# US 2008/0313434 A1 · Rendering Processing Apparatus, Parallel Processing Apparatus, and Exclusive Control Method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2008/0313434 A1
- Document ID: US20080313434A1
- Status: Published application
- Kind: A1
- Application number: 11/793,640
- Filed: 2006-07-25
- Published: 2008-12-18
- Earliest claimed: 2005-10-31
- Assignees: Sony Computer Entertainment Inc.
- Inventors: Junichi Naoi
- Web page: https://patentgrep.com/patent/US20080313434A1

## CPC

- G06T 15/005: {General purpose rendering architectures}
- G06F 9/526: {Mutual exclusion algorithms}
- G06F 12/0875: with dedicated cache, e.g. instruction or stack
- G09G 5/001: {Arbitration of resources in a display system, e.g. control of access to frame buffer by video controller and/or main processor}
- G09G 5/393: Arrangements for updating the contents of the bit-mapped memory
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G09G: ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION (arrangements for transferring data between digital computers and displays G06F3/14; static indicating arrangements comprising an association of a number of separate sources or light control cells G09F9/00; static indicating arrangements comprising integral associations of a number of light sources H01J, H01K, H10H, H05B33/12; scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission, details thereof H04N1/00)

## Abstract

A DDA 34 notifies the coordinates of a rasterized pixel to an exclusive control part 40, acquires a unique identification number associated with the pixel position from the exclusive control part 40, and adds the identification number to pixel data and supplies it to a shader 20. A plurality of shader pipes 22 in the shader 20 perform arithmetic processing for rendering pixels in parallel, and writes the processing results to a frame buffer 50. When performing an arithmetic instruction that requires exclusive control over a pixel, each shader pipe 22 issues a request to lock the pixel by notifying the identification information added to that pixel to the exclusive control part 40. If the lock request is accepted, the shader pipe 22 performs the arithmetic processing on that pixel. If the lock request is rejected, the shader pipe suspends and puts the arithmetic processing on that pixel into a wait state, and executes arithmetic processing on another pixel in the interim.

## Claims

### Claim 1

A rendering processing apparatus comprising:
a rasterizer which rasterizes a rendering primitive in succession to generate a plurality of pixels;
a shader having a plurality of arithmetic processing parts which perform arithmetic processing on the plurality of pixels in parallel; and
an exclusive control part which exercises exclusive control over arithmetic processing on an identical pixel performed by the plurality of arithmetic processing parts, the exclusive control part being operable to:
update a value of an execution order counter indicating order of execution of arithmetic processing each time the rasterizer generates a pixel;
assign the updated value of the execution order counter to the generated pixel; and
accept, upon receiving from the plurality of arithmetic processing parts requests to lock a target pixel subject to arithmetic processing, the lock requests in order of the value of the execution order counter assigned to the target pixel, whereby the order of execution of the arithmetic processing on the identical pixel performed by the plurality of arithmetic processing parts is controlled.

### Claim 2 (depends on 1)

The rendering processing apparatus as recited in claim 1, wherein the exclusive control part has a cache memory which stores the execution order counter and an execution counter in association with pixel coordinates, the value of the execution order counter being updated each time the pixel is generated, and the value of the execution counter being updated each time the pixel is actually processed by the arithmetic processing part.

### Claim 3 (depends on 2)

The rendering processing apparatus as recited in claim 2, wherein the exclusive control part is operable to accept the lock request from the arithmetic processing part if the value of the execution order counter assigned to the target pixel coincides with the value of the execution counter stored in the cache memory.

### Claim 4

A parallel processing apparatus comprising:
a plurality of arithmetic processing parts which perform arithmetic processing on a plurality of processing units in parallel;
an assignment part which assigns to each processing unit in advance a value of an execution order counter indicating order of execution of arithmetic processing expected to be performed by the plurality of arithmetic processing parts; and
an exclusive control part which is operable to:
receive requests to lock a target processing unit subject to arithmetic processing from any of the arithmetic processing parts for actually performing the arithmetic processing; and
accept the lock requests in order of the value of the execution order counter assigned to the target processing unit, whereby the order of execution of the arithmetic processing on the identical processing unit performed by the plurality of arithmetic processing parts is controlled.

### Claim 5 (depends on 4)

The parallel processing apparatus recited in claim 4, further comprising a cache memory which stores the value of the execution order counter with respect to each processing unit.

### Claim 6 (depends on 5)

The parallel processing apparatus recited in claim 5, wherein the cache memory stores the execution order counter and an execution counter in association with the processing unit, the value of the execution counter being updated each time arithmetic processing expected to be performed on the processing unit by an arithmetic processing part is identified, and the value of the execution counter being updated each time the arithmetic processing on the processing unit is actually performed by the arithmetic processing part.

### Claim 7 (depends on 6)

The parallel processing apparatus recited in claim 6, wherein the exclusive control part is operable to accept, upon receiving from the arithmetic processing part a request to lock a target processing unit subject to arithmetic processing, the lock request if the value of the execution order counter assigned to the target processing unit coincides with the value of the execution counter stored in the cache memory.

### Claim 8

An exclusive control method for providing exclusive control in an apparatus which performs arithmetic processing on a plurality of processing units in parallel, the method comprising:
assigning to each processing unit in advance a value of an execution order counter indicating order of execution of arithmetic processing expected to be performed on the respective processing units;
receiving requests to lock a target processing unit subject to arithmetic processing when the arithmetic processing on the processing units is actually performed; and
accepting the lock requests in order of the value of the execution order counter assigned to the target processing unit, whereby the order of execution of the arithmetic processing on the identical processing unit is controlled.

### Claim 9 (depends on 8)

The exclusive control method recited in claim 8, wherein:
an execution counter is stored in a cache memory, the value of the execution counter being updated each time arithmetic processing is performed on each processing unit;
if the value of the execution order counter assigned to the target pixel coincides with the value of the execution counter stored in the cache memory, the lock request is accepted and the arithmetic processing on the processing unit is put into operation; and
if not coincide, the lock request is rejected and the arithmetic processing on the processing unit is put into a wait state.
