WO2016138785A1 - 一种运行进程的方法及装置 - Google Patents

一种运行进程的方法及装置 Download PDF

Info

Publication number
WO2016138785A1
WO2016138785A1 PCT/CN2015/099051 CN2015099051W WO2016138785A1 WO 2016138785 A1 WO2016138785 A1 WO 2016138785A1 CN 2015099051 W CN2015099051 W CN 2015099051W WO 2016138785 A1 WO2016138785 A1 WO 2016138785A1
Authority
WO
WIPO (PCT)
Prior art keywords
page
code page
memory
code
virtual address
Prior art date
Application number
PCT/CN2015/099051
Other languages
English (en)
French (fr)
Inventor
刘铎
王雅坤
钟刊
Original Assignee
华为技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP15883835.9A priority Critical patent/EP3252595A4/en
Publication of WO2016138785A1 publication Critical patent/WO2016138785A1/zh
Priority to US15/693,955 priority patent/US10261918B2/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/12—Replacement control
    • G06F12/121—Replacement control using replacement algorithms
    • G06F12/122—Replacement control using replacement algorithms of the least frequently used [LFU] type, e.g. with individual count value
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
    • G06F12/023—Free address space management
    • G06F12/0238—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
    • G06F12/0246—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory in block erasable memory, e.g. flash memory
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/10—Address translation
    • G06F12/1009—Address translation using page tables, e.g. page table structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/461—Saving or restoring of program or task context
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/15—Use in a specific computing environment
    • G06F2212/152—Virtualized environment, e.g. logically partitioned system
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/17—Embedded application
    • G06F2212/171—Portable consumer electronics, e.g. mobile phone
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/65—Details of virtual memory and virtual address translation
    • G06F2212/657—Virtual address space management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/72—Details relating to flash memory management
    • G06F2212/7201—Logical to physical mapping or translation of blocks or pages

Definitions

  • the present invention relates to the field of computer application technologies, and in particular, to a method and apparatus for running a process.
  • DRAM random random access memory
  • DRAM dynamic random access memory
  • Swap partition a swap partition
  • the Android operating system is based on the Linux kernel operating system, and the Linux kernel provides relief through swap partitions. Insufficient memory resources.
  • the existing Swap mechanism can use a new byte-addressable non-volatile memory (Non-Volatile Memory, NVM for short) as a Swap partition, and the Swap partition is usually divided into page slots (Page Slot) for management, each The size of the page slot is equal to the size of the memory page to store memory pages that are replaced from memory.
  • NVM Non-Volatile Memory
  • Page Slot page slots
  • An existing wear leveling technique maintains two data structures in the operating system: a linked list that holds free page slots and a small top heap that holds all page slots and sorts by age.
  • a linked list that holds free page slots
  • a small top heap that holds all page slots and sorts by age.
  • the embodiment of the invention provides a method and a device for running a process, which are used to solve the problem that the running process is slow, the service life of the swap partition is short, the system resources are wasted, and the system performance is poor.
  • a method for running a process including: scanning a process in a memory, selecting a candidate process from all processes; and obtaining a code page from the selected candidate process; wherein the code page is used for storing a binary code of the process or a shared library file; writing the code page to the swap partition, and releasing the storage space occupied by the code page in the memory; and obtaining the physics of the code page saved in the page table entry An address, and updating a physical address of the code page to a physical address in the swap partition where the code page is stored; wherein the page table entry is used to store a virtual address of a code page and a physical address of the code page Corresponding relationship between the virtual page and the physical address of the updated code page stored in the page table entry, when the candidate process belongs to the coded page, and executed in the swap partition Said code page.
  • the method before scanning a process in the memory, the method further includes: determining that a capacity value of the available storage space included in the memory is less than a preset threshold.
  • the score level of each process is separately obtained; wherein the score level of any one process is according to any one of the processes
  • the priority of the process and the size of the memory space occupied by the process are determined. When the priority of the process is lower, and the occupied memory space is larger, the score of the process is higher. When the priority of any one of the processes is higher, and the occupied memory space is smaller, the score level of any one of the processes is lower; from all processes, the process of ranking from high to low is selected in turn until selected After the available storage space contained in the memory is greater than or equal to the preset threshold, each selected process is determined as a candidate process.
  • the first possible implementation manner of the first aspect, or the second possible implementation manner of the first aspect, in the third possible implementation manner performing the following operations for each of the candidate processes Scanning the virtual address space of the process corresponding to the candidate process, and acquiring all virtual address regions in the virtual address space of the process according to the process descriptor of the candidate process; traversing all virtual addresses in the virtual address space of the process And determining, according to the flag bit of the virtual address area, whether the virtual address area is an executable area; and determining that any virtual address area is an executable area, Included in any virtual address area The page is determined to be the code page.
  • the first possible implementation of the first aspect to any one of the third possible implementations of the first aspect in the fourth possible implementation, for each of the code pages, Performing the following operations: applying a page slot in the swap partition; storing the code page in the page slot obtained by the application; updating the physical address of the code page to the physical address of the page slot .
  • the swap partition is a non-volatile memory NVM swap partition, and the NVM swap partition is directly connected to the memory controller of the memory. And storing the code page through the memory controller into the page slot obtained by the application.
  • the code page belongs to And obtaining a virtual address of the updated code page saved in the page table entry, and determining, from the page table entry, that the obtained virtual address exists Corresponding physical address of the relationship; and acquiring the code page in the exchange area according to the physical address; and executing the code page.
  • a second aspect provides an apparatus for running a process, including: a scanning unit, configured to scan a process in an internal memory; and a candidate process selecting unit, configured to select, according to a scan result of the scanning unit to a process in the memory, select all processes a candidate page; a code page obtaining unit, configured to obtain a code page from the candidate process selected by the candidate process selecting unit; wherein the code page is used to store a binary code of a process or a shared library file; And a code space release unit is configured to release a storage space occupied by the code page acquired by the code page obtaining unit in the memory; physical address acquisition a unit, configured to acquire a physical address of the code page saved in the page table entry, and an update unit, configured to update a physical address of the code page acquired by the physical address obtaining unit to store the code page in the swap partition Physical address; wherein the page table entry is used to store the virtual address of the code page Correspondence between the physical address of said page code; execution unit, for, when run When the candidate process belongs
  • the determining unit is further configured to: before scanning a process in the memory, determining that a capacity value of the available storage space included in the memory is less than a preset threshold.
  • the candidate process selecting unit is specifically configured to: separately obtain a rating level of each process;
  • the score level of a process is determined according to the priority of any one of the processes and the size of the memory space occupied by the any one of the processes.
  • the priority of any one of the processes is lower, and the occupied memory space is larger,
  • the process from high to low ranks until the available storage space included in the memory is greater than or equal to the preset threshold, and the selected processes are determined as candidate processes.
  • the code page obtaining unit is specifically configured to: Each of the candidate processes performs the following operations: scanning a process virtual address space corresponding to the candidate process, and acquiring all virtual address regions in the process virtual address space according to the process descriptor of the candidate process; traversing Determining whether the virtual address area is an executable area according to a flag bit of the virtual address area; and determining any one of the virtual address areas of the virtual address area; When the virtual address area is an executable area, the page included in any one of the virtual address areas is determined as a code page.
  • the writing unit is specifically used For each of the code pages, performing the following operations: applying a page slot in the swap partition; storing the code page in the page slot obtained by the application; the update unit, Specifically, the physical address of the code page is updated to a physical address of the page slot.
  • the writing unit is specifically configured to: store the code page by using the memory controller to obtain the application In the page slot; the swap partition is a non-volatile memory NVM swap partition, and the NVM swap partition is directly connected to a memory controller of the memory.
  • the first possible implementation manner of the second aspect the fifth possible implementation manner of the second aspect, in the sixth possible implementation manner, Obtaining, when it is determined that the code page needs to be executed, obtaining a virtual address of the updated code page saved in the page table item; from the page table item Determining a physical address corresponding to the acquired virtual address; and acquiring the code page in the exchange area according to the physical address; and executing the code page.
  • the code page is selected from the candidate processes, and only the code page in the candidate process is stored in the swap partition, and the memory space occupied by the code page is released; and the code page saved in the page table item is updated. Physical address; when it is determined that the candidate process to which the code page belongs needs to be executed, if the code page needs to be executed, the code page can be directly executed in the swap partition. According to the technical solution of the present invention, only the code page having the read-only feature is stored in the swap partition. When the code page needs to be executed, the code page is directly executed in the swap partition, and the code page is not required to be replaced into the memory for execution. Improve the running speed of the process, effectively avoid the frequent exchange of data, reduce the number of write operations caused by frequent data exchange; and, without saving the age of each page slot locally, saving system space, reducing system load and improving System performance.
  • FIG. 2 is a schematic diagram of a layer level of a terminal software according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of a virtual address space in an embodiment of the present invention.
  • FIGS. 4a and 4b are schematic diagrams of memory and switching partitions in an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a process of a swap partition write operation according to an embodiment of the present invention.
  • FIG. 6 is a schematic diagram 1 of a comparison result in an embodiment of the present invention.
  • FIG. 7 is a second schematic diagram of a comparison result according to an embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of a device for running a process according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of a process running device according to an embodiment of the present invention.
  • the code page is selected from the candidate processes, and only the code page in the candidate process is stored in the swap partition, and the memory space occupied by the code page is released; and the code page saved in the page table item is updated. Physical address; when it is determined that the candidate process to which the code page belongs needs to be executed, if the code page needs to be executed, the code page can be directly executed in the swap partition. According to the technical solution of the present invention, only the code page having the read-only feature is stored in the swap partition.
  • the code page When the code page needs to be executed, the code page is directly executed in the swap partition, and the code page is not required to be replaced into the memory for execution. Improve the running speed of the process, effectively avoid the frequent exchange of data, reduce the number of write operations caused by frequent data exchange; and, without saving the age of each page slot locally, saving system space, reducing system load and improving System performance.
  • a method for running a process in a terminal includes:
  • Step 100 Scan the processes in the memory and select candidate processes from all processes.
  • an application may be run at one time in the terminal, or multiple applications may be run at the same time.
  • Each application needs to apply for a corresponding memory resource and use the memory resource during the running process. Since the process is the smallest unit in the running process of the application, when the terminal is running the application, the memory is insufficient to support the running of the application, that is, when the storage space contained in the memory of the terminal is less than the preset threshold.
  • the terminal needs to scan the process corresponding to the running application, and selects the candidate process from all processes and dumps the file to the swap partition to solve the problem of insufficient memory resources; wherein the terminal may be a mobile terminal.
  • the terminal includes a Central Processing Unit (CPU), a memory (such as DRAM), a swap partition, and the like, and the terminal
  • CPU Central Processing Unit
  • memory such as DRAM
  • swap partition and the like
  • the terminal The operating system is installed in the system; the preset thresholds can be dynamically adjusted according to specific application scenarios.
  • the terminal selects a candidate process from all the processes according to the scoring level of each process, where the terminal obtains the scoring level of each process separately, wherein the scoring level of any one process is based on the priority of the any process.
  • determining the size of the memory space occupied by the arbitrary process for example, assigning a corresponding weight value to the priority and the size of the occupied memory space, by the following formula: “Priority ⁇ Priority Weight Value+Occupy Memory Space ⁇ Memory Space
  • the weight value determines the comprehensive score of any process, and searches the comprehensive score and score level comparison table stored locally by the terminal to determine the score level corresponding to the comprehensive score.
  • the comprehensive score and the score level comparison table are used to save the comprehensive score and the rating scale.
  • each rating level may correspond to a comprehensive score segment, and the comprehensive rating and rating level comparison table may be preset according to a specific application scenario; in addition, the rating level is determined by a specific component in the operating system, for example, In the Android system, the rating can be Android low memory manager (Low Memory Killer, referred to as LMK) is determined; according to rating scale, the selected candidates from all processes in the process memory.
  • LMK Android low memory manager
  • the process of selecting a candidate process from all processes in the memory according to the rating level specifically: selecting a process with a rating from high to low in all processes, until selecting to run all applications. All the selected processes are determined as candidate processes until the requested memory space is less than or equal to the preset threshold position.
  • the scoring level of the arbitrary process when the priority of any process is lower and the memory space is larger, the scoring level of the arbitrary process is higher, and the priority of the arbitrary process is higher, and the memory space is occupied. The smaller the score of the arbitrary process is, the lesser the case is, and the acquisition process of the candidate process is introduced.
  • the mobile terminal can also set the process to be higher when the priority of any process is higher and the memory space is smaller.
  • the operating system installed by the terminal may include multiple types, such as an Android operating system, an IOS operating system, a Microsoft operating system, and the like.
  • the software level installed in the terminal is: the top layer is an application; under the application is the Dalvik virtual machine of the Android system and the running library.
  • Android system under Dalvik is the Linux kernel, Linux kernel contains memory management, process management, device drivers, etc.; the bottom layer is the specific hardware, including memory (such as DRAM), swap partition (such as NVM and flash Flash) Wait.
  • Step 110 Obtain a code page from the selected candidate process; wherein the code page is used to store a binary code of the process, or a shared library file.
  • each process in the memory is composed of multiple pages (hereinafter referred to as memory pages), and each memory page occupies a certain storage space, and the size of the memory space occupied by one process is in the process.
  • the sum of the storage space occupied by all the memory pages included; the pages in memory contain two types, namely Code Page and Data Page, where the code page is used to store the binary code of the process. (Binary Code), or shared library file (Shared Library), and the code page has a read-only feature, and the data page is used to store application data and variables, etc., the data page does not have read-only characteristics.
  • each candidate process performing the following operations: scanning the process virtual address space corresponding to the candidate process, where the process virtual address space corresponding to each candidate process is the memory space allocated by the terminal for each process;
  • Each virtual address space includes multiple virtual address areas, and the virtual address area is mapped to the physical storage space by using a page table entry, and is used to save data of the application runtime. Therefore, the terminal obtains according to the process descriptor of the candidate process.
  • All virtual address areas in the virtual address space of the process traversing all the virtual address areas in the virtual address space of the process acquired; and obtaining the flag bits of each virtual address area traversed, when there is any flag of the virtual address area
  • the bit is an executable area flag, it is determined that the arbitrary one of the virtual address areas is an executable area, and the page included in the arbitrary one of the virtual address areas is determined as a code page.
  • the process descriptor of the candidate process and the flag bit of the virtual address area are operated according to the operation. It is determined by the system that when the operating system installed by the terminal is different, the process descriptors of the candidate processes are different, and the flag bits of the virtual address area are also different.
  • the virtual address space of a process in the Android system includes multiple virtual address areas in the virtual address space of the process, and each virtual address area is represented by vm_area_struct, and each virtual address is represented.
  • the vm_next connection is used between the areas.
  • the virtual address area is the end position of the virtual address space to which the virtual address area belongs.
  • the flag of the virtual address area is VM_READ.
  • VM_EXEC the part shown in the box in Figure 3 (the part shown in the first box is the shared library file /etc/1d.so area, and the part shown in the second box is the binary code /bin/gcc area ), for the flag bits jointly determined by VM_READ and VM_EXEC, the page pointed to by VM_READ and VM_EXEC is the code page, that is, "Memory mapping" and "Text (file backed)" in FIG.
  • the code page is selected from all the pages of the candidate process saved in the memory, and the code of each candidate process can be obtained more quickly and accurately.
  • the page improves system processing efficiency.
  • Step 120 Write the code page into the swap partition, and release the storage space occupied by the code page in the memory.
  • the terminal writes the obtained code page into the swap partition. After the code page is all written to the swap partition, the terminal releases the storage space occupied by the code page in the memory to reduce the occupation of the memory;
  • the swap partition can be implemented by flash memory.
  • the data transfer between the memory and the flash memory is the responsibility of the I/O driver, that is, the data transfer between the memory and the flash memory requires I/O scheduling, as shown in FIG. 4b.
  • the swap partition can also be implemented by NVM. Both the memory and the NVM are connected to the memory controller. The memory controller controls the data exchange between the memory and the NVM, and the operating system installed by the terminal treats the memory and the NVM as nothing.
  • the operating system can access the data stored in the memory and the swap partition through the virtual address, and the data exchange between the swap partition and the memory interface can be through the memory interface, and the data on the NVM can be copied into the memory through the memory, and vice versa.
  • Memory and NVM unified encoding when the swap partition is implemented in flash, the data transfer between memory and flash requires I /O scheduling, and I/O scheduling speed is relatively slow, from However, there is a problem of long read and write delay time; when the swap partition is implemented by NVM, since the data transfer between the memory and the NVM is directly scheduled by the memory controller, the data can be transmitted through the memory bus without the need for I/O scheduling.
  • the swap partition is implemented by NVM, and the read and write delay time is short.
  • the terminal performs the following operations for each code page: since the basic unit of the swap partition is a page slot, the size of the storage space occupied by each code page is equal to the size of the storage space occupied by one page slot,
  • the terminal determines that the memory space needs to be released, a page slot is requested for the code page in the swap partition; the terminal directly stores the code page through the memory controller into the page slot obtained by the application.
  • the terminal stores the acquired code page in the swap partition, thereby releasing the storage space occupied by the code page replaced in the memory, providing more memory resources for other processes to run, effectively solving the problem of insufficient memory resources. problem.
  • Step 130 Acquire a physical address of the above code page saved in the page table entry, and update the physical address of the code page to a physical address in the swap partition where the code page is stored; wherein the page table entry is used to store the virtual page of the code page. The correspondence between the address and the physical address of the code page.
  • the terminal locally stores a page table entry, where the page table entry stores a correspondence between a virtual address of the code page and a physical address of the code page, wherein the virtual address of the code page cannot be changed.
  • the terminal updates the physical address of the code page in the page table entry to the physical address of the swap page in which the code page is stored.
  • the terminal stores the above code page into a corresponding page slot in the swap partition, so the physical address of the code page is updated to the physical address of the code page in the swap partition, that is, the physical address of the code page. , updated to the physical address of the above page slot.
  • FIG. 5 a schematic diagram of data exchange between a page table entry, a memory, and a swap partition when a code page is written in a swap partition is performed in an embodiment of the present invention.
  • the physical address of the code page in the page table item is updated, thereby avoiding subsequent execution of the data contained in the code page in the swap partition, according to the original Some physical addresses can't get the code page problem, save Proven the feasibility of code page execution.
  • Step 140 When the candidate process to which the code page belongs is run, the code page is executed in the swap partition according to the corresponding relationship between the virtual address and the physical address of the updated code page stored in the page table entry.
  • the terminal determines that the candidate process belongs to the running code page, if it is determined that any code page needs to be executed, and the arbitrary code page is located in the swap partition, the updated one of the page table items is obtained. a virtual address of the code page; determining, from the page table entry, a physical address corresponding to the obtained virtual address; and obtaining, according to the physical address, the arbitrary code page in the exchange area; and executing the arbitrary code page The data contained.
  • the application environment is set as a Google Nexus 5 smart phone.
  • different types of applications are selected and grouped into six categories: Browser, Social networking, Multimedia, Gaming, Online Shopping, and News.
  • the technical solutions provided by the present invention are respectively adopted, and the prior art is operated for 20 minutes.
  • the write operation of the Swap partition changes with time.
  • the abscissa represents time
  • the ordinate represents the total number of write operations to the swap partition
  • the first curve represents the operation of the technical solution of the present invention.
  • the result of the operation at the time of the application, and the second curve shows the result of the operation when the application was run using the prior art. It can be seen from FIG. 6 that when the application program is run by using the technical solution provided by the present invention, the write operation is reduced by 14% to 51% compared with when the application is run by the prior art. Moreover, when the write operation is performed on the swap partition by using the technical solution of the present invention, if there is no new application running in the terminal, the number of write operations for the swap partition will not increase after a period of time, and when the application is run by using the prior art The write operation to the swap partition will continue to increase.
  • the switching time of the application is compared in the scenario using the swap partition and the scenario in which the swap partition is not used.
  • a total of 30 applications are set to run in the terminal, and the 30 applications are continuously switched.
  • the comparison result is shown in Figure 7.
  • the horizontal coordinate indicates the application program
  • the ordinate indicates the average switching time of the application
  • the first cylinder indicates the application scenario using the swap partition
  • the second cylinder indicates the application without the swap partition. Scenes.
  • the scenario in which the switching scheme is used in the technical solution of the present invention reduces the switching time of the application by 19% to 41% compared to the scenario in which the switching partition is not used.
  • the mechanism adopted by the terminal is to forcibly stop the application corresponding to the page.
  • the terminal needs to reload the application.
  • the application program takes a long time to execute; when the memory space is insufficient, the code page in the candidate process is stored in the swap partition, and when it is necessary to execute any code page in the swap partition, it is not necessary to
  • the arbitrary code page is recalled into the memory, and only needs to obtain the arbitrary code page according to the physical address of the arbitrary code page, and then directly execute the arbitrary code page, without forcibly stopping the application, and further Improve the execution speed of the application.
  • the embodiment of the present invention further provides an apparatus for running a process, including a scanning unit 80, a candidate process selecting unit 81, a code page obtaining unit 82, a writing unit 83, and a storage space releasing unit. 84.
  • a scanning unit 80 configured to scan a process in the memory
  • a candidate process selecting unit 81 configured to scan the in-memory process according to the scanning unit 80 Select candidate processes from all processes;
  • the code page obtaining unit 82 is configured to obtain a code page from the candidate process selected by the candidate process selecting unit 81; wherein the code page is used to store a binary code of a process or a shared library file;
  • a writing unit 83 configured to write the code page acquired by the code page obtaining unit 82 into the swap partition
  • a storage space release unit 84 configured to release a storage space occupied by the code page acquired by the code page obtaining unit 82 in the memory
  • the physical address obtaining unit 85 is configured to obtain a physical address of the code page saved in the page table entry
  • An update unit 86 configured to update a physical address of the code page acquired by the physical address obtaining unit 85 to a physical address in the swap partition where the code page is stored; wherein the page table entry is used to store a code page a correspondence between a virtual address and a physical address of the code page;
  • the execution unit 87 is configured to: when the candidate process to which the code page belongs is executed, according to the correspondence between the virtual address and the physical address of the code page updated by the update unit 86 saved in the page table entry, The code page is executed in the swap partition.
  • the foregoing apparatus further includes a determining unit 88, configured to: before scanning a process in the memory, determining that a capacity value of the available storage space included in the memory is less than a preset threshold.
  • the candidate process selecting unit 81 is specifically configured to: obtain a rating level of each process separately, where a rating level of any one process is occupied according to a priority of the any one process and any one of the processes The size of the memory space is determined. When the priority of any one of the processes is lower, and the occupied memory space is larger, the scoring level of the arbitrary process is higher, and the priority of any one of the processes is higher. And the smaller the occupied memory space is, the lower the rating level of any one of the processes is; the process of selecting the highest level to the lowest level is selected from all the processes until the available storage space included in the memory is greater than or equal to the pre- After the threshold is set, each selected process is determined as a candidate process.
  • the code page obtaining unit 82 is specifically configured to: feed each candidate And performing the following operations: scanning the process virtual address space corresponding to the candidate process, and acquiring all virtual address regions in the virtual address space of the process according to the process descriptor of the candidate process; traversing the virtual address of the process All the virtual address areas in the space; and each of the virtual address areas traversed, according to the flag bit of the virtual address area, determine whether the virtual address area is an executable area; when determining any virtual address area is When the area is executed, the page included in any of the virtual address areas is determined as a code page.
  • the writing unit 83 is configured to: for each of the code pages, perform an operation of: applying a page slot in the swap partition; storing the code page to an application obtained In the page slot, the updating unit 86 is specifically configured to: update a physical address of the code page to a physical address of the page slot.
  • the writing unit 83 is configured to: store the code page by using the memory controller into the page slot obtained by the application; and the swap partition is a non-volatile memory NVM swap partition. And the NVM swap partition is directly connected to the memory controller of the memory.
  • the executing unit 87 is specifically configured to: when the candidate process to which the code page belongs is run, if it is determined that the code page needs to be executed, obtain the updated information saved in the page table entry a virtual address of the code page; determining, from the page table entry, a physical address corresponding to the obtained virtual address; and acquiring the code page in the exchange area according to the physical address; Said code page.
  • an embodiment of the present invention further provides a process running device, including a memory 90, and a processor 91, where:
  • the memory 90 is configured to store an application and a page table item, including a swap partition and a memory;
  • the processor 91 is configured to run an application in the memory 90 and perform the following operations:
  • Scanning a process in the memory selecting a candidate process from all processes according to a scan result of the scanning unit to a process in the memory; obtaining a code page from the candidate process selected by the candidate process selection unit; wherein the code
  • the page is used to store the binary code of the process or the shared library file; the code page acquired by the code page obtaining unit is written into the swap partition; and the storage space occupied by the code page acquired by the code page obtaining unit in the memory is released.
  • the processor 91 is further configured to: before scanning a process in the memory, determining that a capacity value of the available storage space included in the memory is less than a preset threshold.
  • the processor 91 is specifically configured to: obtain a rating level of each process separately, where a rating level of any one process is based on a priority of the any one of the processes and a memory space occupied by the any one of the processes.
  • the size of the process is determined to be higher, and the higher the occupied memory space is, the higher the scoring level of any one of the processes is, the higher the priority of any one of the processes is, and The smaller the occupied memory space is, the lower the scoring level of any one of the processes is; the processes with high and low scoring levels are selected from all processes until the available storage space included in the memory is greater than or equal to a preset threshold. After that, each selected process is determined as a candidate process.
  • the processor 91 is configured to: perform, for each of the candidate processes, an operation of scanning a process virtual address space corresponding to the candidate process, and according to a process descriptor of the candidate process, Obtaining all the virtual address areas in the virtual address space of the process; traversing all the virtual address areas in the virtual address space of the process; and according to the flag bits of the virtual address area, each of the virtual address areas traversed Determining whether the virtual address area is an executable area; when determining that any virtual address area is an executable area, determining a page included in any one of the virtual address areas as a code page.
  • the processor 91 is specifically configured to: for each of the code pages, perform an operation of: applying a page slot in the swap partition; storing the code page to the application obtained In the page slot; update the physical address of the code page to the physical address of the page slot.
  • the process running device further includes a memory controller 92 for connecting the memory and the intersection
  • the processor 91 is configured to: store the code page through the memory controller 92 into the page slot obtained by the application; and the swap partition is an NVM switching partition.
  • the processor 91 is configured to: when the candidate process to which the code page belongs is run, if it is determined that the code page needs to be executed, obtain the updated information saved in the page table entry a virtual address of the code page; determining, from the page table entry, a physical address corresponding to the obtained virtual address; and acquiring the code page in the exchange area according to the physical address; Said code page.
  • a process in a memory is scanned, and a candidate process is selected from all processes; a code page is obtained from the selected candidate process; wherein the code page is used to store a binary code of the process, or share a library file; writing the code page to the swap partition, and freeing the storage space occupied by the code page in the memory; obtaining the physical address of the above code page saved in the page table entry, and updating the physical address of the code page to the exchange
  • the physical address of the code page is stored in the partition; wherein, the page table entry is used to store the correspondence between the virtual address of the code page and the physical address of the code page; when the candidate process to which the code page belongs is run, according to the page table entry The corresponding relationship between the virtual address and the physical address of the updated code page saved in the execution, and the code page is executed in the swap partition.
  • the code page having the read-only feature is stored in the swap partition.
  • the code page is directly executed in the swap partition, and the code page is not required to be replaced into the memory for execution.
  • Improve the running speed of the process effectively avoid the frequent exchange of data, reduce the number of write operations caused by frequent data exchange; and, without saving the age of each page slot locally, saving system space, reducing system load and improving System performance.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • the present invention is directed to a method, apparatus (system), and computer program according to an embodiment of the present invention.
  • the flow chart and/or block diagram of the product is described. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Memory System Of A Hierarchy Structure (AREA)
  • Memory System (AREA)

Abstract

一种运行进程的方法及装置，方法为：从候选进程中选取代码页面（110），仅将候选进程中的代码页面存储至交换分区中，并释放该代码页面所占用的内存空间（120）；并更新页表项中保存的代码页面的物理地址；当确定需要运行代码页面所属的候选进程时，若需要执行该代码页面，直接在交换分区中执行该代码页面即可。采用该方法，仅将具有只读特性的代码页面存储至交换分区中，当需要执行该代码页面时，直接在交换分区中执行该代码页面，无须将代码页面替换至内存中执行，在提高进程运行速度的同时，有效避免了数据的频繁交换，减少了数据频繁交换造成的写操作次数；并且，无须在本地保存每一个页槽的年龄，节约了系统空间，提高了系统性能。

Description

一种运行进程的方法及装置
本申请要求在2015年03月04日提交中国专利局、申请号为201510097009.3、发明名称为“一种运行进程的方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及计算机应用技术领域，尤其涉及一种运行进程的方法及装置。
背景技术
随着微处理器和电子移动终端操作系统的发展，电子移动终端(例如智能手机，平板电脑等)得到了广泛的应用，针对电子移动终端设计的应用功能也越来越强大。电子移动终端中任何一个应用程序的运行，均需要占用一定的内存资源，因此，功能丰富的应用在提高电子移动终端的用户体验的同时，却需要占用更多的内存资源。在现在的所有的电子移动终端中，通常使用随机存储器(Dynamic Random Access Memory，简称DRAM)作为内存。通过增加DRAM的容量，在一定程度上可以满足应用程序对内存资源的需求。但是，DRAM容量的增加将导致电子移动终端成本的增加；并且，DRAM耗电量大，其最多可占整个操作系统功耗的30％，采用DRAM作为内存的电子移动终端时，将存在电子移动终端能耗高的问题。
基于上述问题，目前，还可以通过在操作系统中设置交换分区(Swap分区)的方式，来缓解内存压力，例如，Android操作系统是基于Linux kernel的操作系统，Linux kernel提供了通过交换分区来缓解内存资源不足的问题。现有的Swap机制可以使用新型可字节寻址的非易失性存储器(Non-Volatile Memory，简称NVM)作为Swap分区，Swap分区通常被划分为页槽(Page Slot)来进行管理，每个页槽的大小等于内存页面的大小，以便于存储从内存中替换出的内存页面。当电子移动终端中的内存资源不足时，操作系统会将一些 不经常使用的内存页面存放至交换分区中，使该不经常使用的内存页面不再占用内存，从而减少对内存的占用。
由于NVM具有有限的擦写次数，需要采取损耗均衡等技术尽可能使写操作均匀地分布在每个NVM页槽上，以实现NVM寿命的最大化。现有的一种损耗均衡技术在操作系统中维护了两个数据结构：保存空闲页槽的链表和保存所有页槽并按其年龄进行排序的小顶堆。当移动终端内存不足，需要将不活跃的内存页面存储至NVM交换分区时，需要将链表头页槽的年龄和小顶堆堆顶页槽的年龄相比较，从而将内存中不活跃的页面存储至年龄较小的页槽中。然而，当小顶堆堆顶页槽已经被使用且年龄较小的时候，还需要将小顶堆堆顶页槽的数据拷贝到链表头页槽中，然后才能使用该小顶堆堆顶页槽。该方法总是保证使用年龄较小的NVM页槽来存储内存页面，进而达到损耗均衡的目的。
采用上述技术方案，需要在电子移动终端中开辟额外的存储空间，用于保存每一个NVM页槽的年龄，造成了系统资源的浪费以及系统性能的下降；并且，任意一NVM页槽的年龄变化，均将导致小顶推中所有NVM页槽的重新排序，进而进一步增加了系统负担；此外，当上述不经常使用的内存页面需要被重新执行时，操作系统将该不经常使用的内存页面重新替换至内存中，从而存在数据的频繁交换，以及数据的多次写入，进而造成影响进程运行时间，以及影响交换分区使用寿命的问题。
由此可见，目前在运行进程的过程中，存在进程运行速度慢，交换分区使用寿命短，系统资源浪费以及系统性能差的问题。
发明内容
本发明实施例提供一种运行进程的方法及装置，用以解决目前在运行进程的过程中，存在进程运行速度慢，交换分区使用寿命短，系统资源浪费以及系统性能差的问题。
本发明实施例提供的具体技术方案如下：
第一方面，提供一种运行进程的方法，包括：扫描内存中的进程，从所有进程中选取候选进程；并从选取的所述候选进程中获取代码页面；其中，所述代码页面用于存放进程的二进制代码或者共享库文件；将所述代码页面写入交换分区中，并释放所述内存中所述代码页面所占用的存储空间；以及获取页表项中保存的所述代码页面的物理地址，并将所述代码页面的物理地址更新为所述交换分区中存储所述代码页面的物理地址；其中，所述页表项用于存储代码页面的虚拟地址与所述代码页面的物理地址之间的对应关系；当运行所述代码页面所属的候选进程时，根据页表项中保存的更新后的所述代码页面的虚拟地址和物理地址的对应关系，在所述交换分区中执行所述代码页面。
结合第一方面，在第一种可能的实现方式中，扫描内存中的进程之前，还包括：确定所述内存中包含的可用存储空间的容量值小于预设阈值。
结合第一方面，或者第一方面第一种可能的实现方式，在第二种可能的实现方式中，分别获取每一个进程的评分等级；其中，任意一进程的评分等级根据所述任意一进程的优先级以及所述任意一进程所占用内存空间的大小确定，当所述任意一进程的优先级越低，且所述占用内存空间越大时，所述任意一进程的评分等级越高，当所述任意一进程的优先级越高，且所述占用内存空间越小时，所述任意一进程的评分等级越低；从所有进程中依次选取评分等级从高到低的进程，直至选择至内存中包含的可用存储空间大于或等于预设阈值为止后，将选取出的各个进程确定为候选进程。
结合第一方面，第一方面第一种可能的实现方式，或者第一方面第二种可能的实现方式，在第三种可能的实现方式中，针对每一个所述候选进程，均执行如下操作：扫描所述候选进程对应的进程虚地址空间，并根据所述候选进程的进程描述符，获取所述进程虚地址空间中所有的虚地址区域；遍历所述进程虚地址空间中所有的虚地址区域；并对遍历到的每一个所述虚地址区域，根据该虚地址区域的标志位，判断该虚地址区域是否为可执行区域；当判定任意一虚地址区域为可执行区域时，将所述任意一虚地址区域中包含 的页面确定为代码页面。
结合第一方面，第一方面第一种可能的实现方式至第一方面第三种可能的实现方式中的任一项，在第四种可能的实现方式中，针对每一个所述代码页面，均执行如下操作：在所述交换分区中申请一个页槽；将所述代码页面存储至申请得到的所述页槽中；将所述代码页面的物理地址，更新为所述页槽的物理地址。
结合第一方面第四种可能的实现方式，在第五种可能的实现方式中，所述交换分区为非易失性存储器NVM交换分区，且所述NVM交换分区直接与内存的内存控制器连接；将所述代码页面通过所述内存控制器存储至申请得到的所述页槽中。
结合第一方面，第一方面第一种可能的实现方式至第一方面第五种可能的实现方式中的任一项，在第六种可能的实现方式中，当运行所述代码页面所属的候选进程时，若确定需要执行所述代码页面，则获取所述页表项中保存的更新后的所述代码页面的虚拟地址；从所述页表项中确定与获取的所述虚拟地址存在对应关系的物理地址；并根据所述物理地址，在所述交换区域中获取所述代码页面；以及执行所述代码页面。
第二方面，提供一种运行进程的装置，包括：扫描单元，用于扫描内存中的进程；候选进程选取单元，用于根据所述扫描单元对内存中进程的扫描结果，从所有进程中选取候选进程；代码页面获取单元，用于从所述候选进程选取单元选取的所述候选进程中获取代码页面；其中，所述代码页面用于存放进程的二进制代码或者共享库文件；写入单元，用于将所述代码页面获取单元获取的代码页面写入交换分区中；存储空间释放单元，用于释放所述内存中所述代码页面获取单元获取的代码页面所占用的存储空间；物理地址获取单元，用于获取页表项中保存的所述代码页面的物理地址；更新单元，用于将所述物理地址获取单元获取的代码页面的物理地址更新为所述交换分区中存储所述代码页面的物理地址；其中，所述页表项用于存储代码页面的虚拟地址与所述代码页面的物理地址之间的对应关系；执行单元，用于当运 行所述代码页面所属的候选进程时，根据页表项中保存的所述更新单元更新后的所述代码页面的虚拟地址和物理地址的对应关系，在所述交换分区中执行所述代码页面。
结合第二方面，在第一种可能的实现方式中，还包括确定单元，用于：扫描内存中的进程之前，确定所述内存中包含的可用存储空间的容量值小于预设阈值。
结合第二方面，或者第二方面第一种可能的实现方式，在第二种可能的实现方式中，所述候选进程选取单元，具体用于：分别获取每一个进程的评分等级；其中，任意一进程的评分等级根据所述任意一进程的优先级以及所述任意一进程所占用内存空间的大小确定，当所述任意一进程的优先级越低，且所述占用内存空间越大时，所述任意一进程的评分等级越高，当所述任意一进程的优先级越高，且所述占用内存空间越小时，所述任意一进程的评分等级越低；从所有进程中依次选取评分等级从高到低的进程，直至选择至内存中包含的可用存储空间大于或等于预设阈值为止后，将选取出的各个进程确定为候选进程。
结合第二方面，第二方面第一种可能的实现方式，或者第二方面第二种可能的实现方式，在第三种可能的实现方式中，所述代码页面获取单元，具体用于：针对每一个所述候选进程，均执行如下操作：扫描所述候选进程对应的进程虚地址空间，并根据所述候选进程的进程描述符，获取所述进程虚地址空间中所有的虚地址区域；遍历所述进程虚地址空间中所有的虚地址区域；并对遍历到的每一个所述虚地址区域，根据该虚地址区域的标志位，判断该虚地址区域是否为可执行区域；当判定任意一虚地址区域为可执行区域时，将所述任意一虚地址区域中包含的页面确定为代码页面。
结合第二方面，第二方面第一种可能的实现方式至第二方面第三种可能的实现方式中的任一项，在第四种可能的实现方式中，所述写入单元，具体用于：针对每一个所述代码页面，均执行如下操作：在所述交换分区中申请一个页槽；将所述代码页面存储至申请得到的所述页槽中；所述更新单元， 具体用于：将所述代码页面的物理地址，更新为所述页槽的物理地址。
结合第二方面第四种可能的实现方式，在第五种可能的实现方式中，所述写入单元，具体用于：将所述代码页面通过所述内存控制器存储至申请得到的所述页槽中；所述交换分区为非易失性存储器NVM交换分区，且所述NVM交换分区直接与内存的内存控制器连接。
结合第二方面，第二方面第一种可能的实现方式至第二方面第五种可能的实现方式中的任一项，在第六种可能的实现方式中，所述执行单元，具体用于：当运行所述代码页面所属的候选进程时，若确定需要执行所述代码页面，则获取所述页表项中保存的更新后的所述代码页面的虚拟地址；从所述页表项中确定与获取的所述虚拟地址存在对应关系的物理地址；并根据所述物理地址，在所述交换区域中获取所述代码页面；以及执行所述代码页面。
本发明实施例中，从候选进程中选取代码页面，仅将候选进程中的代码页面存储至交换分区中，并释放该代码页面所占用的内存空间；并更新页表项中保存的代码页面的物理地址；当确定需要运行代码页面所属的候选进程时，若需要执行该代码页面，直接在交换分区中执行该代码页面即可。采用本发明技术方案，仅将具有只读特性的代码页面存储至交换分区中，当需要执行该代码页面时，直接在交换分区中执行该代码页面，无须将代码页面替换至内存中执行，在提高进程运行速度的同时，有效避免了数据的频繁交换，减少了数据频繁交换造成的写操作次数；并且，无须在本地保存每一个页槽的年龄，节约了系统空间，减少了系统负担，提高了系统性能。
附图说明
图1为本发明实施例中运行进程流程图；
图2为本发明实施例中终端软件层级示意图；
图3为本发明实施例中虚拟地址空间示意图；
图4a和图4b为本发明实施例中内存和交换分区示意图；
图5为本发明实施例中交换分区写操作过程示意图；
图6为本发明实施例中比较结果示意图一；
图7为本发明实施例中比较结果示意图二；
图8为本发明实施例中运行进程装置结构示意图；
图9为本发明实施例中进程运行设备结构示意图。
具体实施方式
为了解决目前在运行进程的过程中，存在进程运行速度慢，交换分区使用寿命短，系统资源浪费以及系统性能差的问题。本发明实施例中，从候选进程中选取代码页面，仅将候选进程中的代码页面存储至交换分区中，并释放该代码页面所占用的内存空间；并更新页表项中保存的代码页面的物理地址；当确定需要运行代码页面所属的候选进程时，若需要执行该代码页面，直接在交换分区中执行该代码页面即可。采用本发明技术方案，仅将具有只读特性的代码页面存储至交换分区中，当需要执行该代码页面时，直接在交换分区中执行该代码页面，无须将代码页面替换至内存中执行，在提高进程运行速度的同时，有效避免了数据的频繁交换，减少了数据频繁交换造成的写操作次数；并且，无须在本地保存每一个页槽的年龄，节约了系统空间，减少了系统负担，提高了系统性能。
下面结合说明书附图，对本发明实施例作进一步详细描述。
参阅图1所示，本发明实施例中，在终端中运行进程的方法，包括：
步骤100：扫描内存中的进程，从所有进程中选取候选进程。
本发明实施例中，终端中可以在一个时刻运行一个应用程序，也可以在一个时刻同时运行多个应用程序，每一个应用程序在运行过程中均需要申请相应的内存资源，并使用该内存资源；由于进程为应用程序在运行过程中的最小单位，因此，当终端在运行应用程序的过程中，内存不足以支撑应用程序的运行时，即终端的内存中包含的存储空间小于预设阈值时，终端需要扫描正在运行的应用程序对应的进程，并从所有进程中选取候选进程，转存至交换分区中，以解决内存资源不足的问题；其中，上述终端可以为移动终端 (如手机等)，也可以为个人电脑(Personal Computer，简称PC)，且该终端包含中心处理器(Central Processing Unit，简称CPU)，内存(如DRAM)，交换分区等硬件设备，且该终端中安装有操作系统；上述预设阈值可以根据具体应用场景动态调整。
可选的，终端根据每一个进程的评分等级，从所有进程中选取候选进程，具体为：终端分别获取每一个进程的评分等级；其中，任意一进程的评分等级根据该任意一进程的优先级以及该任意一进程所占用内存空间的大小确定，如为优先级和占用内存空间的大小分别赋予相应的权重值，通过如下公式“优先级×优先级权重值+占用内存空间的大小×内存空间权重值”确定任意一进程的综合评分，并查找终端本地保存的综合评分和评分等级对照表，以确定该综合评分对应的评分等级，综合评分和评分等级对照表用于保存综合评分和评分等级之间的对应关系，每一个评分等级可以对应一个综合评分段，且该综合评分和评分等级对照表可以根据具体应用场景预先设置；此外，上述评分等级由操作系统中的特定部件确定，例如，在Android系统中，该评分等级可以由Android低内存管理器(Low Memory Killer，简称LMK)确定；根据评分等级，从内存中的所有进程中选定候选进程。
可选的，终端根据评分等级，从内存中的所有进程中选定候选进程的过程，具体为：从所有进程中依次选取评分等级从高到低的进程，直至选择至运行所有应用程序时所申请的内存空间小于或等于上述预设阈值位置为止，将选取出的所有进程确定为候选进程。本发明实施例中，以当任意一进程的优先级越低，且占用内存空间越大时，该任意一进程的评分等级越高，当该任意一进程的优先级越高，且占用内存空间越小时，该任意一进程的评分等级越低为例，介绍候选进程的获取过程；此外，移动终端还可以设置当任意一进程的优先级越高，且占用内存空间越小时，该任意一进程的评分等级越高，当该任意一进程的优先级越低，且占用内存空间越大时，该任意一进程的评分等级越低，此时，移动终端需要从所有进程中依次选取评分等级最低的进程，直至运行所有应用程序时所申请的内存空间小于或等于上述预设阈 值位置，将选取的上述进程确定为候选进程。
本发明实施例中，终端安装的操作系统可以包含多种类型，如Android操作系统，IOS操作系统，微软操作系统等。例如，参阅图2所示，当终端安装的操作系统为Android操作系统时，终端中所安装的软件层级为：最顶层为应用程序；在应用程序之下为Android系统的Dalvik虚拟机以及运行库；Android系统的Dalvik之下为Linux kernel，Linux kernel中包含了内存管理，进程管理，设备驱动程序等；最下层为具体的硬件，包括内存(如DRAM)、交换分区(如NVM以及闪存Flash)等。
步骤110：从选取的候选进程中获取代码页面；其中，该代码页面用于存放进程的二进制代码，或者共享库文件。
本发明实施例中，内存中的每一个进程均由多个页面(以下称为内存页面)组成，每一个内存页面均占用一定的存储空间，一个进程所占用内存空间的大小即为该进程中所包含的所有内存页面所占用的存储空间的总和；内存中的页面包含两种类型，即为代码页面(Code Page)和数据页面(Data Page)，其中，代码页面用于存放进程的二进制代码(Binary Code)，或者共享库文件(Shared Library)，且该代码页面拥有只读特性，而数据页面用于存放应用程序的数据和变量等，该数据页面不具备只读特性。
可选的，针对每一个候选进程，均执行如下操作：扫描候选进程对应的进程虚地址空间，其中，每一个候选进程对应的进程虚拟地址空间即为终端为每一个进程分配的内存空间；由于每一个虚拟地址空间中包含多个虚地址区域，且该虚地址区域通过页表项映射到物理存储空间，用于保存应用程序运行时的数据，因此，终端根据候选进程的进程描述符，获取该进程虚地址空间中所有的虚地址区域；遍历上述获取的进程虚地址空间中所有的虚地址区域；并获取遍历到的每一个虚地址区域的标志位，当存在任意一虚地址区域的标志位为可执行区域标志位时，确定该任意一虚地址区域为可执行区域，将该任意一虚地址区域中包含的页面确定为代码页面。
在上述过程中，候选进程的进程描述符和虚地址区域的标志位根据操作 系统而确定，即终端安装的操作系统不同时，候选进程的进程描述符不同，且虚地址区域的标志位也不同。例如，参阅图3所示，为Android系统中一个进程的虚拟地址空间，在该进程的虚拟地址空间中，包含多个虚地址区域，每一个虚地址区域都使用vm_area_struct表示，且每一个虚地址区域之间用vm_next连接，当虚地址区域之后不存在vm_next时，表示该虚地址区域为该该虚地址区域所属的虚拟地址空间的结束位置；在Android系统中，虚地址区域的标志位为VM_READ和VM_EXEC，在图3中方框内所示部分(第一个方框内所示部分为共享库文件/etc/1d.so区域，第二个方框所示部分为二进制代码/bin/gcc区域)，为由VM_READ和VM_EXEC共同确定的标志位，该VM_READ和VM_EXEC所指向的页面即为代码页面，即为图3中的“Memory mapping”和“Text(file backed)”。
采用上述技术方案，根据代码页面的特性，以及进程描述符和标志位，从内存中所保存的候选进程的所有页面中，选取代码页面，能够更为快速并准确地获取每一个候选进程的代码页面，提高了系统处理效率。
步骤120：将代码页面写入交换分区中，并释放内存中该代码页面所占用的存储空间。
本发明实施例中，终端将获取到的代码页面写入交换分区中，当代码页面全部写入到交换分区之后，终端释放内存中代码页面所占用的存储空间，以减少对内存的占用；其中，参阅图4a所示交换分区可以通过闪存实现，内存和闪存之间的数据传输由I/O驱动程序负责，即内存和闪存之间的数据传输需要I/O调度，参阅图4b所示，交换分区还可以通过NVM实现，内存和NVM均连接至内存控制器(Memory Controller)，由内存控制器控制内存和NVM之间的数据交换，且终端安装的操作系统将内存和NVM视为没有任何区别，操作系统可以和通过虚拟地址访问内存和交换分区中保存的数据，且交换分区和之间的数据交换通过内存接口，NVM上的数据可以通过内存复制到内存中，反之亦可，此外，内存和NVM统一编码；当交换分区通过闪存实现时，由于内存和闪存之间的数据传输需要I/O调度，且I/O调度速度相对较慢，从 而存在读写延迟时间长的问题；当交换分区通过NVM实现时，由于内存和NVM之间的数据传输由内存控制器直接调度，数据通过内存总线传输即可，无须通过I/O调度，相较于通过闪存实现交换分区的方案，采用NVM实现交换分区，读写延迟时间短。
可选的，终端针对每一个代码页面，均执行如下操作：由于交换分区的基本单位为页槽，每一个代码页面所占用存储空间的大小与一个页槽所占用的存储空间的大小相等，因此，当终端确定需要释放内存空间时，在交换分区中为代码页面申请一个页槽；终端直接将代码页面通过内存控制器存储至申请得到的页槽中。
采用上述技术方案，终端将获取的代码页面存储至交换分区中，进而释放内存中替换出的代码页面所占用的存储空间，为其他进程运行提供更多的内存资源，有效解决了内存资源不足的问题。
步骤130：获取页表项中保存的上述代码页面的物理地址，并将该代码页面的物理地址更新为交换分区中存储该代码页面的物理地址；其中，页表项用于存储代码页面的虚拟地址与该代码页面的物理地址之间的对应关系。
本发明实施例中，终端本地保存有页表项，该页表项中保存有代码页面的虚拟地址与该代码页面的物理地址之间的对应关系，其中，代码页面的虚拟地址不可更改。在将代码页面存储至申请得到的页槽之后，终端将页表项中该代码页面的物理地址更新为交换分区中存储该代码页面的物理地址。基于上述过程，终端将上述代码页面存储至交换分区中的相应页槽中，因此，将代码页面的物理地址更新为交换分区中存储该代码页面的物理地址，即为将该代码页面的物理地址，更新为上述页槽的物理地址。参阅图5所示，为本发明实施例中在交换分区中写入代码页面时页表项、内存和交换分区之间数据交换示意图。
采用上述技术方案，在终端将代码页面存储至申请得到的页槽之后，更新页表项中上述代码页面的物理地址，避免了后续在交换分区中执行上述代码页面中包含的数据时，根据原有的物理地址无法获取代码页面的问题，保 证了代码页面执行的可行性。
步骤140：当运行代码页面所属的候选进程时，根据页表项中保存的更新后的该代码页面的虚拟地址和物理地址的对应关系，在交换分区中执行该代码页面。
本发明实施例中，当终端确定运行代码页面所属的候选进程时，若确定需要执行任意一代码页面，且该任意一代码页面位于交换分区中，则获取页表项中更新后的上述任意一代码页面的虚拟地址；从该页表项中确定与获取的虚拟地址存在对应关系的物理地址；并根据该物理地址，在交换区域中获取上述任意一代码页面；以及执行该任意一代码页面中包含的数据。
基于上述技术方案，将本发明技术方案和现有技术中选择不经常使用的内存页面替换至交换分区的方法相比较，设定应用环境为Google Nexus 5智能手机。在比较过程中，选取不同类型的应用程序，并将这些应用程序归为六类：Browser，Social networking，Multimedia，Gaming，Online Shopping和News。对于每一类应用程序，分别采用本发明提供的技术方案，以及采用现有技术运行20分钟。参阅图6所示，为交换(Swap)分区的写操作随时间变化曲线，该图中横坐标表示时间，纵坐标表示对交换分区写操作的总次数；第一曲线表示采用本发明技术方案运行应用程序时的运行结果，第二曲线表示采用现有技术运行应用程序时的运行结果。由图6可知，采用本发明提供的技术方案运行应用程序时相较于采用现有技术运行应用程序时，减少了14％～51％的写操作。并且，采用本发明技术方案对交换分区执行写操作时，若终端不存在新的应用程序运行，在一段时间后对于交换分区的写操作数量将不再增加，而采用现有技术运行应用程序时，对交换分区的写操作会持续增加。
由此可见，相较于现有技术在交换分区中执行写操作的方式，代码页面替换至交换分区后，当需要该代码页面时，对交换分区的写操作会持续的增加。采用本发明技术方案，仅将候选进程中的代码页面替换至交换分区中，当需要执行交换分区中的任意一代码页面时，无须将该任意一代码页面重新 调用至内存中，而直接在交换分区中执行该任意一代码页面即可，随着代码页面的不断换出，可用内存空间也越来越大，最后可用内存空间趋于稳定，不再需要代码页由内存替换至交换分区，使交换分区的写操作会趋于稳定，在一定时间之后不再增加，从而减少了对交换分区的写操作数量，有效延长了交换分区的使用寿命。
基于上述技术方案，将采用本发明技术方案时，应用程序的切换时间在使用交换分区的场景下与没有使用交换分区的场景下进行对比。在本次对比过程中，在上述两种场景中，均设定在终端中共运行了30个应用程序，并在该30个应用程序之间不停切换。比较结果参阅图7所示，图中横坐标表示应用程序，纵坐标表示该应用程序的平均切换时间，第一柱体表示采用交换分区的应用场景，第二柱体表示没有采用交换分区的应用场景。由图7可知，采用本发明技术方案使用交换分区的场景，相较于没有采用交换分区的场景，应用程序的切换时间减少了19％～41％。
由此可见，在现有技术中，当不存在交换分区时，若内存空间不足，终端所采用的机制为强制停止该页面对应的应用程序，当内存空间充足后，终端需要重新加载上述应用程序，导致应用程序执行时间较长；采用本发明技术方案，当内存空间不足时，将候选进程中的代码页面存储至交换分区中，并当需要执行交换分区中的任意一代码页面时，无须将该任意一代码页面重新调用至内存中，而仅需要根据该任意一代码页面的物理地址，获取该任意一代码页面，进而直接执行该任意一代码页面即可，无须将应用程序强制停止，进而提高了应用程序的执行速度。
基于上述技术方案，参阅图8所示，本发明实施例还提供一种运行进程的装置，包括扫描单元80，候选进程选取单元81，代码页面获取单元82，写入单元83，存储空间释放单元84，物理地址获取单元85，更新单元86，以及执行单元87，其中：
扫描单元80，用于扫描内存中的进程；
候选进程选取单元81，用于根据所述扫描单元80对内存中进程的扫描结 果，从所有进程中选取候选进程；
代码页面获取单元82，用于从所述候选进程选取单元81选取的所述候选进程中获取代码页面；其中，所述代码页面用于存放进程的二进制代码或者共享库文件；
写入单元83，用于将所述代码页面获取单元82获取的代码页面写入交换分区中；
存储空间释放单元84，用于释放所述内存中所述代码页面获取单元82获取的代码页面所占用的存储空间；
物理地址获取单元85，用于获取页表项中保存的所述代码页面的物理地址；
更新单元86，用于将所述物理地址获取单元85获取的代码页面的物理地址更新为所述交换分区中存储所述代码页面的物理地址；其中，所述页表项用于存储代码页面的虚拟地址与所述代码页面的物理地址之间的对应关系；
执行单元87，用于当运行所述代码页面所属的候选进程时，根据页表项中保存的所述更新单元86更新后的所述代码页面的虚拟地址和物理地址的对应关系，在所述交换分区中执行所述代码页面。
进一步的，上述装置还包括确定单元88，用于：扫描内存中的进程之前，确定所述内存中包含的可用存储空间的容量值小于预设阈值。
可选的，所述候选进程选取单元81，具体用于：分别获取每一个进程的评分等级；其中，任意一进程的评分等级根据所述任意一进程的优先级以及所述任意一进程所占用内存空间的大小确定，当所述任意一进程的优先级越低，且所述占用内存空间越大时，所述任意一进程的评分等级越高，当所述任意一进程的优先级越高，且所述占用内存空间越小时，所述任意一进程的评分等级越低；从所有进程中依次选取评分等级从高到低的进程，直至选择至内存中包含的可用存储空间大于或等于预设阈值为止后，将选取出的各个进程确定为候选进程。
可选的，所述代码页面获取单元82，具体用于：针对每一个所述候选进 程，均执行如下操作：扫描所述候选进程对应的进程虚地址空间，并根据所述候选进程的进程描述符，获取所述进程虚地址空间中所有的虚地址区域；遍历所述进程虚地址空间中所有的虚地址区域；并对遍历到的每一个所述虚地址区域，根据该虚地址区域的标志位，判断该虚地址区域是否为可执行区域；当判定任意一虚地址区域为可执行区域时，将所述任意一虚地址区域中包含的页面确定为代码页面。
可选的，所述写入单元83，具体用于：针对每一个所述代码页面，均执行如下操作：在所述交换分区中申请一个页槽；将所述代码页面存储至申请得到的所述页槽中；所述更新单元86，具体用于：将所述代码页面的物理地址，更新为所述页槽的物理地址。
可选的，所述写入单元83，具体用于：将所述代码页面通过所述内存控制器存储至申请得到的所述页槽中；所述交换分区为非易失性存储器NVM交换分区，且所述NVM交换分区直接与内存的内存控制器连接。
可选的，所述执行单元87，具体用于：当运行所述代码页面所属的候选进程时，若确定需要执行所述代码页面，则获取所述页表项中保存的更新后的所述代码页面的虚拟地址；从所述页表项中确定与获取的所述虚拟地址存在对应关系的物理地址；并根据所述物理地址，在所述交换区域中获取所述代码页面；以及执行所述代码页面。
基于上述技术方案，参阅图9所示，本发明实施例还提供一种进程运行设备，包括存储器90，处理器91，其中：
存储器90，用于存储应用程序以及页表项，包含交换分区以及内存；
处理器91，用于运行存储器90中的应用程序，执行如下操作：
扫描内存中的进程；根据所述扫描单元对内存中进程的扫描结果，从所有进程中选取候选进程；从所述候选进程选取单元选取的所述候选进程中获取代码页面；其中，所述代码页面用于存放进程的二进制代码或者共享库文件；将所述代码页面获取单元获取的代码页面写入交换分区中；释放所述内存中所述代码页面获取单元获取的代码页面所占用的存储空间；获取页表项 中保存的所述代码页面的物理地址；将所述物理地址获取单元获取的代码页面的物理地址更新为所述交换分区中存储所述代码页面的物理地址；其中，所述页表项用于存储代码页面的虚拟地址与所述代码页面的物理地址之间的对应关系；当运行所述代码页面所属的候选进程时，根据页表项中保存的所述更新单元更新后的所述代码页面的虚拟地址和物理地址的对应关系，在所述交换分区中执行所述代码页面。
进一步的，所述处理器91，还用于：扫描内存中的进程之前，确定所述内存中包含的可用存储空间的容量值小于预设阈值。
可选的，所述处理器91，具体用于：分别获取每一个进程的评分等级；其中，任意一进程的评分等级根据所述任意一进程的优先级以及所述任意一进程所占用内存空间的大小确定，当所述任意一进程的优先级越低，且所述占用内存空间越大时，所述任意一进程的评分等级越高，当所述任意一进程的优先级越高，且所述占用内存空间越小时，所述任意一进程的评分等级越低；从所有进程中依次选取评分等级从高到低的进程，直至选择至内存中包含的可用存储空间大于或等于预设阈值为止后，将选取出的各个进程确定为候选进程。
可选的，所述处理器91，具体用于：针对每一个所述候选进程，均执行如下操作：扫描所述候选进程对应的进程虚地址空间，并根据所述候选进程的进程描述符，获取所述进程虚地址空间中所有的虚地址区域；遍历所述进程虚地址空间中所有的虚地址区域；并对遍历到的每一个所述虚地址区域，根据该虚地址区域的标志位，判断该虚地址区域是否为可执行区域；当判定任意一虚地址区域为可执行区域时，将所述任意一虚地址区域中包含的页面确定为代码页面。
可选的，所述处理器91，具体用于：针对每一个所述代码页面，均执行如下操作：在所述交换分区中申请一个页槽；将所述代码页面存储至申请得到的所述页槽中；将所述代码页面的物理地址，更新为所述页槽的物理地址。
进一步的，所述进程运行设备还包括内存控制器92，用于连接内存和交 换分区；可选的，所述处理器91，具体用于：将所述代码页面通过所述内存控制器92存储至申请得到的所述页槽中；所述交换分区为NVM交换分区。
可选的，所述处理器91，具体用于：当运行所述代码页面所属的候选进程时，若确定需要执行所述代码页面，则获取所述页表项中保存的更新后的所述代码页面的虚拟地址；从所述页表项中确定与获取的所述虚拟地址存在对应关系的物理地址；并根据所述物理地址，在所述交换区域中获取所述代码页面；以及执行所述代码页面。
综上所述，本发明实施例中，扫描内存中的进程，从所有进程中选取候选进程；从选取的候选进程中获取代码页面；其中，该代码页面用于存放进程的二进制代码，或者共享库文件；将代码页面写入交换分区中，并释放内存中该代码页面所占用的存储空间；获取页表项中保存的上述代码页面的物理地址，并将该代码页面的物理地址更新为交换分区中存储该代码页面的物理地址；其中，页表项用于存储代码页面的虚拟地址与该代码页面的物理地址之间的对应关系；当运行代码页面所属的候选进程时，根据页表项中保存的更新后的该代码页面的虚拟地址和物理地址的对应关系，在交换分区中执行该代码页面。采用本发明技术方案，仅将具有只读特性的代码页面存储至交换分区中，当需要执行该代码页面时，直接在交换分区中执行该代码页面，无须将代码页面替换至内存中执行，在提高进程运行速度的同时，有效避免了数据的频繁交换，减少了数据频繁交换造成的写操作次数；并且，无须在本地保存每一个页槽的年龄，节约了系统空间，减少了系统负担，提高了系统性能。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产 品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本发明的优选实施例，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例作出另外的变更和修改。所以，所附权利要求意欲解释为包括优选实施例以及落入本发明范围的所有变更和修改。显然，本领域的技术人员可以对本发明实施例进行各种改动和变型而不脱离本发明实施例的精神和范围。这样，倘若本发明实施例的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。

Claims (14)

  1. 一种运行进程的方法，其特征在于，包括：
    扫描内存中的进程，从所有进程中选取候选进程；并
    从选取的所述候选进程中获取代码页面；其中，所述代码页面用于存放进程的二进制代码或者共享库文件；
    将所述代码页面写入交换分区中，并释放所述内存中所述代码页面所占用的存储空间；以及
    获取页表项中保存的所述代码页面的物理地址，并将所述代码页面的物理地址更新为所述交换分区中存储所述代码页面的物理地址；其中，所述页表项用于存储代码页面的虚拟地址与所述代码页面的物理地址之间的对应关系；
    当运行所述代码页面所属的候选进程时，根据页表项中保存的更新后的所述代码页面的虚拟地址和物理地址的对应关系，在所述交换分区中执行所述代码页面。
  2. 如权利要求1所述的方法，其特征在于，扫描内存中的进程之前，还包括：
    确定所述内存中包含的可用存储空间的容量值小于预设阈值。
  3. 如权利要求1或2所述的方法，其特征在于，从所有进程中选取候选进程，具体包括：
    分别获取每一个进程的评分等级；其中，任意一进程的评分等级根据所述任意一进程的优先级以及所述任意一进程所占用内存空间的大小确定，当所述任意一进程的优先级越低，且所述占用内存空间越大时，所述任意一进程的评分等级越高，当所述任意一进程的优先级越高，且所述占用内存空间越小时，所述任意一进程的评分等级越低；
    从所有进程中依次选取评分等级从高到低的进程，直至选择至内存中包含的可用存储空间大于或等于预设阈值为止后，将选取出的各个进程确定为 候选进程。
  4. 如权利要求1-3任一项所述的方法，其特征在于，从选取的所述候选进程中获取代码页面，具体包括：
    针对每一个所述候选进程，均执行如下操作：
    扫描所述候选进程对应的进程虚地址空间，并根据所述候选进程的进程描述符，获取所述进程虚地址空间中所有的虚地址区域；
    遍历所述进程虚地址空间中所有的虚地址区域；并
    对遍历到的每一个所述虚地址区域，根据该虚地址区域的标志位，判断该虚地址区域是否为可执行区域；
    当判定任意一虚地址区域为可执行区域时，将所述任意一虚地址区域中包含的页面确定为代码页面。
  5. 如权利要求1-4任一项所述的方法，其特征在于，将所述代码页面写入交换分区中，具体包括：
    针对每一个所述代码页面，均执行如下操作：
    在所述交换分区中申请一个页槽；
    将所述代码页面存储至申请得到的所述页槽中；
    将所述代码页面的物理地址更新为所述交换分区中存储所述代码页面的物理地址，具体包括：
    将所述代码页面的物理地址，更新为所述页槽的物理地址。
  6. 如权利要求5所述的方法，其特征在于，所述交换分区为非易失性存储器NVM交换分区，且所述NVM交换分区直接与内存的内存控制器连接；
    将所述代码页面存储至申请得到的所述页槽中，具体包括：
    将所述代码页面通过所述内存控制器存储至申请得到的所述页槽中。
  7. 如权利要求1-6任一项所述的方法，其特征在于，当运行所述代码页面所属的候选进程时，根据页表项中保存的更新后的所述代码页面的虚拟地址和物理地址的对应关系，在所述交换分区中执行所述代码页面，具体包括：
    当运行所述代码页面所属的候选进程时，若确定需要执行所述代码页面， 则获取所述页表项中保存的更新后的所述代码页面的虚拟地址；
    从所述页表项中确定与获取的所述虚拟地址存在对应关系的物理地址；并
    根据所述物理地址，在所述交换区域中获取所述代码页面；以及
    执行所述代码页面。
  8. 一种运行进程的装置，其特征在于，包括：
    扫描单元，用于扫描内存中的进程；
    候选进程选取单元，用于根据所述扫描单元对内存中进程的扫描结果，从所有进程中选取候选进程；
    代码页面获取单元，用于从所述候选进程选取单元选取的所述候选进程中获取代码页面；其中，所述代码页面用于存放进程的二进制代码或者共享库文件；
    写入单元，用于将所述代码页面获取单元获取的代码页面写入交换分区中；
    存储空间释放单元，用于释放所述内存中所述代码页面获取单元获取的代码页面所占用的存储空间；
    物理地址获取单元，用于获取页表项中保存的所述代码页面的物理地址；
    更新单元，用于将所述物理地址获取单元获取的代码页面的物理地址更新为所述交换分区中存储所述代码页面的物理地址；其中，所述页表项用于存储代码页面的虚拟地址与所述代码页面的物理地址之间的对应关系；
    执行单元，用于当运行所述代码页面所属的候选进程时，根据页表项中保存的所述更新单元更新后的所述代码页面的虚拟地址和物理地址的对应关系，在所述交换分区中执行所述代码页面。
  9. 如权利要求8所述的装置，其特征在于，还包括确定单元，用于：
    扫描内存中的进程之前，确定所述内存中包含的可用存储空间的容量值小于预设阈值。
  10. 如权利要求8或9所述的装置，其特征在于，所述候选进程选取单 元，具体用于：
    分别获取每一个进程的评分等级；其中，任意一进程的评分等级根据所述任意一进程的优先级以及所述任意一进程所占用内存空间的大小确定，当所述任意一进程的优先级越低，且所述占用内存空间越大时，所述任意一进程的评分等级越高，当所述任意一进程的优先级越高，且所述占用内存空间越小时，所述任意一进程的评分等级越低；从所有进程中依次选取评分等级从高到低的进程，直至选择至内存中包含的可用存储空间大于或等于预设阈值为止后，将选取出的各个进程确定为候选进程。
  11. 如权利要求8-10任一项所述的装置，其特征在于，所述代码页面获取单元，具体用于：
    针对每一个所述候选进程，均执行如下操作：扫描所述候选进程对应的进程虚地址空间，并根据所述候选进程的进程描述符，获取所述进程虚地址空间中所有的虚地址区域；遍历所述进程虚地址空间中所有的虚地址区域；并对遍历到的每一个所述虚地址区域，根据该虚地址区域的标志位，判断该虚地址区域是否为可执行区域；当判定任意一虚地址区域为可执行区域时，将所述任意一虚地址区域中包含的页面确定为代码页面。
  12. 如权利要求8-11任一项所述的装置，其特征在于，所述写入单元，具体用于：
    针对每一个所述代码页面，均执行如下操作：在所述交换分区中申请一个页槽；将所述代码页面存储至申请得到的所述页槽中；
    所述更新单元，具体用于：
    将所述代码页面的物理地址，更新为所述页槽的物理地址。
  13. 如权利要求12所述的装置，其特征在于，所述写入单元，具体用于：
    将所述代码页面通过所述内存控制器存储至申请得到的所述页槽中；所述交换分区为非易失性存储器NVM交换分区，且所述NVM交换分区直接与内存的内存控制器连接。
  14. 如权利要求8-13任一项所述的装置，其特征在于，所述执行单元， 具体用于：
    当运行所述代码页面所属的候选进程时，若确定需要执行所述代码页面，则获取所述页表项中保存的更新后的所述代码页面的虚拟地址；从所述页表项中确定与获取的所述虚拟地址存在对应关系的物理地址；并根据所述物理地址，在所述交换区域中获取所述代码页面；以及执行所述代码页面。
PCT/CN2015/099051 2015-03-04 2015-12-25 一种运行进程的方法及装置 WO2016138785A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP15883835.9A EP3252595A4 (en) 2015-03-04 2015-12-25 Method and device for running process
US15/693,955 US10261918B2 (en) 2015-03-04 2017-09-01 Process running method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510097009.3 2015-03-04
CN201510097009.3A CN105988875B (zh) 2015-03-04 2015-03-04 一种运行进程的方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/693,955 Continuation US10261918B2 (en) 2015-03-04 2017-09-01 Process running method and apparatus

Publications (1)

Publication Number Publication Date
WO2016138785A1 true WO2016138785A1 (zh) 2016-09-09

Family

ID=56849212

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/099051 WO2016138785A1 (zh) 2015-03-04 2015-12-25 一种运行进程的方法及装置

Country Status (4)

Country Link
US (1) US10261918B2 (zh)
EP (1) EP3252595A4 (zh)
CN (1) CN105988875B (zh)
WO (1) WO2016138785A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106502924B (zh) * 2016-10-27 2020-02-07 深圳创维数字技术有限公司 一种内存优化方法及系统
CN110865884A (zh) * 2019-09-30 2020-03-06 华为技术有限公司 一种内存管理方法及装置
US12086447B2 (en) * 2019-12-18 2024-09-10 Advanced Micro Devices, Inc. Systems and methods for reducing instruction code memory footprint for multiple processes executed at a coprocessor
CN111597124B (zh) * 2020-04-21 2023-05-05 重庆大学 持久化内存文件系统数据组织方法、系统及存储介质
CN113590500B (zh) 2020-04-30 2024-10-11 华为技术有限公司 一种内存管理方法及终端设备
CN114461375B (zh) * 2021-07-30 2023-01-20 荣耀终端有限公司 内存资源管理方法及电子设备
CN113971091B (zh) * 2021-10-25 2024-05-14 重庆大学 一种考虑制程差异的持久内存的分配方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102226894A (zh) * 2011-05-23 2011-10-26 中兴通讯股份有限公司 一种移动设备内存管理方法及装置
WO2014059613A1 (zh) * 2012-10-17 2014-04-24 华为技术有限公司 降低内存系统功耗的方法和内存控制器
CN104182350A (zh) * 2013-05-28 2014-12-03 中国银联股份有限公司 一种针对包括多个进程的应用的内存管理方法和装置
CN104281528A (zh) * 2013-07-09 2015-01-14 浙江大华技术股份有限公司 一种数据存储方法及装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101038167B1 (ko) * 2008-09-09 2011-05-31 가부시끼가이샤 도시바 프로세서로부터 메모리로의 액세스를 관리하는 메모리 관리 장치를 포함하는 정보 처리 장치 및 메모리 관리 방법
US8296496B2 (en) * 2009-09-17 2012-10-23 Hewlett-Packard Development Company, L.P. Main memory with non-volatile memory and DRAM
KR20120066198A (ko) * 2010-12-14 2012-06-22 삼성전자주식회사 휘발성 메모리를 포함하는 저장 장치의 데이터 저장 방법
CN103713882A (zh) * 2013-12-20 2014-04-09 华为技术有限公司 一种数据换入内存的方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102226894A (zh) * 2011-05-23 2011-10-26 中兴通讯股份有限公司 一种移动设备内存管理方法及装置
WO2014059613A1 (zh) * 2012-10-17 2014-04-24 华为技术有限公司 降低内存系统功耗的方法和内存控制器
CN104182350A (zh) * 2013-05-28 2014-12-03 中国银联股份有限公司 一种针对包括多个进程的应用的内存管理方法和装置
CN104281528A (zh) * 2013-07-09 2015-01-14 浙江大华技术股份有限公司 一种数据存储方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3252595A4 *

Also Published As

Publication number Publication date
EP3252595A1 (en) 2017-12-06
US20170364449A1 (en) 2017-12-21
EP3252595A4 (en) 2018-03-28
CN105988875A (zh) 2016-10-05
CN105988875B (zh) 2020-08-14
US10261918B2 (en) 2019-04-16

Similar Documents

Publication Publication Date Title
WO2016138785A1 (zh) 一种运行进程的方法及装置
US10552337B2 (en) Memory management and device
US7653799B2 (en) Method and apparatus for managing memory for dynamic promotion of virtual memory page sizes
US11698757B2 (en) Memory system and method of controlling nonvolatile memory
US8645612B2 (en) Information processing device and information processing method
CN105843748B (zh) 一种对内存中内存页的处理方法及装置
US11151052B2 (en) Reading sequential data from memory using a pivot table
US11762598B2 (en) Memory system and method of controlling nonvolatile memory
CN109508301A (zh) 终端、应用数据的处理方法、数据处理设备及存储介质
US9304946B2 (en) Hardware-base accelerator for managing copy-on-write of multi-level caches utilizing block copy-on-write differential update table
US12020065B2 (en) Hierarchical processor selection
US11520699B2 (en) Using a common pool of blocks for user data and a system data structure
US9552295B2 (en) Performance and energy efficiency while using large pages
CN110531928B (zh) DRAM-Less固态硬盘的数据随机写入方法、装置及DRAM-Less固态硬盘
US10073851B2 (en) Fast new file creation cache
CN102880553A (zh) 一种基于mcu的片外flash文件系统的读写方法
CN116149554B (zh) 一种基于risc-v及其扩展指令的数据存储处理系统及其方法
US9405470B2 (en) Data processing system and data processing method
CN112654965A (zh) 动态模块的外部分页和交换
US9146859B2 (en) Information processing apparatus, method, and program for improving use efficiency of a storage apparatus
US20120011330A1 (en) Memory management apparatus, memory management method, program therefor
WO2022021337A1 (zh) 闪存控制方法和装置
US20240302997A1 (en) Memory system and method of controlling memory chips
CN116719609A (zh) 一种JavaScript引擎的性能优化方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15883835

Country of ref document: EP

Kind code of ref document: A1

REEP Request for entry into the european phase

Ref document number: 2015883835

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE