CN101706754A - Method and device for allocating continuous physical memory in embedded system - Google Patents
Method and device for allocating continuous physical memory in embedded system Download PDFInfo
- Publication number
- CN101706754A CN101706754A CN200910192199A CN200910192199A CN101706754A CN 101706754 A CN101706754 A CN 101706754A CN 200910192199 A CN200910192199 A CN 200910192199A CN 200910192199 A CN200910192199 A CN 200910192199A CN 101706754 A CN101706754 A CN 101706754A
- Authority
- CN
- China
- Prior art keywords
- physical memory
- continuous physical
- memory
- embedded system
- parameter
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a method and a device for allocating continuous physical memory in an embedded system. The method comprises the following steps: in the boost loader of the embedded system, taking the difference of the quantity of the total physical memory and the quantity of the continuous physical memory required to be allocated as a parameter value to set a kernel boot parameter of the quantity of the kernel management memory; and allocating the continuous physical memory according to the quantity of the continuous physical memory required to be allocated in the drive of the embedded system. The invention can ensure successful physical memory allocation in the drive of the embedded system, and can avoid system exception caused by the failure of applying for continuous physical memory.
Description
Technical field
The present invention relates to field of computer technology, particularly continuous physical memory distribution method and device in the embedded system.
Background technology
In computer technology, Memory Allocation comprises that the continuous physical memory distribution technique is very crucial for the influence of computing power.Wherein, in the embedded system driving, there are the occasions that much must distribute continuous physical memories, as: DMA Memory Allocation or distribution contiguous memory are to improve system performance etc.And if distribution may cause system exception or system performance to reduce greatly less than continuous physical memory.But,, after it moves a bit of time, then be difficult in the continuous physical memory that finds to specify size in the internal memory that spreads all over memory fragmentation for embedded computer system.
For guaranteeing that the DMA contiguous memory is allocated successfully, a kind of prior art is to adopt the memory management module of revising the embedded system kernel to solve this problem.Such as, on July 1st, 2009, No. the 200710306110.0th, disclosed Chinese invention patent application.This patented claim that name is called " method of assigned address range reason internal memory on the linux system platform " discloses the method for distributing specified address area and specifying the physical memory of size, but owing to must revise the memory management module of embedded system kernel, exist the development time long, than complicated problems, and may reduce system performance; Again such as on July 6th, 2005 disclosed Chinese invention patent a kind of new memory management scheme has been proposed for 200310110344.X number, can reduce the generation of memory fragmentation, improve the success ratio of big internal memory application, but equally also exist the development time long, complicated and can not guarantee to distribute the defective of big physical memory success; A kind of method of managing embedded system memory is proposed for No. 200710121449.3 such as disclosed Chinese invention patent application on March 11st, 2009 again, to reclaim memory fragmentation, improve the success ratio of big internal memory application, but exist the development time long, complicated and can not guarantee to distribute the defective of big physical memory success too.
And in the occasion of non-DMA Memory Allocation, after distributing the continuous physical memory failure, prior art proposes to adopt the continuous memory field distribution method of non-physics, but can cause the decline slightly of computing power like this.Because adopt the continuous memory field distribution of non-physics can reduce the probability that page table buffering (TLB) is hit greatly, increase the time of linear address to physical address translations.
Summary of the invention
The technical matters that the present invention mainly solves provides continuous physical memory distribution method and device in a kind of embedded system, can guarantee that continuous physical memory is allocated successfully in the embedded system driving, and avoid occurring the situation of system exception owing to the failure of application continuous physical memory.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is: continuous physical memory distribution method in a kind of embedded system is provided, comprise: in the bootstrap loader of described embedded system, the whole physical memory quantity and the difference of the continuous physical memory quantity that needs to distribute are provided with the kernel start-up parameter of inner core managing amount of memory as parameter value; In the driving of described embedded system, distribute described continuous physical memory by the described continuous physical memory quantity that needs to distribute.
Wherein, described embedded system is the Linux embedded system; The described step that the kernel start-up parameter of inner core managing amount of memory is set comprises: add following information: mem=C in the kernel start-up parameter that is provided with in the bootstrap loader of Linux embedded system, wherein said C is the poor of whole physical memory quantity and the continuous physical memory quantity that needs distribution.
Wherein, in the driving of described embedded system, distribute the step of described continuous physical memory to comprise: the physical address of in the driving of described embedded system, specifying described continuous physical memory by the described continuous physical memory quantity that need distribute by the described continuous physical memory quantity that needs to distribute; The physical address that remaps described continuous physical memory in the driving of described embedded system is a virtual address.
Wherein, in the driving of described embedded system, specify the step of the physical address of described continuous physical memory to comprise: the physical address of in the driving of described embedded system, specifying described continuous physical memory by following statement by the described continuous physical memory quantity that needs to distribute:
#define?REAL_MEM_SIZE 0x2000000；
u8 *pp_virt_buf＝NULL；
dma_addr_t?pp_phys_buf；
unsigned?int t_size＝0x100000；
pp_phys_buf＝PHYS_OFFSET+(REAL_MEM_SIZE-t_size)；
The physical address that remaps described continuous physical memory in the driving of described embedded system is that the step of virtual address comprises: the physical address that remaps described continuous physical memory by following statement in the driving of described embedded system is a virtual address:
pp_virt_buf＝ioremap_nocache(pp_phys_buf，t_size)；
Wherein, after the step of distributing described continuous physical memory, comprising: the described parameter value that the kernel start-up parameter of inner core managing amount of memory is set is revised in the order by described bootstrap loader, to change the described continuous physical memory quantity that needs distribution.
Wherein, after distributing the step of described continuous physical memory, comprising: remove the described kernel start-up parameter that the inner core managing amount of memory is set by the order of described bootstrap loader, when discharging the continuous physical memory of described distribution.
Wherein, after the step of distributing described continuous physical memory, comprising: pp_virt_buf is sent to described driving as the DMA memory address.
For solving the problems of the technologies described above, another technical solution used in the present invention is: continuous physical memory distributor in a kind of embedded system is provided, comprise: parameter set unit, be used for bootstrap loader, the whole physical memory quantity and the difference of the continuous physical memory quantity that needs to distribute be provided with the kernel start-up parameter of inner core managing amount of memory as parameter value in described embedded system; The Memory Allocation unit is used for distributing described continuous physical memory in the driving of described embedded system by the described continuous physical memory quantity that needs to distribute.
Wherein, further comprise: internal memory changes the unit, is used for revising the described parameter value that the kernel start-up parameter of inner core managing amount of memory is set by the order of described bootstrap loader, to change the described continuous physical memory quantity that needs distribution.
Wherein, further comprise: the internal memory releasing unit is used for removing the described kernel start-up parameter that the inner core managing amount of memory is set by the order of described bootstrap loader, when discharging the continuous physical memory of described distribution.
The invention has the beneficial effects as follows: be different from prior art and adopt the memory management module of revising the embedded system kernel, or the continuous memory field distribution method of non-physics obtains continuous physical memory and causes the development time longer, than complicated problems, and may reduce the situation of system performance, the embodiment of the invention mainly utilizes the interior nuclear energy of most of embedded systems to receive the kernel start-up parameter that its bootstrap loader imports into, and according to the characteristics of this parameter configuration kernel, the kernel start-up parameter of inner core managing amount of memory is set in the bootstrap loader of described embedded system, distribute described continuous physical memory with convenient by the described continuous physical memory quantity that needs to distribute, because the free memory outside the inner core managing amount of memory that is provided with in the kernel start-up parameter can not be embedded into the kernel of formula system and use, therefore can guarantee the success that continuous physical memory distributes, dexterously, solve prior art at all and can't guarantee to carry out the technical matters that continuous physical memory is allocated successfully for a long time, make the applicant can apply for specifying the continuous physical memory of size at any time, and avoid occurring the decline of the situation and the system performance of system exception owing to the failure of application continuous physical memory.
In addition, for various embedded systems, its most of bootstrap loader all has provides the order that the kernel start-up parameter is set, that is to say, pass through one embodiment of the invention, the order that the kernel start-up parameter is set of using bootstrap loader to provide is provided, does not need to revise the source code of any bootstrap loader or embedded system kernel, just can realize distributing at any time continuous physical memory, the change memory size of specifying size or discharge the purpose of storage allocation.
Description of drawings
Fig. 1 is the process flow diagram of continuous physical memory distribution method embodiment one in the embedded system of the present invention;
Fig. 2 is the process flow diagram of continuous physical memory distribution method embodiment two in the embedded system of the present invention;
Fig. 3 is the theory diagram of continuous physical memory distributor embodiment one in the embedded system of the present invention;
Fig. 4 is the theory diagram of continuous physical memory distributor embodiment two in the embedded system of the present invention.
Embodiment
In some embedded systems are used, distribute less than continuous physical memory owing to drive, its possibility of result is exactly that system exception or system performance reduce greatly.Traditionally, system developer may adopt other methods that reduce performances to evade this problem or adopt and revise that the memory management module code settles the matter once and for all in the kernel.Method provided by the invention has following advantage: can thoroughly address this problem, can not reduce system performance, the more important thing is on embedded system when implementing simplely, and have more dirigibility.
Be that example is elaborated below with the embedded Linux system, but the alleged embedded system of the embodiment of the invention is not limited to linux system, in other embedded systems as the Windows type, also in like manner can realize, its difference is aspects such as system's setting, software design of each embedded system, and spirit is widely used in the embodiment of the invention in these embedded systems according to the present invention but these differences do not influence those skilled in the art.
Embedded Linux system generally is made of bootstrap loader (Bootloader), linux kernel (kernel), driving (part that also belongs to kernel after the loading), root file system and several parts of application program.
In one embodiment of the invention, mainly utilize linux kernel can receive the kernel start-up parameter that Bootloader imports into, and, comprise that the physical memory size of use is carried out system's setting according to the feature of this parameter configuration kernel.But it is reported, prior art generally all adopts this feature to import information such as root file system place subregion, path, initialize routine place into, technical matters in the face of the distribution that can't guarantee to carry out continuous physical memory for a long time, prior art is failed the ingenious above-mentioned feature of utilizing, to solve the problem of distributing continuous physical memory to make mistakes.
For various embedded systems, its most of Bootloader has provides the order that the kernel start-up parameter is set, that is to say, pass through one embodiment of the invention, do not need to revise the source code of any Bootloader or linux kernel, just can realize distributing at any time the purpose of the continuous physical memory of specifying size.When the memory size of needs changed, the only order that need provide by Bootloader was revised the kernel start-up parameter and is revised to drive and just can achieve the goal.
Below describe each embodiment of the present invention in detail:
Consult Fig. 1, continuous physical memory distribution method embodiment comprises in the embedded system of the present invention:
Step 101: in the bootstrap loader of described embedded system, the whole physical memory quantity and the difference of the continuous physical memory quantity that needs to distribute are provided with the kernel start-up parameter of inner core managing amount of memory as parameter value;
Described whole physical memory quantity can be the total amount of memory of system, and the difference of the continuous physical memory quantity that whole physical memory quantity and need distribute is the amount of memory that the embedded system kernel can be managed; That is to say, when storage allocation, the kernel start-up parameter of inner core managing amount of memory is not set to full memory quantity, but establishes a less value, and the difference of whole physical memory quantity and inner core managing amount of memory is the continuous physical memory quantity that needs distribution;
Step 102: in the driving of described embedded system, distribute described continuous physical memory by the described continuous physical memory quantity that needs to distribute.
Here utilize the interior nuclear energy of embedded system to receive the characteristics of the kernel start-up parameter that its bootstrap loader imports into, according to this kernel start-up parameter configuration continuous physical memory.
Be different from prior art and adopt the memory management module of revising the embedded system kernel, or the continuous memory field distribution method of non-physics obtains continuous physical memory and causes the development time longer, than complicated problems, and may reduce the situation of system performance, the embodiment of the invention mainly utilizes the interior nuclear energy of most of embedded systems to receive the kernel start-up parameter that its bootstrap loader imports into, and according to the characteristics of this parameter configuration kernel, the kernel start-up parameter of inner core managing amount of memory is set in the bootstrap loader of described embedded system, distribute described continuous physical memory with convenient by the described continuous physical memory quantity that needs to distribute, because the free memory outside the inner core managing amount of memory that is provided with in the kernel start-up parameter can not be embedded into the kernel of formula system and use, therefore can guarantee the success that continuous physical memory distributes, dexterously, solve prior art at all and can't guarantee to carry out the technical matters that continuous physical memory is allocated successfully for a long time, make the applicant can apply for specifying the continuous physical memory of size at any time, and avoid occurring the decline of the situation and the system performance of system exception owing to the failure of application continuous physical memory.
In an embodiment of the present invention, described embedded system is the Linux embedded system; The described step that the kernel start-up parameter of inner core managing amount of memory is set comprises: add following information: mem=C in the kernel start-up parameter that is provided with in the bootstrap loader of Linux embedded system, wherein said C is the poor of whole physical memory quantity and the continuous physical memory quantity that needs distribution.
As the kernel start-up parameter is set is: " noinitrd root=/dev/mtdblock3 init=/linuxrcmem=31M console=ttySAC0 ".Wherein 31M is the inner core managing amount of memory.
In another embodiment of the present invention, in the driving of described embedded system, distribute the step of described continuous physical memory to comprise by the described continuous physical memory quantity that needs to distribute:
1) the continuous physical memory quantity of distributing by described need in the driving of described embedded system is specified the physical address of described continuous physical memory;
2) physical address that remaps described continuous physical memory in the driving of described embedded system is a virtual address.
As follows in detail:
1) in the driving of described embedded system, specify the physical address of described continuous physical memory by following statement:
#define?REAL_MEM_SIZE 0x2000000；
u8 *pp_virt-buf＝NULL；
dma_addr_t?pp_phys_buf；
unsigned?int t_size＝0x100000；
pp_phys_buf＝PHYS_OFFSET+(REAL_MEM_SIZE-t_size)；
2) physical address that remaps described continuous physical memory by following statement in the driving of described embedded system is a virtual address:
pp_virt_buf＝ioremap_nocache(pp_phys_buf，t_size)；
In an embodiment of the present invention, after the step of distributing described continuous physical memory, may further include step:
The described parameter value that the kernel start-up parameter of inner core managing amount of memory is set is revised in order by described bootstrap loader, to change the described continuous physical memory quantity that needs distribution.
Be not revise kernel and driving under the situation of source code, changing the size of continuous physical memory simply and easily equally.
In an embodiment of the present invention, after the step of distributing described continuous physical memory, may further include step:
Remove the described kernel start-up parameter that the inner core managing amount of memory is set by the order of described bootstrap loader, when discharging the continuous physical memory of described distribution.
Be not revise kernel and driving under the situation of source code, equally simply and easily to distributing continuous physical memory to discharge.
Distributed after the continuous physical memory, just can utilize for driving supervisor, such as: after the step of distributing described continuous physical memory, comprise step:
Pp_virt_buf is sent to described driving as the DMA memory address.
Consult Fig. 2, below to realize that on the internal memory of a 32MByte a kind of can both to become the method for the continuous physical memory of a 1MByte of the distribution of work at any time be that example is described one of them embodiment of the present invention.
At first, add following information: mem=31M in the kernel start-up parameter that in embedded bootstrap loaders such as u-boot, vivi, is provided with, as " noinitrd root=/dev/mtdblock3init=/linuxrcmem=31M console=ttySAC0 ".Kernel receives after this start-up parameter the preceding 31M internal memory in will 32M internal memory only good at managing, and last 1M internal memory will keep for the program that needs always;
Secondly, in driving, distribute the internal memory of above-mentioned 1Mbyte, owing to this 1M internal memory linux kernel can not use, so can guarantee the success of distribution by following statement.
#define?REAL_MEM_SIZE 0x2000000//32M
u8 *pp_virt_buf＝NULL；
dma_addr_t?pp_phys_buf；
unsigned?int t_size＝0x100000；//32M
Pp_phys_buf=PHYS_OFFSET+ (REAL_MEM_SIZE-t_size); //PHYS_OFFSET is the initial physical address of physical memory, the physical address of pp_phys_buf for distributing
Pp_virt_buf=ioremap_nocache (pp_phys_buf, t_size); // remap and be virtual address
At last, pp_virt_buf can be used for driving the place of the contiguous memory that needs 1M as the DMA memory address.
By whole implementation process as seen, the suitable simple and flexible of the method.When not needing to distribute continuous physical memory, can revise kernel start-up parameter (removing mem=*) by the order of Bootloader, and not need to revise any code, and not influence the performance of system.
Consult Fig. 3, the present invention also provides continuous physical memory distributor embodiment in a kind of embedded system, comprising:
Parameter set unit is used for the bootstrap loader in described embedded system, the whole physical memory quantity and the difference of the continuous physical memory quantity that needs to distribute is provided with the kernel start-up parameter of inner core managing amount of memory as parameter value;
The Memory Allocation unit is used for distributing described continuous physical memory in the driving of described embedded system by the described continuous physical memory quantity that needs to distribute.
In like manner, utilize the interior nuclear energy of most of embedded systems to receive kernel start-up parameter that its bootstrap loader imports into and according to the characteristics of this parameter configuration kernel, continuous physical memory distributor embodiment can carry out the continuous physical memory distribution simply and flexibly in the embedded system of the present invention under the prerequisite of the performance that does not need to revise any code and do not influence system, because the free memory outside the inner core managing amount of memory that is provided with in the kernel start-up parameter can not be embedded into the kernel of formula system and use, therefore guarantee that it is allocated successfully.
Consult Fig. 4, in other embodiments, can further include:
Internal memory changes the unit, is used for revising the described parameter value that the kernel start-up parameter of inner core managing amount of memory is set by the order of described bootstrap loader, to change the described continuous physical memory quantity that needs distribution.
And further comprise:
The internal memory releasing unit is used for removing the described kernel start-up parameter that the inner core managing amount of memory is set by the order of described bootstrap loader, when discharging the continuous physical memory of described distribution.
The continuous physical memory distributor can be the main frame of embedded computer or the processor in the main frame in the described embedded system.
The above only is embodiments of the invention; be not so limit claim of the present invention; every equivalent structure or equivalent flow process conversion that utilizes instructions of the present invention and accompanying drawing content to be done; or directly or indirectly be used in other relevant technical fields, all in like manner be included in the scope of patent protection of the present invention.
Claims (10)
1. continuous physical memory distribution method in the embedded system is characterized in that, comprising:
In the bootstrap loader of described embedded system, the whole physical memory quantity and the difference of the continuous physical memory quantity that needs to distribute are provided with the kernel start-up parameter of inner core managing amount of memory as parameter value;
In the driving of described embedded system, distribute described continuous physical memory by the described continuous physical memory quantity that needs to distribute.
2. method according to claim 1 is characterized in that:
Described embedded system is the Linux embedded system;
The described step that the kernel start-up parameter of inner core managing amount of memory is set comprises: add following information: mem=C in the kernel start-up parameter that is provided with in the bootstrap loader of Linux embedded system, wherein said C is the poor of whole physical memory quantity and the continuous physical memory quantity that needs distribution.
3. method according to claim 2 is characterized in that:
In the driving of described embedded system, distribute the step of described continuous physical memory to comprise by the described continuous physical memory quantity that needs to distribute:
In the driving of described embedded system, specify the physical address of described continuous physical memory by the described continuous physical memory quantity that need distribute;
The physical address that remaps described continuous physical memory in the driving of described embedded system is a virtual address.
4. method according to claim 3 is characterized in that:
In the driving of described embedded system, specify the step of the physical address of described continuous physical memory to comprise: the physical address of in the driving of described embedded system, specifying described continuous physical memory by following statement by the described continuous physical memory quantity that needs to distribute:
#define?REAL_MEM_SIZE 0x2000000；
u8 *pp_virt_buf＝NULL；
dma_addr_t pp_phys_buf；
unsigned?int t_size＝0x100000；
pp_phys_buf＝PHYS_OFFSET+(REAL_MEM_SIZE-t_size)；
The physical address that remaps described continuous physical memory in the driving of described embedded system is that the step of virtual address comprises: the physical address that remaps described continuous physical memory by following statement in the driving of described embedded system is a virtual address:
pp_virt_buf＝ioremap_nocache(pp_phys_buf，t_size)。
5. according to each described method of claim 1 to 4, it is characterized in that, after the step of distributing described continuous physical memory, comprising:
The described parameter value that the kernel start-up parameter of inner core managing amount of memory is set is revised in order by described bootstrap loader, to change the described continuous physical memory quantity that needs distribution.
6. according to each described method of claim 1 to 4, it is characterized in that, after the step of distributing described continuous physical memory, comprising:
Remove the described kernel start-up parameter that the inner core managing amount of memory is set by the order of described bootstrap loader, when discharging the continuous physical memory of described distribution.
7. method according to claim 4 is characterized in that, after the step of distributing described continuous physical memory, comprising:
Pp_virt_buf is sent to described driving as the DMA memory address.
8. continuous physical memory distributor in the embedded system is characterized in that, comprising:
Parameter set unit is used for the bootstrap loader in described embedded system, the whole physical memory quantity and the difference of the continuous physical memory quantity that needs to distribute is provided with the kernel start-up parameter of inner core managing amount of memory as parameter value;
The Memory Allocation unit is used for distributing described continuous physical memory in the driving of described embedded system by the described continuous physical memory quantity that needs to distribute.
9. device according to claim 8 is characterized in that, further comprises:
Internal memory changes the unit, is used for revising the described parameter value that the kernel start-up parameter of inner core managing amount of memory is set by the order of described bootstrap loader, to change the described continuous physical memory quantity that needs distribution.
10. device according to claim 9 is characterized in that, further comprises:
The internal memory releasing unit is used for removing the described kernel start-up parameter that the inner core managing amount of memory is set by the order of described bootstrap loader, when discharging the continuous physical memory of described distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910192199A CN101706754A (en) | 2009-09-08 | 2009-09-08 | Method and device for allocating continuous physical memory in embedded system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910192199A CN101706754A (en) | 2009-09-08 | 2009-09-08 | Method and device for allocating continuous physical memory in embedded system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101706754A true CN101706754A (en) | 2010-05-12 |
Family
ID=42376980
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910192199A Pending CN101706754A (en) | 2009-09-08 | 2009-09-08 | Method and device for allocating continuous physical memory in embedded system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101706754A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799536A (en) * | 2011-05-24 | 2012-11-28 | 中兴通讯股份有限公司 | Self-adapted method and device of storage device capacity in embedded type system |
| CN105243029A (en) * | 2015-11-24 | 2016-01-13 | 无锡江南计算技术研究所 | NUMA memory management fault tolerance method based on TLB-MISS |
| CN105490985A (en) * | 2014-09-15 | 2016-04-13 | 腾讯科技（深圳）有限公司 | Implementation method and device for TCP long connection based on signal transmission |
| CN106484632A (en) * | 2015-09-01 | 2017-03-08 | 华耀(中国)科技有限公司 | The method distributing continuous physical memory in linux system user's space |
| CN110032446A (en) * | 2019-03-27 | 2019-07-19 | 中国科学院微电子研究所 | A kind of method and device applied to storage allocation space in embedded system |
| CN111240765A (en) * | 2020-01-15 | 2020-06-05 | 深圳英飞拓智能技术有限公司 | LINUX compression application program loading method |
-
2009
- 2009-09-08 CN CN200910192199A patent/CN101706754A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799536A (en) * | 2011-05-24 | 2012-11-28 | 中兴通讯股份有限公司 | Self-adapted method and device of storage device capacity in embedded type system |
| CN102799536B (en) * | 2011-05-24 | 2016-11-16 | 天津中兴智联科技有限公司 | The adaptive method and device of capacity of memory device in a kind of embedded system |
| CN105490985A (en) * | 2014-09-15 | 2016-04-13 | 腾讯科技（深圳）有限公司 | Implementation method and device for TCP long connection based on signal transmission |
| CN105490985B (en) * | 2014-09-15 | 2019-12-13 | 腾讯科技（深圳）有限公司 | Method and device for realizing TCP long connection based on signaling transmission |
| CN106484632A (en) * | 2015-09-01 | 2017-03-08 | 华耀(中国)科技有限公司 | The method distributing continuous physical memory in linux system user's space |
| CN105243029A (en) * | 2015-11-24 | 2016-01-13 | 无锡江南计算技术研究所 | NUMA memory management fault tolerance method based on TLB-MISS |
| CN105243029B (en) * | 2015-11-24 | 2018-01-19 | 无锡江南计算技术研究所 | NUMA memory management fault-tolerance approaches based on TLB MISS |
| CN110032446A (en) * | 2019-03-27 | 2019-07-19 | 中国科学院微电子研究所 | A kind of method and device applied to storage allocation space in embedded system |
| CN111240765A (en) * | 2020-01-15 | 2020-06-05 | 深圳英飞拓智能技术有限公司 | LINUX compression application program loading method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101599046B (en) | Method and device for detecting memory | |
| CN101706754A (en) | Method and device for allocating continuous physical memory in embedded system | |
| CN102096639B (en) | Method for trimming data on non-volatile flash media | |
| CN108369520B (en) | Protecting basic input/output (BIOS) code | |
| CN115344521A (en) | Virtual device composition in extensible input/output (I/O) virtualization (S-IOV) architecture | |
| US8700879B2 (en) | Concurrent memory operations | |
| CN103984599B (en) | Method for improving utilization rate of large pages of operating system | |
| US20120005445A1 (en) | Memory allocation with identification of requesting loadable kernel module | |
| CN102306126A (en) | Memory management method, device and system | |
| CN108351826B (en) | Monitoring operation of a processor | |
| CN103995784A (en) | Flash memory controller, storage device and flash memory control method | |
| US8473460B2 (en) | Driver model for replacing core system hardware | |
| CN101739283A (en) | Computer and method for virtual system to directly access computer hardware | |
| US9262332B2 (en) | Memory management with priority-based memory reclamation | |
| CN103927130B (en) | Based on external memory framework in the unification of memory management unit MMU | |
| CN100594481C (en) | Method and system for allowing program operated on logical partitions to access resources | |
| CN106293781A (en) | The method and apparatus of online upgrading machine virtual machine | |
| US10268592B2 (en) | System, method and computer-readable medium for dynamically mapping a non-volatile memory store | |
| CN103455438A (en) | Internal memory management method and equipment | |
| CN102096594B (en) | Method and device for using system resources in memory | |
| CN104182352A (en) | Device and method for accessing to physical memory address spaces larger than 4GB | |
| US20050216539A1 (en) | Integration of mark bits and allocation bits | |
| CN102541654A (en) | Efficient memory management method and device | |
| CN102789384B (en) | Method for implementing server operating system applied to Loongson 3B processor | |
| Lu et al. | Using hugetlbfs for mapping application text regions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100512 |