公开(公告)号：US09092345B2

公开(公告)日：2015-07-28

申请号：US13962905

申请日：2013-08-08

Applicant: ARM Limited

Inventor： Jorn Nystad ,  Andreas Engh-Halstvedt

IPC: G06F12/08 ,  G06F9/30 ,  G06F9/38 ,  G06T1/20 ,  G06T1/60

CPC classification number: G06F12/0875 ,  G06F9/3004 ,  G06F9/3824 ,  G06F9/3828 ,  G06F9/3834 ,  G06F9/3851 ,  G06F9/3887 ,  G06F12/0815 ,  G06F12/0842 ,  G06T1/60

Abstract: A data processing system includes one or more processors 4, 5, 6, 7 operable to initiate atomic memory requests for execution threads and plural data caches 8, 9, 10, 11 that are used to store data needed to perform an atomic memory operation when an atomic memory operation is to be performed.When atomic operations are to be performed against a data cache, the results of atomic operations that are to access the same memory location are accumulated in a temporary cache line in the data cache pending the arrival in the cache of the “true” cache line from memory. The accumulated results of the atomic operations stored in the temporary cache line are then combined with the cache line from memory when the cache line arrives in the cache. Individual atomic values can also be reconstructed once the cache line arrives at the cache.

Abstract translation: 数据处理系统包括一个或多个处理器4,5,6,7，其可操作以发起用于执行线程的原子存储器请求以及用于存储执行原子存储器操作所需的数据的多个数据高速缓存8,9,10,11， 将执行原子存储器操作。 当要针对数据高速缓存执行原子操作时，访问同一内存位置的原子操作的结果将累积在数据高速缓存中的临时高速缓存行中，等待到达“真”高速缓存行的高速缓存中 记忆。 然后，当高速缓存行到达高速缓存时，存储在临时高速缓存行中的原子操作的累积结果与来自存储器的高速缓存行组合。 一旦高速缓存行到达高速缓存，也可以重构单个原子值。

公开(公告)号：US20150002524A1

公开(公告)日：2015-01-01

申请号：US13929599

申请日：2013-06-27

Applicant: ARM Limited

Inventor： Jorn Nystad ,  Sean Tristram Ellis

IPC: G06T1/60

CPC classification number: G06F17/17

Abstract: A data processor of a processing system, such as a graphics processing system, converts an input data value into an output data value by approximating a function which maps input values to output values. The data processor approximates the function using first and second predetermined ranges of values which are quantised into plural corresponding pairs of range sections, a predetermined gradient for each pair of range sections, and predetermined section end values for each pair of range sections. By using these predetermined parameters, the approximation of the function can be implemented efficiently by the data processor of the processing system.

Abstract translation: 诸如图形处理系统的处理系统的数据处理器通过近似将输入值映射到输出值的函数将输入数据值转换为输出数据值。 数据处理器使用被量化为多个对应的范围段对的第一和第二预定范围的值，每对范围段的预定梯度以及每对范围段的预定段结束值来近似该函数。 通过使用这些预定参数，功能的近似可以由处理系统的数据处理器有效地实现。

公开(公告)号：US08891886B2

公开(公告)日：2014-11-18

申请号：US13630040

申请日：2012-09-28

Applicant: ARM Limited

Inventor： Jorn Nystad

IPC: G06K9/36 ,  G06T9/00

CPC classification number: G06K9/36 ,  G06T9/00 ,  H04N19/119 ,  H04N19/136 ,  H04N19/176

Abstract: A graphics texture data encoding arrangement in which the texels in a texel block 30 to be encoded are divided into different partitions within the block. A reference partitioning pattern for a texel block to be encoded is generated by using a partitioning function 32 to partition the data values for the texels into a number of data value partitions, and then sorting the individual texels in the texel block into respective partitions 33 based on their values. A set of predefined partitioning patterns 35 that the encoding scheme supports is then compared 36 to the generated reference partitioning pattern. The predefined partitioning pattern that best matches 39 the generated reference partitioning pattern is then used 42 to encode the block of texels.

Abstract translation: 图形纹理数据编码装置，其中要编码的纹素块30中的纹理细分被分割成块内的不同分区。 通过使用分区函数32来生成要编码的纹素块的参考分割模式，以将纹理数据的数据值分割成多个数据值分区，然后将纹理块中的单个纹素分类为基于各个分区33 对他们的价值观。 然后将编码方案支持的一组预定义分区模式36与生成的参考分区模式进行比较36。 最终匹配39所生成的参考分区模式的预定义分区模式然后用42来编码纹素块。

公开(公告)号：US20140168220A1

公开(公告)日：2014-06-19

申请号：US13716952

申请日：2012-12-17

Applicant: ARM Limited

Inventor： Jorn Nystad ,  Edvard Sorgard ,  Frode Heggelund

IPC: G06T15/40

CPC classification number: G06T15/40 ,  G06T1/20

Abstract: The early depth test stages 4, 13 of a graphics processing pipeline 1 are configured to broadcast information 9, 10, 11, 14 about fragments, etc., that pass those early depth tests to other stages 3, 4, 6, 12 in the pipeline. The other stages in the pipeline then use the early depth test pass information to determine if the processing of any fragments that they are currently processing can be stopped.

Abstract translation: 图形处理流水线1的早期深度测试阶段4,13被配置为将关于通过这些早期深度测试的片段等的信息9,10,11,14广播到其他阶段3,4,6,12中 管道。 管道中的其他阶段然后使用早期深度测试通过信息来确定它们当前处理的任何片段的处理是否可以被停止。