CN106709861A - Dye device drive static reconstruction method - Google Patents
Dye device drive static reconstruction method Download PDFInfo
- Publication number
- CN106709861A CN106709861A CN201611140690.6A CN201611140690A CN106709861A CN 106709861 A CN106709861 A CN 106709861A CN 201611140690 A CN201611140690 A CN 201611140690A CN 106709861 A CN106709861 A CN 106709861A
- Authority
- CN
- China
- Prior art keywords
- module
- optimization
- atomic
- code
- atomic section
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
Abstract
The invention belongs to the field of computer graphics, and in particular relates to a dye device drive static reconstruction method. The method comprises the steps that a drive program atomic segment division module (1) divides a drive code into atomic segments and issues the atomic segments to an atomic segment program reconstruction module (3); the atomic segment program reconstruction module (3) extracts the corresponding atomic segment according to configured function parameters to reconstruct and generate a needed software code, and sends the software code to an instruction optimization module (4); the instruction optimization module (4) receives the software code sent by the atomic segment program reconstruction module (3), optimizes the software code, and sends a drive program to a machine code generation module (5); and the machine code generation module (5) generates a corresponding machine code. According to the method, dye device drive code atomic segment division is carried out; for the need of different scenes configured by a user, static reconstruction is carried out to generate the relevant code; invalid codes are deleted; a dye device drive program is optimized; and the operation performance of a dye device is improved.
Description
Technical field
Static reconfiguration method is driven the invention belongs to area of computer graphics, more particularly to a kind of stainer.
Background technology
Stainer is driven as the core of graphic process unit, and its operational efficiency directly determines the property of graphic process unit
Energy.Existing graphic process unit realized using programmable stainer array format on a large scale mostly, do not carry out Module Division and
Related Optimization Work, causes stainer driver complexity, redundancy, as the bottleneck for improving graphic process unit performance.
The content of the invention
The purpose of the present invention is：
Static reconfiguration method is driven present invention generally provides a kind of stainer, stainer driver is optimized, so as to carry
The performance of graphic process unit high.
Solution of the invention is：
A kind of stainer drives static reconfiguration method, including：
Step 1, driver atomic section division module (1) will drive code to be divided into most basic atomic section, will generate
Atomic section be handed down to atomic section program restructuring module (3)；
Step 2, in user function configuration module (2), the static specified functional parameter to be used of user；
The functional parameter that step 3, atomic section program restructuring module (3) are configured according to user function configuration module (2), carries
Corresponding atomic section in driver atomic section division module (1), the software code needed for reconstruct, generation are taken, and is sent to finger
Make optimization module (4)；
Step 4, optimization module (4) receive the software code that atomic section program restructuring module (3) sends, and carry out data
Relevance optimization, structural dependence optimization, the driver after the completion of optimization is sent to machine code generation module (5)；
Step 5, machine code generation module (5) receive the driver of optimization module (4), generate corresponding machine
Code.
Data dependence described in step 4 optimization module (4) optimizes：Do not exist in close a plurality of instruction
Write after write, read-after-write, the limitation of writeafterread, in multi-emitting mechanism, a plurality of instruction can be performed in synchronization.
Structural dependence described in step 4 optimization module (4) optimizes：Close a plurality of instruction is with for the moment
Quarter can perform in different arithmetic elements.
It is an advantage of the invention that：A kind of stainer that the present invention is provided drives static reconfiguration method, is driven by by stainer
Dynamic program is divided into atomic section, further according to the functional parameter of user's static configuration, extracts corresponding atomic section, completes static weight
Structure, so as to reject redundant code, realizes the optimization of stainer driving, improves the performance of graphic process unit.
Brief description of the drawings
Fig. 1 is method of the present invention module map.
Specific embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
Technical scheme is described in further detail with specific embodiment below in conjunction with the accompanying drawings.
As shown in figure 1, a kind of stainer of the embodiment of the present invention drives static reconfiguration method, including：
Step 1, driver atomic section division module (1) will drive code to be divided into most basic atomic section, will generate
Atomic section be handed down to atomic section program restructuring module (3)；
Step 2, in user function configuration module (2), the static specified functional parameter to be used of user；
The functional parameter that step 3, atomic section program restructuring module (3) are configured according to user function configuration module (2), carries
Corresponding atomic section in driver atomic section division module (1), the software code needed for reconstruct, generation are taken, and is sent to finger
Make optimization module (4)；
Step 4, optimization module (4) receive the software code that atomic section program restructuring module (3) sends, and carry out data
Relevance optimization, structural dependence optimization, the driver after the completion of optimization is sent to machine code generation module (5)；
Step 5, machine code generation module (5) receive the driver of optimization module (4), generate corresponding machine
Code.
Data dependence described in step 4 optimization module (4) optimizes：Do not exist in close a plurality of instruction
Write after write, read-after-write, the limitation of writeafterread, in multi-emitting mechanism, a plurality of instruction can be performed in synchronization.
Structural dependence described in step 4 optimization module (4) optimizes：Close a plurality of instruction is with for the moment
Quarter can perform in different arithmetic elements.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
The present invention is explained with reference to the foregoing embodiments, it will be understood by those within the art that：It still may be used
Modified with to the technical scheme described in foregoing embodiments, or equivalent is carried out to which part technical characteristic；
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (3)
1. a kind of stainer drives static reconfiguration method, it is characterised in that including：
Step 1, driver atomic section division module (1) will drive code to be divided into most basic atomic section, the original that will be generated
Subsegment is handed down to atomic section program restructuring module (3)；
Step 2, in user function configuration module (2), the static specified functional parameter to be used of user；
The functional parameter that step 3, atomic section program restructuring module (3) are configured according to user function configuration module (2), extracts and drives
Corresponding atomic section in dynamic program atomic section division module (1), the software code needed for reconstruct, generation, and it is excellent to be sent to instruction
Change module (4)；
Step 4, optimization module (4) receive the software code that atomic section program restructuring module (3) sends, and carry out data related
Property optimization, structural dependence optimization, the driver after the completion of optimization is sent to machine code generation module (5)；
Step 5, machine code generation module (5) receive the driver of optimization module (4), generate corresponding machine code.
2. a kind of stainer as claimed in claim 1 drives static reconfiguration method, it is characterised in that step 4 optimization mould
Data dependence described in block (4) optimizes：In the absence of write after write, read-after-write, writeafterread in close a plurality of instruction
Limitation, in multi-emitting mechanism, a plurality of instruction can be performed in synchronization.
3. a kind of stainer as claimed in claim 1 drives static reconfiguration method, it is characterised in that step 4 optimization mould
Structural dependence described in block (4) optimizes：Close a plurality of instruction can be in different arithmetic elements in synchronization
Perform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611140690.6A CN106709861B (en) | 2016-12-12 | 2016-12-12 | Stainer driving static reconstruction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611140690.6A CN106709861B (en) | 2016-12-12 | 2016-12-12 | Stainer driving static reconstruction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106709861A true CN106709861A (en) | 2017-05-24 |
| CN106709861B CN106709861B (en) | 2020-08-11 |
Family
ID=58936875
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611140690.6A Active CN106709861B (en) | 2016-12-12 | 2016-12-12 | Stainer driving static reconstruction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106709861B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109726816A (en) * | 2018-12-12 | 2019-05-07 | 中国航空工业集团公司西安航空计算技术研究所 | A kind of assembly level stainer program chains optimization method |
| CN109800088A (en) * | 2018-11-14 | 2019-05-24 | 西安翔腾微电子科技有限公司 | Based on trained GPU configuring management method, device, storage medium and GPU |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8595439B1 (en) * | 2007-09-28 | 2013-11-26 | The Mathworks, Inc. | Optimization of cache configuration for application design |
| CN105549932A (en) * | 2015-12-11 | 2016-05-04 | 中国航空工业集团公司西安航空计算技术研究所 | Graphic processor host driver software structure |
| CN105574807A (en) * | 2015-12-11 | 2016-05-11 | 中国航空工业集团公司西安航空计算技术研究所 | Development platform of programmable stainer embedded into graphics processor |
-
2016
- 2016-12-12 CN CN201611140690.6A patent/CN106709861B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8595439B1 (en) * | 2007-09-28 | 2013-11-26 | The Mathworks, Inc. | Optimization of cache configuration for application design |
| CN105549932A (en) * | 2015-12-11 | 2016-05-04 | 中国航空工业集团公司西安航空计算技术研究所 | Graphic processor host driver software structure |
| CN105574807A (en) * | 2015-12-11 | 2016-05-11 | 中国航空工业集团公司西安航空计算技术研究所 | Development platform of programmable stainer embedded into graphics processor |
Non-Patent Citations (1)
| Title |
|---|
| 邓军勇 等: "MIGPU-9多核交互式图形处理器的设计", 《计算机辅助设计与图形学学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109800088A (en) * | 2018-11-14 | 2019-05-24 | 西安翔腾微电子科技有限公司 | Based on trained GPU configuring management method, device, storage medium and GPU |
| CN109800088B (en) * | 2018-11-14 | 2023-06-20 | 西安翔腾微电子科技有限公司 | Training-based GPU configuration management method and device, storage medium and GPU |
| CN109726816A (en) * | 2018-12-12 | 2019-05-07 | 中国航空工业集团公司西安航空计算技术研究所 | A kind of assembly level stainer program chains optimization method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106709861B (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103135976B (en) | code automatic generation method and device | |
| CN1313923C (en) | A software distribution method and system | |
| CN104461484B (en) | The implementation method and device of front-end template | |
| CN105512425B (en) | A kind of IO PAD layout construction method based on graphical interfaces | |
| Nicolae et al. | Deepfreeze: Towards scalable asynchronous checkpointing of deep learning models | |
| CN111666071B (en) | Method and system for resisting network transplantation and optimization based on target many-core | |
| CN104160397A (en) | Location independent files | |
| CN104375875B (en) | Method and compiler for the compiling optimization of application program | |
| CN103885942B (en) | A kind of rapid translation device and method | |
| CN106709861A (en) | Dye device drive static reconstruction method | |
| CN107122190A (en) | A kind of multilingual static page generation method, server and storage medium | |
| CN106446019A (en) | Software function processing method and device | |
| US11934866B2 (en) | Operator operation scheduling method and apparatus to determine an optimal scheduling policy for an operator operation | |
| JP2008186244A (en) | Program development device, program development method and program | |
| CN102629204A (en) | Generating method and device for webpage source code | |
| WO2022142106A1 (en) | Text analysis method and apparatus, electronic device, and readable storage medium | |
| US9235388B2 (en) | Multi-dimensional, multi-configuration compilation phase output visualization technique | |
| CN101169711A (en) | Data conversion system and method | |
| CN103605514B (en) | A kind of processing method and processing device of front-end template | |
| CN110210010A (en) | PSD file editing method, electronic equipment based on demixing technology realization | |
| CN103593277A (en) | Log processing method and system | |
| CN102999488B (en) | Transcription platform system and its implementation | |
| CN102541824B (en) | A kind of method and system in order to realize document amendment | |
| CN104574451A (en) | Method and device for character replacing by graph in AutoCAD | |
| CN105867847A (en) | Memory access control method, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |