Parallel processing image data having top-left dependent pixels

Patent metadata

Field	Value
Publication number	US-9547880-B2
Application number	US-201213997842-A
Country	US
Kind code	B2
Filing date	Apr 9, 2012
Priority date	Apr 9, 2012
Publication date	Jan 17, 2017
Grant date	Jan 17, 2017

Abstract

Methods and systems may include logic to identify a plurality of blocks in image data having one or more top-left dependent pixels, and select the plurality of blocks in a wavefront order for processing. In addition, the logic may process a plurality of pixels in each block in the wavefront order. The system may also include a display device to output a result associated with processing the plurality of pixels.

First claim

I claim:

1. A method comprising:
receiving image data in a graphics processing unit, wherein the image data is associated with a graphics application and includes one or more top-left dependent pixels;
identifying a plurality of blocks in the image data;
selecting the plurality of blocks in a wavefront order for processing;
partitioning each block into an upper left section and a lower right section of equal portions, wherein the upper left section and the lower right section each use a vector reference; and
processing a plurality of pixels in each block in the wavefront order, wherein processing the plurality of pixels includes dispatching one or more parallel processing instructions to parallel process the upper left section and the lower right section using the vector reference.

2. The method of claim 1 , wherein the one or more parallel processing instructions include one or more single instruction multiple data (SIMD) instructions.

3. The method of claim 1 , wherein processing the plurality of pixels includes:
altering an offset parameter and a mask parameter to select pixels in the upper left section and the lower right section of each block for processing, wherein the pixels are selected starting from the offset parameter.

4. The method of claim 1 , wherein a media kernel is used to select the plurality of blocks and process the plurality of pixels.

5. The method of claim 4 , wherein the media kernel is a C for media (CM) kernel.

6. A method comprising:
identifying a plurality of blocks in image data having one or more top-left dependent pixels;
selecting the plurality of blocks in a wavefront order for processing;
partitioning each block into an upper left section and a lower right section of equal portions, wherein the upper left section and the lower right section each use a vector reference; and
processing a plurality of pixels in each block in the wavefront order, wherein processing the plurality of pixels includes dispatching one or more parallel processing instructions to parallel process the upper left section and the lower right section using the vector reference.

7. The method of claim 6 , wherein the one or more parallel processing instructions include one or more single instruction multiple data (SIMD) instructions.

8. The method of claim 6 , wherein processing the plurality of pixels includes:
altering an offset parameter and a mask parameter to select pixels in the upper left section and the lower right section of each block for processing, wherein the pixels are selected starting from the offset parameter.

9. The method of claim 6 , wherein a media kernel is used to select the plurality of blocks and process the plurality of pixels.

10. The method of claim 9 , wherein the media kernel is a C for media (CM) kernel.

11. The method of claim 6 , further including receiving the image data in a graphics processing unit.

12. The method of claim 6 , wherein the image data is associated with a graphics application.

13. A system comprising:
logic to,
identify a plurality of blocks in image data having one or more top-left dependent pixels,
select the plurality of blocks in a wavefront order for processing,
partition each block into an upper left section and a lower right section of equal portions, wherein the upper left section and the lower right section each use a vector reference,
process a plurality of pixels in each block in the wavefront order, by dispatching one or more parallel processing instructions to parallel process the upper left section and the lower right section using the vector reference, and
a display device to output a result associated with processing the plurality of pixels.

14. The system of claim 13 , wherein the one or more parallel processing instructions are to include one or more single instruction multiple data (SIMD) instructions.

15. The system of claim 13 , wherein the logic is to,
alter an offset parameter and a mask parameter to select pixels in the upper left section and the lower right section of each block for processing, wherein the pixels are selected starting from the offset parameter.

16. The system of claim 13 , wherein the logic includes media kernel logic.

17. The system of claim 16 , wherein the media kernel logic is C for media (CM) kernel logic.

18. The system of claim 13 , further including:
a central processing unit (CPU); and
a graphics processing unit (GPU), wherein the logic is to receive the image data in the GPU.

19. The system of claim 13 , wherein the image data is to be associated with a graphics application.

20. At least one non-transitory computer readable storage medium comprising a set of instructions which, if executed by a processor, cause a computer to:
identify a plurality of blocks in image data having one or more top-left dependent pixels;
select the plurality of blocks in a wavefront order for processing;
partition each block into an upper left section and a lower right section of equal portions, wherein the upper left section and the lower right section each use a vector reference; and
process a plurality of pixels in each block in the wavefront order, by dispatching one or more parallel processing instructions to parallel process the upper left section and the lower right section using the vector reference.

21. The medium of claim 20 , wherein the one or more parallel processing instructions are to include one or more single instruction multiple data (SIMD) instructions.

22. The medium of claim 20 , wherein the set of instructions, if executed, cause a computer to:
alter an offset parameter and a mask parameter to select pixels in the upper left section and the lower right section of each block for processing, wherein the pixels are selected starting from the offset parameter.

23. The medium of claim 20 , wherein the set of instructions include media kernel instructions.

24. The medium of claim 23 , wherein the media kernel instructions include C for media (CM) kernel instructions.

25. The medium of claim 20 , wherein the set of instructions, if executed, cause a computer to receive the image data in a graphics processing unit (GPU).

26. The medium of claim 20 , wherein the image data is to be associated with a graphics application.

Assignees

• Yuan Hao
• Intel Corp

Inventors

• Yuan Hao

Classifications

• G06T1/20 Primary

  Processor architectures; Processor configuration, e.g. pipelining · CPC title

• G06F15/8007

  single instruction multiple data [SIMD] multiprocessors · CPC title

• G06T2200/28

  involving image processing hardware · CPC title