, 211B, 212A, 212B, 221A, 221B, 222A, 222B, 231A, 231B, 232A, 232B.
As a result, the memory I / F circuit 144 can simultaneously write, for example, data for 16 pixels of 8 (X direction) × 2 pixels (Y direction) to the DRAM 147.
For example, data can be read simultaneously for 8 pixels of 8 (X direction) × 1 pixel (Y direction).
Note that the memory I / F circuit 144 accesses (writes) the DRAM 147 based on so-called predetermined interleaving addressing, as will be described later.
[0034]
FIG. 3 is a schematic diagram illustrating a configuration example of the DRAM 147 as a buffer (for example, a texture buffer).
As shown in FIG. 3, data accessed in a memory area of 2 × 8 pixels (pixels) is stored in an area called a page (row) or a block (column).
Each row ROW0 to ROWn + 1 is divided into four columns (blocks) M0A, M0B, M1A, and M1B as shown in FIG.
Then, at the time of writing, an even number of boundaries in the X direction and the Y direction is accessed, and at the time of reading, an access is made in a boundary that is a multiple of 8 in the X direction and an arbitrary boundary in the Y direction.
[0035]
Banks 201C, 201D, 202C, 202D, 211C, 211D, 212C, 212D, 221C, 221D, 222C, 222D, 231C, 231D, 232C, and 232D are included in banks 201A, 201B, 202A, 202B, 211A, and 211B, respectively. , 212A, 212B, 221A, 221B, 222A, 222B, 231A, 231B, 232A, and 232B are stored.
[0036]
Next, the texture data storage pattern in the texture buffer 147a based on the interleaved addressing will be described in more detail with reference to FIGS.
4 is a diagram for explaining color data (pixel data) included in the texture data and accessed simultaneously, FIG. 5 is a diagram for explaining unit blocks constituting the texture data, and FIG. 6 is an address of the texture buffer. It is a figure for demonstrating space.
[0037]
In the case of the present embodiment, as shown in FIG. 4, color data pix indicating the color data of pixels arranged in a 2 × 8 matrix included in the texture data.₀~ Pix₁₅Are accessed simultaneously.
[0038]
Color data pix₀~ Pix₁₅Need to be stored in different banks of the SRAM 148 that make up the texture buffer 147a.
In this embodiment, the color data pix₀, Pix₁, Pix₈, Pix₉Are stored in the banks 201C and 201D of the memory 201 and the banks 202C and 202D of the memory 202 shown in FIG. Also, color data pix₂, Pix_Three, Pix_Ten, Pix₁₁Are stored in the banks 211C and 211D of the memory 211 and the banks 212C and 212D of the memory 212 shown in FIG. Also, color data pix_Four, Pix_Five, Pix₁₂, Pix₁₃Are stored in the banks 221C and 221D of the memory 221 and the banks 222C and 222D of the memory 222 shown in FIG. Furthermore, color data pix₆, Pix₇, Pix₁₄, Pix₁₅Are stored in the banks 231C and 231D of the memory 231 and the banks 232C and 232D of the memory 232 shown in FIG.
[0039]
In the present embodiment, color data pix of pixels located in a rectangular area that is processed simultaneously.₀~ Pix₁₅Unit block R_iFor example, texture data indicating one image is represented by unit blocks R arranged in a B × A matrix as shown in FIG.₀~ R_BA-1Consists of.
Unit block R₀~ R_BA-1Is stored in the DRAM 147 constituting the texture buffer 147a so as to have continuous addresses in a one-dimensional address space, as shown in FIG. Each unit block R₀~ R_BA-1Color data pix in₀~ Pix₁₅Are stored in different banks of the SRAM 148 so as to have consecutive addresses in a one-dimensional address space.
That is, the texture buffer 147a stores unit blocks composed of color data that are accessed simultaneously so as to have continuous addresses in a one-dimensional address space.
[0040]
Host interface circuit 149
The host interface circuit 149 is connected to the main bus 15 outside the rendering circuit 14, and is connected to the DDA setup circuit 141 and the memory I / F circuit 144 inside the rendering circuit 14.
As will be described later, the host interface circuit 149 serves as an interface for performing communication between the memory I / F circuit 144 and the main processor 11 and the I / O interface circuit 13.
[0041]
DDA setup circuit 141
Prior to obtaining the color and depth information of each pixel inside the triangle by linearly interpolating the value of each vertex of the triangle on the physical coordinate system in the triangle DDA circuit 142 in the subsequent stage, the DDA setup circuit 141 is polygon rendering data. For the (z, R, G, B, s, t, q) data indicated by S11, a setup calculation is performed to obtain the difference between the sides of the triangle and the horizontal direction.
Specifically, this set-up calculation uses the start point value, end point value, and distance between the start point and end point to calculate the variation of the value to be obtained when the unit length is moved. .
The DDA setup circuit 141 outputs the calculated variation data S141 to the triangle DDA circuit 142.
[0042]
Triangle DDA circuit 142
The triangle DDA circuit 142 uses the variation data S141 input from the DDA setup circuit 141 to calculate linearly interpolated (z, R, G, B, s, t, q) data for each pixel inside the triangle. .
The triangle DDA circuit 11 converts (x, y) data of each pixel and (z, R, G, B, s, t, q) data in the (x, y) coordinates into DDA data (interpolation data). In step S142, the data is output to the texture engine circuit 143.
For example, the triangle DDA circuit 142 outputs, to the texture engine circuit 143, DDA data S142 for 8 (= 2 × 4) pixels located in a rectangle to be processed in parallel.
[0043]
Texture engine circuit 143
The texture engine circuit 143 calculates “s / q” and “t / q”, calculates texture coordinate data (u, v), reads (R, G, B) data from the texture buffer 147a, etc. Is performed in a pipeline manner.
Note that the texture engine circuit 143 simultaneously performs, for example, processing on eight pixels located within a predetermined rectangle.
[0044]
The texture engine circuit 143 performs an operation for dividing the s data by the q data and an operation for dividing the t data by the q data for the (s, t, q) data indicated by the DDA data S142.
The texture engine circuit 143 is provided with, for example, eight division circuits (not shown), and divisions “s / q” and “t / q” for eight pixels are performed simultaneously.
[0045]
Also, the texture engine circuit 143 multiplies the division results “s / q” and “t / q” by the texture sizes USIZE and VSIZE to generate texture coordinate data (u, v).
Further, the texture engine circuit 143 outputs a read request including the generated texture coordinate data (u, v) to the SRAM 148 or the DRAM 147 via the memory I / F circuit 144 and passes through the memory I / F circuit 144. By reading the texture data stored in the SRAM 148 or the texture buffer 147a, (R, G, B) data S148 stored at the texture address corresponding to the (s, t) data is obtained.
Here, the texture data stored in the texture buffer 147a as described above is stored in the SRA 5148.
The texture engine circuit 143 reads the (R, G, B) data of the read (R, G, B) data S148 and the (R, G, B) data included in the DDA data S142 from the triangle DDA circuit 142 in the previous stage. And color data S143 is generated.
The texture engine circuit 143 outputs the color data S143 to the memory I / F circuit 144.
[0046]
The texture buffer 147a stores texture data corresponding to a plurality of reduction ratios such as MIPMAP (multiple resolution texture). Here, which reduction rate of texture data is used is determined in units of triangles using a predetermined algorithm.
[0047]
The texture engine circuit 143 directly uses the (R, G, B) data read from the texture buffer 147a in the case of the full color system.
On the other hand, in the case of the index color method, the text 7 engine circuit 143 reads a color lookup table (CLUT) created in advance from the texture CLUT buffer 147d, transfers and stores it in the built-in SRAM, and this color lookup table. Is used to obtain (R, G, B) data corresponding to the color index read from the texture buffer 147a.
[0048]
Memory I / F circuit 144
The memory I / F circuit 144 compares the z data corresponding to the color data S143 input from the texture engine circuit 143 with the z data stored in the z buffer 147c, and is rendered by the input color data S143. It is determined whether or not the image is positioned on the near side (viewpoint side) with respect to the previous image written in the display buffer 147b. If the image is positioned on the near side, the z buffer 147c is used with z data corresponding to the image data S143. The z data stored in is updated.
Further, the memory I / F circuit 144 writes (R, G, B) data to the display buffer 147b.
Further, when the memory I / F circuit 144 receives a read request including the generated texture coordinate data (u, v) from the texture engine circuit 143 to the SRAM 148, the memory I / F circuit 144 stores the (R, G) in the SRAM 148. , B) Read data S148.
Further, when the memory I / F circuit 144 receives a request to read display data from the CRT control circuit 145, the memory I / F circuit 144 displays a certain amount of data from the display buffer 147b, for example, in units of 8 pixels or 16 pixels in response to the request. Read data.
[0049]
Further, the memory I / F circuit 144 outputs the data read from the DRAM 147 to the host interface circuit 149.
[0050]
The memory I / F circuit 144 accesses (writes or reads) the DRAM 147 and the SRAM 148, but the write path and the read path are configured as separate paths.
That is, in the case of writing, the write address ADRW and the color data cdw are processed by the writing system circuit and written to the DRAM 147, and in the case of reading, the reading address circuit is processed and read from the DRAM 147 or the SRAM 148.
Then, the memory I / F circuit 144 performs access to the DRAM 147 based on a predetermined interleave type addressing, writing in units of 16 pixels, and reading out in units of 8 pixels.
[0051]
A specific configuration example of the memory I / F circuit 144 will be described below with reference to FIG.
[0052]
As shown in FIG. 2, the memory I / F circuit 144 includes a distributor 300, address converters 310, 320, 330, and 340, memory controllers 350, 360, 370, and 380, a read controller 390, and a data rearrangement unit 400.
[0053]
[Distributor 300]
At the time of writing, the distributor 300 outputs the color data cwd and write address ADRW for 8 pixels from the data rearrangement unit 400.₂And color data for 16 pixels are generated from these, and then divided into four image data S301, S302, S303, and S304 each consisting of data for four pixels, and the image data and the write address are respectively converted into address converters. It outputs to 310,320,330,340.
Here, (R, G, B) data for one pixel is 32 bits each, and z data is 32 bits.
[0054]
[Address converters 310, 320, 330, 340]
Address converters 310, 320, 330, and 340 use addresses corresponding to (R, G, B) data and z data input from distributor 300 at the time of writing as addresses in memory modules 200, 210, 220, and 230, respectively. The converted addresses S310, S320, S330, and S340 and the divided image data are output to the memory controllers 350, 360, 370, and 380, respectively.
[0055]
FIG. 7 is a diagram schematically showing image data processing (pixel processing) of the distributor 300.
This figure corresponds to FIG. 3 to FIG. 6 described above, and the distributor 300 performs image data processing so that data for 16 pixels of 2 × 8 pixels, for example, can be simultaneously accessed to the DRAM 147. Do.
Then, in the read operation, access is made in an area of a boundary that is a multiple of 8 in the X direction and in an arbitrary boundary in the Y direction, and in the write operation, access is made in an even boundary area in the X direction and the Y direction. The image data is processed so as to achieve proper addressing.
As a result, the DRAM 147 does not start with the memory cell numbers MCN “1”, “2”, “3”, but always has the memory cell number MCN “0”, thereby preventing the occurrence of page violation.
The distributor 300 processes the color data for each of the DRAM modules 220 to 230 so that the adjacent portions in the display area are arranged in different DRAM modules.
As a result, when a plane such as a triangle is drawn, processing can be performed simultaneously on the plane, so that the operation probability of each DRAM module is very high.
[0056]
[Memory controllers 350, 360, 370, 380]
The memory controllers 350, 360, 370, and 380 include write system wiring groups 401W, 402W, 411W, 412W, 421W, 422W, 431W, and 432W, and read system wiring groups 401R, 402R, 411R, 412R, 421R, 422R, and 431R, respectively. , 432R to the memory modules 200, 210, 220, and 230, and controls access to the memory modules 200, 210, 220, and 230 during writing and reading.
[0057]
Specifically, at the time of writing, the memory controllers 350, 360, 370, 380 are output from the distributor 300, and (R, G, B) data for four pixels input from the address converters 350, 360, 370, 380 and The z data is simultaneously written into the memory modules 200, 210, 220, and 230 via the write system wiring groups 401W, 402W, 411W, 412W, 421W, 422W, 431W, and 432W.
At this time, for example, in the memory module 200, as described above, (R, G, B) data and z data for one pixel are stored in each of the banks 201A, 201B, 202A, 202B. The same applies to the memory modules 210, 220, and 230.
[0058]
Further, each of the memory controllers 350, 360, 370, and 380 actively outputs an idle signal S350, S360, S370, and S380 to the controller 390 when its state machine is in a so-called idle (IDLE) state. In response to the read address and read request signal S391 from the read controller 390 in response to the idle signals S350, S360, S370, and S380, the data is transmitted via the read system wiring groups 401R, 402R, 411R, 412R, 421R, 422R, 431R, and 432R. Are read out and output to the read controller 390 via the read system wiring groups 351, 361, 371, 381 and the wiring group 440.
[0059]
In the present embodiment, the number of wirings of the write system wiring groups 401W, 402W, 411W, 412W, 421W, 422W, 431W, 432W and the read system wiring groups 401R, 402R, 411R, 412R, 421R, 422R, 431R, 432R. Is 128 (128 bits), the number of wirings of the readout system wiring groups 351, 361, 371, 381 is 256 (256 bits), and the number of wirings of the readout system wiring group 440 is 1024 (1024 bits).
[0060]
[Read controller 390]
The read controller 390 includes an address converter 391 and a data operation processing unit 392.
The address converter 391 reads the read address ADRR₂If the idle signals S350, S360, S370, and S380 from the memory controllers 350, 360, 370, and 380 are all received in an active state, in response to the idle signals S350, S360, S370, and S380, the unit is 8 pixels. The read address and the read request signal S391 are output to each of the memory controllers 350, 360, 370, and 380 so that the data is read at the same time.
The data operation unit 392 is a texture data (R, G, B) in units of 8 pixels or 16 pixels read by each of the memory controllers 350, 360, 370, 380 in response to the read address and the read request signal S391. Data, z data, and texture color look-up table data are input via the wiring group 440, subjected to predetermined arithmetic processing, and after data rearrangement by the data rearrangement unit 400, a request destination, for example, a host interface circuit To 149.
Even when data is read from one rectangular area in the storage area of the DRAM 147 and written to another rectangular area, data transfer is performed via the read controller 390, the data rearrangement unit 400, and the distributor 300.
[0061]
As described above, the read controller 390 outputs the read address and the read request signal S391 to the memory controllers 350, 360, 370, and 380 when all of the memory controllers 350, 360, 370, and 380 are in the idle state. Since the read data is received, the read data can be synchronized.
Therefore, the read controller 390 does not need to have a holding circuit such as a FIFO (First In First Out) circuit that temporarily holds data, and the circuit scale is reduced.
[0062]
[Data rearrangement unit 400]
The data rearrangement unit 400 has three transfer modes of “Host to Local”, “Local to Host”, and “Local to Local”, and rearranges data according to each transfer mode.
For example, the data rearrangement unit 400 determines the transfer mode and the data length of one pixel based on the transfer mode instruction signal and the pixel data length instruction signal input from the texture engine circuit 143.
The data length of one pixel includes, for example, 32 bits, 16 bits, 8 bits, and 4 bits.
Here, the data length of 32 bits and 16 bits per pixel is used for texture data, and the data length of 8 bits and 4 bits per pixel is used for index data.
[0063]
In the “Host to Local” transfer mode, the data rearrangement unit 400 transfers the 64-bit width transfer data hwd (color data S143) input from the texture engine circuit 143 and the write address ADRW.₁And an address ADRW including 256-bit width color data cwd, representative point coordinates (btrx, btry) and a valid flag btrvld.₂Are output to the distributor 300.
[0064]
In addition, the data rearrangement unit 400 receives the address ADRR input from the host interface circuit 149 in the “Local to Host” transfer mode.₁Address ADRR according to₂And the address ADRR₂By the reading operation performed based on the above, the 256-bit color data mc_0dtr to mc_7dtr for 8 pixels input from the controller 390 are packed to generate the transfer data hrd having a 64-bit width, which is transferred to the host interface circuit 149. Output.
[0065]
Furthermore, in the “Local to Local” transfer mode, the data rearrangement unit 400 reads data from one rectangular storage area in the storage area of the DRAM 147 and transfers (writes) data to another rectangular storage area. 390-bit 256-bit color data mc_0dtr to mc_7dtr is input from 390 and shifted according to the start point coordinates of the rectangular storage area of the transfer destination to generate color data cwd. The color data cwd and the transfer destination Address ADRW including representative point coordinates (btrx, btry) and valid flag btrvld used in the write operation₂Are output to the distributor 300.
[0066]
FIG. 8 is a configuration diagram of the data rearrangement unit 400 shown in FIG.
As shown in FIG. 8, the data rearrangement unit 400 includes a FIFO (First In First Out) circuit 410.₁~ 410_Three, Latch circuit 410_FourIt has a data rearrangement circuit 420, an address generation unit 430, and a controller 440.
[0067]
FIFO circuit 410₁Is a FIFO circuit having a 64-bit width and a three-stage depth, and a control signal S440 from the controller 440, for example.₁The 64-bit width transfer data hwd input from the texture engine circuit 143 is output to the data rearrangement circuit 420 by the FIFO method.
FIFO circuit 410₂Is a FIFO circuit having a 64-bit width and a three-stage depth, and a control signal S440 from the controller 440, for example.₂The 64-bit width transfer data hrd input from the data rearrangement circuit 420 is output to the host interface circuit 149 by the FIFO method.
[0068]
FIFO circuit 410_ThreeIs, for example, a 64-bit FIFO circuit having a depth of 16 stages, and a control signal S440 from the controller 440._ThreeBased on the above, color data mc_0dtr to mc_7dtr for 8 pixels each having 32 bits input from the read controller 390 are output to the data rearrangement circuit 420 by the FIFO method.
FIFO circuit 410_ThreeThe color data mc_0dtr to mc_7dtr are input to the controller 440 as will be described later.
Latch circuit 410_FourIs, for example, 256 bits wide and a control signal S440 from the controller 440._FourThe color data cwd_a having a 256-bit width input from the data rearrangement circuit 420 is output to the distributor 300 shown in FIG.
[0069]
FIG. 9 is a configuration diagram of the data rearrangement circuit 420 shown in FIG.
As shown in FIG. 9, the data rearrangement circuit 420 includes a shift circuit 450.₁, 450₂, Selector 451₀~ 451₇, Latch circuit 452₀~ 452₇And latch circuit 453₀~ 453₇Have
[0070]
Shift circuit 450₁The control signal S440 input from the controller 440 when performing “Host to Local” transfer.₆Based on the above, the 64-bit wide transfer data hwd, that is, 256 bits (8 pixels) of color data in which 2 pixels of color data are arranged on the LSB side is input from the texture engine circuit 143, and the color data is 32 bits. The 256-bit color data mc_0a to mc_7a generated by the shift processing in units is output.
[0071]
Selector 451₀~ 451₇Control signal S440₆Are selected from among the 32-bit color data mc_0a to mc_7a input via the input terminal a and the 32-bit color data mc_0a to mc_7a input via the input terminal b, respectively, and the output terminal a to latch circuit 452₀~ 452₇Output to.
[0072]
Latch circuit 452₀~ 452₇Control signal S440₆Based on the selectors 451 respectively.₀~ 451₇The 32-bit color data output from the output terminal c of the FIFO circuit 410 shown in FIG.₂Alternatively, the latch circuit 410_FourOutput to.
Latch circuit 453₀~ 453₇Control signal S440₆Based on the shift circuit 450 respectively.₁The color data mc_0a to mc_7a input from the shift circuit 450₂And the FIFO circuit 410 shown in FIG.₂Alternatively, the latch circuit 410_FourOutput to.
Here, the latch circuit 452₀~ 452₇Each of the 32-bit output lines is connected to a latch circuit 453, respectively.₀~ 453₇Are the same as the 32-bit output lines of the latch circuit 452 respectively.₀~ 452₇And latch circuit 453₀~ 453₇Control is performed so that 32-bit color data is output from one of the two.
[0073]
Shift circuit 450₂The latch circuit 453₀~ 453₇Each of the 32-bit color data mc_0b to mc_7b generated by shifting the 256-bit color data input from is shifted in units of 32 bits.₀~ 451₇To the input terminal b.
[0074]
Hereinafter, the data rearrangement process in the data rearrangement unit 400 shown in FIG. 8 and the address generation process in the address generation unit 430 will be described separately for each transfer mode.
As described above, writing to the DRAM 147 is performed in units of a rectangular storage area in which 16 pixels of color data of 8 pixels (X direction) × 2 pixels (Y direction) with reference to the representative point are stored. As done. On the other hand, reading from the DRAM 147 is performed in units of rectangular storage areas in which 8 color data of 8 pixels (X direction) × 1 pixel (Y direction) with reference to the representative point are stored.
[0075]
As the representative point address, an arbitrary address can be specified in the Y direction in the read operation, but only an address that is a multiple of 8 in the X direction can be specified. On the other hand, the address of the representative point can specify only an address that is a multiple of 2 in both the X direction and the Y direction in the write operation.
[0076]
In the rearrangement processing shown below, the color for the rectangular storage area 500 having the width “3”, the height “3”, and the origin coordinates (5, 5) as shown in FIG. A case involving data reading and writing operations will be described as an example.
[0077]
<"Host to Local" transfer mode>
Hereinafter, the 64-bit width transfer data hwd shown in FIG. 11 input from the texture engine circuit 143₀~ Hwd_FourColor data pix of 32 bits each (data length of one pixel is 32 bits)₀~ Pix₈Will be described in the case of transferring and writing to the rectangular storage area 500 shown in FIG.
[0078]
In this case, the transfer data hwd is given from the texture engine circuit 143 in the order in which the X coordinate increases in the scan line direction from the upper left end point in FIG. 10 of the rectangular storage area 500 in FIG. Line data is given in the order of increasing X-coordinates. At this time, data having different scan lines is also packed in 64 bits.
[0079]
First, the 64-bit-wide transfer data hwd shown in FIG. 11 from the texture engine circuit 143 is shown.₀~ Hwd_FourIs the FIFO circuit 410 shown in FIG.₁And sequentially output to the data rearrangement circuit 420.
And transfer data hwd₀Color data pix₀, Pix₁Are arranged in order from the LSB, and the 256-bit color data is converted into the shift circuit 450 shown in FIG.₁As shown in FIG. 12A, the color data cwd_a after the shift is shifted by “1” toward the MSB in units of 32 bits.₀Is the latch circuit 453 shown in FIG.₀~ 453₇To color data cwd shown in FIG.₀Output as color data cwd₀Is a latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
[0080]
Next, transfer data hwd₁Color data pix₂, Pix_ThreeAre arranged in order from the LSB, and the 256-bit color data is converted into the shift circuit 450 shown in FIG.₁In FIG. 12B, as shown in FIG. 12B, the color data cwd_a is shifted by “1” toward the MSB in units of 32 bits.₁Is the latch circuit 453 shown in FIG.₀~ 453₇To color data cwd shown in FIG.₁Output as color data cwd₁Is a latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
In addition, the color data cwd_a after the shift₁The selector 451₀~ 451₇And latch circuit 452₀~ 452₇Via the shift circuit 450₂Is output.
[0081]
Next, transfer data hwd₂Color data pix_Four, Pix_FiveAre arranged in order from the LSB, and the 256-bit color data is converted into the shift circuit 450 shown in FIG.₁12C, after being shifted by “2” toward the MSB in 32-bit units, the selector 451 shown in FIG.₀~ 451₇Through the latch circuit 453₀~ 453₇Is remembered.
At this time, the color data pix_Four, Pix_FiveHowever, the latch circuit 453₂453_ThreeIs remembered.
Further, the shift circuit 450 described above is used.₂The 256-bit color data cwd_a shown in FIG.₁Is shifted by “1” toward the LSB, then the selector 451₀~ 451₇Via the latch circuit 452₀~ 452₇Is remembered. At this time, the color data pix_ThreeIs latch circuit 452₁Is remembered.
Then, the latch circuit 453₀, 452₁453₂~ 453₇As shown in FIG. 13C, the color data pix_Three, Pix_Four, Pix_FiveColor data cwd including₂Is the latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
[0082]
Next, transfer data hwd_ThreeColor data pix₆, Pix₇Are arranged in order from the LSB, and the 256-bit color data is converted into the shift circuit 450 shown in FIG.₁As shown in FIG. 12D, the color data cwd_a after the shift is shifted by “1” toward the MSB in units of 32 bits._ThreeIs the latch circuit 453 shown in FIG.₀~ 453₇To color data cwd shown in FIG._ThreeOutput as color data cwd_ThreeIs a latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
[0083]
Next, transfer data hwd_FourColor data pix₈Are arranged in order from the LSB, and the 256-bit color data is converted into the shift circuit 450 shown in FIG.₁In FIG. 12E, as shown in FIG. 12E, the color data cwd_a is shifted by “1” toward the MSB in units of 32 bits._FourIs the latch circuit 453 shown in FIG.₀~ 453₇To color data cwd shown in FIG._FourOutput as color data cwd_FourIs a latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
[0084]
In the data rearrangement unit 400, the color data cwd described above.₀~ Cwd_FourThe address generation unit 430 generates the representative point coordinates (btrx, btry) and the valid flag btrvld corresponding to.
The representative point coordinates (btrx, btry) and the valid flag btrvld are stored in the address ADRW.₂And output to the distributor 300.
Here, the color data cwd₀~ Cwd_FourThe representative point coordinates corresponding to (btrx₀, Btry₀) To (btrx_Four, Btry_Four) And the valid flag is btrvld₀~ Btrvld_FourAnd
The address ADRW in the address generator 430₂Is generated by the control signal S440 from the controller 440 shown in FIG._FiveBased on.
[0085]
The valid flag btrvld is composed of 16 bits and, as described above, writes to the DRAM 147 in units of a rectangular storage area (256 × 2 bits) of 16 pixels of 8 pixels (X direction) × 2 pixels (Y direction). Whether or not to rewrite the stored contents when performing is shown for each storage area of each pixel. Each bit of the valid flag btr coordinate vld corresponds to a storage area for one pixel in a storage area for 16 pixels which is a unit for writing. Each bit of the valid flag btrvld indicates a logical value “0” when the corresponding storage area is masked (when rewriting is not performed), and indicates a logical value “1” when not masked (when rewriting). Yes.
[0086]
Here, since the color data used in one writing is 256-bit data for 8 pixels, the color data is written in a storage area where the address in the Y direction is an even number among the storage areas to be written. When the upper 8 bits of the 16-bit valid flag btrvld are set to the logical value “0” and the address in the Y direction is written to the odd-numbered storage area among the storage areas to be written, the 16-bit valid flag btrvld The lower 8 bits are set to the logical value “0”.
[0087]
As shown in FIGS. 13A to 13E, the color data cwd₀~ Cwd_FourFor each of the valid flags btrvld₀~ Btrvld_FourIs generated by the address generation unit 430.
Here, for example, a valid flag btrvld₀Since the lower 8 bits are the logical value “0”, the color data cwd₀Is written in a memory area where the address in the Y direction is odd among the memory areas to be written.
[0088]
Further, the address generation unit 430 performs color data cwd shown in FIGS.₀~ Cwd_FourAre written in the DRAM 147 in units of a rectangular storage area of 16 pixels of 8 pixels (X direction) × 2 pixels (Y direction) described above, as shown in FIG. btrx₀, Btry₀) To (btrx_Four, Btry_Four) Respectively.
[0089]
As described above, the color data cwd shown in FIGS. 13A to 13E generated by the data rearrangement circuit 420 shown in FIG. 8 of the data rearrangement unit 400 shown in FIG.₀~ Cwd_FourIs the address ADRW generated in the address generation unit 430 shown in FIG.₂Representative point coordinates (btrx) shown in FIGS.₀, Btry₀) To (btrx_Four, Btry_Four) And the valid flag to btrvld₀~ Btrvld_Four2 is processed by the distributor 300, the address converter 310, and the memory controllers 350, 360, 370, and 380 shown in FIG. 2, and written in the DRAM 147 shown in FIG.
[0090]
First, as shown in FIG. 14A, the representative point coordinates (btrx₀, Btry₀) = (4, 4), the storage area of the rectangular storage area 500 corresponding to 16 pixels has a Y coordinate of “5” (odd number), and the X coordinate in the storage area has the smallest address in FIG. Color data cwd shown in A)₀Color data cwd so that the LSB of₀Is written. At this time, as described above, the valid flag btrvld₀When the lower 8 bits of all are logical values “0”, the color data cwd is stored in the storage area where the Y coordinate is an odd number.₀Is written and the valid flag btrvld₀When the upper 8 bits of all are logical values “0”, the color data cwd is stored in the storage area where the Y coordinate is an even number.₀Is written.
Also, color data cwd₀Is written as a valid flag btrvld.₀Of these, only the 32-bit storage area corresponding to the bit of the logical value “1” is performed. In this case, the valid flag btrvld₀Is “0x0600”, that is, “000001000000000”, and the bits of the logical value “1” correspond to the storage areas with the (X, Y) coordinates of (5, 5) and (6, 5). As shown in FIG. 14A, color data pix only for the storage area.₀, Pix₁Is written.
[0091]
Next, as shown in FIG. 14B, the representative point coordinates (btrx₁, Btry₁) = (6, 4) The color data cwd shown in FIG. 13B is stored in the storage area where the Y coordinate in the rectangular storage area 500 for 16 pixels is “5” (odd number).₁Color data pix₂Is written. The writing method is the same as that shown in FIG.
[0092]
Next, similarly, as shown in FIG. 14C, the representative point coordinates (btrx₂, Btry₂) = (4, 6) The color data cwd shown in FIG. 13C is stored in the storage area where the Y coordinate in the rectangular storage area 500 for 16 pixels is “6” (even number).₂Color data pix_Three, Pix_Four, Pix_FiveIs written.
[0093]
Next, similarly, as shown in FIG. 15D, the representative point coordinates (btrx_Three, Btry_Three) = (4, 6), the color data cwd shown in FIG. 13D is stored in the storage area where the Y coordinate in the rectangular storage area 500 for 16 pixels is “7” (odd number)._ThreeColor data pix₆, Pix₇Is written.
[0094]
Next, similarly, as shown in FIG. 15 (F), the representative point coordinates (btrx_Four, Btry_Four) = (6, 6), the color data cwd shown in FIG. 13E is stored in the storage area where the Y coordinate in the rectangular storage area 500 for 16 pixels is “7” (odd number)._FourColor data pix₈Is written.
[0095]
As a result, the 64-bit width transfer data hwd shown in FIG. 11 input from the texture engine circuit 143 is obtained.₀~ Hwd_FourColor data pix for 8 pixels of 32 bits each₀~ Pix₈Is written in the rectangular storage area 500 shown in FIG.
[0096]
In the example shown in FIGS. 11 to 15 described above, color data pix having a data length of one pixel of 32 bits.₀~ Pix₈Is input from the texture engine circuit 143 and written in the rectangular storage area 500 shown in FIG. 10, but the data length of one pixel is arbitrary, and may be 16 bits, for example.
Hereinafter, color data pix having a data length of 16 bits per pixel₀~ Pix₈Is input from the texture engine circuit 143 and written into the rectangular storage area 500 shown in FIG.
In this case, as shown in FIG. 16, the 64-bit width transfer data input from the texture engine circuit 143 includes color data for 4 pixels each.
In this case, the transfer data hwd shown in FIG.₀Color data pix included in₀~ Pix_Three256-bit color data including the shift data 450 of the data rearrangement circuit 420 shown in FIG.₁Is shifted by 1 bit in the MSB direction, and 256-bit color data cwd_a shown in FIG.₀Is generated and the color data cwd_a₀In response to the latch circuit 453₀~ 453₇To color data cwd shown in FIG.₀Is output to the distributor 300.
Next, the transfer data hwd shown in FIG.₁Color data pix included in_Four~ Pix₇256-bit color data including the shift data 450 of the data rearrangement circuit 420 shown in FIG.₁Is shifted by 2 bits in the MSB direction, and the transfer data hwd₀Color data pix included in_ThreeTo add 256-bit color data cwd_a shown in FIG.₁Is generated and the color data cwd_a₁Color data cwd shown in FIG.₁Is output to the distributor 300.
[0097]
Next, the transfer data hwd shown in FIG.₂Color data pix included in₈256-bit color data including the shift data 450 of the data rearrangement circuit 420 shown in FIG.₁Is shifted by 3 bits in the MSB direction, and the transfer data hwd₁Color data pix included in₆, Pix₇To add 256-bit color data cwd_a shown in FIG.₂Is generated and the color data cwd_a₂Color data cwd shown in FIG.₂Is output to the distributor 300.
[0098]
Also, color data cwd₀~ Cwd_ThreeThe representative point coordinates (btrx, btry) and the valid flag btrvld shown in FIGS. 18A to 18C corresponding to are generated in the address generation unit 430.
[0099]
As a result, the 64-bit width transfer data hwd shown in FIG. 16 input from the texture engine circuit 143 is obtained.₀~ Hwd₂Color data pix for 8 pixels of 16 bits each₀~ Pix₈Is written in the rectangular storage area 500 shown in FIG.
[0100]
<"Local to Host" transfer mode>
Hereinafter, an operation when the color data mc_0dtr to mc_7dtr for 8 pixels read from the DRAM 147 is packed into 64-bit transfer data and output to the host interface circuit 149 will be described.
As described above, reading from the DRAM 147 is performed in units of rectangular storage areas in which 8 color data of 8 pixels (X direction) × 1 pixel (Y direction) with reference to the representative point are stored. . As the representative point address, an arbitrary address can be specified in the Y direction in the read operation, but only an address that is a multiple of 8 in the X direction can be specified.
[0101]
In this case, the address generation unit 430 of the data rearrangement unit 400 illustrated in FIG. 8 receives the address ADRR input from the host interface circuit 149.₁For the rectangular storage area to be read out in the DRAM 147, the representative point coordinates (sbx, sby) in the order in which the X coordinate increases in the scan line direction from the upper left end in FIG. 19 in the X, Y coordinate system shown in FIG. Is calculated. Then, an address ADRR including the representative point coordinates (sbx, sby)₂Is output to the distributor 300.
[0102]
In addition, the data rearrangement circuit 420 has an address ADRR.₂The color data mc_0dtr to mc_7dtr for 8 pixels read from the DRAM 147 in response to the read data from the read controller 390 shown in FIG. 2 to the FIFO circuit 410 of the data rearrangement unit 400 shown in FIG._ThreeIs packed into 64-bit wide transfer data hwd, and the FIFO circuit 410₁To the host interface circuit 149.
At this time, the color data mc_0dtr is the color data stored in the representative point coordinates of the rectangular storage area to be read out, and the color data mc_7dtr is stored in the coordinates having the maximum X coordinate in the rectangular storage area. Color data.
[0103]
Hereinafter, as an example, the color data pix read from the rectangular storage area 600 in the DRAM 147 with the width 13, the height 3, and the starting point coordinates (5, 5) shown in FIG.₀~ Pix₃₈Will be described when the data is packed into 64-bit width transfer data including color data for four pixels and output to the host interface circuit 149.
[0104]
First, the address ADRR is sent to the address generator 430.₁Is input from the host interface circuit 149.
Then, in order to perform the read operation of the rectangular storage area 600 shown in FIG. 19 in the address generation unit 430, the representative point coordinates (sbx shown in FIGS. 19 and 20A to 20I) are used.₀, Sby₀) To (sbx₈, Sby₈) Including address ADRR₂2 are sequentially output to the distributor 300 shown in FIG. 2, and after the processing of the address converter 310 and the memory controllers 350, 360, 370, and 380, the color data mc_0dtr to mc_7dtr shown in FIGS. The read circuit 390 shown in FIG. 2 to the FIFO circuit 410 of the data rearrangement unit 400 shown in FIG._ThreeAre sequentially output to the data rearrangement circuit 420.
[0105]
In the data rearrangement circuit 420 shown in FIG. 9, the control signal S440 from the controller 440 is displayed.₆Based on the above, the following processing is performed.
Note that the shift circuit 450 shown in FIG.₁The selector 451₀~ 451₇A 256 (= 32 × 8) -bit output storage area connected to the input terminal “a”, and a 224 (= 32 × 8) -bit internal storage area located in the upper bits of the output storage area. The shift operation is performed in units of 32 bits.
[0106]
First, 256-bit color data mc_0dtr to mc_7dtr shown in FIG.₁Color data pix after being shifted by “5” toward the LSB in units of 32 bits.₀~ Pix₂However, the latch circuit 453₀~ 453₂Is remembered.
[0107]
Next, 256-bit color data mc_0dtr to mc_7dtr shown in FIG.₁The color data pix that is input to the MSB, shifted by “3” toward the MSB in units of 32 bits, and stored in the output storage area_Three~ Pix₇Is selector 451_Three~ 451₇Through the input terminal a and the output terminal c of the latch circuit 452._Three~ 452₇Is written to.
As a result, the latch circuit 453₀~ 453₇The color data cwd_b shown in the portion corresponding to the output storage area in FIG.₁Is memorized.
[0108]
Then, the latch circuit 453₀~ 453_ThreeColor data pix stored in₀~ Pix_ThreeIs the 64-bit transfer data hrd shown in FIG.₀As shown in FIG.₂Is output to the host interface circuit 149.
Next, the latch circuit 453_Four~ 453₇Color data pix stored in_Four~ Pix₇Is the 64-bit transfer data hrd shown in FIG.₁As shown in FIG.₂Is output to the host interface circuit 149.
[0109]
The shift circuit 450₁Color data pix stored in the internal storage area₈~ Pix_TenIs shifted by “8” toward the LSB in units of 32 bits, and then the latch circuit 453₀~ 453₂Is written to.
Next, the color data mc_0dtr to mc_7dtr shown in FIG.₁Is output after being shifted by “3” toward the MSB in units of 32 bits, and the color data pix₁₁, Pix₁₂However, the latch circuit 453_Three453_FourIs written to.
Next, the color data mc_0dtr to mc_7dtr shown in FIG.₁Is output without being shifted, and the color data pix₁₃~ Pix₁₅However, the latch circuit 453_Five~ 453₇Is written to.
As a result, the latch circuit 453₀~ 453₇The color data cwd_b shown in the portion corresponding to the output storage area in FIG._ThreeIs memorized.
[0110]
Then, the latch circuit 453₀~ 453_ThreeColor data pix stored in₈~ Pix₁₁Is the 64-bit transfer data hrd shown in FIG.₂As shown in FIG.₂Is output to the host interface circuit 149.
Next, the latch circuit 453_Four~ 453₇Color data pix stored in₁₂~ Pix₁₅Is the 64-bit transfer data hrd shown in FIG._ThreeAs shown in FIG.₂Is output to the host interface circuit 149.
[0111]
Next, 256-bit color data mc_0dtr to mc_7dtr shown in FIG.₁Is output without being shifted, and the color data pix₁₆~ Pix_{twenty three}However, the latch circuit 453₀~ 453₇Is written to.
As a result, the latch circuit 453₀~ 453₇The color data cwd_b shown in the portion corresponding to the output storage area in FIG._FourIs memorized.
[0112]
Then, the latch circuit 453₀~ 453_ThreeColor data pix stored in₁₆~ Pix₁₉Is the 64-bit transfer data hrd shown in FIG._FourAs shown in FIG.₂Is output to the host interface circuit 149.
Next, the latch circuit 453_Four~ 453₇Color data pix stored in₂₀~ Pix_{twenty three}Is the 64-bit transfer data hrd shown in FIG._FiveAs shown in FIG.₂Is output to the host interface circuit 149.
[0113]
Next, 256-bit color data mc_0dtr to mc_7dtr shown in FIG.₁Color data pix without being shifted to_{twenty four}, Pix_{twenty five}However, the latch circuit 453₀453₁Is remembered.
[0114]
Next, 256-bit color data mc_0dtr to mc_7dtr shown in FIG.₁Is shifted by “3” toward the LSB in units of 32 bits, and the color data pix₂₆~ Pix₂₈However, the latch circuit 453₂~ 453_FourIs written to.
[0115]
Next, 256-bit color data mc_0dtr to mc_7dtr shown in FIG.₁Is shifted by “5” toward the MSB in units of 32 bits, and the color data pix₂₉~ Pix₃₆Color data pix stored in the output storage area₂₉~ Pix₃₁However, the latch circuit 453_Five~ 453₇Written to
As a result, the latch circuit 453₀˜453, the color data cwd_b shown in the portion corresponding to the output storage area of FIG.₇Is memorized.
[0116]
Then, the latch circuit 453₀~ 453_ThreeColor data pix stored in_{twenty four}~ Pix₂₇Is the 64-bit transfer data hrd shown in FIG.₆As shown in FIG.₂Is output to the host interface circuit 149.
Next, the latch circuit 453_Four~ 453₇Color data pix stored in₂₈~ Pix₃₁Is the 64-bit transfer data hrd shown in FIG.₇As shown in FIG.₂Is output to the host interface circuit 149.
[0117]
Next, the shift circuit 450₁Color data pix stored in the internal storage area₃₂~ Pix₃₆Is shifted by “8” toward the LSB in units of 32 bits, and then the latch circuit 453₀~ 453_FourIs written to.
Next, the color data mc_0dtr to mc_7dtr shown in FIG.₁Is output after being shifted by “5” toward the MSB in units of 32 bits, and the color data pix₃₇, Pix₃₈However, the latch circuit 453_Five453₆Is written to.
As a result, the latch circuit 453₀~ 453₆In addition, the color data cwd_b shown in FIG.₈Is memorized.
[0118]
Then, the latch circuit 453₀~ 453_ThreeColor data pix stored in₃₂~ Pix₃₅Is the 64-bit transfer data hrd shown in FIG.₈As shown in FIG.₂Is output to the host interface circuit 149.
Next, the latch circuit 453_Four~ 453₆Color data pix stored in₃₆~ Pix₃₈Is the 64-bit transfer data hrd shown in FIG.₉As shown in FIG.₂Is output to the host interface circuit 149.
[0119]
As described above, the color data pix read from the rectangular storage area 600 in the DRAM 147 having the width 13, the height 3, and the starting point coordinates (5, 5) shown in FIG.₀~ Pix₃₈However, as shown in FIG. 22, 64-bit wide transfer data hrd including color data for four pixels.₀~ Hrd₉And output to the host interface circuit 149.
[0120]
<"Local to Local" transfer mode>
In this case, the 8-pixel color data mc_0dtr to mc_7dtr read out from the transfer source storage area in the DRAM 147 with reference to the representative point coordinates (sbx, sby) is stored in the write destination (transfer destination) storage area. A point coordinate (btrx, btry) is written as a reference.
At this time, the calculation order of the representative point coordinates of the rectangular storage area of the transfer source is performed in a designated direction instead of one direction as in the “Host to Local” transfer mode described above.
[0121]
Hereinafter, the representative point coordinates (sbx, sby) are generated in order from right to left and from bottom to top, and the color data pix read out from the rectangular storage area 600 shown in FIG.₀~ Pix₃₈Is described when the representative point coordinates (btrx, btry) are generated and written in order from right to left and from bottom to top in a rectangular storage area having a width of 13 and a height of 3 and an origin coordinate of (10, 7). To do.
[0122]
First, in order to perform the read operation of the rectangular storage area 600 shown in FIG. 19 in the address generator 430 shown in FIG. 8, the representative point coordinates (sbx) shown in FIGS.₈, Sby₈) To (sbx₀, Sby₀) Including address ADRR₂2 are sequentially output to the distributor 300 shown in FIG. 2, and after the processing of the address converter 310 and the memory controllers 350, 360, 370, and 380, the color data mc_0dtr to mc_7dtr shown in FIGS. The read circuit 390 shown in FIG. 2 to the FIFO circuit 410 of the data rearrangement unit 400 shown in FIG._ThreeAre sequentially output to the data rearrangement circuit 420.
That is, the representative point coordinates are output to the distributor 300 in the reverse order compared to the case shown in FIG.
[0123]
Next, in the data rearrangement circuit 420 shown in FIG. 9, the control signal S440 from the controller 440 is obtained.₆Based on the above, the following processing is performed.
First, the color data mc_0dtr to mc_7dtr shown in FIG. 23A is converted into the shift circuit 450 shown in FIG.₁9 is shifted by “1” toward the MSB in units of 32 bits, and the color data after the shift is the latch circuit 453 shown in FIG.₀~ 453₇To color data cwd shown in FIG.₀Output as color data cwd₀Is a latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
[0124]
Next, the color data mc_0dtr to mc_7dtr shown in FIG. 23B is converted into the shift circuit 450 shown in FIG.₁Without being shifted in the latch circuit 453₀~ 453₇Is written to the shift circuit 450.₂Output to the shift circuit 450₂, The data is shifted by “1” toward the LSB in units of 32 bits, and the color data after the shift is the selector 451.₀~ 451₇Through the latch circuit 452₀~ 452₇Is written to.
The latch circuit 452₀~ 452₇To the color data cwd shown in FIG.₁Is the latch circuit 410 shown in FIG._FourIs output to the distributor 300.
[0125]
Next, the latch circuit 453₀~ 453₇The color data shown in FIG. 23B stored in FIG.₂Output to the shift circuit 450₂In FIG. 5, the shift is performed by “3” toward the MSB in units of 32 bits, and the color data after the shift is converted to the selector 451.₀~ 451₇Through the latch circuit 452₀~ 452₇Is written to. Thus, the latch circuit 452_ThreeAnd color data pix₂₉Is memorized.
Further, the color data mc_0dtr to mc_7dtr shown in FIG. 23C is converted into the shift circuit 450 shown in FIG.₁, The latch circuit 453 is shifted by “5” toward the LSB.₀~ 453₇Is written to. As a result, the latch circuit 453₀~ 453₂And color data pix₂₆~ Pix₂₈Is memorized.
The latch circuit 452_Three453₀~ 453₂To color data cwd shown in FIG.₂Is the latch circuit 410 shown in FIG._FourIs output to the distributor 300.
[0126]
Next, the color data mc_0dtr to mc_7dtr shown in FIG. 23D is converted into the shift circuit 450 shown in FIG.₁9 is shifted by “1” toward the MSB in units of 32 bits, and the color data after the shift is the latch circuit 453 shown in FIG.₀~ 453₇To color data cwd shown in FIG._ThreeOutput as color data cwd_ThreeIs a latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
[0127]
Next, the color data mc_0dtr to mc_7dtr shown in FIG. 23E is converted into the shift circuit 450 shown in FIG.₁Without being shifted in the latch circuit 453₀~ 453₇Is written to the shift circuit 450.₂Output to the shift circuit 450₂, The data is shifted by “1” toward the LSB in units of 32 bits, and the color data after the shift is the selector 451.₀~ 451₇Through the latch circuit 452₀~ 452₇Is written to.
The latch circuit 452₀~ 452₇To the color data cwd shown in FIG._FourIs the latch circuit 410 shown in FIG._FourIs output to the distributor 300.
[0128]
Next, the latch circuit 453₀~ 453₇The color data shown in FIG. 23E stored in FIG.₂Output to the shift circuit 450₂In FIG. 5, the shift is performed by “3” toward the MSB in units of 32 bits, and the color data after the shift is converted to the selector 451.₀~ 451₇Through the latch circuit 452₀~ 452₇Is written to. Thus, the latch circuit 452_ThreeAnd color data pix₁₆Is memorized.
Further, the color data mc_0dtr to mc_7dtr shown in FIG. 23F is converted into the shift circuit 450 shown in FIG.₁, The latch circuit 453 is shifted by “5” toward the LSB.₀~ 453₇Is written to. As a result, the latch circuit 453₀~ 453₂And color data pix₁₃~ Pix₁₅Is memorized.
The latch circuit 452_Three453₀~ 453₂To the color data cwd shown in FIG._FiveIs the latch circuit 410 shown in FIG._FourIs output to the distributor 300.
[0129]
Next, the color data mc_0dtr to mc_7dtr shown in FIG. 23G is converted into the shift circuit 450 shown in FIG.₁9 is shifted by “1” toward the MSB in units of 32 bits, and the color data after the shift is the latch circuit 453 shown in FIG.₀~ 453₇To color data cwd shown in FIG.₆Output as color data cwd₆Is a latch circuit 410 shown in FIG._FourTo the distributor 300 shown in FIG.
[0130]
Next, the color data mc_0dtr to mc_7dtr shown in FIG. 23 (H) is converted into the shift circuit 450 shown in FIG.₁Without being shifted in the latch circuit 453₀~ 453₇Is written to the shift circuit 450.₂Output to the shift circuit 450₂, The data is shifted by “1” toward the LSB in units of 32 bits, and the color data after the shift is the selector 451.₀~ 451₇Through the latch circuit 452₀~ 452₇Is written to.
The latch circuit 452₀~ 452₇To color data cwd shown in FIG.₇Is the latch circuit 410 shown in FIG._FourIs output to the distributor 300.
[0131]
Next, the latch circuit 453₀~ 453₇The color data shown in FIG. 23 (H) stored in FIG.₂Output to the shift circuit 450₂In FIG. 5, the shift is performed by “3” toward the MSB in units of 32 bits, and the color data after the shift is converted to the selector 451.₀~ 451₇Through the latch circuit 452₀~ 452₇Is written to. Thus, the latch circuit 452_ThreeAnd color data pix_ThreeIs memorized.
Further, the color data mc_0dtr to mc_7dtr shown in FIG. 23 (I) is converted into the shift circuit 450 shown in FIG.₁, The latch circuit 453 is shifted by “5” toward the LSB.₀~ 453₇Is written to. As a result, the latch circuit 453₀~ 453₂And color data pix₀~ Pix_ThreeIs memorized.
The latch circuit 452_Three453₀~ 453₂To the color data cwd shown in FIG.₈Is the latch circuit 410 shown in FIG._FourIs output to the distributor 300.
[0132]
In the data rearrangement unit 400, as shown in FIGS. 24A to 24I, the color data cwd described above is used.₀~ Cwd₈Representative point coordinates (btrx₀, Btry₀) To (btrx₈, Btry₈) And valid flag btrvld₀~ Btrvld₈Is generated by the address generation unit 430. Representative point coordinates (btrx₀, Btry₀) To (btrx₈, Btry₈) And valid flag btrvld₀~ Btrvld₈Is the address ADRW₂And output to the distributor 300.
[0133]
And the representative point coordinates (btrx₀, Btry₀) To (btrx₈, Btry₈) And valid flag btrvld₀~ Btrvld₈24A, the color data cwd shown in FIGS. 24A to 24I with 16 pixels of 8 pixels (X direction) × 2 pixels (Y direction) as a unit.₀~ Cwd₈Color data pix included in₀~ Pix₃₈Is written into the rectangular storage area 700 in the DRAM 147.
As a result, the color data pix read from the rectangular storage area 600 shown in FIG.₀~ Pix₃₈Are transferred to the rectangular storage area shown in FIG.
The above is the description of the data rearrangement circuit 420 and the address generation unit 430 shown in FIG.
[0134]
The controller 440 controls the control signal S440 based on the transfer mode instruction signal S800 and the pixel data length instruction signal S801 input from the texture engine circuit 143.₁~ S440₆Is generated.
The controller 440 includes a FIFO circuit 410.₁~ 410_ThreeAnd latch circuit 410_FourControl signal S440 for controlling₁~ S440_FourIs output.
Further, the controller 440 controls the control signal S440 that controls the address generation unit 430._FiveIs output.
In addition, the controller 440 controls the data rearrangement circuit 420 to control signal S440.₆Is output.
[0135]
The controller 440 also includes a FIFO circuit 410._ThreeInput, that is, reading of the color data mc_0dtr to mc_7dtr from the DRAM 147 is controlled as follows.
The controller 440 includes a FIFO circuit 410._ThreeAnd when it is detected that half of the 16-bit storage circuits each having a 64-bit width are empty (effective data is not stored), the pixel data length instruction signal S801 is detected. Control signal S440 for instructing to read continuously for the number of times corresponding to_FiveIs output to the address generation unit 430.
Here, 256-bit color data mc_0dtr to mc_7dtr for 8 pixels is read from the DRAM 147 by one reading.
[0136]
Specifically, the controller 440 controls the control signal S440 indicating the number of read instructions shown in FIG. 26 according to the pixel data length indicated by the pixel data length instruction signal S801._FiveIs output to the address generation unit 430.
That is, when the data length of one pixel is 32 bits, 16 bits, 8 bits, and 4 bits, the controller 440 controls the control signal S440 indicating the number of read instructions twice, four times, eight times, and 16 times, respectively._FiveIs output to the address generation unit 430.
[0137]
A control signal S440 indicating the number of read instructions shown in FIG._FiveIs output from the FIFO circuit 410._ThreeThe color data mc_0dtr to mc_7dtr read from the DRAM 147 can be stored in the empty (unstored) 8-stage storage circuit. At this time, the FIFO circuit 410_ThreeThe 8-stage storage circuit stores 8 × 64-bit color data.
[0138]
That is, when the pixel data length is 32 bits, the color data mc_0dtr to mc_7dtr read out in one readout operation includes 32 × 8-bit color data as shown in FIG. Therefore, it is possible to read out 32 × 8 × 2 (= 8 × 64) bits by continuously reading this twice.
Further, when the pixel data length is 16 bits, the color data mc_0dtr to mc_7dtr read out in one readout operation includes 16 × 8-bit color data as shown in FIG. Therefore, it is possible to read out 16 × 8 × 4 (= 8 × 64) bits by continuously reading this four times.
Further, when the pixel data length is 8 bits, the color data mc_0dtr to mc_7dtr read out in one readout operation includes 8 × 8-bit color data as shown in FIG. Therefore, it is possible to read out 8 × 8 × 8 (= 8 × 64) bits by continuously reading this out eight times.
Further, when the pixel data length is 4 bits, the color data mc_0dtr to mc_7dtr read out by one readout operation includes 4 × 8-bit color data as shown in FIG. Therefore, 4 × 8 × 16 (= 8 × 64) bits can be read by continuously reading this 16 times.
[0139]
As described above, the controller 440 includes the FIFO circuit 410._ThreeWhen the half of the 16-bit storage circuits each having a 64-bit width are emptied, the DRAM 147 continuously receives the number of times corresponding to the pixel data length instruction signal S801. By performing the reading, the FIFO circuit 410 by the controller 440 is obtained._ThreeThe control of the controller 440 can be reduced. Therefore, the other processing performed by the controller 440 is the FIFO circuit 410._ThreeIt is possible to avoid a state waiting for the control.
That is, in a general conventional FIFO circuit that is not controlled by the controller 440 as described above, for example, while there is an empty (unstored) storage circuit, reading from the memory is continuously performed, Input control is performed so that reading is stopped when all memory circuits are in a memory state. However, when the control is performed in this way, when the reading is stopped, a state in which only a part of the color data for 8 pixels read by one reading is written in the FIFO circuit occurs. Therefore, next, the same 8-pixel color data as before is read, and among the read 8-pixel color data, the color data of the pixels not previously written in the FIFO circuit is specified, and only the specified color data is FIFO. It is necessary to perform control so as to write to the circuit, and the control becomes complicated.
[0140]
CRT control circuit 145
The CRT control circuit 145 generates an address to be displayed on a CRT (not shown) in synchronization with the applied horizontal and vertical synchronization signals, and outputs a request to read display data from the display buffer 147b to the memory I / F circuit 144. In response to this request, the memory I / F circuit 144 reads the display data from the display buffer 147b in a certain chunk. The CRT controller circuit 145 includes a FIFO circuit that stores display data read from the display buffer 147b, and outputs RGB index values to the RAMDAC circuit 146 at regular time intervals.
[0141]
RAMDAC circuit 146
The RAMDAC circuit 146 stores R, G, B data corresponding to each index value, and digital R, G, B data corresponding to the RGB index value input from the CRT controller circuit 145 is not shown. The data is transferred to a D / A converter (Digital / Analog Converter) to generate analog R, G, B data. The RAMDAC circuit 146 outputs the generated R, G, B data to the CRT.
[0142]
Hereinafter, the overall operation of the above-described three-dimensional computer graphic system 10 will be described.
In the three-dimensional computer graphic system 10, data such as graphic drawing is given to the rendering circuit 14 via the main bus 15 from the main memory 12 of the main processor 11 or the I / O interface circuit 13 that receives graphics data from the outside. It is done.
If necessary, data such as graphics drawing is subjected to geometry processing such as coordinate conversion, clip processing, and lighting processing in the main processor 11 or the like.
The graphics data that has been subjected to the geometry processing includes vertex coordinates x, y, and z of each of the three vertices of the triangle, luminance values R, G, and B, and texture coordinates s, t, and q corresponding to the pixel to be rendered. This is polygon rendering data S11.
[0143]
The polygon rendering data S11 is input to the DDA setup circuit 141 of the rendering circuit 14.
In the DDA setup circuit 141, variation data S141 indicating the difference between the sides of the triangle and the horizontal direction is generated based on the polygon rendering data S11. Specifically, using the starting point value and the ending point value, and the distance between them, a variation that is a change in the value to be obtained when the unit length is moved is calculated, and the variation data The data is output to the triangle DDA circuit 142 as S141.
[0144]
In the triangle DDA circuit 142, linearly interpolated (z, R, G, B, s, t, q) data for each pixel inside the triangle is calculated using the variation data S141.
Then, the calculated (z, R, G, B, s, t, q) data and (x, y) data of each vertex of the triangle are used as DDA data S142 from the triangle DDA circuit 142 to the texture engine. It is output to the circuit 143.
[0145]
In the texture engine circuit 143, for the (s, t, q) data indicated by the DDA data S142, an operation for dividing the s data by the q data and an operation for dividing the t data by the q data are performed. The division results “s / q” and “t / q” are multiplied by the texture sizes USIZE and VSIZE, respectively, to generate texture coordinate data (u, v).
[0146]
Next, the address ADRR is sent from the texture engine circuit 143 to the memory I / F circuit 144.₁Is output, and (R, G, B) data S148 stored in the DRAM 147 (SRAM 148) is read via the memory I / F circuit 144.
[0147]
Next, in the texture engine circuit 143, the (R, G, B) data of the read (R, G, B) data S148 and the DDA data S142 from the triangle DDA circuit 142 in the preceding stage are included (R, G, B). B) The data is multiplied and generated as color data S143.
The color data S143 is output from the texture engine circuit 143 to the memory I / F circuit 144.
[0148]
In the case of full color, the data (R, G, B) from the texture buffer 147a may be used directly. However, in the case of index color, the data of the color index table (Color Index Table) created in advance is used. Then, the texture is transferred from a texture CLUT (Color Look Up Table) buffer 147d to a temporary storage buffer composed of SRAM or the like, and the actual R, G, B colors are obtained from the color index using the CLUT of the temporary storage buffer.
When the CULT is configured by SRAM, when the color index is input to the SRAM address, the actual R, G, B color is output as the output.
[0149]
Then, the memory I / F circuit 144 compares the z data corresponding to the color data S143 input from the texture engine circuit 143 with the z data stored in the z buffer 147c, and uses the input color data S12. It is determined whether or not the image to be drawn is positioned closer to the front (viewpoint side) than the image previously written in the display buffer 21.
As a result of the determination, if it is located on the near side, the z data stored in the z buffer 147c is updated with the z data corresponding to the image data S143.
[0150]
Next, in the memory I / F circuit 144, (R, G, B) data is written to the display buffer 147b.
These data to be written (including update) are memory controllers 350, 360, 370, and 370 via the data rearrangement unit 400, distributor 300, and address decoders 310, 320, 330, and 340 shown in FIG. 380, and written in parallel to a predetermined memory by the memory controllers 350, 360, 370, and 380 via the write system wiring groups 401W, 402W, 411W, 412W, 421W, 422W, 431W, and 432W, respectively.
[0151]
In the memory I / F circuit 144, a memory block storing a texture corresponding to a texture address in a pixel to be drawn is calculated based on the texture address, and a read request is issued only to the memory block. Texture data is read out.
In this case, in the memory block that does not hold the corresponding texture data, access for texture reading is not performed, so that it is possible to provide more access time for drawing.
[0152]
At this time, the process of writing the image data S143 from the texture engine circuit 143 to the DRAM 147 in the memory I / F circuit 144 is, for example, as described above with reference to FIGS. Is converted into color data cwd having a 256-bit width.
[0153]
Similarly, in drawing, the color data is read from the corresponding address to modify write (Modify Write) to the memory block storing the color data corresponding to the pixel address to be drawn. And after modification, it is written back to the same address.
[0154]
When performing hidden surface processing, the depth data is modified and written from the corresponding address to the memory block storing the depth data corresponding to the pixel address to be drawn. It is read to do, and if necessary, after modification, it is written back to the same address.
[0155]
In such data exchange with the DRAM 147 based on the memory I / F circuit 144, drawing performance can be improved by performing a plurality of processes up to that time in parallel.
In particular, the triangle DDA circuit 142 and the texture engine 143 are provided in the same circuit in parallel execution format (spatial parallel), or the pipeline is finely inserted (time parallel) to partially increase the operating frequency. The simultaneous calculation of a plurality of pixels is performed by the means of causing them.
[0156]
The color data is arranged so that adjacent portions in the display area are different DRAM modules under the control of the memory I / F circuit 144.
As a result, when drawing a plane such as a triangle, the plane is processed simultaneously. For this reason, the operation probability of each DRAM module is very high.
[0157]
When an image is displayed on a CRT (not shown), the CRT control circuit 145 generates a display address in synchronization with a given horizontal / vertical synchronization frequency, and requests a display data transfer to the memory I / F circuit 144. Is issued.
In the memory I / F circuit 144, the display data is transferred to the CRT control circuit 145 in a certain set according to the request.
In the CRT control circuit 145, the display data is stored in a display FIFO (First In First Out) or the like (not shown), and the RGB index values are transferred to the RAMDAC 146 at regular intervals.
[0158]
When there is a request for reading data stored in the DRAM 147 or the SRAM 148 from the main processor 11 to the memory I / F circuit 144 via the host interface circuit 149 shown in FIG. Read address ADRR 391₂Is entered.
At this time, the address converter 391 checks whether or not the idle signals S350, S360, S370, and S380 from the memory controllers 350, 360, 370, and 380 are all input in an active state. When the idle signals S350, S360, S370, and S380 are all input in an active state, in response to the idle signals S350, S360, S370, and S380, the readout address and A read request signal S391 is output to each memory controller 350, 360, 370, 380.
[0159]
Upon receipt of the read address and read request signal S391, the texture data, (R, G, B) data, z data, and texture color lookup table in units of 8 pixels or 16 pixels in each of the memory controllers 350, 360, 370, 380 Data is read out in parallel through the read system wiring groups 401R, 402R, 411R, 412R, 421R, 422R, 431R, and 432R, and further through the read system wiring groups 351, 361, 371, 381, and the wiring group 440. The data is input to the data calculation unit 392.
Then, after a predetermined calculation process is performed by the data calculation unit 392, the data rearrangement unit 400 converts the 256-bit width color data mc_0dtr to mc_7dtr to the 64-bit width transfer data hrd, and requests the request destination, for example, the main The data is output to the processor 11 via the host interface circuit 149.
[0160]
In the RAMDAC 146, RGB values for RGB indexes are stored in the RAM, and the RGB values for the index values are transferred to a D / A converter (not shown).
Then, the RGB signal converted into an analog signal by the D / A converter is transferred to the CRT.
[0161]
As described above, according to the present embodiment, even if the data format between the host interface circuit 149 and the texture engine circuit 143 shown in FIG. 1 and the DRAM 147 is different, the data rearrangement unit 400 shown in FIG. By rearranging the data in, data transfer between the texture engine circuit 143 and the host interface circuit 149 and the DRAM 147 can be performed accurately.
In addition, even when there is a restriction on the address pattern for accessing the DRAM 147 according to the present embodiment, the data rearrangement unit 400 performs data rearrangement in consideration of the restriction, thereby Color data can be written and read accurately.
[0162]
Further, according to the present embodiment, in the data rearrangement unit 400 shown in FIG._ThreeWhen the half of the 16-bit storage circuits each having a 64-bit width are emptied, the DRAM 147 continuously receives the number of times corresponding to the pixel data length instruction signal S801. By performing the reading, the FIFO circuit 410 by the controller 440 is obtained._ThreeThe control of the controller 440 can be reduced. Therefore, the other processing performed by the controller 440 is the FIFO circuit 410._ThreeIt is possible to avoid a state of waiting by the control process.
[0163]
Further, according to the present embodiment, in the memory I / F circuit 144 that accesses (writes or reads) the DRAM 147 and the SRAM 148, the write path and the read path are configured as separate paths, and in the case of writing, the write address When ADRW and color data cwd are processed by the data rearrangement unit 400, the distributor 300, the address converters 310, 320, 330, and 340 and the memory controllers 350, 360, 370, and 380, which are writing circuits, and written to the DRAM 147 and read out For example, the data rearrangement unit 400, the read controller 390, and the memory controllers 350, 360, 370, and 380, which are read system circuits, are processed and read from the DRAM 147 or the SRAM 148. Can be switched at an earlier timing in a time like that contains the interruption of the reading, there is an advantage that attained the performance improvement of the time of reading.
[0164]
Further, the memory I / F circuit 144 performs access to the DRAM 147 based on predetermined interleaving addressing, for example, in units of 16 pixels or in units of 8 pixels, and restricts the area to be accessed. An image processing apparatus that does not cause a violation can be realized.
[0165]
Furthermore, according to the present embodiment, the DRAM 147 built in the semiconductor chip has a configuration in which display data and texture data required by at least one graphic element are stored. Since data can be stored, the built-in DRAM can be used effectively, and an image processing apparatus capable of paralleling high-speed processing operation and low power consumption can be realized.
And, a single memory system can be realized, and processing can be performed only when everything is built in. As a result, the architecture is a big paradigm shift.
In addition, since the memory can be used effectively, the processing can be performed only with the internal DRAM, and the large bandwidth between the memory and the drawing system can be fully used because it is inside. Also, special processing can be incorporated in the DRAM.
[0166]
Furthermore, in the display address space, display elements at adjacent addresses are arranged so as to be different DRAM blocks, so that bit lines can be used more effectively, and a relatively solid display as in graphics drawing. When there are many accesses to the area, the probability that each module can be processed simultaneously increases, and the drawing performance can be improved.
[0167]
In addition, since the DRAM is built in the chip, the high-speed interface part is completed only inside the chip, so that it is an I / O buffer with a large additional capacity, or an inter-chip wiring capacity. It is no longer necessary to drive the drive, and the power consumption is smaller than when not built in.
Therefore, a mechanism that can do everything in one chip using various technologies is an indispensable technical element for familiar digital devices such as portable information terminals in the future.
[0168]
The present invention is not limited to the embodiment described above.
For example, the access pattern restrictions on the DRAM 147 are not limited to those described above, and the present invention can be applied even when there are other access pattern restrictions.
Further, the configuration of the data rearrangement unit 400 shown in FIG. 8 and the configuration of the data rearrangement circuit 420 shown in FIG. 9 are not limited to these.
[0169]
Further, the FIFO circuit 410 shown in FIG.₁~ 410_ThreeThe number of stages is arbitrary.
The controller 440 also includes a FIFO circuit 410._ThreeThe timing of issuing the read request after starting the storage state may be other than when the empty area is halved.
[0170]
Further, in the above-described three-dimensional computer graphics system 10 shown in FIG. 1, the configuration using the SRAM 148 is illustrated, but a configuration without the SRAM 148 may be used.
[0171]
Further, in the three-dimensional computer graphics system 10 shown in FIG. 1, the case where the geometry processing for generating polygon rendering data is performed by the main processor 11 is exemplified, but the configuration may be such that it is performed by the rendering circuit 14.
[0172]
【The invention's effect】
As described above, according to the image processing apparatus of the present invention, data transmission / reception between the image processing circuit and the storage circuit can be performed accurately even when there is a limitation in addressing when accessing the storage circuit. .
Further, according to the image processing apparatus of the present invention, even when there is a restriction on address designation when accessing the storage circuit, data transfer within the storage area of the storage circuit can be performed accurately.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a three-dimensional computer graphics system according to the present invention.
FIG. 2 is a block diagram showing a specific configuration example of a DRAM, SRAM, and a memory I / F circuit for accessing the DRAM and SRAM in the rendering circuit according to the present invention.
FIG. 3 is a schematic diagram showing a configuration example of a DRAM buffer according to the present invention.
FIG. 4 is a diagram for explaining color data included in texture data and accessed simultaneously.
FIG. 5 is a diagram for explaining unit blocks constituting texture data.
FIG. 6 is a diagram for explaining an address space of a texture buffer.
FIG. 7 is a diagram for explaining image data processing of a distributor in the memory I / F circuit according to the present invention.
FIG. 8 is a configuration diagram of the data rearrangement unit 400 shown in FIG. 2;
9 is a configuration diagram of the data rearrangement circuit 420 shown in FIG. 8. FIG.
FIG. 10 is a diagram showing a storage area in the DRAM shown in FIG. 1;
11 is a diagram for explaining an operation in a “Host to Local” transfer mode when the data length of one pixel in the data rearrangement unit shown in FIG. 2 is 32 bits.
12 is a diagram for explaining the operation in “Host to Local” transfer mode when the data length of one pixel in the data rearrangement unit shown in FIG. 2 is 32 bits. FIG.
13 is a diagram for explaining an operation in a “Host to Local” transfer mode in a case where the data length of one pixel in the data rearrangement unit shown in FIG. 2 is 32 bits.
14 is a diagram for explaining an operation of writing the color data shown in FIG. 13 to the DRAM shown in FIG. 1;
15 is a diagram for explaining an operation of writing the color data shown in FIG. 13 to the DRAM shown in FIG. 1;
16 is a diagram for explaining an operation in a “Host to Local” transfer mode when the data length of one pixel in the data rearrangement unit shown in FIG. 2 is 16 bits.
17 is a diagram for explaining an operation in a “Host to Local” transfer mode when the data length of one pixel in the data rearrangement unit shown in FIG. 2 is 16 bits.
18 is a diagram for explaining an operation in “Host to Local” transfer mode when the data length of one pixel in the data rearrangement unit shown in FIG. 2 is 16 bits.
FIG. 19 is a diagram for explaining an operation in a “Local to Host” transfer mode in the data rearrangement unit shown in FIG. 2;
FIG. 20 is a diagram for explaining the operation in the “Local to Host” transfer mode in the data rearrangement unit shown in FIG. 2;
FIG. 21 is a diagram for explaining the operation in the “Local to Host” transfer mode in the data rearrangement unit shown in FIG. 2;
FIG. 22 is a diagram for explaining an operation in a “Local to Host” transfer mode in the data rearrangement unit shown in FIG. 2;
FIG. 23 is a diagram for explaining an operation in a “Local to Loacl” transfer mode in the data rearrangement unit shown in FIG. 2;
24 is a diagram for explaining the operation in the “Local to Local” transfer mode in the data rearrangement unit shown in FIG. 2; FIG.
FIG. 25 is a diagram for explaining the operation in the “Local to Local” transfer mode in the data rearrangement unit shown in FIG. 2;
FIG. 26 is a diagram for explaining the relationship between the number of read instructions indicated in the read request for the control signal output from the controller shown in FIG. 8 to the address generation unit and the data length of one pixel.
FIG. 27 is a diagram for explaining color data read from a DRAM by one read operation when the data length of one pixel is 32 bits, 16 bits, 8 bits, and 4 bits;
FIG. 28 is a system configuration diagram showing a basic concept of a three-dimensional computer graphics system.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... Three-dimensional computer graphics system, 11 ... Main processor, 12 ... Main memory, 13 ... I / O interface circuit, 14 ... Rendering circuit, 141 ... DDA setup circuit, 142 ... Triangle DDA circuit, 143 ... Texture engine circuit, 144 ... Memory I / F circuit, 145 ... CRT controller circuit, 146 ... RAMDAC circuit, 147 ... DRAM, 147a ... Texture buffer, 147b ... Display buffer, 147c ... z buffer, 147d ... Texture CLUT buffer, 148 ... SRAM, 200, 210, 220, 230 ... Memory module, 300 ... Distributor, 310, 320, 330, 340 ... Address decoder, 350, 360, 370, 380 ... Memory controller 390 ... read controller, 391 ... address decoder, 392 ... data processing unit, 400 ... data rearrangement unit, 410₁~ 410_Three... FIFO circuit, 410_Four... Latch circuit, 420 ... Data rearrangement circuit, 430 ... Address generation unit, 440 ... Controller

Claims (15)

Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic An image processing circuit that performs rendering processing using data and generates image data including a plurality of pixels;
A plurality of storage circuits to which image data composed of the plurality of pixels generated by the rendering process in units of a predetermined size block is written , wherein the writing is based on a representative point as a first A plurality of storage circuits which are performed in units of rectangular areas and have a restriction that only addresses that are multiples of the first number can be specified in the first and second directions orthogonal to each other in the first rectangular area ;
When the address of the representative point of the block including the image data composed of the plurality of pixels generated by the rendering process input from the image processing circuit is specified based on the restriction, the input from the image processing circuit above so that the image data is written in the first rectangular area corresponding to the appropriate memory circuits of said plurality of memory circuits, based on the constraints were, previous SL image data, rearranged by the pixel data unit, data A reordering circuit;
; And a written can lump circuit,
The rearrangement circuit, in the first transfer mode for storing image data composed of the plurality of pixels generated by the rendering processing input from the image processing circuit is transferred to the plurality of memory circuits,
Placing the rendered first image data of the first bit width input from the image processing circuit in the second image data of the second bit width ;
When the address of the representative point of the block including the second image data is designated based on the restriction, the first image data corresponds to the corresponding storage circuit of the plurality of storage circuits . The second image data is rearranged so as to be written in the rectangular storage area by repeating shift processing, selection processing, and holding processing based on a control signal ,
The writing circuit distributes and writes the blocks including the rearranged second image data to corresponding storage circuits of the plurality of storage circuits;
Image processing device. In each of the storage circuits, the address of the representative point is designated at a predetermined address interval for each of two orthogonal directions in the two-dimensional address space.
The image processing apparatus according to claim 1. The block is rectangular;
The second bit width of the second image data is the same as the length of one side of the block;
When the address in the one side direction of the block in the storage area of the storage circuit is the address in the X direction and the address in the direction orthogonal to the X direction is the address in the Y direction, the data rearrangement circuits are the same Among the plurality of pieces of pixel data included in the first image data, pixel data written at different Y-direction addresses in the predetermined storage area are arranged in the second image data different from each other.
The image processing apparatus according to claim 1. The data rearrangement circuit generates validity instruction data for specifying valid pixel data in the block, and includes the validity instruction data in the generated address.
The writing circuit is configured to store the block including the image data on the basis of pixel data on which the validity in the block is instructed based on the validity instruction data included in the address. To distribute and write
The image processing apparatus according to claim 1 . Wherein the storage circuitry, the image processing apparatus according to claim 1, and data rearrangement circuit are mixed in one semiconductor chip. Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic plurality data are stored respectively distributed and was generated by the rendering processing stored in the storage area in units of predetermined size of the block, the writing and reading of image data composed of a plurality of pixel data is performed The writing is performed in units of the first rectangular area with the representative point as a reference, and is a multiple of the first number in the first and second directions orthogonal to the first rectangular area. There is a first restriction that only an address can be specified, and the reading is performed in units of a second rectangular area with a representative point as a reference, and in the first direction perpendicular to the second rectangular area. Can specify any address In its for the second direction is a second limitation that can address only in multiples of the second number, and a plurality of storage circuits,
An image processing circuit for inputting first image data having a first bit width and performing a rendering process on the first image data;
A readout circuit for designating a representative point based on the second restriction and reading out the image data from the plurality of storage circuits in the block unit;
Data rearrangement for outputting the first image data of the first bit width generated by rearranging the second image data of the second bit width obtained by the reading in units of pixel data to the image processing circuit Circuit and
In the second transfer mode in which the rearrangement circuit transfers image data from the storage circuit to the image processing circuit,
An address corresponding to the address input from the image processing circuit is generated and output to the reading circuit to read out corresponding image data from the plurality of storage circuits,
Converting to image data of the first bit width defined by the image processing circuit and outputting to the image processing circuit;
Image processing device. In each of the storage circuits, the address of the representative point is designated at a predetermined address interval for each of two orthogonal directions in the two-dimensional address space.
The image processing apparatus according to claim 6. The image processing apparatus according to claim 6, wherein the storage circuit, the readout circuit, and the data rearrangement circuit are mixedly mounted in one semiconductor chip. Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic The data is distributed, the image data generated by the rendering process is written in units of a predetermined size block, and the image data generated by the distributed writing process is read. In the plurality of memory circuits, the writing is performed in units of a first rectangular area with a representative point as a reference, and the first and second directions orthogonal to the first rectangular area are first described. There is a first restriction that only addresses that are multiples of the number can be specified, and the reading is performed in units of a second rectangular area with the representative point as a reference. Although for the first direction of the orthogonal of the rectangular area may be set to any address for the second direction is a second limitation that can address only in multiples of the second number, and a plurality of storage circuits,
A readout circuit for designating a representative point based on the second restriction and reading out the image data from the plurality of storage circuits in units of the block;
Processing is performed according to either a first transfer mode in which image data is transferred from the image processing circuit to the storage circuit, or a second transfer mode in which image data is transferred from the storage circuit to the image processing circuit. A rearrangement circuit that stores the corresponding image data when the address of the representative point of the block including the read image data is designated based on the first restriction. A data rearrangement circuit for rearranging the read image data in units of pixel data so as to be written in a corresponding first rectangular area in the circuit;
A writing circuit that designates a representative point based on the first restriction and writes the block including the rearranged image data to the storage circuit;
In the first transfer mode, the rearrangement circuit transfers and stores the image data from the image processing circuit to the plurality of storage circuits.
Placing the rendered first image data of the first bit width input from the image processing circuit in the second image data of the second bit width;
When the address of the representative point of the block including the second image data is designated based on the first restriction , the first image data corresponds to the corresponding storage circuit of the plurality of storage circuits. The second image data is rearranged by repeating the shift process, the selection process, and the holding process based on the control signal so as to be written in the rectangular area .
The writing circuit distributes and writes the blocks including the rearranged second image data to corresponding storage circuits of the plurality of storage circuits,
In the second transfer mode in which the rearrangement circuit transfers image data from the storage circuit to the image processing circuit,
An address corresponding to the address input from the image processing circuit is generated and output to the reading circuit to read out corresponding image data from the plurality of storage circuits,
Arranged in the image data of the first bit width defined by the image processing circuit and outputting to the image processing circuit;
Image processing device. The data rearrangement circuit generates validity instruction data for specifying valid pixel data in the block, and includes the validity instruction data in the generated address.
The writing circuit is configured to store the block including the image data on the basis of pixel data on which the validity in the block is instructed based on the validity instruction data included in the address. To distribute and write
The image processing apparatus according to claim 9. In each of the storage circuits, the address of the representative point is designated at a predetermined address interval for each of two orthogonal directions in the two-dimensional address space.
The image processing apparatus according to claim 9 or 10. Each of the storage circuits is different in the size of the block and the designation of the representative point address between the reading and the writing,
The image processing apparatus according to claim 9. The image processing apparatus according to claim 9, wherein the storage circuit, the read circuit, the data rearrangement circuit, and the write circuit are mounted together in one semiconductor chip. Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic An image processing apparatus that performs rendering processing using data,
A plurality of storage circuits, wherein display data consisting of a plurality of pixel bits and texture data required by at least one graphic element are distributed and stored in the plurality of storage circuits, and a block of a predetermined size is unit A plurality of storage circuits to which the image data generated by the rendering process is written , wherein the writing is performed in units of a first rectangular area with a representative point as a reference, and the first rectangular area In the first and second directions orthogonal to each other, there is a first restriction that only an address that is a multiple of the first number can be specified. An arbitrary address can be specified for the first direction orthogonal to the second rectangular area, but an address can only be specified by a multiple of the second number for the second direction. Cormorants there is a second constraint, a plurality of storage circuits,
An interpolation data generation circuit that interpolates polygon rendering data at the vertices of the unit graphic and generates interpolation data of pixels located in the unit graphic;
The texture homogeneous coordinates (s, t) included in the interpolation data are divided by the homogeneous term q to generate “s / q” and “t / q”, and the “s / q” and “t / q”. Texture processing that reads texture data from the plurality of storage circuits using a texture address according to the image, and pastes the read texture data on the surface of the graphic element to generate display data composed of a plurality of pixel bits Circuit,
A data rearrangement circuit for performing data rearrangement processing according to a first transfer mode for transferring image data from the texture processing circuit to the storage circuit, and comprising a plurality of pixel bits input from the texture processing circuit When the address of the representative point of the block including display data is designated based on the first restriction , the display data is written in a distributed manner in the corresponding storage circuits in the plurality of storage circuits. A data rearrangement circuit for rearranging the display data in units of image data;
A writing circuit that distributes and writes the blocks including the rearranged display data to the plurality of storage circuits, and
In the first transfer mode, the rearrangement circuit transfers and stores the image data from the image processing circuit to the plurality of storage circuits.
Placing the rendered first image data of the first bit width input from the image processing circuit in the second image data of the second bit width;
When the address of the representative point of the block including the second image data is designated based on the first restriction , the first image data is written to a predetermined storage area of the plurality of storage circuits. As described above, the second image data is rearranged by repeating shift processing, selection processing, and holding processing based on the control signal ,
The writing circuit distributes and writes the blocks including the rearranged second image data to corresponding storage circuits of the plurality of storage circuits;
Image processing device. Polygon rendering including three-dimensional coordinates (x, y, z), R (red), G (green), B (blue) data, texture homogeneous coordinates (s, t) and homogeneous term q for the vertices of the unit graphic An image processing apparatus that performs rendering processing using data,
A plurality of storage circuits, wherein display data and texture data required by at least one graphic element are distributed and stored in the plurality of storage circuits, and generated by the above rendering process in units of a predetermined size block. a plurality of storage circuits writing Ru performed of the image data, the writing is performed a first rectangular area relative to the representative point as a unit, the first and perpendicular to the first rectangular region There is a first restriction that only addresses that are multiples of the first number can be specified in the second direction, and the reading is performed in units of the second rectangular area based on the representative point, and the second rectangular area Although in the first direction perpendicular can specify any address for the second direction is a second limitation that can address only in multiples of the second number, a plurality A memory circuit,
An interpolation data generation circuit that interpolates polygon rendering data at the vertices of the unit graphic and generates interpolation data of pixels located in the unit graphic;
The texture homogeneous coordinates (s, t) included in the interpolation data are divided by the homogeneous term q to generate “s / q” and “t / q”, and the “s / q” and “t / q”. A texture processing circuit that generates display data by applying a texture data read from the plurality of storage circuits to a surface of a graphic element using a texture address according to
A readout circuit for designating a representative point based on the second restriction and reading the texture data from the plurality of storage circuits in units of blocks;
An interface circuit;
A data rearrangement circuit for rearranging data according to a second transfer mode for transferring display data read from the storage circuit to the texture processing circuit, wherein the read texture data is transferred to the interface. A data rearrangement circuit that rearranges the output according to the output format of the circuit and outputs the data to the interface circuit,
The rearrangement circuit, in the second transfer mode, via to generate and said plurality of storage and output to the read circuit an address corresponding to an address inputted from the image processing circuit from said texture circuit the interface reads out the image data corresponding the circuit, and outputs the arranged in the image data of the first bit width specified by the image processing circuit to the image processing circuit,
Image processing device.

Images