CN101625656A - Method and device for processing abnormity of PCI system - Google Patents
Method and device for processing abnormity of PCI system Download PDFInfo
- Publication number
- CN101625656A CN101625656A CN200910089083A CN200910089083A CN101625656A CN 101625656 A CN101625656 A CN 101625656A CN 200910089083 A CN200910089083 A CN 200910089083A CN 200910089083 A CN200910089083 A CN 200910089083A CN 101625656 A CN101625656 A CN 101625656A
- Authority
- CN
- China
- Prior art keywords
- target device
- accessed
- pci
- pci bus
- 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.)
- Granted
Links
Images
Abstract
The invention relates to a method and a device for processing the abnormity of a peripheral component interconnect (PCI) system. An abnormity monitoring and isolating device arranged between host equipment in the PCI system and PCI equipment implements the following steps: A. monitoring the access of the host equipment to the PCI equipment used as target equipment through a PCI bus; B. isolating the accessed target equipment when monitoring that abnormity occurs in access; and C. replacing the accessed target equipment to send a target abolition response to the host equipment when the isolated target equipment is accessed. The invention can prevent the whole PCI system from collapsing probably caused by the long-time occupation of the PCI bus, enhance the reliability of the PCI system and reduce the performance loss of the system when the PCI equipment has abnormity to the maximum extent.
Description
Technical field
The present invention relates to network communications technology field, particularly a kind of unusual method and apparatus of pci system of handling.
Background technology
Peripheral component interconnect (PCI, Peripheral Component Interconnect) bus is used widely because of it has advantages such as open architecture, ease for use and extensibility are good.Fig. 1 is a kind of structural representation of pci system, main process equipment is the built-in pci controller of CPU, usually also become the host pci bridge controller, can articulate a plurality of PCI equipment by pci bus in the pci system, this PCI equipment can be that fixed equipment also can be a card insert type equipment, can be that bootstrapping equipment also can be conventional equipment.After main process equipment powers on, at first finish after the basic configuration of self, can scan each PCI equipment by pci bus, and each PCI equipment is disposed accordingly, after configuration was finished, main process equipment just can carry out data interaction by pci bus and each PCI equipment.
In pci system, if certain piece PCI equipment breaks down during as target device, then may cause whole pci system fault and can't operate as normal, for example following problem may appear: one, during main equipment access destination equipment, the target device fault causes after statement transaction and delays to be ready to Data Receiving, after promptly in the 3rd clock period, sending effective DEVSEL signal, after reaching, the 4th cycle do not send effective TRDY signal always, thereby make main process equipment be in the wait always, the while target device can not send retry or signal is abrogated in transaction, thereby cause pci system to be monopolized for a long time by this transaction, bus is hung dead.If two, target device is a bootstrapping equipment, it does not rely on the configuration of main process equipment after powering on, but oneself carries out initial configuration, before finishing initial configuration, this target device is in locking (Locked) state, main process equipment can't conduct interviews to it, will be by retry to its access transaction.If should occur can't finishing initial configuration unusually in initialization procedure by bootstrapping equipment, then main process equipment will be by unlimited retry to its visit of carrying out, thereby causes bus to be taken for a long time.In addition other unusual condition also may occur and cause bus to be taken for a long time, thereby cause the collapse of whole pci system.
Summary of the invention
In view of this, the invention provides and a kind ofly handle the unusual method and apparatus of pci system, thereby avoid causing the whole pci system collapse that bus is taken for a long time to be caused unusually because of pci system.
A kind of unusual method of peripheral component interconnect pci system of handling is provided with the abnormal monitoring spacer assembly between main process equipment in pci system and the PCI equipment, and this abnormal monitoring spacer assembly is carried out following steps:
A, monitor main process equipment by pci bus to visit as the PCI equipment of target device;
B, listen to visit and occur isolating accessed target device when unusual;
C, when segregate target device is accessed, replaces this accessed target device to send target and abrogate and reply to described main process equipment.
A kind of unusual device of pci system of handling, this device is applied to comprise the pci system of main process equipment and PCI equipment, and this device comprises: control module and isolated location;
Described control module, be used to monitor main process equipment by pci bus to visit as the PCI equipment of target device; When the target device of being isolated by described isolated location is accessed, replaces this accessed target device to send target and abrogate and reply to described main process equipment;
Described isolated location is used for listening to visit at described control module and occurs isolating accessed target device when unusual.
As can be seen from the above technical solutions, the present invention makes this abnormal monitoring spacer assembly monitor the visit of target device by the CPI bus main process equipment by between main process equipment in pci system and the PCI equipment abnormal monitoring spacer assembly being set; Listening to visit occurs isolating accessed target device when unusual; When segregate target device is accessed, replaces this accessed target device to abrogate and reply to main process equipment transmission target.This mode can make main process equipment that the access exception of target device in time is found, and by replacing segregate accessed target device to abrogate the mode of replying to main process equipment transmission target, make main process equipment can jump out visit to the unusual target device of this generation, thereby avoid pci bus taken for a long time the whole pci system collapse that may cause, improved the reliability of pci system, and the performance loss of system when having reduced the PCI unit exception to greatest extent.
Description of drawings
Fig. 1 is a kind of structural representation of pci system;
The main method process flow diagram that Fig. 2 provides for the embodiment of the invention;
The detailed method process flow diagram that Fig. 3 provides for the embodiment of the invention;
The pci system structural drawing that comprises the abnormal monitoring spacer assembly that Fig. 4 provides for the embodiment of the invention;
The concrete structure figure of the abnormal monitoring spacer assembly that Fig. 5 provides for the embodiment of the invention.
Embodiment
In order to make the purpose, technical solutions and advantages of the present invention clearer, describe the present invention below in conjunction with the drawings and specific embodiments.
The present invention is provided with the abnormal monitoring spacer assembly between main process equipment and the PCI equipment in pci system, the performed main method of this abnormal monitoring spacer assembly can mainly may further comprise the steps as shown in Figure 2:
Step 201: monitor main process equipment by pci bus to visit as the PCI equipment of target device.
Step 202: listen to main process equipment and the visit of target device is occurred isolating this target device when unusual by pci bus.
Step 203: when the target device that is in isolation is visited by main process equipment, replace accessed target device to carry out target and abrogate and reply to main process equipment.
The abnormal monitoring spacer assembly can comprise the monitoring that main frame carries out the visit of target device by pci bus: to the monitoring of pci bus incident, or to target device dont answer duration, or simultaneously both are monitored.
Below in conjunction with specific embodiment said method is described in detail.The detailed method process flow diagram that Fig. 3 provides for the embodiment of the invention, as shown in Figure 3, this method can may further comprise the steps:
Step 301: abnormal monitoring spacer assembly monitoring pci bus incident and target device be response time not, if listen to the pci bus incident, execution in step 302; If listen to target device not response time reach default duration threshold value, execution in step 314.
Step 302: determine the pci bus event type that listens to, if be transaction beginning incident, execution in step 303; If for main process equipment is ready to incident, execution in step 304; If for target device is replied incident, execution in step 308.
In this step, can if listen to effective FRAME signal, determine that then the event type that listens to is transaction beginning incident for the signal on the pci bus is monitored to the monitoring of pci bus event type; If listen to the effective I rdy signal, determine that then the event type that listens to is that main process equipment is ready to incident.
Step 303: latch the data on the pci bus, latched data is deciphered the target device of determining this visit, go to execution in step 301.
When listening to transaction beginning incident, can latch address on the pci bus and data, be used for decoding target equipment.
Step 304: whether the target device of judging this visit is in isolation, if, execution in step 305; Otherwise execution in step 306.
Will relate in subsequent descriptions how target-marking equipment is in isolation, a status register can be set in the present embodiment to be stored each PCI equipment and whether is in isolation, the status register of 32 bits for example is set, the corresponding equipment of each bit, the corresponding PCI equipment of 0 sign is in non-isolation, the corresponding PCI equipment of 1 sign is in isolation, and reset values is 0.
Step 305: replace target device to carry out target and abrogate and reply, go to execution in step 301 to main process equipment.
In this step, the abnormal monitoring spacer assembly is in case determine that accessed target device is in isolation, illustrate that then this target device has been confirmed as unusually, then, causes this unusual target device the pci system collapse for fear of being conducted interviews, the abnormal monitoring spacer assembly replaces target device to carry out target to main process equipment abrogating and reply, make that main process equipment can continue other PCI equipment is conducted interviews.Concrete condition and process that target device is isolated will occur in subsequent descriptions.
Step 306: start timing to target device dont answer duration.
After listening to the ready bus events of main equipment, the abnormal monitoring spacer assembly begins to wait for replying of target device that if target device can not be replied, it is unusual to illustrate that then target device exists in setting duration.
Step 307: if current transaction is the configuration destination device address, then the destination device address information with visit records in the address realm table, goes to execution in step 301.
Store the address of accessed PCI equipment in this address realm table, promptly each destination device address is used for target device is discerned.List item in this address realm table can generate by this step 307, also can be configured by the collocation channel between abnormal monitoring spacer assembly and the main process equipment.
In addition, also may exist a kind of situation to be, main process equipment may repeatedly be configured destination device address, therefore, determining current transaction is when disposing destination address, can at first judge the address information that whether has had this accessed target device in the address realm table, if there is no, then with the address information recording of target device of visit in the address realm table; If exist, then directly go to step 301, thereby avoid the destination device address information that storage repeats in the address realm table.
Step 308: will be to the timing zero clearing of target device dont answer duration.
Target device is replied and is shown that bus is hung extremely, therefore, can stop the timing to target device dont answer duration.
Step 309: whether the type of judging the incident of replying is retry, if execution in step 310; Otherwise execution in step 313.
Target device is replied and may be had multiple situation, and retry is a kind of legal replying for standard, but can may bring unusual condition in the practical application, therefore, need in the present invention retry is carried out special processing, promptly number of retries is counted, and when transfiniting, isolate this target device.
Step 310: number of retries is added 1.
Step 311: judge whether number of retries reaches default retry threshold value, if, execution in step 312, otherwise, go to step 301.
Can preestablish the number of retries that main process equipment can carry out target device, when reaching this number of retries, it is unusual to think that this target device occurs, and can't finish initial configuration, need isolate this target device, make main process equipment finish visit this target device.
Step 312: target-marking equipment is in isolation, with the number of retries zero clearing, goes to step 301.
Step 313: with the number of retries zero clearing, judge whether the incident of replying belongs to the abnormal patterns that the user is provided with, if, execution in step 314; Otherwise, go to step 301.
If the incident of replying is not a retry,, just be considered to unusual because have only the number of times of continuous retry to reach the retry threshold value then with the number of retries zero clearing.
Because except retry, the incident of replying also may be other unusual condition, the user can more pre-configured unusual conditions, main process equipment disposes these unusual conditions to the abnormal monitoring spacer assembly by collocation channel, therefore, can further judge whether other abnormal patterns of being provided with into the user in this step.If do not consider other abnormal patterns, perhaps the user is not provided with other abnormal patterns, then after with the number of retries zero clearing, directly goes to step 301.
Can be stored in bus abnormal patterns table with being used for pre-configured unusual condition, it is unusual that each list item defines a kind of bus, when carrying out this step, judges by the mode of tabling look-up whether the incident of replying belongs to the abnormal patterns that the user is provided with.
Step 314: the potential unusual number of times of target device is added 1.
Step 315: judge whether the potential unusual number of times of target device surpasses default potential unusual threshold value, if, execution in step 316; Otherwise, go to execution in step 301.
Step 316: whether the transaction of judging this visit finishes, if, execution in step 317; Otherwise execution in step 318.
When the DEVSEL invalidating signal, when the TRDY signal was also invalid, promptly target device finished statement and is ready to data transmission, and during the statement closing the transaction, can determine the closing the transaction of this visit, otherwise the transaction of this visit did not finish.
Step 317: target-marking equipment is in isolation, goes to step 301.
Step 318: target-marking equipment is in isolation, and replaces target device to carry out target to main process equipment abrogating and reply, go to step 301.
If transaction finishes as yet, then the abnormal monitoring spacer assembly notifies main process equipment to finish this transaction by abrogating to reply to main process equipment transmission target.
In this step, before target-marking equipment was in isolation, the abnormal monitoring spacer assembly can disconnect pci bus, and can further send error reporting to main process equipment, replace target device to main process equipment carry out target abrogate reply after, open pci bus, allow the pci signal transparent transmission.
In addition, determine that at some target device unusual situation takes place might not be fit to isolate immediately target device, situation about transfiniting for number of retries for example is preferably in that the accessed moment of target device is isolated target device more next time.Therefore, can in step 312, be in isolation by direct target-marking equipment, and target device is labeled as the need isolation, after the number of retries zero clearing, go to step 301; And, between step 306 and step 307, increase a step: judge whether target device is in the need isolation, if, will be to the timing zero clearing of target device dont answer duration, execution in step 318; Otherwise continue execution in step 307.
Whether target device is in needs isolation, also can come mark by a status register is set, for example, the status register of one 32 bit is set, the corresponding PCI equipment of each bit, the corresponding PCI equipment of 0 sign is in normal condition, and the corresponding PCI equipment of 1 sign is in and needs isolation, wait for nextly when accessed, isolate this equipment.
Can the abnormal monitoring spacer assembly replacing taking place unusual target device by above-mentioned flow process carries out target to main process equipment and abrogates and reply, feasible visit can both be finished in the cycle at the pci transaction of regulation, too many bandwidth can be do not taken, influence can be eliminated to greatest extent pci system.
More preferably, the abnormal monitoring spacer assembly can send error reporting to main process equipment by collocation channel when isolating target device, and this error reporting sends to main process equipment with the form of interrupting usually.After main process equipment receives this error reporting, can be further in the abnormal monitoring spacer assembly the segregate target device information of inquiry, also can select not carry out inquiring about and carry out other operation, this can carry out according to concrete strategy.
In addition, main process equipment can be removed isolation to unusual target device by collocation channel, promptly by collocation channel notice abnormal monitoring spacer assembly target device is labeled as normal condition.And after target device was removed isolation, main process equipment can reset to the target device of removing isolation or reinitialize; In addition, consider that it might not be physical damage unusually that some equipment occurs, and therefore also can be resetted or hot plug to the target device of removing after isolating by the user.
In above-mentioned flow process, when target device not response time transfinite or when listening to user-defined other abnormal patterns, it is potential unusual to think that target device exists, and potential unusual number of times is counted, after potential unusual number of times transfinites, again target device is isolated.This is a kind of fault-tolerant processing by the mode that potential unusual number of times is set, the processing of potential unusual number of times also can be set, when target device not response time transfinite or listen to user-defined other abnormal patterns, then directly target device is isolated, the limit value that this situation can be considered as being provided with potential unusual number of times is 1.
More than be the detailed description that method provided by the present invention is carried out, below device provided by the present invention be described in detail.The pci system structural drawing that comprises the abnormal monitoring spacer assembly that Fig. 4 provides for the embodiment of the invention, as shown in Figure 4, this abnormal monitoring spacer assembly is arranged in the pci system between the main process equipment and PCI equipment, and this abnormal monitoring spacer assembly can comprise: control module 400 and isolated location 410.
The concrete structure figure of the abnormal monitoring spacer assembly that Fig. 5 provides for the embodiment of the invention, as shown in Figure 5, control module 400 comprises: pci interface unit 500 and core processing unit 510.
Wherein, pci interface unit 500 can specifically comprise: intercept unit 501 and pci bus driver element 502.
Pci bus driver element 502 when being used for determining that the target device of being isolated by isolated location 410 is accessed, replacing this accessed target device to abrogate to main process equipment transmission target and replys.
Particularly, intercept unit 501 the pci bus reporting events that listens to is supplied with core processing unit 510.
Unusual definite subelement 512, be used in the recognin unit 511 and determine that the pci bus incidents are that retry is replied incident and number of retries when reaching default retry threshold value, determine that perhaps the pci bus incident is the incident of replying and when belonging to default exception response pattern, when the not response time of perhaps determining accessed target device reaches default duration threshold value, it is unusual to determine that visit occurs, and notice isolated location 410 is isolated accessed target device.
For the different situations that recognin unit 511 identifies, definite unusually subelement 512 can be made different processing, specifically can comprise following several situation:
First kind of situation: recognin unit 511 determines that the pci bus incidents are that main process equipment is when being ready to incident, unusual definite subelement 512 judges whether accessed target device is isolated, if the accessed target device of triggering pci bus driver element 502 replacements is abrogated to main process equipment transmission target and is replied; Otherwise, start timing to accessed target device dont answer duration.
Second kind of situation: recognin unit 511 determines that the pci bus incidents are main process equipment when being ready to incident, and core processing unit 510 also comprises: retry count subelement 513.
Unusual definite subelement 512 will judge whether reply event type is retry to the timing zero clearing of target device dont answer duration, if, send the counting notice to retry count subelement 513, otherwise, the zero clearing notice sent to retry count subelement 513; After receiving retry thresholding judgement notice, judge whether the number of retries of retry count subelement 513 records reaches default retry threshold value, if send first quarantine notification to isolated location 410; And to retry count subelement 513 transmission zero clearing notices; After receiving answer-mode judgement notice,, judge then whether the transaction of this visit finishes, if send first quarantine notification to isolated location 410 if the incident of replying belongs to default exception response pattern; Otherwise send second quarantine notification to isolated location 410, and trigger the pci bus driver element and replace accessed target device to send target abrogating and reply to main process equipment.
Retry count subelement 513, be used to receive counting notice after, the number of retries of record is added 1, send the retry thresholdings to unusual definite subelement 512 and judge notice; After receiving the zero clearing notice, send answer-mode to unusual definite subelement 512 and judge notice.
The third situation: when definite unusually subelement 512 is determined accessed target device response time is not reached default duration threshold value, judge whether the transaction of this visit finishes, if send first quarantine notification to isolated location 410; Otherwise send second quarantine notification 410 to isolated location, and trigger pci bus driver element 502 and replace accessed target device to send target abrogating and reply to main process equipment.
Based on second kind of situation and the third situation, unusual definite subelement 512, can also be used for before carrying out the operation whether transaction judge this visit finish, the potential unusual number of times of target device is added 1, judge whether potential unusual number of times transfinites, if continue the operation that execution judges whether the transaction of this visit finishes; Otherwise end process.
In addition, control module 400 can also comprise: local bus unit 520, be used to receive the unusual condition configuration information that main computer unit sends by collocation channel or remove the target device information of isolating, and with this unusual condition configuration information or remove the target device information of isolating and offer core processing unit 510.
In addition, core processing unit 510 can also send error reporting by local bus unit 520 to main process equipment when isolating target device.
Further, control module 400 can also comprise: clock synthesis unit 530 is used to utilize the outside needed clock of reference clock resulting anomaly monitoring spacer assembly.
By above description as can be seen, the present invention makes this abnormal monitoring spacer assembly monitor the visit of target device by the CPI center line main process equipment by between main process equipment in pci system and the PCI equipment abnormal monitoring spacer assembly being set; Listening to visit occurs isolating accessed target device when unusual; When segregate target device is accessed, replaces this accessed target device to abrogate and reply to main process equipment transmission target.This mode can make main process equipment that the access exception of target device in time is found, and by replacing segregate accessed target device to abrogate the mode of replying to main process equipment transmission target, make main process equipment can jump out visit to the unusual target device of this generation, thereby pci bus taken for a long time the whole pci system collapse that may cause, improved the reliability of pci system, and the performance loss of system when having reduced the PCI unit exception to greatest extent.
In addition, at several unusual conditions commonly used, provide concrete solution among the present invention, as can be seen, the present invention is a kind of general unusual method of solution pci system, can realize monitoring and isolation that specific pci bus is unusual easily and effectively, the extensibility height.
The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being made, is equal to replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (16)
1, a kind of unusual method of peripheral component interconnect pci system of handling is characterized in that, between main process equipment in pci system and the PCI equipment abnormal monitoring spacer assembly is set, and this abnormal monitoring spacer assembly is carried out following steps:
A, monitor main process equipment by pci bus to visit as the PCI equipment of target device;
B, listen to visit and occur isolating accessed target device when unusual;
C, when segregate target device is accessed, replaces this accessed target device to send target and abrogate and reply to described main process equipment.
2, method according to claim 1 is characterized in that, described steps A comprises: described abnormal monitoring spacer assembly is monitored the not response time of pci bus incident and accessed target device.
3, method according to claim 2 is characterized in that, listens to visit described in the step B and occurs comprising unusually:
When listening to the pci bus incident is that retry is replied incident and number of retries when reaching default retry threshold value, determines to listen to visit and occurs unusual; Perhaps,
When the pci bus incident that listens to is the incident of replying and when belonging to default exception response pattern, determine to listen to visit and occur unusual; Perhaps,
When the not response time that listens to accessed target device reaches default duration threshold value, determine to listen to visit and occur unusual.
4, method according to claim 2, it is characterized in that, described step C comprises: when listening to the pci bus incident is that main process equipment is when being ready to incident, judge whether accessed target device is isolated, if, described abnormal monitoring spacer assembly replaces described accessed target device to send target to described main process equipment abrogating and reply, go to and carry out described steps A; Otherwise, start timing to described accessed target device dont answer duration, go to and carry out described steps A.
5, method according to claim 4 is characterized in that, described main process equipment is in advance in the address realm table of address configuration in described abnormal monitoring spacer assembly with each target device; Perhaps,
After the timing that starts described accessed target device dont answer duration, and also comprise before going to the described steps A of execution: if current transaction is the accessed destination address of configuration, then described abnormal monitoring spacer assembly obtains accessed target address information, and the destination device address information of visit is recorded in the address realm table.
6, method according to claim 2 is characterized in that, described step B comprises:
B11, be target device when replying incident, will judge whether reply event type is retry the timing zero clearing of target device dont answer duration when listening to the pci bus incident, if, execution in step B12, otherwise, execution in step B13;
B12, number of retries is added 1, judge that whether number of retries reaches default retry threshold value, if the accessed target device of mark is in isolation with the number of retries zero clearing, goes to described steps A; Otherwise, go to described steps A;
B13, with the number of retries zero clearing, judge whether the incident of replying belongs to default exception response pattern, if execution in step B14, otherwise go to described steps A;
B14, judge whether the transaction of this visit finishes, if, the accessed target device of mark is in isolation, go to described steps A, otherwise the disconnection pci bus, the accessed target device of mark is in isolation, and replaces described accessed target device to send target to described main process equipment abrogating and reply, open pci bus, go to described steps A.
7, method according to claim 2 is characterized in that, described step B comprises:
When the not response time that listens to accessed target device reaches default duration threshold value, whether the transaction of judging this visit finishes, if, the accessed target device of mark is in isolation, goes to described steps A, otherwise disconnects pci bus, the accessed target device of mark is in isolation, and replace described accessed target device to send target abrogating and reply, open pci bus, go to described steps A to described main process equipment.
8, according to claim 6 or 7 described methods, it is characterized in that, before whether the described transaction of judging this visit finishes, also comprise: the potential unusual number of times of target device is added 1, judge whether potential unusual number of times transfinites, if continue to carry out the step whether described transaction of judging this visit finishes; Otherwise go to described steps A.
9, a kind of unusual device of pci system of handling, this device is applied to comprise the pci system of main process equipment and PCI equipment, it is characterized in that, and this device comprises: control module and isolated location;
Described control module, be used to monitor main process equipment by pci bus to visit as the PCI equipment of target device; When the target device of being isolated by described isolated location is accessed, replaces this accessed target device to send target and abrogate and reply to described main process equipment;
Described isolated location is used for listening to visit at described control module and occurs isolating accessed target device when unusual.
10, device according to claim 9 is characterized in that, described control module comprises: pci interface unit and core processing unit;
Wherein, the pci interface unit comprises: intercept unit and pci bus driver element;
The described unit of intercepting is used to monitor main process equipment by the visit of pci bus to target device, and the situation that listens to is reported described core processing unit;
Described pci bus driver element when being used for determining that the target device of being isolated by described isolated location is accessed, replacing this accessed target device to abrogate to described main process equipment transmission target and replys;
Described core processing unit is used for listening to visit and occurring notifying described isolated location to isolate accessed target device when unusual in the described unit of intercepting.
11, device according to claim 10 is characterized in that, described core processing unit is supplied with the pci bus reporting events that listens in the described unit of intercepting;
Described core processing unit comprises:
The recognin unit is used to determine to intercept the not response time of pci bus incident that the unit reports or accessed target device;
Unusual definite subelement, be used for determining that the pci bus incident is that retry is replied incident and number of retries when reaching default retry threshold value in the recognin unit, determine that perhaps the pci bus incident is the incident of replying and when belonging to default exception response pattern, when the not response time of perhaps determining accessed target device reaches default duration threshold value, it is unusual to determine that visit occurs, and notifies described isolated location to isolate accessed target device.
12, device according to claim 11, it is characterized in that, described recognin unit determines that the pci bus incident is that main process equipment is when being ready to incident, described unusual definite subelement judges whether accessed target device is isolated, if trigger described pci bus driver element and replace described accessed target device to send target abrogating and reply to described main process equipment; Otherwise, start timing to described accessed target device dont answer duration.
13, device according to claim 11 is characterized in that, described core processing unit also comprises: the retry count subelement;
Described recognin unit determines that the pci bus incident is that main process equipment is when being ready to incident, described unusual definite subelement will be to the timing zero clearing of target device dont answer duration, judge whether reply event type is retry, if, send the counting notice to the retry count subelement, otherwise, send the zero clearing notice to described retry count subelement; After receiving retry thresholding judgement notice, judge whether the number of retries of described retry count subelement record reaches default retry threshold value, if send first quarantine notification to described isolated location; And to described retry count subelement transmission zero clearing notice; After receiving answer-mode judgement notice,, judge then whether the transaction of this visit finishes, if send first quarantine notification to described isolated location if the incident of replying belongs to default exception response pattern; Otherwise send second quarantine notification to described isolated location, and trigger described pci bus driver element and replace described accessed target device to send target abrogating and reply to described main process equipment;
Described retry count subelement, be used to receive counting notice after, the number of retries of record is added 1, send the retry thresholding to described unusual definite subelement and judge notice; After receiving the zero clearing notice, send answer-mode to described unusual definite subelement and judge notice;
Described isolated location, when being used to receive described first quarantine notification, the accessed target device of mark is in isolation; When receiving described second quarantine notification, disconnect pci bus, the accessed target device of mark is in isolation, described pci bus driver element send target abrogate reply after, open pci bus.
14, device according to claim 11 is characterized in that, described unusual definite subelement, when determining accessed target device response time not reaching default duration threshold value, whether the transaction of judging this visit finishes, if send first quarantine notification to described isolated location; Otherwise send second quarantine notification to described isolated location, and trigger described pci bus driver element and replace described accessed target device to send target abrogating and reply to described main process equipment;
Described isolated location, when being used to receive described first quarantine notification, the accessed target device of mark is in isolation; When receiving described second quarantine notification, disconnect pci bus, the accessed target device of mark is in isolation, described pci bus driver element send target abrogate reply after, open pci bus.
15, according to claim 13 or 14 described devices, it is characterized in that, described unusual definite subelement, also be used for before carrying out the operation whether described transaction of judging this visit finish, the potential unusual number of times of target device is added 1, judge whether potential unusual number of times transfinites, if continue to carry out the operation whether described transaction of judging this visit finishes; Otherwise end process.
16, device according to claim 10, it is characterized in that, described control module also comprises: the local bus unit, be used to receive the unusual condition configuration information that described main computer unit sends by collocation channel or remove the target device information of isolating, and with this unusual condition configuration information or remove the target device information of isolating and offer described core processing unit;
Described core processing unit determines according to described unusual condition configuration information whether visit occurs unusually, perhaps will remove the target device information of isolating and offer described isolated location;
Described isolated location also is used for the target device information according to described releasing isolation, removes the isolation of corresponding target device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910089083A CN101625656B (en) | 2009-07-28 | 2009-07-28 | Method and device for processing abnormity of PCI system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910089083A CN101625656B (en) | 2009-07-28 | 2009-07-28 | Method and device for processing abnormity of PCI system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101625656A true CN101625656A (en) | 2010-01-13 |
| CN101625656B CN101625656B (en) | 2012-09-19 |
Family
ID=41521511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910089083A Expired - Fee Related CN101625656B (en) | 2009-07-28 | 2009-07-28 | Method and device for processing abnormity of PCI system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101625656B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103309762A (en) * | 2013-06-21 | 2013-09-18 | 杭州华三通信技术有限公司 | Equipment exception handling method and device |
| WO2016127600A1 (en) * | 2015-02-12 | 2016-08-18 | 中兴通讯股份有限公司 | Exception handling method and apparatus |
| CN106326151A (en) * | 2016-08-19 | 2017-01-11 | 浪潮(北京)电子信息产业有限公司 | Method and device for unplugging PCIe equipment |
| CN107577550A (en) * | 2017-08-31 | 2018-01-12 | 北京奇安信科技有限公司 | A kind of whether abnormal method and device of response for determining access request |
| CN113127287A (en) * | 2019-12-31 | 2021-07-16 | 北京车和家信息技术有限公司 | Processor control method and device and electronic equipment |
| WO2022165790A1 (en) * | 2021-02-07 | 2022-08-11 | 华为技术有限公司 | Power-down isolation device and related method |
| CN117076169A (en) * | 2023-08-04 | 2023-11-17 | 阿波罗智联(北京)科技有限公司 | Method and device for detecting interruption abnormality of operating system and electronic equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6223299B1 (en) * | 1998-05-04 | 2001-04-24 | International Business Machines Corporation | Enhanced error handling for I/O load/store operations to a PCI device via bad parity or zero byte enables |
| US6182182B1 (en) * | 1998-10-28 | 2001-01-30 | Adaptec, Inc. | Intelligent input/output target device communication and exception handling |
| CN100498723C (en) * | 2006-12-31 | 2009-06-10 | 华为技术有限公司 | Method for preventing bus fault, communication equipment and bus monitoring device |
-
2009
- 2009-07-28 CN CN200910089083A patent/CN101625656B/en not_active Expired - Fee Related
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103309762A (en) * | 2013-06-21 | 2013-09-18 | 杭州华三通信技术有限公司 | Equipment exception handling method and device |
| CN103309762B (en) * | 2013-06-21 | 2015-12-23 | 杭州华三通信技术有限公司 | Unit exception disposal route and device |
| WO2016127600A1 (en) * | 2015-02-12 | 2016-08-18 | 中兴通讯股份有限公司 | Exception handling method and apparatus |
| CN106326151A (en) * | 2016-08-19 | 2017-01-11 | 浪潮(北京)电子信息产业有限公司 | Method and device for unplugging PCIe equipment |
| CN107577550A (en) * | 2017-08-31 | 2018-01-12 | 北京奇安信科技有限公司 | A kind of whether abnormal method and device of response for determining access request |
| CN107577550B (en) * | 2017-08-31 | 2021-02-09 | 奇安信科技集团股份有限公司 | Method and device for determining whether response of access request is abnormal |
| CN113127287A (en) * | 2019-12-31 | 2021-07-16 | 北京车和家信息技术有限公司 | Processor control method and device and electronic equipment |
| WO2022165790A1 (en) * | 2021-02-07 | 2022-08-11 | 华为技术有限公司 | Power-down isolation device and related method |
| CN117076169A (en) * | 2023-08-04 | 2023-11-17 | 阿波罗智联(北京)科技有限公司 | Method and device for detecting interruption abnormality of operating system and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101625656B (en) | 2012-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101625656B (en) | Method and device for processing abnormity of PCI system | |
| US7240130B2 (en) | Method of transmitting data through an 12C router | |
| US20070070885A1 (en) | Methods and structure for detecting SAS link errors with minimal impact on SAS initiator and link bandwidth | |
| CN106502814B (en) | Method and device for recording error information of PCIE (peripheral component interface express) equipment | |
| CN103514173B (en) | The method and node device of data processing | |
| US7630304B2 (en) | Method of overflow recovery of I2C packets on an I2C router | |
| CN101635652B (en) | Method and equipment for recovering fault of multi-core system | |
| CN102227131A (en) | Hot backup system of NVR and method thereof | |
| JP2006072717A (en) | Disk subsystem | |
| CN107517110A (en) | Veneer configuration self-recovery method and device in a kind of distributed system | |
| CN107678994A (en) | PCIe device hot drawing method and device | |
| CN102244600A (en) | Method and device for detecting and processing link failure in RRPP (Rapid Ring Protect Protocol) ring network | |
| US5363493A (en) | Token ring network test device using finite state machine | |
| CN101854263B (en) | Method, system and management server for analysis processing of network topology | |
| CN110659147B (en) | Self-repairing method and system based on module self-checking behavior | |
| US20070055913A1 (en) | Facilitating detection of hardware service actions | |
| CN110609762A (en) | Method and device for preventing advanced high performance bus (AHB) from deadlock | |
| CN110008681A (en) | Access control method, equipment and system | |
| US11704180B2 (en) | Method, electronic device, and computer product for storage management | |
| CN110912760B (en) | Link state detection method and device | |
| CN114448689A (en) | Method, device and equipment for determining boundary equipment of industrial control network and storage medium | |
| CN103281209B (en) | The processing method of a kind of warning information and equipment | |
| CN106506074A (en) | A kind of method and apparatus of detection optical port state | |
| CN101605052A (en) | A kind of SDH equipment boss frame obtains the method and system of extended subrack information on the throne | |
| CN110795263B (en) | Hard disk link protection method and related 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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No. Patentee after: Xinhua three Technology Co., Ltd. Address before: 310053 Hangzhou hi tech Industrial Development Zone, Zhejiang province science and Technology Industrial Park, No. 310 and No. six road, HUAWEI, Hangzhou production base Patentee before: Huasan Communication Technology Co., Ltd. |
|
| 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: 20120919 Termination date: 20200728 |