This application is related to, and incorporates by reference, the following commonly assigned U.S. Patent Application, A GPU HAVING RASTER COMPONENTS CONFIGURED FOR USING NESTED BOUSTROPHEDONIC PATTERNS TO TRAVERSE SCREEN AREAS, by Crow, et al., filed on Dec. 15, 2005, Ser. No. 11/304,904.
The present invention is generally related to hardware accelerated graphics computer systems.
Recent advances in computer performance have enabled graphic systems to provide more realistic graphical images using personal computers, home video game computers, handheld devices, and the like. In such graphic systems, a number of procedures are executed to “render” or draw graphic primitives to the screen of the system. A “graphic primitive” is a basic component of a graphic picture, such as a vertex, polygon, or the like. Rendered images are formed with combinations of these graphic primitives. Many procedures may be utilized to perform 3-D graphics rendering.
Specialized graphics processing units (e.g., GPUs, etc.) have been developed to optimize the computations required in executing the graphics rendering procedures. The GPUs are configured for high-speed operation and typically incorporate one or more rendering pipelines. Each pipeline includes a number of hardware-based functional units that are optimized for high-speed execution of graphics instructions/data, where the instructions/data are fed into the front end of the pipeline and the computed results emerge at the back end of the pipeline. The hardware-based functional units, cache memories, firmware, and the like, of the GPU are optimized to operate on the low-level graphics primitives (e.g., comprising “points”, “lines”, “triangles”, etc.) and produce real-time rendered 3-D images.
The real-time rendered 3-D images are generated using raster display technology. Raster display technology is widely used in computer graphics systems, and generally refers to the mechanism by which the grid of multiple pixels comprising an image are influenced by the graphics primitives. For each primitive, a typical rasterization system generally steps from pixel to pixel and determines whether or not to “render,” or write a given pixel into a frame buffer or pixel map, as per the contribution of the primitive. This, in turn, determines how to write the data to the display buffer representing each pixel.
Various traversal algorithms have been developed for moving from pixel to pixel in a way such that all pixels within the primitive are covered. For example, some solutions involve generating the pixels in a unidirectional manner. Such traditional unidirectional solutions involve generating the pixels row-by-row in a constant direction. This requires that the sequence shift across the primitive to a starting location on a first side of the primitive upon finishing at a location on an opposite side of the primitive. Each time this shift is executed, pixels or texture values are stored which were not positioned adjacent to pixels or texture values processed immediately beforehand. Therefore, such distant pixels or texture values have a greater chance of belonging to different memory access blocks, making such access inefficient.
Thus, a need exists for a rasterization process that can ensure needed graphics rendering data (e.g., texture values, normal maps, etc.) can be maintained in memory for an efficient access by the GPU.
Embodiments of the present invention provide a method and system for ensuring needed graphics rendering data (e.g., texture values, normal maps, etc.) can be maintained in low latency memory for an efficient access by the GPU. Embodiments of the present invention provide fast and efficient real-time 3-D graphics rendering by increasing the efficiency of cache memory access and by limiting the performance penalties resulting from accessing higher latency memory.
In one embodiment, the present invention is implemented as a method for traversing pixels of an area, including the steps of traversing a plurality of pixels of an image using a first boustrophedonic pattern along a predominant axis, and, during the traversal using the first boustrophedonic pattern, traversing a plurality of pixels of the image using a second boustrophedonic pattern. The second boustrophedonic pattern is nested within the first boustrophedonic pattern. The predominant axis of the first boustrophedonic pattern can be a horizontal axis, the second boustrophedonic pattern can be along a vertical axis. Similarly, the predominant axis of the first boustrophedonic pattern can be a vertical axis, the second boustrophedonic pattern can be along a horizontal axis.
In one embodiment, the plurality of pixels of the image are grouped as tiles and the tiles are traversed using the first boustrophedonic pattern and the second boustrophedonic pattern (e.g., nested within the first pattern). Additionally, in one embodiment, the first boustrophedonic pattern and the second boustrophedonic pattern are configured to maintain locality with respect to cache memory accesses.
The present invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements.
Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings. While the invention will be described in conjunction with the preferred embodiments, it will be understood that they are not intended to limit the invention to these embodiments. On the contrary, the invention is intended to cover alternatives, modifications and equivalents, which may be included within the spirit and scope of the invention as defined by the appended claims. Furthermore, in the following detailed description of embodiments of the present invention, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be recognized by one of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, components, and circuits have not been described in detail as not to unnecessarily obscure aspects of the embodiments of the present invention.
Notation and Nomenclature:
Some portions of the detailed descriptions, which follow, are presented in terms of procedures, steps, logic blocks, processing, and other symbolic representations of operations on data bits within a computer memory. These descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. A procedure, computer executed step, logic block, process, etc., is here, and generally, conceived to be a self-consistent sequence of steps or instructions leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated in a computer system. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.
It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussions, it is appreciated that throughout the present invention, discussions utilizing terms such as “processing” or “accessing” or “executing” or “storing” or “rendering” or the like, refer to the action and processes of a computer system (e.g., computer system 100 of
Computer System Platform:
It should be appreciated that the GPU 110 can be implemented as a discrete component, a discrete graphics card designed to couple to the computer system 100 via a connector (e.g., AGP slot, PCI-Express slot, etc.), a discrete integrated circuit die (e.g., mounted directly on a motherboard), or as an integrated GPU included within the integrated circuit die of a computer system chipset component (not shown). Additionally, a local graphics memory 112 can be included for the GPU 110 for high bandwidth graphics data storage.
In one embodiment, as depicted in diagram 200 of
Thus, as depicted in
Referring still to
One such memory is cache memory. Cache memory is optimized for high speed, low latency access. However the amount of cache memory available to the GPU is typically much less than the total frame buffer memory needed. Thus, the boustrophedonic sequence more often finds the pixels or texture values already loaded into cache memory, and therefore avoiding costly high latency accesses to slower frame buffer memory (e.g., local graphics memory 112) or even the much lower system memory (e.g., memory 115).
It should be noted that in one embodiment, instead of implementing a boustrophedonic pattern traversal of a grid of pixels, the boustrophedonic pattern traversal can be implemented on a grid of tiles. In such an embodiment, each tile comprises a block of pixels. The size of the tile (e.g., number of pixels per tile) is generally tailored in accordance with the rendering power of the GPU and/or the size of the cache memory or frame buffer memory. In one embodiment, the size of the tile is programmable and can be dynamically selected to be, for example, 4×4, 8×8, 16×16, 32×32, 64×64, 128×128, or the like, including any rectangular (as well as square) array of pixels or subpixels per tile).
As described above, a horizontal boustrophedonic pattern (e.g., a horizontal predominant axis), refers to a serpentine pattern that folds back and forth along the horizontal axis. In the present embodiment, a second boustrophedonic pattern is embedded within this first horizontal boustrophedonic pattern. As depicted in
In this manner, embodiments of the present invention can embed a second vertical boustrophedonic pattern into a first horizontal boustrophedonic pattern. The embedding of the second boustrophedonic pattern into the first boustrophedonic pattern serves to further localize data accesses and limit cache misses. For example, the multiple embedded boustrophedonic patterns can be configured to align closely with the size and configuration of the cache memory of the GPU (e.g., cache line sizes, etc.). This feature helps to maintain locality with respect to cache memory accesses. The multiple embedded boustrophedonic patterns can also be configured to maintain locality with respect to memory bank accesses of the frame buffer memory 112, where multiple DRAM banks (not shown) comprise the frame buffer memory 112. Furthermore, the multiple boustrophedonic patterns can be configured to align with a prefetching process for accessing memory, where the prefetching process relies upon a predictable memory access pattern so that it can look ahead to fetch data from memory into the GPU's cache.
It should be noted that, as described above, embodiments of the present invention can operate on individual pixels, or even arrays of samples within pixels, in addition to tiles of pixels, and the size of the tiles can be customized in accordance with the needs of a particular application (e.g., programmable by a graphics driver, etc.). Additionally, it should be noted that the configuration of the boustrophedonic patterns can be customized in accordance with the needs of a particular application (e.g., programmable by a graphics driver, etc.). For example, the first and second boustrophedonic patterns can be vertical and horizontal, horizontal and vertical, or the like.
In one embodiment, the rasterizer unit 502 includes a fine raster component 503 and a coarse raster component 504. The coarse raster component 504 implements the tile based boustrophedonic pattern rasterizing as described above, as it rapidly searches a grid of tiles to identify tiles of interest (e.g., tiles that are covered by a primitive). Once the tiles of interest are identified, the fine raster component 503 individually identifies the pixels that are covered by the primitive. Hence, in such an embodiment, the coarse raster component 504 rapidly searches a grid of pixels by using tiles, and the fine raster component 503 uses the information generated by the coarse raster component 504 and implements fine granularity rasterization by individually identifying pixels covered by the primitive. In both cases, both the coarse raster component 504 and the fine raster component 503 can utilize one or more boustrophedonic patterns (e.g., nested or non-nested) during their rasterization.
Referring still to
In one embodiment, the present invention is implemented as a GPU having a raster unit configured to implement boustrophedonic pattern rasterization, whereby a number of performance attributes of the boustrophedonic pattern rasterization process are programmable. In such an embodiment, for example, the number of pixels per tile is programmable in accordance with a size of a cache memory of the GPU. Similarly, the configuration of the first boustrophedonic pattern, the second boustrophedonic pattern, and the like, are programmable (e.g., the predominant axis, tile size, initiating order, finishing order, and the like).
The foregoing descriptions of specific embodiments of the present invention have been presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed, and many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application, to thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims appended hereto and their equivalents.
Number | Name | Date | Kind |
---|---|---|---|
4208810 | Rohner et al. | Jun 1980 | A |
4918626 | Watkins et al. | Apr 1990 | A |
5081594 | Horsley | Jan 1992 | A |
5287438 | Kelleher | Feb 1994 | A |
5313287 | Barton | May 1994 | A |
5432898 | Curb et al. | Jul 1995 | A |
5446836 | Lentz et al. | Aug 1995 | A |
5452104 | Lee | Sep 1995 | A |
5452412 | Johnson, Jr. et al. | Sep 1995 | A |
5483258 | Cornett et al. | Jan 1996 | A |
5543935 | Harrington | Aug 1996 | A |
5570463 | Dao | Oct 1996 | A |
5594854 | Baldwin et al. | Jan 1997 | A |
5623692 | Priem et al. | Apr 1997 | A |
5633297 | Valko et al. | May 1997 | A |
5664162 | Dye | Sep 1997 | A |
5815162 | Levine | Sep 1998 | A |
5854631 | Akeley et al. | Dec 1998 | A |
5854637 | Sturges | Dec 1998 | A |
5872902 | Kuchkuda et al. | Feb 1999 | A |
5977987 | Duluk, Jr. | Nov 1999 | A |
6028608 | Jenkins | Feb 2000 | A |
6034699 | Wong et al. | Mar 2000 | A |
6072500 | Foran et al. | Jun 2000 | A |
6104407 | Aleksic et al. | Aug 2000 | A |
6104417 | Nielsen et al. | Aug 2000 | A |
6115049 | Winner et al. | Sep 2000 | A |
6118394 | Onaya | Sep 2000 | A |
6128000 | Jouppi et al. | Oct 2000 | A |
6137918 | Harrington et al. | Oct 2000 | A |
6160557 | Narayanaswami | Dec 2000 | A |
6160559 | Omtzigt | Dec 2000 | A |
6188394 | Morein et al. | Feb 2001 | B1 |
6201545 | Wong et al. | Mar 2001 | B1 |
6204859 | Jouppi et al. | Mar 2001 | B1 |
6219070 | Baker et al. | Apr 2001 | B1 |
6249853 | Porterfield | Jun 2001 | B1 |
6259460 | Gossett et al. | Jul 2001 | B1 |
6323874 | Gossett | Nov 2001 | B1 |
6359623 | Larson | Mar 2002 | B1 |
6362819 | Dalal et al. | Mar 2002 | B1 |
6366289 | Johns | Apr 2002 | B1 |
6429877 | Stroyan | Aug 2002 | B1 |
6437780 | Baltaretu et al. | Aug 2002 | B1 |
6452595 | Montrym et al. | Sep 2002 | B1 |
6469707 | Voorhies | Oct 2002 | B1 |
6480205 | Greene et al. | Nov 2002 | B1 |
6501564 | Schramm et al. | Dec 2002 | B1 |
6504542 | Voorhies et al. | Jan 2003 | B1 |
6522329 | Ihara et al. | Feb 2003 | B1 |
6525737 | Duluk, Jr. et al. | Feb 2003 | B1 |
6529207 | Landau et al. | Mar 2003 | B1 |
6606093 | Gossett et al. | Aug 2003 | B1 |
6611272 | Hussain et al. | Aug 2003 | B1 |
6614444 | Duluk, Jr. et al. | Sep 2003 | B1 |
6614448 | Garlick et al. | Sep 2003 | B1 |
6624823 | Deering | Sep 2003 | B2 |
6633197 | Sutardja | Oct 2003 | B1 |
6633297 | McCormack et al. | Oct 2003 | B2 |
6646639 | Greene et al. | Nov 2003 | B1 |
6671000 | Cloutier | Dec 2003 | B1 |
6693637 | Koneru et al. | Feb 2004 | B2 |
6693639 | Duluk, Jr. et al. | Feb 2004 | B2 |
6697063 | Zhu | Feb 2004 | B1 |
6717576 | Duluk, Jr. et al. | Apr 2004 | B1 |
6717578 | Deering | Apr 2004 | B1 |
6734861 | Van Dyke et al. | May 2004 | B1 |
6741247 | Fenney | May 2004 | B1 |
6747057 | Ruzafa et al. | Jun 2004 | B2 |
6765575 | Voorhies et al. | Jul 2004 | B1 |
6778177 | Furtner | Aug 2004 | B1 |
6788301 | Thrasher | Sep 2004 | B2 |
6798410 | Redshaw et al. | Sep 2004 | B1 |
6803916 | Ramani et al. | Oct 2004 | B2 |
6819332 | Baldwin | Nov 2004 | B2 |
6833835 | van Vugt | Dec 2004 | B1 |
6906716 | Moreton et al. | Jun 2005 | B2 |
6938176 | Alben et al. | Aug 2005 | B1 |
6940514 | Wasserman et al. | Sep 2005 | B1 |
6947057 | Nelson et al. | Sep 2005 | B2 |
6956579 | Diard et al. | Oct 2005 | B1 |
6961057 | Van Dyke et al. | Nov 2005 | B1 |
6978317 | Anantha et al. | Dec 2005 | B2 |
7002591 | Leather et al. | Feb 2006 | B1 |
7009607 | Lindholm et al. | Mar 2006 | B2 |
7009615 | Kilgard et al. | Mar 2006 | B1 |
7061495 | Leather | Jun 2006 | B1 |
7064771 | Jouppi et al. | Jun 2006 | B1 |
7075542 | Leather | Jul 2006 | B1 |
7081902 | Crow et al. | Jul 2006 | B1 |
7119809 | McCabe | Oct 2006 | B1 |
7126600 | Fowler et al. | Oct 2006 | B1 |
7154066 | Talwar et al. | Dec 2006 | B2 |
7158148 | Toji et al. | Jan 2007 | B2 |
7170515 | Zhu | Jan 2007 | B1 |
7184040 | Tzvetkov | Feb 2007 | B1 |
7224364 | Yue et al. | May 2007 | B1 |
7307628 | Goodman et al. | Dec 2007 | B1 |
7307638 | Leather et al. | Dec 2007 | B2 |
7382368 | Molnar et al. | Jun 2008 | B1 |
7453466 | Hux et al. | Nov 2008 | B2 |
7483029 | Crow et al. | Jan 2009 | B2 |
7548996 | Baker et al. | Jun 2009 | B2 |
7551174 | Iourcha et al. | Jun 2009 | B2 |
7633506 | Leather et al. | Dec 2009 | B1 |
7634637 | Lindholm et al. | Dec 2009 | B1 |
7791617 | Crow et al. | Sep 2010 | B2 |
7965902 | Zelinka et al. | Jun 2011 | B1 |
8063903 | Vignon et al. | Nov 2011 | B2 |
20010005209 | Lindholm et al. | Jun 2001 | A1 |
20020050979 | Oberoi et al. | May 2002 | A1 |
20020097241 | McCormack et al. | Jul 2002 | A1 |
20020130863 | Baldwin | Sep 2002 | A1 |
20020140655 | Liang et al. | Oct 2002 | A1 |
20020158885 | Brokenshire et al. | Oct 2002 | A1 |
20020196251 | Duluk, Jr. et al. | Dec 2002 | A1 |
20030058244 | Ramani et al. | Mar 2003 | A1 |
20030067468 | Duluk, Jr. et al. | Apr 2003 | A1 |
20030076325 | Thrasher | Apr 2003 | A1 |
20030122815 | Deering | Jul 2003 | A1 |
20030163589 | Bunce et al. | Aug 2003 | A1 |
20030194116 | Wong et al. | Oct 2003 | A1 |
20030201994 | Taylor et al. | Oct 2003 | A1 |
20040046764 | Lefebvre et al. | Mar 2004 | A1 |
20040085313 | Moreton et al. | May 2004 | A1 |
20040130552 | Duluk, Jr. et al. | Jul 2004 | A1 |
20040183801 | Deering | Sep 2004 | A1 |
20040196285 | Rice et al. | Oct 2004 | A1 |
20040207642 | Crisu et al. | Oct 2004 | A1 |
20040246251 | Fenney et al. | Dec 2004 | A1 |
20050030314 | Dawson | Feb 2005 | A1 |
20050041037 | Dawson | Feb 2005 | A1 |
20050066148 | Luick | Mar 2005 | A1 |
20050122338 | Hong et al. | Jun 2005 | A1 |
20050134588 | Aila et al. | Jun 2005 | A1 |
20050134603 | Iourcha et al. | Jun 2005 | A1 |
20050179698 | Vijayakumar et al. | Aug 2005 | A1 |
20050259100 | Teruyama | Nov 2005 | A1 |
20060044317 | Bourd et al. | Mar 2006 | A1 |
20060170690 | Leather | Aug 2006 | A1 |
20060203005 | Hunter | Sep 2006 | A1 |
20060245001 | Lee et al. | Nov 2006 | A1 |
20060267981 | Naoi | Nov 2006 | A1 |
20070139440 | Crow et al. | Jun 2007 | A1 |
20070268298 | Alben et al. | Nov 2007 | A1 |
20070273689 | Tsao | Nov 2007 | A1 |
20070296725 | Steiner et al. | Dec 2007 | A1 |
20080024497 | Crow et al. | Jan 2008 | A1 |
20080024522 | Crow et al. | Jan 2008 | A1 |
20080100618 | Woo et al. | May 2008 | A1 |
20080273218 | Kitora et al. | Nov 2008 | A1 |
Number | Date | Country |
---|---|---|
101093578 | Dec 2007 | CN |
06180758 | Jun 1994 | JP |
10134198 | May 1998 | JP |
11195132 | Jul 1999 | JP |
2005182547 | Jul 2005 | JP |
0013145 | Mar 2000 | WO |
Entry |
---|
A VLSI architecture for updating raster-scan displays Satish Gupta, Robert F. Sproull, Ivan E. Sutherland Aug. 1981 ACM SIGGRAPH Computer Graphics, Proceedings of the 8th annual conference on Computer graphics and interactive techniques SIGGRAPH '81, vol. 15 Issue 3 Publisher: ACM Press. |
A parallel algorithm for polygon rasterization Juan Pineda Jun. 1988 ACM. |
A hardware assisted design rule check architecture Larry Seiler Jan. 1982 Proceedings of the 19th conference on design automation DAC '82 Publisher: IEEE Press. |
Non-Final OA Dated Nov. 27, 2007; U.S. Appl. No. 11/304,904. |
Non-Final OA Dated May 1, 2007; U.S. Appl. No. 11/304,904. |
Crow; “The Use of Grayscale for Improves Raster Display of Vectors and Characters;” University of Texas, Austin, Texas; Work supported by the National Science Foundation unser Grants MCS 76-83889; pp. 1-5: ACM Press. |
Boyer, et al.; “Discrete Analysis for Antialiased Lines;” Eurographics 2000; 3 Pages. |
Blythe, OpenGL Section 3.4.1, “Basic Line Segment Rasterization”, Mar. 29, 1997, pp. 1-3. |
Foley, J. “Computer Graphics: Principles and Practice”, 1987, Addison-Wesley Publishing, 2nd Edition, p. 545-546. |
Fuchs; “Fast Spheres Shadow, Textures, Transparencies, and Image Enhancements in Pixel-Planes”; ACM; 1985; Department of Computer Science, University of North Carolina at Chapel Hill, Chapel Hill, NC 27514. |