CN107168848B - Multithreading analysis and operation visualization method - Google Patents

Multithreading analysis and operation visualization method Download PDF

Info

Publication number
CN107168848B
CN107168848B CN201710295198.4A CN201710295198A CN107168848B CN 107168848 B CN107168848 B CN 107168848B CN 201710295198 A CN201710295198 A CN 201710295198A CN 107168848 B CN107168848 B CN 107168848B
Authority
CN
China
Prior art keywords
calling
analysis
sub
function
thread
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.)
Active
Application number
CN201710295198.4A
Other languages
Chinese (zh)
Other versions
CN107168848A (en
Inventor
林禹
林畅辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujian Sinoregal Software Co ltd
Original Assignee
Fujian Sinoregal Software Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Fujian Sinoregal Software Co ltd filed Critical Fujian Sinoregal Software Co ltd
Priority to CN201710295198.4A priority Critical patent/CN107168848B/en
Publication of CN107168848A publication Critical patent/CN107168848A/en
Application granted granted Critical
Publication of CN107168848B publication Critical patent/CN107168848B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
    • G06F11/3017—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is implementing multitasking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3051—Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/32—Monitoring with visual or acoustical indication of the functioning of the machine
    • G06F11/324—Display of status information

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Stored Programmes (AREA)

Abstract

The invention provides a multithreading analysis and operation visualization method, which comprises the following steps: running a source program, and collecting a thread ID and a running stack of the source program; separating the collected runtime stacks according to the thread IDs, so that each thread ID corresponds to one sub runtime stack; counting calling data of each function in each sub-runtime stack, wherein the calling data comprises an attribution module, calling times, calling depth and calling span; identifying and filtering the basic class function in each sub runtime stack according to an attribution module; further filtering each sub-runtime stack according to a preset calling number value, a preset calling depth value and a preset calling span value; and visually displaying each sub-runtime stack to obtain a function call relation diagram corresponding to each thread. The invention realizes the screening analysis and visual display of the multithread operation data, provides an effective source program analysis basis for engineers, and improves the software development efficiency.

Description

Multithreading analysis and operation visualization method
Technical Field
The invention relates to the technical field of data analysis, in particular to a multithread analysis and operation visualization method.
Background
At present, many projects have more source code lines, and are executed in parallel by multiple threads, and if the source code needs to be changed on the basis, a required function needs to be searched for and modified in a targeted manner, but searching for a certain function or analyzing a code framework thereof under the condition of more source code lines is not an easy matter, and is mainly realized by artificial look-up and analysis.
The existing tool for analyzing stack records in runtime mainly relates to single-thread program analysis, cannot realize analysis of multi-thread programs, and is only suitable for analysis of a small number of code lines. The existing analysis method for the single-thread program commonly utilizes open-source software grapeviz to realize function call visualization, and provides an engineer with an intuitive function call relation diagram, which specifically comprises the following steps: the method comprises the steps of firstly obtaining a source program running log through a compiler tool chain (such as a GNU compiler tool chain, and forming a proxy entry for all function calls of a source code through a parameter definition-function), wherein the running log comprises function tracking data (function addresses and running stacks) of the source program, then processing the function tracking data by using a customized intermediate code (such as a pvtrace tool), generating a matrix format, mapping the obtained function addresses into function names by using an address mapping tool (such as an Addr2line tool) so as to facilitate reading of a user, and finally performing graph drawing on the function call relation of the whole source program by using an open-source grapeviz tool. However, in practical applications, this method has limited contribution to log analysis, and when a multi-threaded program or a large project with more than twenty-ten thousand lines is involved, the data size is too large and complex, and effective data cannot be extracted for highlighting, so that engineers are very arduous to analyze. In view of this, it is very important to develop a method for multi-thread analysis and visualization.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a multithreading analysis and operation visualization method, which is used for realizing the screening analysis and visualization display of multithreading operation data, providing an effective source program analysis basis for engineers and improving the software development efficiency.
The invention is realized by the following steps: a multithread analysis and operation visualization method comprises the following steps:
step 1, running a source program, and collecting a thread ID and a running stack of the source program;
step 2, separating the collected runtime stacks according to the thread IDs, so that each thread ID corresponds to one sub runtime stack;
step 3, counting the calling data of each function in each sub-runtime stack, wherein the calling data comprises an attribution module, calling times, calling depth and calling span;
step 4, identifying and filtering the basic class function in each sub-runtime stack according to the attribution module;
step 5, further filtering each sub-runtime stack according to a preset calling number value, a preset calling depth value and a preset calling span value;
and 6, visually displaying each sub-runtime stack to obtain a function call relation diagram corresponding to each thread for analysis and review by an engineer.
Further, the thread ID is obtained by directly calling the operating system thread api by an external program.
Further, the runtime stack is obtained by adding a flag at the compiler, followed by a proxy function.
Furthermore, the attribution module comprises a memory allocation module, a file operation module and a lock module.
Further, the step 6 specifically includes: and drawing a required function call analysis graph by using open source grapeviz software by extracting a file name, a method name, a directory, call times and a function call sequence corresponding to the function in each sub-runtime stack.
Further, important function nodes are identified through colors in the function call analysis graph, and the important function nodes are function nodes with span larger than a threshold value.
The invention has the following advantages: the method comprises the steps of separating multithreading runtime stacks, screening and filtering functions which do not need to be displayed according to user requirements by setting filtering parameters (an attribution module, a calling number value, a calling depth value and a calling span value), obtaining filtered sub runtime stacks, and visually displaying the simplified runtime stacks, so that clear and concise analysis graphs are provided for users, the users can conveniently look up the analysis graphs.
Drawings
The invention will be further described with reference to the following examples with reference to the accompanying drawings.
Fig. 1 is an execution flow chart of a multithread analysis and operation visualization method according to the present invention.
FIG. 2 is a diagram illustrating analysis of function calls according to an embodiment of the present invention.
Detailed Description
As shown in fig. 1, a method for visualizing multithread analysis and operation includes the following steps:
step 1, running a source program, and acquiring a thread ID and a runtime stack of the source program, wherein the thread ID is obtained by directly calling an operating system thread api through an external program, and the runtime stack is obtained by adding a mark in a compiler (such as a GNU compiler) and then through a proxy function;
step 2, separating the collected runtime stacks according to the thread IDs, so that each thread ID corresponds to one sub runtime stack;
step 3, counting calling data of each function in each sub-runtime stack, wherein the calling data comprises an attribution module, calling times, calling depth and calling span, and the attribution module comprises a memory allocation module, a file operation module and a lock module;
step 4, identifying and filtering the basic class function in each sub-runtime stack according to the attribution module;
step 5, further filtering each sub-runtime stack according to a preset calling number value, a preset calling depth value and a preset calling span value;
and 6, visually displaying each sub-runtime stack to obtain a function call relation diagram corresponding to each thread for an engineer to analyze and look up, wherein the step 6 specifically comprises the following steps: the file name, the method name, the directory, the calling times and the function calling sequence corresponding to the function in each sub-runtime stack are extracted, the required function calling analysis graph is drawn by utilizing open source grapeviz software, important function nodes are identified in the function calling analysis graph through colors, and the important function nodes are function nodes with span (namely the number of the included primary sub-functions) larger than a threshold value.
The invention is further illustrated below with reference to a specific embodiment:
the invention discloses a multithreading analysis and operation visualization method, which is used for analyzing a source program comprising three mutually switched threads, and comprises the following specific steps:
step 1, running a source program, adding a mark in a compiler (such as a GNU compiler), then obtaining a run-time stack through an agent, and directly calling an operating system thread api through an external program to obtain a thread ID of the source program, for example, in the embodiment, the source program has 4 threads in total, that is, 4 thread IDs are obtained, which are 1 main thread ID (000) and 3 sub-thread IDs (001, 002 and 003), respectively;
step 2, separating the collected runtime stacks according to the thread IDs, so that each thread ID corresponds to one sub runtime stack, namely the runtime stack of the source program can be separated into 4 sub runtime stacks;
step 3, counting calling data of each function in each sub-runtime stack, wherein the calling data comprises an attribution module, calling times, calling depth and calling span, and the attribution module mainly comprises a memory allocation module, a file operation module, a lock module and the like and is a frequently called system basic module;
step 4, identifying and filtering the basic function in each sub-runtime stack according to the attribution module, namely identifying and filtering frequently called system basic modules such as a memory allocation module, a file operation module and a lock module;
step 5, further filtering each sub-runtime stack according to a preset calling number value, a preset calling depth value and a preset calling span value;
and 6, extracting a file name, a method name, a directory, calling times and a function calling sequence corresponding to the function in each sub-runtime stack after the two times of filtering, and drawing a required function calling analysis graph by using open source grapeviz software, for example, taking a calling time value for filtering as 100, a calling depth value as 4, and a calling span value as 0, so as to obtain the function calling analysis graphs of the four threads as shown in fig. 2. In the function call analysis graph, function nodes are represented by file names, method names and directories, the call relation between the two function nodes is represented by line segments with arrows, and the call sequence is visually identified by marking numbers on the line segments so that a user can quickly search and locate, wherein the function call analysis graph also identifies important function nodes, and the important function nodes are function nodes (generally 6) with span (namely the number of the included primary sub-functions) larger than a threshold value.
The attribution module, the calling times, the calling depth, the calling span and the threshold value in the step 6 can be adjusted according to the needs of the user, so that functions which do not need to be displayed or are redundant are selectively hidden, a clear and key function calling analysis graph is provided for the user, and the positioning query and analysis efficiency of functions in the later stage of an engineer is improved.
The method comprises the steps of collecting multithreading runtime stacks, separating the runtime stacks according to thread IDs to obtain a plurality of sub runtime stacks, setting filtering parameters (an attribution module, a calling number value, a calling depth value and a calling span value) according to user requirements to screen and filter functions which do not need to be displayed to obtain the filtered sub runtime stacks, and respectively displaying the threads in a visualized mode after filtering to provide clear and concise analysis graphs for users so as to facilitate the users to read.
Although specific embodiments of the invention have been described above, it will be understood by those skilled in the art that the specific embodiments described are illustrative only and are not limiting upon the scope of the invention, and that equivalent modifications and variations can be made by those skilled in the art without departing from the spirit of the invention, which is to be limited only by the appended claims.

Claims (5)

1. A multithread analysis and operation visualization method is characterized in that: the method comprises the following steps:
step 1, running a source program, and collecting a thread ID and a running stack of the source program;
step 2, separating the collected runtime stacks according to the thread IDs, so that each thread ID corresponds to one sub runtime stack;
step 3, counting the calling data of each function in each sub-runtime stack, wherein the calling data comprises an attribution module, calling times, calling depth and calling span;
step 4, identifying and filtering the basic class function in each sub-runtime stack according to the attribution module, wherein the attribution module comprises a memory allocation module, a file operation module and a lock module;
step 5, further filtering each sub-runtime stack according to a preset calling number value, a preset calling depth value and a preset calling span value;
and 6, visually displaying each sub-runtime stack to obtain a function call relation diagram corresponding to each thread for analysis and review by an engineer.
2. A multithread analysis and execution visualization method as recited in claim 1, wherein: the thread ID is obtained by an external program directly calling the operating system thread api.
3. A multithread analysis and execution visualization method as recited in claim 1, wherein: the runtime stack is obtained by adding a flag at the compiler, followed by a proxy function.
4. A multithread analysis and execution visualization method as recited in claim 1, wherein: the step 6 specifically comprises the following steps: and drawing a required function call analysis graph by using open source grapeviz software by extracting a file name, a method name, a directory, call times and a function call sequence corresponding to the function in each sub-runtime stack.
5. A multithread analysis and execution visualization method as recited in claim 4, wherein: and identifying important function nodes through colors in the function call analysis graph, wherein the important function nodes are function nodes with span larger than a threshold value.
CN201710295198.4A 2017-04-28 2017-04-28 Multithreading analysis and operation visualization method Active CN107168848B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710295198.4A CN107168848B (en) 2017-04-28 2017-04-28 Multithreading analysis and operation visualization method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710295198.4A CN107168848B (en) 2017-04-28 2017-04-28 Multithreading analysis and operation visualization method

Publications (2)

Publication Number Publication Date
CN107168848A CN107168848A (en) 2017-09-15
CN107168848B true CN107168848B (en) 2020-06-09

Family

ID=59812573

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710295198.4A Active CN107168848B (en) 2017-04-28 2017-04-28 Multithreading analysis and operation visualization method

Country Status (1)

Country Link
CN (1) CN107168848B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109710396B (en) * 2017-10-26 2023-08-22 华为技术有限公司 Method and device for information acquisition and memory release
CN109831351B (en) * 2018-12-29 2021-03-05 深圳云天励飞技术有限公司 Link tracking method, device, terminal and storage medium
CN116661987A (en) * 2022-12-29 2023-08-29 荣耀终端有限公司 Memory application method and electronic equipment

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105630471A (en) * 2014-11-03 2016-06-01 广州市动景计算机科技有限公司 Code tracking method and system based on GCC and Chromium

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120180053A1 (en) * 2011-01-10 2012-07-12 International Business Machines Corporation Call stack aggregation and display
US20150199262A1 (en) * 2014-01-16 2015-07-16 Vivek Bhavsar Runtime code visualization

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105630471A (en) * 2014-11-03 2016-06-01 广州市动景计算机科技有限公司 Code tracking method and system based on GCC and Chromium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《基于推测多线程技术的程序可视化方法与实现》;赵永刚，李美蓉，李保红，韩博;《西安邮电学院学报》;20100930;第15卷(第5期);69-74页 *

Also Published As

Publication number Publication date
CN107168848A (en) 2017-09-15

Similar Documents

Publication Publication Date Title
US8627290B2 (en) Test case pattern matching
US8359292B2 (en) Semantic grouping for program performance data analysis
US9747190B2 (en) Analysis system, analysis method, and computer program product
US9003240B2 (en) Blackbox memory monitoring with a calling context memory map and semantic extraction
CN107045478B (en) lightweight thread analysis and operation visualization method
US9176840B2 (en) Tool for analyzing and resolving errors in a process server
CN107168848B (en) Multithreading analysis and operation visualization method
US8601442B2 (en) Marker correlation of application constructs with visualizations
Adamoli et al. Trevis: A context tree visualization & analysis framework and its use for classifying performance failure reports
CN109284102B (en) Information acquisition method and device
US10365995B2 (en) Composing future application tests including test action data
CN113688288A (en) Data association analysis method and device, computer equipment and storage medium
CN108399321B (en) Software local plagiarism detection method based on dynamic instruction dependence graph birthmark
EP2951680B1 (en) Acquiring identification of an application lifecycle management entity associated with similar code
JP4928848B2 (en) Message converter in computer system integrated management environment.
Revelle et al. Understanding concerns in software: insights gained from two case studies
US8701086B2 (en) Simplifying analysis of software code used in software systems
US11119899B2 (en) Determining potential test actions
JP2007241426A (en) Specification operation support system for failure factor of application
US20130307854A1 (en) Method and System for Visualising a System Model
US10042638B2 (en) Evaluating documentation coverage
CN107330031B (en) Data storage method and device and electronic equipment
US8819626B2 (en) Sharable development environment bookmarks for functional/data flow
Gobert et al. Understanding schema evolution as a basis for database reengineering
CN115510361A (en) Data chart page generation method based on visual configuration and related equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP02 Change in the address of a patent holder

Address after: 350000 21 / F, building 5, f District, Fuzhou Software Park, 89 software Avenue, Gulou District, Fuzhou City, Fujian Province

Patentee after: FUJIAN SINOREGAL SOFTWARE Co.,Ltd.

Address before: 350000, No. 5, building F, zone 20-21, Fuzhou Software Park, 89 software Avenue, Gulou District, Fujian, Fuzhou

Patentee before: FUJIAN SINOREGAL SOFTWARE Co.,Ltd.

CP02 Change in the address of a patent holder