CN1786926A - Method for using buffer memory area in processor - Google Patents
Method for using buffer memory area in processor Download PDFInfo
- Publication number
- CN1786926A CN1786926A CN 200410097102 CN200410097102A CN1786926A CN 1786926 A CN1786926 A CN 1786926A CN 200410097102 CN200410097102 CN 200410097102 CN 200410097102 A CN200410097102 A CN 200410097102A CN 1786926 A CN1786926 A CN 1786926A
- Authority
- CN
- China
- Prior art keywords
- buffer area
- buffer
- information
- descriptor
- error
- 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
Abstract
The invention discloses a method for using buffer areas in a processor, firstly setting a buffer area descriptor comprising buffer area state and user information; as applying for a buffer area, obtaining an idle buffer area as the current buffer area according to the buffer area state; then modifying the buffer area descriptor in the current buffer area; as releasing a buffer area, releasing the current buffer area according to the buffer area descriptor. By setting buffer area state and user information in the buffer area descriptor, as the user applies for and releases the buffer area, it firstly checks if the buffer area state is an error state and dynamically records the course of the user using the buffer area resources. At the time of failure, i.e. the buffer area state is an error state, it can making accurate location according to the user information.
Description
Technical field
The present invention relates to computer realm, relate in particular to the using method in processor cache district.
Background technology
Processor is used widely today, and other is serving as different tasks in using in diverse network business and different performance level for it.Network technology and application and development are more and more higher to the requirement of the processing power of the capacity of the network equipment and veneer.Usually need the concurrent working simultaneously of a plurality of coprocessors in processor, the processing of a packet will be passed through a plurality of modules, finishes by several coprocessors are collaborative.In order to relax the unmatched contradiction of speed between disparate modules, reduce interruption frequency, and improve the concurrency between CPU and the I/O unit CPU in each module, need to introduce buffering, the transceive data of each module is carried out buffer memory.
Usually, to the management of communication facilities internal buffer according to following three kinds of modes:
(1) way to manage of static allocation.When design to equipment in the buffer area of each the module assignment fixed size dedicated buffering when sending and receiving as data, in the data transmit-receive process, use repeatedly.
Though this static allocation mode is handled simple, but because each resume module function is different with processing speed, its data traffic also is among the continuous variation, therefore may make some buffer area space idle for a long time, and some buffer area space can crowd, even overflow, also can cause losing of normal data when serious.When mistakes such as generation loss of data,, can't review the reason that causes loss of data owing to there is not relevant information.
(2) way to manage of dynamic assignment.In order to improve the utilization factor of buffer zone, design a buffer zone pond, will put into this buffer zone pond between all available buffer empties, the buffer zone in the pond can be shared for a plurality of modules.
When certain module sends data, at first to buffer zone pond application spatial cache, from the empty buffer formation, be its distribution spatial cache, and this spatial cache is linked in the output queue; When next module receives these data, to buffer zone pond application spatial cache, from the empty buffer formation, distribute spatial cache for it, and read these data from the corresponding cache space and write spatial cache into its distribution according to output queue, simultaneously, corresponding cache space in the output queue is discharged and be linked in the empty buffer formation, the spatial cache that writes after the data is linked in the input queue.
Because the buffer area fragment, might be caused because of certain module breaks down in a plurality of module shared buffers pond therefore, also may exhaust the resource of system when serious.And, when the equipment inner module more for a long time, the management of buffer zone is realized complicated, going wrong is difficult to the location.
(3) static state and the way to manage that dynamically combines.This method adopts the quantity of dynamic assignment buffer zone when configuration, the mode of static state operation in the processing.Required buffer zone between the dynamic-configuration adjacent block at first; Then each described buffer zone is divided into the sub-buffer zone of predetermined number, each sub-buffer zone corresponding buffer region descriptor is set, and set up annular chain meter according to the information of described buffer descriptor.After this buffer configuration is good, be static, do not need to apply for dispose procedure, specifically, when adjacent block is received/sent out data and needs buffer zone, from described annular chain meter, obtain corresponding sub-buffer zone in turn and read/write data its accessing operation that carries out.
Owing to do not need to apply for can not have influence on total system in case certain module goes wrong by dispose procedure, and by disparate modules and buffer zone are set up corresponding relation, improvement to a certain degree the difficulty of location when going wrong.When but mistake appears in certain sub-buffer area in this method chained list, will be owing to forming the confusion that complete buffer memory chain causes whole buffer area, and, owing to lack relevant information, can't accurately navigate to the problem code, therefore, this method still has problems when mistake appears in the buffer area resource and is difficult to location defect.
Summary of the invention
The technical problem to be solved in the present invention is: the method for using buffer zone in a kind of processor is provided, can reliably manages buffer zone simply and effectively, and can accurately navigate to the buffer area resource that breaks down.
For solving the problems of the technologies described above, the invention provides the method for using buffer zone in a kind of processor, comprise step:
(1) the buffer area descriptor is set, the information of described buffer area descriptor comprises buffer state and user's information;
When asking buffer area (2), obtain the free buffer district according to described buffer state, and as the current cache district;
(3) information of buffer area descriptor in the described current cache of the modification district;
(4) when discharging buffer area, according to the information release current cache district of described buffer area descriptor.
Wherein, described buffer state comprises: normal condition and error condition.
Wherein, acquisition process step 2) is specially: obtaining described buffer state is the free buffer district of normal condition.
Wherein, the modification process of step 3) is specially:
The described user's information of dynamically recording;
If fault processing appears in the current cache district, then described buffer state is set to error condition.
Wherein, the dispose procedure of step 5) is specially:
If buffer state is a normal condition, then discharges the current cache district, and empty the information in the buffer area descriptor;
If buffer state is an error condition, then discharges the current cache district, and keep described buffer state and user's information.
Wherein, described step also comprises: if buffer state is an error condition, then according to described user's information location error source.
Wherein, the information of the descriptor of buffer area described in the step 1) also comprises: user mode information is used to identify buffer area and whether is in idle condition;
During described application buffer area and during described release buffer area, also comprise the described user mode information of upgrading.
Wherein, the information of the described buffer area descriptor of step 1) also comprises: type of error information;
If buffer state is an error condition, then fills in described fault type information, and when discharging the current cache district, keep described type of error information.
Wherein, the information of the described buffer area descriptor of step 1) also comprises: applicant's information and releaser's information;
During described application buffer area, also comprise: fill in described applicant's information;
If when described buffer state is normal condition, when then discharging described current cache district, also comprise the described releaser's information that empties;
When if described buffer state is normal condition, when then discharging described current cache district, also comprise keeping described releaser's information.
Compared with prior art, the invention has the beneficial effects as follows: by buffer state and user's information are set in the buffer area descriptor, when the user checks at first whether buffer state is error condition when applying for and discharge buffer area, and dynamically recording is carried out in the use of buffer area resource with user's information, when breaking down, thereby when buffer state is error condition, can accurately locate according to user's information.
Whether further, the present invention also is provided with buffer area user mode information, can regularly detect buffer area according to this information and exist fault to leak, and when breaking down leakage, can accurately navigate to the buffer area that leakage occurs according to user's information.
Further, the present invention also is provided with applicant's information, releaser's information and type of error information, can review the source of fault, and the misregistration reason, is convenient to positioning problems and tracking.
In addition, because the way to manage of the present invention and buffer area is irrelevant, therefore all be suitable for for static management, dynamic management and the static buffer queue management mode that dynamically combines.
Description of drawings
Fig. 1 is the way to manage synoptic diagram of static allocation
Fig. 2 is the way to manage synoptic diagram of dynamic assignment.
Embodiment
Way to manage for cache resources mainly contains static management, dynamic management and the static way to manage that dynamically combines.Core of the present invention is to preserve error condition and buffer area user information in cache resources, memory source is located fast and accurately, so that carry out positioning problems when breaking down.The present invention all is suitable for for various buffer queue management modes.
Specifically, when utilizing the present invention to carry out buffer queue management, the buffer area descriptor is set at first, the information of described buffer area descriptor comprises buffer state and user's information; When the application buffer area, obtain the free buffer district according to described buffer state, and as the current cache district; When the information in the buffer area descriptor changes, revise the information of buffer area descriptor in the described current cache district; When discharging buffer area, according to the information release current cache district of described buffer area descriptor.Below implementation process of the present invention is described in detail.
The buffer area descriptor at first is set, and the information of described buffer area descriptor comprises buffer state and user's information; In the present invention, for the ease of understanding, buffer state is defined as Error, the value that defines Error simultaneously is to represent that buffer area made a mistake at 1 o'clock, and the value of Error is to represent that buffer area was in normal condition at present at 0 o'clock; With buffer area user information definition is Last_user_ID.For the value of Error is that 1 situation comprises: during the application buffer area, if described buffer area is in just in user mode; When discharging buffer area, if described buffer area is in unused state and buffer area when in use occurring handling mistake.
Simultaneously, the present invention also is provided with buffer area user mode sign Using, indicating the user mode of buffer area, and represents that with Using=0 buffer area is in unused state, and Using=1 represents that buffer area is in user mode.The existence of this field can accelerating system treatment scheme.Whether when certain module application buffer area, at first detecting Using is 1, if then directly report is unusual, and needn't reexamine the value of Error, if not, then continues to check the value of Error, and report is unusual when Error is 1; In order to follow the tracks of from the source the use of cache resources, the present invention also is provided with applicant Applier_ID field, but whether this field exists the positioning problems when not influencing fault.When the value of Error and Using is 0, this buffer area is distributed to this module, apply for successfully after, fill in the Applier_ID field;
Buffer area uses the module of this buffer area for all, the sign of this module need be write Last_user_ID, so that can navigate to the module of this buffer area of application when breaking down.Last_user_ID is a field along with the use dynamic change of buffer area, each handles the module of this buffer area, when entering, this field extended this as the sign of self, subsequently, along with the situation of change to the user of this buffer memory block in pilot process and the dispose procedure changes accordingly;
In order better to understand the therefore reason of buffer area mistake, the present invention also is provided with the error code Error_code that is used to show type of error, among the present invention, this error code is supported 64 kinds of mistakes, and for example: cyclic check code mistake etc. is limited, found to long surpassing of bag.In the use of buffer area,, fill in corresponding Error_code by the module of finding mistake simultaneously if abnormality processing then is changed to 1 with Error;
In addition, the present invention also is provided with the releaser and identifies the Releaser_ID field, is filled in by the module that discharges this buffer area when buffer area discharges; When using this buffer area, discharge when not having the buffer area of fault, the whole zero clearings of above-mentioned all fields; When there is the buffer area of fault in release,, keep Error, Error_code, Last_user_ID, Releaser_ID, Applier_ID field, and upgrade the Using field for mistake is positioned;
More than introduced the process of utilizing the present invention to apply for and discharging buffer area.To introduce below when making a mistake and various faults be carried out the process of localization of fault.
One, the location that buffer area repeats to apply for:
When the application buffer area, if it is clear 0 to find that the using field does not have, then the Error with this buffer area is changed to 1, confirms that this buffer area applies for for repeating, and the module that detection is being used according to the Last_user_ID field, thereby judge that there is the mistake that disorderly discharges buffer area in this module.
Two, the detection when treatment scheme is unusual:
The intermediate link of buffer area between applicant and releaser need be checked the Error sign, if be changed to 1, then shows to have fault processing, can learn failure cause according to Error_code then, and navigate to the module that breaks down by Last_user_ID.Can in debugging code, add printing function, failure cause and the module of finding fault are printed, make things convenient for positioning problems and tracking greatly.
Three, the error-detecting that repeats to discharge:
When discharging a block cache district, the value that needs to detect Using whether 0, if, show that then this buffer area is repeated to discharge, navigate to the module that repeats to discharge according to Last_user_ID then.Printing function be can increase equally, module name that is moving and the buffer area that repeats to discharge printed for the location.
Four, the detection that buffer area leaks:
It is that buffer area is forgotten release that buffer area leaks, and causes being continued to use.If leak because the flow process mistake produces buffer area, the time has been grown, and may cause buffer area to exhaust, and lead to system abnormity even crash.
When debugging, can start a task, regularly detect all buffer area resources, if certain buffer area block length phase does not discharge, be that Using is always 1, can judge the final module of using according to Last_user_ID, thereby module arrived in the location of mistake that will forget release.This testing process can also can exist in formal code just as debugging code, as the method for abnormal restoring, regularly forces to discharge for the buffer area that can not discharge, and the assurance system can long-term stable operation.
More than variously detect the methods wrong and location, can also can in formal code, exist just as debugging code, be used for abnormality detection and recovery.
The method of management buffer area provided by the invention and the way to manage of buffer area are irrelevant, that is to say, the present invention is applicable to static allocation, dynamic assignment and the static way to manage that dynamically combines.Below with static allocation and dynamic assignment as describing in detail as two specific embodiments.
First embodiment: the way to manage of static allocation cache resources.
The dedicated cache district of this allocation scheme when when design sends and receive as data the buffer area of each module assignment fixed size uses in the data transmit-receive process repeatedly.Fig. 1 is the synoptic diagram of the buffer queue management under this kind allocation scheme.As shown in the figure, system is divided into N Memory slice with all cache resources, and each Memory slice identifies with a buffer area descriptor, and the information data that comprises in each buffer area descriptor comprises: displacement, length and the present invention increase applicant, releaser, user mode sign, error identification and error code field.By the management of pointer realization to each buffer area descriptor.Two pointers of system held: head pointer headptr, tail pointer tailptr.HeadPtr points to current operable Memory slice, and TailPtr points to the current Memory slice that needs release, when two hands coincide, does not have Memory slice to use.The way to manage requirement of static allocation must be carried out using of buffer area in order and discharge, and each buffer area corresponding cache district descriptor region must be continuous.
When the new buffer memory of needs application, at first the buffer area descriptor that points to according to head pointer HeadPtr finds corresponding Memory slice, checks the value of Using then, the value of Error and Error_code, if be 1, show that there is fault in this Memory slice, then reports error reporting.If be 0, then data are write the Memory slice that HeadPtr points to, HeadPtr moves down then; In use, write down Last_user at any time, if abnormality processing, then the value with Error is changed to 1, fills in corresponding Error_code simultaneously; When using this buffer area need discharge the time, the TailPtr pointed is current to need the Memory slice that discharges, checks then whether the value of Using is 1, and whether the value of Error and Error_code is 0, if, then discharge this Memory slice, and with whole field zero clearings; If not, preserve Error, Error_code, Last_user_ID, Realese_ID, Applier_ID field after then discharging this module, upgrade the Using field.
Localization of fault: because the pairing Memory slice of buffer area descriptor that HeadPtr points to is the Memory slice that is in idle condition, therefore, if the Using value is 1 during the application internal memory, represent that then this Memory slice existence repeats application, in the time need carrying out positioning problems, can detect the module of using this Memory slice according to Last_user_ID to the Memory slice that duplicates application.
Equally, because the pairing Memory slice of buffer area descriptor that TailPtr points to is to be in the Memory slice that seizure condition will discharge, therefore, if the Using value is 0 during the application internal memory, represent that then this Memory slice existence repeats to discharge, in the time need carrying out positioning problems, can detect the module that repeats to discharge according to Last_user_ID, and can review according to Realese_ID, Applier_ID field to the Memory slice that duplicates release.
Processor can setting cycle, regularly travel through all Memory slices, when the value of the Using value that detects certain Memory slice corresponding cache district descriptor surpasses the schedule time, then be judged as and memory overflow occurs, navigate to the module of forgetting the releasing memory sheet according to Last_user_ID and Realese_ID, Applier_ID.
Second embodiment: the way to manage of dynamic assignment cache resources.
Fig. 2 is the synoptic diagram of buffer queue management under this mode.System is divided into all cache resources N Memory slice equally, and each Memory slice identifies with a buffer area descriptor, the information data that comprises in each buffer area descriptor comprises: displacement, length and the present invention increase applicant, releaser, user mode sign, error identification and error code field, different with the way to manage of static allocation buffer memory, there is chain list index NextBufHandle pointer in the buffer area descriptor of the way to manage of dynamic assignment.By the management of pointer realization to each buffer area descriptor.This way to manage adopts the mode of dynamic link table to the free buffer district, 2 pointer head of system held, tail, point to chained list respectively end to end, the NextBufHandle thresholding of the initial buffer area descriptor that points to of tail pointer is 0, when certain Memory slice need discharge, the value in NextBufHandle territory in the buffer area descriptor of needs release is replaced 0 original value.Buffer area descriptor address of the unique sensing of pointer, each buffer area descriptor is unique corresponding to one section buffer area.The structure of buffer area descriptor is as shown in the table:
| 31 | 30 | 29 | 28 | 27 | 26 | 25 | 24 | 23 | 22 | 21 | 20 | 19 | 18 | 17 | 16 | 15 | 14 | 13 | 12 | 11 | 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 | 0 | |
| 0 1 | NextBufHandle | |||||||||||||||||||||||||||||||
| Displacement | Length | |||||||||||||||||||||||||||||||
| Using | Error_code | Applier_ID | Releaser_ID | Last_user_ID | ||||||||||||||||||||||||||||
Wherein, NextBufHandle points to the buffer area descriptor address of next buffer area, the position of first byte of Offset indication bag data, and Length indicates the length of whole bag.
When certain module need be applied for buffer area, with the buffer area taking-up of chained list head, the buffer area descriptor that the NextBufHandle territory is pointed in the buffer area descriptor of while this buffer area of head pointed; Check Using then, the value of Error and Error_code if be 1, shows that there is fault in this Memory slice, then reports error reporting.If be 0, then data are write buffer area descriptor corresponding cache district, NextBufHandle moves down then; In use, write down Last_user at any time, if abnormality processing, then the value with Error is changed to 1, fills in corresponding Error_code simultaneously; When using this buffer area need discharge the time, elder generation changes the NextBufHandle territory of the buffer area descriptor of Tail pointed into the buffer area descriptor address in current cache district, the buffer area descriptor address of the Memory slice of the current release of tail pointed then, fill in the releaser simultaneously and identify Releaser_ID, revise user mode sign Using.
If because some reason, the NextBufHandle territory mistake of certain buffer area descriptor in the chained list, the time, cause some internal memory section to can not get forever using, promptly buffer area leaks, and perhaps repeatedly using is that buffer area is reused.The process of using the present invention to carry out localization of fault is:
One, the location that buffer area repeats to apply for:
When the application buffer area, because the Head pointed is the free memory sheet that desire is distributed, if find that therefore the Using field does not have clear 0, can confirm as buffer area and repeat application, then can detect the module of using, thereby judge that there is the mistake that disorderly discharges buffer area in this module according to the Last_user_ID field.
Two, the detection when treatment scheme is unusual:
The intermediate link of buffer area between applicant and releaser need be checked the Error sign, if be changed to 1, then shows to have fault processing, can learn failure cause according to Error_code then, and navigate to the module that breaks down by Last_user_ID.。
Three, the error-detecting that repeats to discharge:
When discharging a block cache district, because the Tail pointed is the Memory slice that desire discharges,, show that then this buffer area is repeated to discharge if therefore the value of Using is 0, navigate to the module that repeats to discharge according to Last_user_ID then.Printing function be can increase equally, module name that is moving and the buffer area that repeats to discharge printed for the location.
Four, the detection that buffer area leaks:
When debugging, can start a task, regularly detect all buffer area resources, if certain buffer memory block does not discharge in the given time, be that Using is always 1, can judge the final module of using according to Last_user_ID, thereby module arrived in the location of mistake that will forget release.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (9)
1, use the method for buffer zone in a kind of processor, it is characterized in that, comprise step:
(1) the buffer area descriptor is set, the information of described buffer area descriptor comprises buffer state and user's information;
(2) when the application buffer area, obtain the free buffer district according to described buffer state, and as the current cache district;
(3) information of buffer area descriptor in the described current cache of the modification district;
(4) when discharging buffer area, according to the information release current cache district of described buffer area descriptor.
2, use the method for buffer area in the processor according to claim 1, it is characterized in that described buffer state comprises: normal condition and error condition.
3, use the method for buffer area in the processor according to claim 2, it is characterized in that step 2) acquisition process be specially:
Obtaining described buffer state is the free buffer district of normal condition.
4, use the method for buffer area in the processor according to claim 2, it is characterized in that the modification process of step 3) is specially:
The described user's information of dynamically recording;
If fault processing appears in the current cache district, then described buffer state is set to error condition.
5, use the method for buffer area in the processor according to claim 4, it is characterized in that the dispose procedure of step 5) is specially:
If buffer state is a normal condition, then discharges the current cache district, and empty the information in the buffer area descriptor;
If buffer state is an error condition, then discharges the current cache district, and keep described buffer state and user's information.
6, use the method for buffer area in the processor according to claim 5, it is characterized in that described step also comprises: if buffer state is an error condition, then according to described user's information location error source.
7, use the method for buffer area in the processor according to claim 5, it is characterized in that the information of the descriptor of buffer area described in the step 1) also comprises: user mode information is used to identify buffer area and whether is in idle condition;
During described application buffer area and during described release buffer area, also comprise the described user mode information of upgrading.
8, use the method for buffer zone in the processor according to claim 5, it is characterized in that the information of the described buffer area descriptor of step 1) also comprises: type of error information;
If buffer state is an error condition, then fills in described fault type information, and when discharging the current cache district, keep described type of error information.
9, use the method for buffer zone in the processor according to claim 5, it is characterized in that the information of the described buffer area descriptor of step 1) also comprises: applicant's information and releaser's information;
During described application buffer area, also comprise: fill in described applicant's information;
If when described buffer state is normal condition, when then discharging described current cache district, also comprise the described releaser's information that empties;
When if described buffer state is normal condition, when then discharging described current cache district, also comprise keeping described releaser's information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100971026A CN100361093C (en) | 2004-12-07 | 2004-12-07 | Method for using buffer memory area in processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100971026A CN100361093C (en) | 2004-12-07 | 2004-12-07 | Method for using buffer memory area in processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1786926A true CN1786926A (en) | 2006-06-14 |
| CN100361093C CN100361093C (en) | 2008-01-09 |
Family
ID=36784397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100971026A Expired - Fee Related CN100361093C (en) | 2004-12-07 | 2004-12-07 | Method for using buffer memory area in processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100361093C (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101478481B (en) * | 2009-01-07 | 2011-04-06 | 成都市华为赛门铁克科技有限公司 | Buffer management method and apparatus, data forwarding system |
| CN102209016A (en) * | 2010-03-29 | 2011-10-05 | 成都市华为赛门铁克科技有限公司 | Data processing method, device and data processing system |
| CN101562559B (en) * | 2008-04-15 | 2011-12-07 | 大唐移动通信设备有限公司 | Method and device for serial Rapid IO line data transmission |
| CN102790723A (en) * | 2012-08-29 | 2012-11-21 | 福建星网锐捷网络有限公司 | Processing method, device and equipment for buffer objects |
| CN101350744B (en) * | 2008-06-19 | 2013-08-07 | 中兴通讯股份有限公司 | Method for detecting descriptor resource of network processing system and system thereof |
| CN103914355A (en) * | 2014-03-12 | 2014-07-09 | 汉柏科技有限公司 | Method and device for positioning memory repeated release |
| CN104317734A (en) * | 2014-11-28 | 2015-01-28 | 迈普通信技术股份有限公司 | Memory allocation method and device applicable to SLAB |
| CN103286749B (en) * | 2013-06-07 | 2016-04-13 | 彭辉 | A kind of Multifunctional shovel |
| WO2019061270A1 (en) * | 2017-09-29 | 2019-04-04 | 深圳市大疆创新科技有限公司 | Data caching device and control method therefor, data processing chip, and data processing system |
| CN110800008A (en) * | 2019-03-04 | 2020-02-14 | 阿里巴巴集团控股有限公司 | Building block chain world state mercker patricia dictionary tree subtree |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002164940A (en) * | 2000-11-28 | 2002-06-07 | Fujitsu Ltd | System for detecting fault of descriptor |
| KR100403620B1 (en) * | 2001-02-28 | 2003-10-30 | 삼성전자주식회사 | Communication system and method for raising coefficient of utilization of channels |
| CN1189826C (en) * | 2002-05-23 | 2005-02-16 | 华为技术有限公司 | Data receiving and transmitting processing method based on multibusiness and multichannel |
| CN100343826C (en) * | 2003-04-29 | 2007-10-17 | 华为技术有限公司 | Method for implementing memory management |
-
2004
- 2004-12-07 CN CNB2004100971026A patent/CN100361093C/en not_active Expired - Fee Related
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101562559B (en) * | 2008-04-15 | 2011-12-07 | 大唐移动通信设备有限公司 | Method and device for serial Rapid IO line data transmission |
| CN101350744B (en) * | 2008-06-19 | 2013-08-07 | 中兴通讯股份有限公司 | Method for detecting descriptor resource of network processing system and system thereof |
| CN101478481B (en) * | 2009-01-07 | 2011-04-06 | 成都市华为赛门铁克科技有限公司 | Buffer management method and apparatus, data forwarding system |
| CN102209016A (en) * | 2010-03-29 | 2011-10-05 | 成都市华为赛门铁克科技有限公司 | Data processing method, device and data processing system |
| CN102209016B (en) * | 2010-03-29 | 2014-02-26 | 成都市华为赛门铁克科技有限公司 | Data processing method, device and data processing system |
| CN102790723A (en) * | 2012-08-29 | 2012-11-21 | 福建星网锐捷网络有限公司 | Processing method, device and equipment for buffer objects |
| CN103286749B (en) * | 2013-06-07 | 2016-04-13 | 彭辉 | A kind of Multifunctional shovel |
| CN103914355A (en) * | 2014-03-12 | 2014-07-09 | 汉柏科技有限公司 | Method and device for positioning memory repeated release |
| CN104317734A (en) * | 2014-11-28 | 2015-01-28 | 迈普通信技术股份有限公司 | Memory allocation method and device applicable to SLAB |
| WO2019061270A1 (en) * | 2017-09-29 | 2019-04-04 | 深圳市大疆创新科技有限公司 | Data caching device and control method therefor, data processing chip, and data processing system |
| CN110800008A (en) * | 2019-03-04 | 2020-02-14 | 阿里巴巴集团控股有限公司 | Building block chain world state mercker patricia dictionary tree subtree |
| CN110800008B (en) * | 2019-03-04 | 2023-06-30 | 创新先进技术有限公司 | Building a blockchain world state merker patril Xia Zidian tree subtree |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100361093C (en) | 2008-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1181435C (en) | Method and system to obtain optimum utility through resource recovery | |
| CN1801106A (en) | Error monitoring of partitions in a computer system using supervisor partitions | |
| US8417991B2 (en) | Mitigating reduction in availability level during maintenance of nodes in a cluster | |
| CN101036125A (en) | Masking within a data processing system having applicability for a development interface | |
| US7697443B2 (en) | Locating hardware faults in a parallel computer | |
| CN1702625A (en) | Method and apparatus to preserve trace data | |
| US20100017655A1 (en) | Error Recovery During Execution Of An Application On A Parallel Computer | |
| CN1786926A (en) | Method for using buffer memory area in processor | |
| CN1975696A (en) | Method and calculating device for management memory | |
| CN103106134B (en) | A kind of performance deficiency detection method, device and system | |
| US20070242609A1 (en) | Locating Hardware Faults in a Data Communications Network of a Parallel Computer | |
| CN1960261A (en) | Method and system of obtaining physical slot number of single board, and type of slot | |
| CN1959652A (en) | Method and apparatus for debugging computer program in distributed debugger | |
| US9003240B2 (en) | Blackbox memory monitoring with a calling context memory map and semantic extraction | |
| CN1707462A (en) | Method and apparatus for dynamic hosting partition page assignment | |
| KR20130101548A (en) | Improving reliability in distributed environments | |
| CN1908906A (en) | Method for monitoring software operation state | |
| CN102207956A (en) | Database management method, database management system and program thereof | |
| EP2424190A1 (en) | Monitoring connections | |
| CN105446871A (en) | Resource leakage detection method, apparatus and system | |
| CN1838078A (en) | Method and system for swapping code in a digital signal processor | |
| CN101034417A (en) | Multithreaded reachability | |
| US20080201393A1 (en) | Identifying unnecessary synchronization objects in software applications | |
| US8850273B2 (en) | Dynamic hardware watchpoint | |
| CN116027365A (en) | Distributed GNSS deformation monitoring real-time resolving system, device and method |
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: 20080109 Termination date: 20191207 |