CN102388368B - Method and device for monitoring memory - Google Patents
Method and device for monitoring memory Download PDFInfo
- Publication number
- CN102388368B CN102388368B CN201180002063.5A CN201180002063A CN102388368B CN 102388368 B CN102388368 B CN 102388368B CN 201180002063 A CN201180002063 A CN 201180002063A CN 102388368 B CN102388368 B CN 102388368B
- Authority
- CN
- China
- Prior art keywords
- leaf
- debugging
- address
- skipping
- page
- 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
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/362—Software debugging
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a method and device for monitoring a memory. The method includes the steps of placing memory has to be monitered to a debug page, triggering a page-missing abornality commond when CPU reads and writes the memory address of the debug page, determing whether the page-missing abornality commond is caused by a memory address of which accessing is monitored, and processing the page-missing abornality commond according to a preset modulating process. The invention also provides a memory monitoring device. Through the technical solution of the invention, accessing other memory addresses other than the monitoring address can not trigger abnormity, thereby preventing mighty influence on system performance, and effectively improving debugging and maintainance efficiency of asystem.
Description
Technical field
The present invention relates to computer application field, in particular, relate to a kind of internal memory monitoring method and device.
Background technology
In computer application field, whether monitor a certain section internal memory accessed, typically use that some professional internal memory monitoring softwares realize, but after monitoring is destroyed, this internal memory monitoring software can not in time report mistake, can only just can find mistake in internal memory release or Memory Allocation resource, therefore it is ageing that this scheme does not have monitoring, and the debug difficulty of calling program is significantly increased; In addition, existing internal memory monitoring software can not be compatible fully for independently developed operating system, also makes the debug effect quite limited.How can be real-time monitor internal memory and whether read and write and do not destroy the system of current operation, and then shortens the construction cycle, is a major issue that will solve at present.
At present, debug registers can provide the read-write of 4 address registers for the monitoring memory address, and it is fast to have processing speed, can be to the characteristics of the real-time monitoring of internal memory.The detection of data breakpoint utilizes this mechanism to realize, but the realization of this technology will depend on hardware mechanisms, so usable range is very limited; Jian Kong address number is limited simultaneously, has caused developer's debugging bottleneck.
Part CPU support hardware breakpoint in the existing scheme under the X86 framework, has 8 debug registers DR0-DR7, is used for preserving memory address or the I/O port address of breakpoint.Current data breakpoint technology utilizes debug registers to realize, the DR0-DR3 register is put in the address that needs monitor, gives the corresponding set of debugging control register DR7, and the monitor address of depositing among the DR0-DR3 is come into force.
When program is read and write the variable of breakpoint of certain setting, it is unusual to trigger debugging, kernel is just unusual by taking over debugging, the decision operation of being correlated with and other are handled operation, and then carry out breakpoint information and record and export, whole process does not influence the normal operation of program, and can successfully record breakpoint information.
In realizing process of the present invention, the inventor finds that there are the following problems at least in the prior art:
Existing scheme depends on the hardware debug resource, because portion C PU framework do not have debug registers, and therefore support hardware breakpoint not, poor compatibility simultaneously because the number of debug registers is limited, has also limited the usable range that data breakpoint is surveyed quantity.
Summary of the invention
Embodiments of the invention provide a kind of internal memory monitoring method and device, can reduce the limitation of data breakpoint technology in the existing scheme.
Technical scheme of the present invention comprises:
One aspect of the present invention provides a kind of internal memory monitoring method, comprising:
The memory address of needs monitoring is put into the debugging page or leaf;
CPU to described debugging page or leaf in during the memory address read-write, then trigger the exceptional instructions that skips leaf;
Be when causing, to adopt predetermined debugging flow process that the described exceptional instructions that skips leaf is handled by the monitored memory address of visit at definite described exceptional instructions that skips leaf.
The present invention provides a kind of internal memory monitoring device on the other hand, comprising:
Integration module, the data that are used for will needing to monitor are put into the debugging page or leaf;
Monitoring module is used at CPU during to described debugging page or leaf reading and writing data, then the triggering exceptional instructions that skips leaf;
Execution module, being used at definite described exceptional instructions that skips leaf is when being caused by the monitored memory address of visit, adopts the predetermined flow process of debugging that the described exceptional instructions that skips leaf is handled.
The technical scheme that is provided by the embodiment of the invention described above as can be seen, put into the debugging page or leaf by the memory address that will need to monitor, make that the visit of other internal storage datas can not trigger unusually, can not cause excessive influence to system performance except the monitoring memory address; CPU to described debugging page or leaf in during reading and writing data, then trigger the exceptional instructions that skips leaf; Be when causing, to adopt predetermined debugging flow process that the described exceptional instructions that skips leaf is handled, effectively the debugging of Hoisting System and maintenance efficiency by the monitored memory address of visit at definite described exceptional instructions that skips leaf.
Description of drawings
Fig. 1 is the process flow diagram of the described internal memory monitoring method of the embodiment of the invention;
Fig. 2 is the process flow diagram of the embodiment of the invention one;
Fig. 3 is the structural representation of the described internal memory monitoring device of the embodiment of the invention.
Embodiment
A kind of internal memory monitoring method and device that the embodiment of the invention proposes are a kind ofly can increase breakpoint usage quantity and scope based on the breakpoint detecting strategy that skips leaf unusual, the limitation of having avoided existing scheme that number and the monitored length of data breakpoint are brought.The present invention utilizes the protection mechanism of page or leaf to realize the multidata breakpoint in addition, has versatility.
The present invention puts into the debugging page or leaf by the memory address that will need to monitor, and makes that all the other internal storage access can't trigger unusually, thereby are unlikely to system performance is caused excessive influence except the monitoring internal memory; CPU to described debugging page or leaf in during the memory address read-write, then trigger the exceptional instructions that skips leaf; Be when causing, to adopt predetermined debugging flow process that the described exceptional instructions that skips leaf is handled by the monitored memory address of visit at definite described exceptional instructions that skips leaf.Can monitor the state in the physical memory in this way, grasp the person of writing and the related data of reading taker simultaneously, improve the processing speed of EMS memory error location at random, effectively promote the efficient of debugging and the maintenance of system.
For the ease of the understanding to relational language in the embodiment of the invention, below the implication of the term that relates to is carried out brief description:
Data breakpoint: so-called data breakpoint refers to that CPU can suspend this instruction and carry out when memory address of visit, then carries out an interrupt routine, finishes the process that presence states grasps.
Paging: the memory management function that modern CPU provides, it is divided into Physical Page one by one with whole physical memories, and these Physical Page have some controlled attributes, and this technology is called paging.
Physical Page: every internal memory after cutting apart is called a Physical Page.
Mistake skips leaf: it is unusual based on above-mentioned paging mechanism to skip leaf; under paging; linear address and physical address might not be corresponding one by one; thereby may there be the situation of corresponding physical page or leaf in the visit that has the linear address of CPU; CPU can produce a mistake, and this situation just is called as the mistake that skips leaf.
Debugging page or leaf: a Physical Page, to the data in this page operate all need monitored.
As shown in Figure 1, the embodiment of the invention provides a kind of internal memory monitoring method, and technical scheme comprises:
101, the memory address that will need to monitor is put into the debugging page or leaf;
102, CPU to described debugging page or leaf in during the memory address read-write, then trigger the exceptional instructions that skips leaf;
103, be when causing, to adopt predetermined debugging flow process that the described exceptional instructions that skips leaf is handled by the monitored memory address of visit at definite described exceptional instructions that skips leaf.
In an optional embodiment of the present invention, the process of step 101 specifically comprises:
The memory address of needs monitoring is placed in one or more debugging pages or leaves, and with the footers position of described debugging page for skipping leaf or read-only.
Concrete, in the embodiment of the invention all memory addresss that need monitor are placed in the special section; This section is formed one or more Physical Page, and this Physical Page can become the debugging page or leaf; In addition, the footers position of described debugging page or leaf is set to skip leaf or read-only purpose is for the read or write of surveying internal memory.
Wherein, " special section " refers to that compiler is one section binary code that program produces, and different sections can be loaded into different region of memorys, can separate data monitored and that do not monitor like this.
In addition, the memory address of needs monitoring is put into after the debugging page or leaf, still needed certain or some data are carried out the monitoring of data breakpoint, memory address that can dynamically will be scattered adds in the debugging page.
As seen, make except monitor data by step 101 that all the other internal storage access can't trigger and skip leaf unusually, thereby are unlikely to system performance is caused excessive influence.
Optionally, the embodiment of the invention can be looked configuring condition and determine whether that mark should debugging page exist in the internal memory after system is started, so that the unlatching of realization data breakpoint function or close; When determining to need to open debugging (debug) function, be 0 with the present position, CPU execution in step 102 is gone to read and write any data of debugging in the page or leaf and all can be produced the mistake that once skips leaf during this period.
In an optional embodiment of the present invention, described method also comprises:
Find the address of skipping leaf according to described skipping leaf, and judge that the described address of skipping leaf is whether within the address realm of debugging page or leaf;
If the described address of skipping leaf not within the address realm of debugging page or leaf, determines then that the described mistake that skips leaf is not to be caused by the monitored memory address of visit, adopts the wrong treatment scheme that normally skips leaf to handle;
Within the address realm of debugging page or leaf, determine then that the described mistake that skips leaf is to be caused by the monitored memory address of visit as if the described address of skipping leaf, predetermined debugging flow process is handled the described exceptional instructions that skips leaf in the employing step 103.
Concrete, the predetermined debugging of the described employing of step 103 flow process is handled the described exceptional instructions that skips leaf, and comprising:
(1) backup causes the unusual instruction of skipping leaf, and the described unusual instruction that causes skipping leaf is replaced with the unusual instruction of a generation CPU;
(2) described debugging page or leaf existence is set, allows read-write; Wherein, the purpose of doing like this is for the content that reads debugging page or leaf, suppose if described debugging page or leaf is in the state of skipping leaf always, and the data in the described debugging of so each read-write page all can produce one unusually, thereby be absorbed in circulation, the content that causes debugging in the page or leaf can't read;
(3) carry out the unusual instruction of described generation CPU, and switch to CPU abnormality processing flow process;
(4) with the data read in the internal memory and preserve, and context data is preserved, in order to obtain monitor data in this data breakpoint process.
By technique scheme, in predetermined debugging abnormality processing flow process, can grasp the original value that CPU needs the read/write memory address, and record to produce and read and write program stack and process constantly, thereby finish the functional requirement of data breakpoint.
Described step 103 can also comprise:
Backup is carried out in single step describedly causes the unusual instruction of skipping leaf;
In the treatment scheme of the described unusual instruction that causes skipping leaf, described debugging page or leaf is set to not existence, in order to wait for the triggering of data breakpoint next time.
By technique scheme as can be known, the present invention has rationally adjusted the deposit position of monitor data, and the data of needs monitoring are put into the debugging page or leaf, the internal memory that utilizes CPU to provide the simultaneously mechanism that skips leaf, and monitoring CPU is to the read-write motion of particular address; Expanded the breakpoint mechanism that CPU hardware provides, expanded memory address quantity and the length of monitoring, avoided the restriction of hardware, thereby effectively promoted debugging and the maintenance efficiency of system.
Existing CPU substantially supports the paging management mechanism of internal memory, and different CPU has different physical page size, and typically the physical page size as the CPU of x86 framework just has 4KB, 2MB, three kinds of different sizes of 4MB.Physical Page with the 4KB size is example, its effective monitored length is 4KB, the number of its monitoring changes with the size of monitored data individuality, as the int data for 4 byte longs, then can monitor 1024 data item simultaneously, the number of debugging page or leaf also is not limited to 1 debugging page or leaf, makes length and the number of monitoring to be expanded thereby can expand to a plurality of pages or leaves.
The embodiment of the invention one:
Inscribe before having the realization that paging mechanism function is the embodiment of the invention one with CPU; In addition, in embodiment one, the paging size is set at 4KB;
Idiographic flow is as shown in Figure 2:
201, the memory address that will need to monitor is placed in one or more debugging pages or leaves, and with the footers position of described debugging page or leaf for skipping leaf or read-only;
202, after system starts, determine whether that according to configuring condition mark should debugging page exist in the internal memory, so that the unlatching of realization data breakpoint function or close; Debugging (debug) function has been opened in supposition in the embodiment of the invention one, and therefore the present position being set is 0;
203, CPU then can trigger the exceptional instructions that skips leaf in the described debugging page or leaf during memory address read-write, and system carries out and forwards the abnormality processing flow process that skips leaf to;
204, judge whether the described exceptional instructions that skips leaf is caused by the monitored memory address of visit;
Be specially: find the address of skipping leaf according to described skipping leaf, and judge that the described address of skipping leaf is whether within the address realm of debugging page or leaf;
If the described address of skipping leaf not within the address realm of debugging page or leaf, determines then that the described mistake that skips leaf is not to be caused by the monitored memory address of visit, adopts the wrong treatment scheme that normally skips leaf to handle;
205, if the described address of skipping leaf within the address realm of debugging page or leaf, determines then that the described mistake that skips leaf is to be caused by the monitored memory address of visit, adopt predetermined debugging flow process that the described exceptional instructions that skips leaf is handled;
Be specially: backup causes the unusual instruction of skipping leaf, and the described unusual instruction that causes skipping leaf is replaced with the unusual instruction of a generation CPU;
206, the abnormality processing flow process that skips leaf is returned, and carries out the unusual instruction of described generation CPU, and switches to CPU abnormality processing flow process, and described debugging page or leaf existence is set, and allows read-write;
207, with the data read in the internal memory and preserve, and context data is preserved, in order to obtain monitor data in this data breakpoint process;
208, the instruction of backup is carried out in single step, just carries out to cause the unusual instruction of skipping leaf;
209, in the treatment scheme of the described unusual instruction that causes skipping leaf, described debugging page or leaf is set to not existence, in order to wait for the triggering of data breakpoint next time;
210, finish to skip leaf the abnormity processing flow process, finish the function of data breakpoint, return step 201.
So far program is returned former execution stream, has finished the function of data breakpoint simultaneously.
Technical scheme by the invention described above embodiment one proposes can increase breakpoint usage quantity and scope, the limitation of having avoided existing scheme that number and the monitored length of data breakpoint are brought.The present invention utilizes the protection mechanism of page or leaf to realize the multidata breakpoint in addition, has versatility.
As shown in Figure 3, the embodiment of the invention also provides a kind of internal memory monitoring device, comprising:
In an optional embodiment of the present invention, described execution module 33 comprises:
Back-up processing unit 331 is used for backup and causes the unusual instruction of skipping leaf, and the described unusual instruction that causes skipping leaf is replaced with the unusual instruction of a generation CPU; Described debugging page or leaf existence is set, allows read-write;
Replacement instruction performance element 332 is used for carrying out the unusual instruction of described generation CPU, and switches to CPU abnormality processing flow process; Also be used for the data read of internal memory and preserve, and context data is preserved, in order to obtain monitor data in this data breakpoint process.
Further alternative, described execution module 33 can also comprise:
Former instruction execution unit 333, what be used for that single step carries out backup describedly causes the unusual instruction of skipping leaf;
In the treatment scheme of the described unusual instruction that causes skipping leaf, described debugging page or leaf is set to not existence, in order to wait for the triggering of data breakpoint next time.
In an optional embodiment of the present invention, described device can also comprise:
If the described address of skipping leaf not within the address realm of debugging page or leaf, determines then that the described mistake that skips leaf is not to be caused by the monitored memory address of visit, described execution module 33 adopts the wrong treatment scheme that normally skips leaf to handle.
In an optional embodiment of the present invention, described integration module 31 can specifically be used for:
The memory address of needs monitoring is placed in one or more debugging pages or leaves, and with the footers position of described debugging page for skipping leaf or read-only.
Need to prove, the embodiment of the invention is based on the device embodiment of Fig. 1 and the described method embodiment acquisition of Fig. 2, comprised identical with described method embodiment or corresponding technical characterictic, therefore, in embodiments of the present invention the technical scheme of repeatability is not done detailed description, see also the description among the method embodiment of Figure 1 and Figure 2.
One of ordinary skill in the art will appreciate that all or part of flow process that realizes in above-described embodiment method, be to instruct relevant hardware to finish by computer program, described program can be stored in the computer read/write memory medium, this program can comprise the flow process as the embodiment of above-mentioned each side method when carrying out.Wherein, described storage medium can be magnetic disc, CD, read-only storage memory body (Read-Only Memory, ROM) or at random store memory body (Random Access Memory, RAM) etc.
Claims (8)
1. an internal memory monitoring method is characterized in that, comprising:
The memory address of needs monitoring is put into the debugging page or leaf;
CPU to described debugging page or leaf in during the memory address read-write, then trigger the exceptional instructions that skips leaf;
Be when causing, to adopt predetermined debugging flow process that the described exceptional instructions that skips leaf is handled by the monitored memory address of visit at definite described exceptional instructions that skips leaf;
The predetermined debugging of described employing flow process is handled the described exceptional instructions that skips leaf, and comprising:
Backup causes the unusual instruction of skipping leaf, and the described unusual instruction that causes skipping leaf is replaced with the unusual instruction of a generation CPU;
Described debugging page or leaf existence is set, allows read-write;
Carry out the unusual instruction of described generation CPU, and switch to CPU abnormality processing flow process;
With the data read in the internal memory and preserve, and context data is preserved, in order to obtain monitor data in this data breakpoint process.
2. method according to claim 1 is characterized in that, described method also comprises:
That carries out backup describedly causes the unusual instruction of skipping leaf;
In the treatment scheme of the described unusual instruction that causes skipping leaf, described debugging page or leaf is set to not existence, in order to wait for the triggering of data breakpoint next time.
3. method according to claim 1 is characterized in that, described method also comprises:
Find the address of skipping leaf according to described skipping leaf, and judge that the described address of skipping leaf is whether within the address realm of debugging page or leaf;
If the described address of skipping leaf not within the address realm of debugging page or leaf, determines then that the described mistake that skips leaf is not to be caused by the monitored memory address of visit, adopts the wrong treatment scheme that normally skips leaf to handle.
4. method according to claim 1 is characterized in that, the memory address of needs monitoring is put into the debugging page or leaf, comprising:
The memory address of needs monitoring is placed in one or more debugging pages or leaves, and with the footers position of described debugging page for skipping leaf or read-only.
5. an internal memory monitoring device is characterized in that, comprising:
Integration module, the memory address that is used for will needing to monitor is put into the debugging page or leaf;
Monitoring module is used for then triggering the exceptional instructions that skips leaf when CPU reads and writes described debugging page or leaf memory address;
Execution module, being used at definite described exceptional instructions that skips leaf is when being caused by the monitored memory address of visit, adopts the predetermined flow process of debugging that the described exceptional instructions that skips leaf is handled;
Described execution module comprises:
The back-up processing unit is used for backup and causes the unusual instruction of skipping leaf, and the described unusual instruction that causes skipping leaf is replaced with the unusual instruction of a generation CPU; Described debugging page or leaf existence is set, allows read-write;
The replacement instruction performance element is used for carrying out the unusual instruction of described generation CPU, and switches to CPU abnormality processing flow process; Also be used for the data read of internal memory and preserve, and context data is preserved, in order to obtain monitor data in this data breakpoint process.
6. device according to claim 5 is characterized in that, described execution module also comprises:
Former instruction execution unit, what be used for that single step carries out backup describedly causes the unusual instruction of skipping leaf;
In the treatment scheme of the described unusual instruction that causes skipping leaf, described debugging page or leaf is set to not existence, in order to wait for the triggering of data breakpoint next time.
7. device according to claim 5 is characterized in that, described device also comprises:
Determination module is used for finding the address of skipping leaf according to described skipping leaf, and judges that the described address of skipping leaf is whether within the address realm of debugging page or leaf;
If the described address of skipping leaf not within the address realm of debugging page or leaf, determines then that the described mistake that skips leaf is not to be caused by the monitored memory address of visit, described execution module adopts the wrong treatment scheme that normally skips leaf to handle.
8. device according to claim 5 is characterized in that, described integration module specifically is used for:
The data of needs monitoring are placed in one or more debugging pages or leaves, and with the footers position of described debugging page for skipping leaf or read-only.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2011/079888 WO2012119446A1 (en) | 2011-09-20 | 2011-09-20 | Memory monitoring method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102388368A CN102388368A (en) | 2012-03-21 |
| CN102388368B true CN102388368B (en) | 2013-10-02 |
Family
ID=45826492
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201180002063.5A Expired - Fee Related CN102388368B (en) | 2011-09-20 | 2011-09-20 | Method and device for monitoring memory |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102388368B (en) |
| WO (1) | WO2012119446A1 (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239200B (en) * | 2013-06-20 | 2018-10-30 | 上海博达数据通信有限公司 | Memory read-write monitoring method in a kind of hard single step system |
| CN105718357B (en) * | 2016-01-22 | 2018-12-28 | 烽火通信科技股份有限公司 | A kind of method of internal memory monitoring |
| CN109977061B (en) * | 2017-12-28 | 2023-04-11 | 中兴通讯股份有限公司 | Interrupt processing method and interrupt processing device |
| WO2021026938A1 (en) * | 2019-08-15 | 2021-02-18 | 奇安信安全技术(珠海)有限公司 | Shellcode detection method and apparatus |
| CN110764995B (en) * | 2019-09-05 | 2023-06-06 | 北京字节跳动网络技术有限公司 | Method, device, medium and electronic equipment for detecting file access abnormality |
| CN110716747B (en) * | 2019-09-17 | 2022-08-12 | 百富计算机技术（深圳）有限公司 | Program operation efficiency optimization method based on function parameter statistics and terminal equipment |
| CN110764914B (en) * | 2019-10-28 | 2022-09-20 | 锐捷网络股份有限公司 | Memory rewriting positioning method and device |
| CN111400096B (en) * | 2020-03-16 | 2023-05-02 | 杭州涂鸦信息技术有限公司 | Memory mirroring method based on linux page-missing mechanism and system and device thereof |
| CN112269716A (en) * | 2020-06-01 | 2021-01-26 | 中国科学院信息工程研究所 | Flexibly defined processor abnormal access real-time monitoring method and electronic device |
| CN112463544B (en) * | 2020-12-17 | 2023-05-02 | 深圳软牛科技有限公司 | Program running monitoring method, device, equipment and storage medium |
| CN112685212B (en) * | 2021-01-05 | 2024-03-19 | 上海擎昆信息科技有限公司 | Processor exception debugging and tracking method, device and system |
| CN117453495B (en) * | 2023-12-26 | 2024-03-26 | 睿思芯科(成都)科技有限公司 | Chip supporting online error correction and debugging, design method and related equipment |
| CN117648196B (en) * | 2024-01-29 | 2024-04-26 | 苏州元脑智能科技有限公司 | Memory allocation protection method, device, equipment and medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5613063A (en) * | 1994-07-01 | 1997-03-18 | Digital Equipment Corporation | Method and apparatus for checking validity of memory operations |
| US7490117B2 (en) * | 2003-12-31 | 2009-02-10 | Intel Corporation | Dynamic performance monitoring-based approach to memory management |
| US7890804B2 (en) * | 2007-09-28 | 2011-02-15 | Cypress Semiconductor Corporation | Program memory test access collar |
| CN101441600B (en) * | 2008-12-25 | 2010-12-08 | 中国科学院计算技术研究所 | Apparatus and method for rapidly changing page in microprocessor verification |
| CN101561775B (en) * | 2009-05-12 | 2010-09-15 | 华为技术有限公司 | Method and device for monitoring memory |
-
2011
- 2011-09-20 CN CN201180002063.5A patent/CN102388368B/en not_active Expired - Fee Related
- 2011-09-20 WO PCT/CN2011/079888 patent/WO2012119446A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| CN102388368A (en) | 2012-03-21 |
| WO2012119446A1 (en) | 2012-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102388368B (en) | Method and device for monitoring memory | |
| Moraru et al. | Consistent, durable, and safe memory management for byte-addressable non volatile main memory | |
| Lee et al. | A semi-preemptive garbage collector for solid state drives | |
| CN100474268C (en) | Cache integrity apparatus, systems, and methods | |
| US10824362B2 (en) | File migration to persistent memory | |
| CN101458725B (en) | Microcontroller chip and debug method thereof | |
| JP2005339561A5 (en) | ||
| KR102443600B1 (en) | hybrid memory system | |
| Moraru et al. | Persistent, protected and cached: Building blocks for main memory data stores | |
| KR101862094B1 (en) | Software replayer for transactional memory programs | |
| Chen et al. | Unified non-volatile memory and NAND flash memory architecture in smartphones | |
| Xue et al. | Adaptive memory fusion: Towards transparent, agile integration of persistent memory | |
| CN102037448A (en) | Device emulation support within a host data processing apparatus | |
| EP3871093A1 (en) | Processor memory reordering hints in a bit-accurate trace | |
| Joyce et al. | Exploring storage bottlenecks in Linux-based embedded systems | |
| CN105612505A (en) | Method and apparatus for scheduling CPU | |
| Won et al. | Trapfetch: A breakpoint-based prefetcher for both launch and run-time | |
| Spear et al. | Alert-on-update: a communication aid for shared memory multiprocessors | |
| Zhang et al. | Fast persistent heap based on non-volatile memory | |
| US11693725B2 (en) | Detecting execution hazards in offloaded operations | |
| US11782854B2 (en) | Cache architecture for a storage device | |
| 손용석 | Optimizing File Systems for High-Performance Storage Devices | |
| Xu et al. | I/O Transit Caching for PMem-based Block Device | |
| Cockrum et al. | Motivating engineering students to write technical papers | |
| Zhang et al. | On virtual-machine-based windows file reads: a performance study |
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: 20131002 Termination date: 20180920 |