CN100501699C - Method and system for capturing and storing bare nucleus in exception in multi-core processor - Google Patents
Method and system for capturing and storing bare nucleus in exception in multi-core processor Download PDFInfo
- Publication number
- CN100501699C CN100501699C CN 200710181024 CN200710181024A CN100501699C CN 100501699 C CN100501699 C CN 100501699C CN 200710181024 CN200710181024 CN 200710181024 CN 200710181024 A CN200710181024 A CN 200710181024A CN 100501699 C CN100501699 C CN 100501699C
- Authority
- CN
- China
- Prior art keywords
- task
- abnormality processing
- nuclear
- abnormal
- unusual
- 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.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4812—Task transfer initiation or dispatching by interrupt, e.g. masked
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/362—Software debugging
- G06F11/3636—Software debugging by tracing the execution of the program
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/48—Indexing scheme relating to G06F9/48
- G06F2209/481—Exception handling
Abstract
The invention provides a method to capture and save the exception of bare core for a multi-core processor, comprising a plurality of procedures: controlling surface core creates an abnormal protection task; run the abnormal protection task and create a sharing queue between the controlling surface core and a data surface core; run the data core surface and then start the initialization process of abnormal handling and then set an abnormal vector table; mount service routines by setting the abnormal vector table and create an abnormal handling task; in the abnormal event of the data surface core, processor captures abnormal information, calls back the take-over handling of the abnormal service and releases the quantity of abnormal handling signal to inform an abnormal handling task; when the signal quantity is acquired by the abnormal handling task, save and then process the abnormal information and then write the abnormal information together with the result after disposal into the sharing queue; the abnormal protection task polls the sharing queue at the set time; the abnormal information in the sharing queue is taken out and stored in the memory. The invention also provides a system to capture and save the exception of bare core for a multi-core processor.
Description
Technical field
The present invention relates to field of processors, specifically, the method and system of exception catching and storage when unusual take place in its bare nucleus when relating to a kind of polycaryon processor and being applied in the communications field.
Background technology
Along with arrival and various new technology, the new business of information age continues to bring out, the user is for being that the mobile communication or the bandwidth demand of data communication are all more and more higher, in order to satisfy the demand of unique user high bandwidth, the operator requires mobile communication equipment and data communications equipment to have high capacity, the high access and high handling property.In order to reach high capacity, the demand of high access and high handling property, a large amount of polycaryon processors that use high throughput of beginning in mobile communication equipment and the data communications equipment at present.
Polycaryon processor is a processor of concentrating a plurality of nuclears on a slice chip, and each nuclear is equivalent to a CPU.In mobile communication equipment and data communications equipment, the common application model of polycaryon processor is that a part is examined as chain of command and chain of command, and this part is the operation commercial operation system above the nuclear, as the embedded OS Vxworks of Wind River company.Remaining nuclear is handled as data surface and is wrapped processing, in order to improve the bag handling property, as the nuclear of data surface bare nucleus and do not move complete commercial operation system normally.
The chain of command of operation commercial operation system is examined in operational process if unusual, details in the time of can taking place by means of the abnormality processing task catching exception in the operating system, and unusual details of catching are write in the flash storer, for the reason that searches problem behind the chip reset.And as the bare nucleus of data surface because operation system not, do not manage the sheet external equipment and can't direct control flash storer, if take place unusually like this as the bare nucleus of data surface, information that then can't catching exception more can't be saved in abnormal information in the flash storer.
Summary of the invention
The technical matters that the present invention solves has provided in a kind of polycaryon processor bare nucleus exception catching and storage means when unusual has taken place, and solving in the polycaryon processor that the bare nucleus of operation system not takes place when unusual can't catching exception information and can't preserve the problem of abnormal information in the storer.
For addressing the above problem, the method for bare nucleus exception catching and storage may further comprise the steps in a kind of polycaryon processor of the present invention:
The described chain of command of A nuclear its operating system of operation is also created the task of guarding unusually, described guard task run unusually and create chain of command nuclear and data surface nuclear between shared queue;
Carry out the initialization flow process of abnormality processing after the operation of B data surface nuclear, exception vector table is set, articulate unusual service routine, and create the abnormality processing task by exception vector table is set;
After C data surface nuclear took place unusually, processor catching exception information was adjusted back unusual service routine by described exception vector table and is taken over processing, discharged abnormality processing semaphore notice abnormality processing task;
After the described abnormality processing task of D obtains described semaphore, preserve described abnormal information, after handling abnormal information and result are write in the described shared queue;
The described described shared queue of task automatic regular polling of guarding unusually of E, finding has abnormal information in the formation, then takes out to write in the external memory storage.
Above-mentioned method, wherein, in the steps A, the shared queue of guarding unusually between task creation chain of command nuclear and the data surface nuclear specifically comprises:
The described task of guarding is unusually utilized the shared queue between the pre-assigned internal memory establishment of processor chain of command nuclear and the data surface nuclear, and each data surface is checked and answered a shared queue.
Above-mentioned method, wherein, among the step B, the initialization flow process of described abnormality processing specifically comprises:
Initialization one global variable is as the abnormality processing sign;
Initialization one semaphore is as the abnormality processing semaphore.
Above-mentioned method, wherein, among the step C, unusual service routine is taken over to handle and is specifically comprised:
Judge whether the abnormality processing sign is true,, then directly withdraw from processing, otherwise continue to handle if true;
The abnormality processing sign is changed to very;
Preserve the described abnormal information that processor transmits;
Abnormal information is passed to described abnormality processing task.
Above-mentioned method, wherein among the step D, the abnormality processing task is handled specifically and is comprised:
Read described abnormality processing semaphore;
Judge whether the abnormal signal amount is available,, then continue to read if unavailable;
Read unusual task context;
Preserve unusual task context and exception register collection;
Carry out the storehouse analysis, analyze the function calling relationship of unusual task and unusual function name occurs;
The unusual details of analysis and the function name of function calling relationship and appearance are put into shared queue.
Above-mentioned method, wherein, step e specifically comprises:
Unusually guard task automatic regular polling shared queue;
Judge whether abnormal information is arranged in the described shared queue,, then continue automatic regular polling if do not have;
Taking out abnormal information from the formation that abnormal information is arranged writes in the external memory storage.
Above-mentioned method, wherein, the abnormal information in the shared queue comprises the message of unusual mission function call relation, abnormity point function name, unusual time of origin, Exception Type, storehouse, code segment and processing.
The present invention also provides the system of bare nucleus exception catching and storage in a kind of polycaryon processor, and polycaryon processor comprises that chain of command nuclear and data surface examine, and system comprises:
Be used for the operation system and create and to guard the unit of task unusually, described guard task run unusually and create chain of command nuclear and data surface nuclear between shared queue;
Be used for after the operation of data surface nuclear, carrying out the initialization flow process of abnormality processing, exception vector table is set, articulate unusual service routine, and create the unit of abnormality processing task by exception vector table is set;
Be used for after data surface nuclear takes place unusually, processor catching exception information is adjusted back unusual service routine by described exception vector table and is taken over processing, discharges the unit of abnormality processing semaphore notice abnormality processing task;
Be used for after described abnormality processing task obtains described semaphore, preserving described abnormal information, after handling abnormal information and result write the unit in the described shared queue;
Be used in the described described shared queue of task automatic regular polling of guarding unusually, finding has abnormal information in the formation, then takes out the unit that writes in the external memory storage.
Adopt method of the present invention, the abnormal information of catching is write the flash storer and do not need to keep on-the-spot, therefore adopt the product of polycaryon processor of the present invention taking place to continue use by reset mode when non-lethal is unusual, shortening equipment is delayed machine release time so greatly, minimizing improves the satisfaction of user to product to user's influence and user's complaint.After the device reset before unusual details of storing in the flash storer and the function calling relationship that unusual task takes place come analytical equipment to reset unusual basic reason takes place by reading out, therefore also greatly facilitate the maintenance cost of equipment vendors' quick positioning question and minimizing equipment, have excellent economy.
Description of drawings
Fig. 1 is the system diagram of bare nucleus exception catching and storage in the polycaryon processor of the present invention;
Fig. 2 is the method figure of bare nucleus exception catching and storage in the polycaryon processor of the present invention;
Fig. 3 is a unusual service routine processing flow chart among the present invention;
Fig. 4 is an abnormality processing task processing flow chart among the present invention;
Fig. 5 guards the task processing flow chart unusually among the present invention.
Embodiment
Below in conjunction with accompanying drawing the specific embodiment of the present invention is further described.
As shown in Figure 1, be the system diagram of bare nucleus exception catching and storage in the polycaryon processor.The complete commercial operation system of chain of command nuclear operation is created the task of guarding unusually on operating system during the operation of chain of command nuclear.Carry out the initialization flow process of abnormality processing after the operation of data surface nuclear, articulate unusual service routine and create the abnormality processing task by exception vector table is set.Create shared queue between chain of command and the data surface nuclear when guarding task run unusually.Wherein,
Unusually guard task, create the shared queue between chain of command nuclear and the data surface nuclear when being used to move, and be used for the described shared queue of automatic regular polling, finding has unexpected message in the formation, then takes out to write in the external memory storage.
Unusual service routine, be used for processor take place unusual after, described processor is taken over processing, release semaphore notice abnormality processing task;
The abnormality processing task, be used for the picked up signal amount after, preserve abnormal information, abnormity point is handled, abnormal information and result are write in the described shared queue;
Shared queue is used to receive abnormal information and result;
As shown in Figure 2, the method figure of bare nucleus exception catching and storage in the polycaryon processor of the present invention said method comprising the steps of:
Carry out the initialization flow process of abnormality processing after step 203, data surface are examined and moved, articulate unusual service routine and create the abnormality processing task by exception vector table is set.
Step 204, if occur unusually in the polycaryon processor course of work, system acquisition is unusual, enters unusual service routine and carries out.
Step 205, unusual service routine carries out preliminary abnormality processing work, finishes the preservation to unusual task correlation parameter (such as the register context environmental etc.).Then, release semaphore notice abnormality processing task run.
Step 207 is guarded each shared queue of task scan in the chain of command nuclear unusually, if find in the formation abnormal information is arranged, then takes out and writes in the outer flash storer of sheet.
Wherein, in the step 203, the initialization flow process that data surface nuclear carries out abnormality processing comprises:
Initialization one global variable is as the abnormality processing sign, and initial value is false.Be used for preventing introducing new unusual, form nested unusually in abnormality processing;
Initialization one semaphore is used to prevent that as the abnormality processing semaphore abnormality processing task is by other task preemption.
In the step 205, as shown in Figure 3, be unusual service routine processing flow chart in the embodiment of the invention, be specially:
Step 302 judges whether the abnormality processing sign is true, if abnormality processing is masked as very, then directly withdraws from, otherwise changes step 303; Step 303 is changed to the abnormality processing sign very;
Step 304 is preserved the parameter that CPU transmits;
Step 305 discharges the abnormal signal amount, activates the abnormality processing task, and the parameter of preserving in the step 204 is passed to the abnormality processing task;
In the said method step 206, the concrete treatment scheme of abnormality processing task as shown in Figure 4:
Step 401 reads the abnormal signal amount;
Step 402 judges whether the abnormal signal amount is available, if unavailable, step 301 then;
Step 403 reads unusual task context;
Step 404 is preserved unusual task context and exception register collection;
Step 405 is carried out the storehouse analysis, analyzes the function calling relationship of unusual task and unusual function name occurs;
Step 406 is with unusual details and the function calling relationship analyzed and unusual function name occurs and put into shared queue.
Step 202,206 in the said method, the concrete treatment scheme of guarding task unusually be as shown in Figure 5:
Step 502 is guarded task automatic regular polling individual queue unusually
Step 504 is taken out unusual details and the function calling relationship and the abnormity point function name of unusual task occurred and write outside the own sheet in the flash storer from the formation that abnormal information is arranged, go back to step 501;
The above is the preferred embodiments of the present invention only, is not limited to the present invention, and for a person skilled in the art, the present invention can have various changes and variation.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1, the method for bare nucleus exception catching and storage in a kind of polycaryon processor, described polycaryon processor comprise chain of command nuclear and data surface nuclear, it is characterized in that described method comprises:
The described chain of command of A nuclear its operating system of operation is also created the task of guarding unusually, described guard task run unusually and create chain of command nuclear and data surface nuclear between shared queue;
Carry out the initialization flow process of abnormality processing after the operation of B data surface nuclear, exception vector table is set, articulate unusual service routine, and create the abnormality processing task by exception vector table is set;
After C data surface nuclear took place unusually, processor catching exception information was adjusted back unusual service routine by described exception vector table and is taken over processing, discharged abnormality processing semaphore notice abnormality processing task;
After the described abnormality processing task of D obtains described semaphore, preserve described abnormal information, after handling abnormal information and result are write in the described shared queue;
The described described shared queue of task automatic regular polling of guarding unusually of E, finding has abnormal information in the formation, then takes out to write in the external memory storage.
2, method according to claim 1 is characterized in that, in the steps A, the shared queue of guarding unusually between task creation chain of command nuclear and the data surface nuclear specifically comprises:
The described task of guarding is unusually utilized the shared queue between the pre-assigned internal memory establishment of processor chain of command nuclear and the data surface nuclear, and each data surface is checked and answered a shared queue.
3, method according to claim 2 is characterized in that, among the step B, the initialization flow process of described abnormality processing specifically comprises:
Initialization one global variable is as the abnormality processing sign;
Initialization one semaphore is as the abnormality processing semaphore.
4, method according to claim 3 is characterized in that, among the step C, unusual service routine is taken over to handle and specifically comprised:
Judge whether the abnormality processing sign is true,, then directly withdraw from processing, otherwise continue to handle if true;
The abnormality processing sign is changed to very;
Preserve the described abnormal information that processor transmits;
Abnormal information is passed to described abnormality processing task.
5, method according to claim 4 is characterized in that, among the step D, the abnormality processing task is handled specifically and comprised:
Read described abnormality processing semaphore;
Judge whether the abnormal signal amount is available,, then continue to read if unavailable;
Read unusual task context;
Preserve unusual task context and exception register collection;
Carry out the storehouse analysis, analyze the function calling relationship of unusual task and unusual function name occurs;
The unusual details of analysis and the function name of function calling relationship and appearance are put into shared queue.
6, method according to claim 5 is characterized in that, step e specifically comprises:
Unusually guard task automatic regular polling shared queue;
Judge whether abnormal information is arranged in the described shared queue,, then continue automatic regular polling if do not have;
Taking out abnormal information from the formation that abnormal information is arranged writes in the external memory storage.
7, method according to claim 6 is characterized in that, the abnormal information in the described shared queue comprises the message of unusual mission function call relation, abnormity point function name, unusual time of origin, Exception Type, storehouse, code segment and processing.
8, the system of bare nucleus exception catching and storage in a kind of polycaryon processor, described polycaryon processor comprises chain of command nuclear and data surface nuclear, it is characterized in that described system comprises:
Be used for the operation system and create and to guard the unit of task unusually, described guard task run unusually and create chain of command nuclear and data surface nuclear between shared queue;
Be used for after the operation of data surface nuclear, carrying out the initialization flow process of abnormality processing, exception vector table is set, articulate unusual service routine, and create the unit of abnormality processing task by exception vector table is set;
Be used for after data surface nuclear takes place unusually, processor catching exception information is adjusted back unusual service routine by described exception vector table and is taken over processing, discharges the unit of abnormality processing semaphore notice abnormality processing task;
Be used for after described abnormality processing task obtains described semaphore, preserving described abnormal information, after handling abnormal information and result write the unit in the described shared queue;
Be used in the described described shared queue of task automatic regular polling of guarding unusually, finding has abnormal information in the formation, then takes out the unit that writes in the external memory storage.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200710181024 CN100501699C (en) | 2007-09-26 | 2007-09-26 | Method and system for capturing and storing bare nucleus in exception in multi-core processor |
PCT/CN2007/003686 WO2009039696A1 (en) | 2007-09-26 | 2007-12-19 | Method and system for capturing and storing exception of bared core in multi-cores processor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200710181024 CN100501699C (en) | 2007-09-26 | 2007-09-26 | Method and system for capturing and storing bare nucleus in exception in multi-core processor |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101174242A CN101174242A (en) | 2008-05-07 |
CN100501699C true CN100501699C (en) | 2009-06-17 |
Family
ID=39422764
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 200710181024 Expired - Fee Related CN100501699C (en) | 2007-09-26 | 2007-09-26 | Method and system for capturing and storing bare nucleus in exception in multi-core processor |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN100501699C (en) |
WO (1) | WO2009039696A1 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102866951B (en) * | 2012-09-21 | 2015-07-15 | 武汉烽火网络有限责任公司 | Rapid positioning method of internal storage boundary crossing errors of embedded system |
CN111209112A (en) * | 2019-12-31 | 2020-05-29 | 杭州迪普科技股份有限公司 | Exception handling method and device |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7613911B2 (en) * | 2004-03-12 | 2009-11-03 | Arm Limited | Prefetching exception vectors by early lookup exception vectors within a cache memory |
US7536694B2 (en) * | 2004-11-30 | 2009-05-19 | International Business Machines Corporation | Exception handling in a multiprocessor system |
CN100538647C (en) * | 2006-11-13 | 2009-09-09 | 杭州华三通信技术有限公司 | The processing method for service stream of polycaryon processor and polycaryon processor |
-
2007
- 2007-09-26 CN CN 200710181024 patent/CN100501699C/en not_active Expired - Fee Related
- 2007-12-19 WO PCT/CN2007/003686 patent/WO2009039696A1/en active Application Filing
Also Published As
Publication number | Publication date |
---|---|
WO2009039696A1 (en) | 2009-04-02 |
CN101174242A (en) | 2008-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN100543683C (en) | The method and system that process is monitored | |
CN109240886B (en) | Exception handling method, exception handling device, computer equipment and storage medium | |
CN106407031B (en) | A kind of RAM leakage localization method and electronic equipment | |
US8171129B2 (en) | Smart endpoint and smart monitoring system having the same | |
CN101681288A (en) | Providing event-controlled continuous logging for a mobile operating environment | |
CN102203738B (en) | Event queue managing device and event queue managing method | |
CN110442498A (en) | Localization method, device, storage medium and the computer equipment of abnormal data node | |
CN102722431A (en) | Process monitoring method and process monitoring device | |
CN101546279A (en) | Device, system and method for exception processing of embedded device | |
CN100501699C (en) | Method and system for capturing and storing bare nucleus in exception in multi-core processor | |
CN108737786B (en) | Intelligent monitoring system | |
CN103150155A (en) | Data interception method and device | |
CN109388671A (en) | Data processing method, device, equipment and storage medium | |
CN108566528A (en) | A kind of monitor video storage system | |
CN100392606C (en) | Method for leaking memory of positioning virtual operation system | |
CN101645802B (en) | Method and device for controlling contents | |
CN101221512A (en) | Timing method and system used for real-time operating system | |
CN102750202A (en) | Data protection method and device | |
CN101982949B (en) | Detection system of intelligent network equipment and method | |
CN101060683B (en) | Warning information generation method and device for communication system | |
CN108509257B (en) | Message processing method and device based on multithreading | |
CN105930504B (en) | A kind of network management timing file cocurrent processing system and concurrent processing method | |
CN107688481B (en) | Multi-node-supporting KVM virtual machine hiding process detection system | |
CN110958129A (en) | Method, system and device for flow analysis | |
CN109542903B (en) | Data processing method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090617 Termination date: 20160926 |