CN113326213B - Method for realizing address mapping in driver under Feiteng server platform - Google Patents
Method for realizing address mapping in driver under Feiteng server platform Download PDFInfo
- Publication number
- CN113326213B CN113326213B CN202110563969.XA CN202110563969A CN113326213B CN 113326213 B CN113326213 B CN 113326213B CN 202110563969 A CN202110563969 A CN 202110563969A CN 113326213 B CN113326213 B CN 113326213B
- Authority
- CN
- China
- Prior art keywords
- operating system
- linux operating
- address
- kernel
- mapping
- 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.)
- Active
Links
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
- G06F12/06—Addressing a physical block of locations, e.g. base addressing, module addressing, memory dedication
- G06F12/0615—Address space extension
- G06F12/063—Address space extension for I/O modules, e.g. memory mapped I/O
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Memory System Of A Hierarchy Structure (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a method for realizing address mapping in a driver under a Feiteng server platform, belonging to the field of computers. The invention realizes the one-to-one mapping function of the applied virtual address and the appointed physical address in the kernel function module of the Linux operating system, and the self-defined address mapping module realized in the kernel of the Linux operating system additionally provides a set of address mapping interfaces for the kernel function module of the Linux operating system, and realizes the one-to-one mapping function of the virtual address and the physical address by configuring page table items used by the kernel of the Linux operating system. The invention does not influence the address mapping function of the memory management subsystem in the kernel of the Linux operating system, improves the development freedom of engineering driven by specific requirements, can facilitate the modification of the functional module of the Linux operating system, and does not reduce the robustness and portability of the functional module of the Linux operating system.
Description
Technical Field
The invention belongs to the field of computers, and particularly relates to a method for realizing address mapping in a driver under a Feiteng server platform.
Background
The CPU accesses the units in the memory through addresses, the addresses have a division of virtual addresses and physical addresses, if the CPU does not have an MMU (Memory Management Unit ) or has an MMU but is not enabled, the addresses sent by the CPU core when fetching instructions or accessing the memory are directly transmitted to the external address pins of the CPU chip and are directly received by the memory chip, namely the physical addresses. If the CPU has an MMU enabled, the address issued by the CPU core will be intercepted by the MMU, the address from the CPU to the MMU being referred to as the virtual address, and the MMU translating this address into another address to be sent to the external address pins of the CPU chip, i.e. mapping the virtual address to a physical address.
The processors of the current servers are basically mainly 64-bit address lengths, and the Linux operating system on the processors can support 64-bit address space at maximum. But there is far less address space supported on the actual server to the maximum supportable 64-bit length of the Linux operating system because the current application does not have as much memory requirements. Furthermore, the Linux operating system is a multi-process type operating system, each process has a respective user address space, and all processes share the same kernel address space. Each process needs to dynamically map the addresses of the virtual address pages of the user address space used by the process to physical memory pages. This part of the mapping of virtual address pages to physical address memory pages is done by the memory management subsystem of the Linux operating system kernel. The memory management subsystem of the Linux operating system kernel provides a function interface of the subsystem so that other functional modules of the Linux operating system kernel can be called. The block device access mechanism used by the server platform of the FT2000+ home processor is described above.
The memory management subsystem of the Linux operating system completes the management work of all virtual addresses and physical addresses. In the actual engineering project, if the specified physical address and the applied virtual address need to be mapped one by one in the functional module of the kernel of the Linux operating system. For example, in a driving module of a Linux operating system kernel, the implementation driving module uses a specified physical memory to implement a large amount of data exchange in a shared data area with other driving modules. Then it is necessary to implement some functions of the Linux operating system memory management subsystem in the functional module of the Linux operating system kernel. However, the function module of the Linux operating system kernel cannot acquire the base address of the page table, and the applied virtual addresses do not have physical addresses to perform one-to-one correspondence, so that the function module of the Linux operating system kernel cannot use the applied virtual address space. Otherwise, the kernel panic of the Linux operating system is generated to cause the system to crash. Therefore, the function of one-to-one mapping between the designated physical address and the applied virtual address is realized in the functional module of the kernel of the Linux operating system, the modification amount is huge, and the robustness of the modified kernel of the Linux operating system is poor. And the universality and portability of the functional module of the kernel of the Linux operating system are reduced.
Disclosure of Invention
First, the technical problem to be solved
The invention aims to solve the technical problem of providing a method for realizing address mapping in a driver under a Feiteng server platform so as to solve the problems of large change amount and poor robustness of realizing one-to-one mapping function of a designated physical address and an applied virtual address in a functional module of a kernel of a Linux operating system.
(II) technical scheme
In order to solve the technical problems, the invention provides a method for realizing address mapping in a driver under a Feiteng server platform, which comprises the following steps:
s1, powering on a kernel of a Linux operating system, and entering a memory initialization stage; reserving a physical address space to be used, namely, the designated physical address space is not registered in a memory management subsystem of the Linux operating system; if successful, entering step S2, otherwise, alarming to the kernel of the Linux operating system, and then using a default functional module of the kernel of the Linux operating system;
s2, registering the reserved physical address space into a resource linked list of a Linux operating system kernel;
s3, realizing a self-defined address mapping module, wherein the self-defined address mapping module has the function of providing a calling interface for an external module by mapping a virtual address of an incoming interface with a physical address registered in a Linux operating system kernel in the step; the self-defined address mapping module can finish mapping the virtual address to the appointed physical address, and then step S4 is carried out, otherwise, the self-defined address mapping module alarms to the kernel of the Linux operating system, and then the function module defaulted by the kernel of the Linux operating system is used;
s4, entering an initialization stage of a kernel function module of the Linux operating system, wherein the kernel function module of the Linux operating system applies for a virtual memory address space meeting the requirement from a memory management subsystem of a kernel of the Linux operating system;
s5, the kernel function module of the Linux operating system acquires a reserved physical address space from a resource linked list registered in the Linux operating system kernel in the step S2;
s6, calling an interface provided by the self-defined address mapping module and used for being called by an external module, and realizing an interface of an address mapping function in a kernel function module of the Linux operating system;
s7, in the kernel function module of the Linux operating system, mapping between the applied virtual address space and the reserved physical address space is realized by using an interface of an address mapping function.
Further, the Feiteng server platform is a 64-bit machine, and the bit width of the virtual address and the physical address is 64 bits.
Further, the virtual address effective bit width of the operating system operated by the Feiteng server platform is 48 bits.
Further, the size of the reserved physical address space in step S1 is defined according to the requirement.
Further, the step S3 specifically includes the following steps:
s31, a self-defined address mapping module is realized, the head address of the virtual address space and the head address of the physical address space, which are received by the self-defined address mapping module through a calling interface, are transmitted to a standard address mapping interface function provided in a kernel of the Linux operating system, and the size of the virtual address space to be mapped is also transmitted to the standard address mapping interface function provided by the kernel of the Linux operating system, so that page table items of the Linux operating system are configured according to the virtual address and the physical address;
s32, configuring the page table item attribute according to the interface of the page table item attribute of the standard provided by the Linux operating system kernel.
Further, the attributes of the page table entry include no_block_map, which is used to flag the MAPPING of the limit block_map macro page.
Further, the attributes of the page table entries include no_cont_mappings, which are used to flag the constraint map to consecutive physical pages.
Further, the virtual memory address space applied in step S4 is 1GB.
Further, in the step S5, when the reserved physical address space is obtained, the Linux operating system kernel function module also obtains the physical address head address and the physical address size of the reserved physical address space.
Further, in the step S7, when the mapping between the applied virtual address space and the reserved physical address space is implemented by using the interface of the address mapping function, a page table is further configured.
(III) beneficial effects
The invention provides a method for realizing address mapping in a driver under a Feiteng server platform. The self-defined address mapping module realized in the kernel of the Linux operating system additionally provides a set of address mapping interfaces for the kernel functional module of the Linux operating system, and the one-to-one mapping function of the virtual address and the physical address is realized by configuring page table items used by the kernel of the Linux operating system. The mapping between the virtual address and the designated physical address is very complicated in the kernel function module in the Linux operating system kernel, and the stability of the Linux operating system kernel is reduced. The invention realizes a self-defined address mapping module which is specially used in a functional module of the kernel of the Linux operating system in a standard interface for mapping virtual addresses and physical addresses in the kernel of the Linux operating system, and encapsulates the self-defined address mapping module into an interface for other functional modules to use. The invention does not affect the address mapping function of the memory management subsystem in the kernel of the Linux operating system. The development freedom of engineering driven by specific requirements is improved.
Compared with the prior art, the technical method provides a set of functional interfaces for realizing one-to-one correspondence between the appointed virtual address and the physical address for the kernel module of the Linux operating system. When a certain physical address is needed to be used in the kernel of the Linux operating system, the functional module of the Linux operating system can be modified conveniently, and the robustness and portability of the functional module of the Linux operating system are not reduced. And the method can be compatible with the running of the existing application program on the system, and does not influence the existing user space program.
Drawings
FIG. 1 is a flow chart of a method of address mapping according to the present invention.
Detailed Description
To make the objects, contents and advantages of the present invention more apparent, the following detailed description of the present invention will be given with reference to the accompanying drawings and examples.
And an operating system running on the FT2000+ new generation domestic multi-core processor has an effective bit width of 48 bits. And the virtual address space available in the operating system for allocation to block devices is 512TB. The invention mainly realizes the one-to-one mapping function of the appointed virtual address and the appointed physical address in the functional module of the kernel of the Linux operating system. The function module of the Linux operating system kernel can use the designated physical address to perform functions such as data storage, transmission and control. And the robustness of the kernel of the Linux operating system is not reduced after modification, and the universality and portability of the functional module of the kernel of the Linux operating system are not reduced. The user can conveniently develop the kernel of the Linux operating system for the second time.
On a 64-bit machine, the bit width of both the virtual address and the physical address is 64 bits. The 48 bits are valid bit wide, i.e., 0xFFFF xxxx xxxx xxxx, and the upper 16 bits are 1, i.e., four F of the upper address.
The invention realizes that a virtual address space is applied for by a memory management subsystem of the Linux operating system in a functional module of the kernel of the Linux operating system and is mapped with a designated physical address space one by one. Thus, the function module of the kernel of the Linux operating system reads and writes the applied virtual address space, namely the appointed physical address space. The specific implementation steps of the present invention are shown with reference to fig. 1. The specific steps are as follows.
S1, powering on and starting a kernel of a Linux operating system, and entering a memory initialization stage. The physical address space needed to be used is reserved, namely the designated physical address space is not registered in the memory management subsystem of the Linux operating system. The size of the reserved physical address space is defined by itself as needed. If successful, the step S2 is entered, otherwise, the alarm is given to the kernel of the Linux operating system, and then the default function module of the kernel of the Linux operating system is used.
S2, registering the reserved physical address space into a resource linked list of a Linux operating system kernel;
s3, realizing a self-defined address mapping module, wherein the self-defined address mapping module has the function of providing a calling interface for an external module by mapping the virtual address of the incoming interface and the physical address registered in the Linux operating system kernel in the step. The specific implementation steps are as follows:
s31, transmitting the first address of the virtual address space and the first address of the physical address space received by the custom address mapping module through the calling interface to a standard address mapping interface function provided in a Linux operating system kernel, and transmitting the size of the virtual address space to be mapped into the standard address mapping interface function provided by the Linux operating system kernel, so that page table items of the Linux operating system are configured according to the virtual address and the physical address.
S32, configuring the page table item attribute according to the interface of the page table item attribute of the standard provided by the Linux operating system kernel. For example, attribute no_block_map_map_is used to mark the map of the limit block_map_macro page, and attribute no_cont_map_is used to mark the physical pages that are consecutive to the limit map. The present invention uses a page table entry configured by the attribute no_block_mappings.
And (4) the user-defined address mapping module can finish mapping the virtual address to the appointed physical address, and the step (S4) is entered, otherwise, the alarm is given to the kernel of the Linux operating system, and then the function module defaulted by the kernel of the Linux operating system is used.
S4, entering an initialization stage of a kernel function module of the Linux operating system, and applying a virtual memory address space meeting the requirement to a memory management subsystem of the kernel of the Linux operating system by the kernel function module of the Linux operating system.
S5, the kernel function module of the Linux operating system acquires a reserved physical address space from a resource linked list registered in the Linux operating system kernel in the step S2. The kernel function module of the Linux operating system acquires the physical address head address and the physical address size of the reserved physical address space.
S6, calling an interface provided by the self-defined address mapping module and used for being called by an external module, and realizing an interface of an address mapping function in a kernel function module of the Linux operating system. That is, mapping a block of virtual address space applied by the functional module, for example, 1GB, into the physical address space reserved in the step 1 is realized in the kernel of the Linux operating system.
S7, in the kernel function module of the Linux operating system, mapping the applied virtual address space and the reserved physical address space by using an interface of an address mapping function, and configuring a page table.
The invention realizes the function of one-to-one mapping between the applied virtual address and the appointed physical address in the kernel function module of the Linux operating system. The self-defined address mapping module realized in the kernel of the Linux operating system additionally provides a set of address mapping interfaces for the kernel functional module of the Linux operating system, and the one-to-one mapping function of the virtual address and the physical address is realized by configuring page table items used by the kernel of the Linux operating system. The mapping between the virtual address and the designated physical address is very complicated in the kernel function module in the Linux operating system kernel, and the stability of the Linux operating system kernel is reduced. The invention realizes a self-defined address mapping module which is specially used in a functional module of the kernel of the Linux operating system in a standard interface for mapping virtual addresses and physical addresses in the kernel of the Linux operating system, and encapsulates the self-defined address mapping module into an interface for other functional modules to use. The invention does not affect the address mapping function of the memory management subsystem in the kernel of the Linux operating system. The development freedom of engineering driven by specific requirements is improved.
Compared with the prior art, the technical method provides a set of functional interfaces for realizing one-to-one correspondence between the appointed virtual address and the physical address for the kernel module of the Linux operating system. When a certain physical address is needed to be used in the kernel of the Linux operating system, the functional module of the Linux operating system can be modified conveniently, and the robustness and portability of the functional module of the Linux operating system are not reduced. And the method can be compatible with the running of the existing application program on the system, and does not influence the existing user space program.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (10)
1. A method for realizing address mapping in a driver under a Feiteng server platform is characterized by comprising the following steps:
s1, powering on a kernel of a Linux operating system, and entering a memory initialization stage; reserving a physical address space to be used, namely, the designated physical address space is not registered in a memory management subsystem of the Linux operating system; if successful, entering step S2, otherwise, alarming to the kernel of the Linux operating system, and then using a default functional module of the kernel of the Linux operating system;
s2, registering the reserved physical address space into a resource linked list of a Linux operating system kernel;
s3, realizing a self-defined address mapping module, wherein the self-defined address mapping module has the function of providing a calling interface for an external module by mapping a virtual address of an incoming interface with a physical address registered in a Linux operating system kernel in the step; the self-defined address mapping module can finish mapping the virtual address to the appointed physical address, and then step S4 is carried out, otherwise, the self-defined address mapping module alarms to the kernel of the Linux operating system, and then the function module defaulted by the kernel of the Linux operating system is used;
s4, entering an initialization stage of a kernel function module of the Linux operating system, wherein the kernel function module of the Linux operating system applies for a virtual memory address space meeting the requirement from a memory management subsystem of a kernel of the Linux operating system;
s5, the kernel function module of the Linux operating system acquires a reserved physical address space from a resource linked list registered in the Linux operating system kernel in the step S2;
s6, calling an interface provided by the self-defined address mapping module and used for being called by an external module, and realizing an interface of an address mapping function in a kernel function module of the Linux operating system;
s7, in the kernel function module of the Linux operating system, mapping between the applied virtual address space and the reserved physical address space is realized by using an interface of an address mapping function.
2. A method for implementing address mapping in a driver under a Feiteng server platform as in claim 1, wherein the Feiteng server platform is a 64-bit machine and the bit width of both the virtual address and the physical address is 64 bits.
3. A method for implementing address mapping in a driver under a Feiteng server platform as in claim 2, wherein the virtual address valid bit width of an operating system running on the Feiteng server platform is 48 bits.
4. A method for implementing address mapping in a driver under a Feiteng server platform according to claim 2, wherein the size of the physical address space reserved in the step S1 is defined according to the requirement.
5. A method for implementing address mapping in a driver under a Feiteng server platform according to any of claims 1-4, wherein the step S3 specifically includes the steps of:
s31, a self-defined address mapping module is realized, the head address of the virtual address space and the head address of the physical address space, which are received by the self-defined address mapping module through a calling interface, are transmitted to a standard address mapping interface function provided in a kernel of the Linux operating system, and the size of the virtual address space to be mapped is also transmitted to the standard address mapping interface function provided by the kernel of the Linux operating system, so that page table items of the Linux operating system are configured according to the virtual address and the physical address;
s32, configuring the page table item attribute according to the interface of the page table item attribute of the standard provided by the Linux operating system kernel.
6. A method for implementing address MAPPING in a driver under a Feiteng server platform as in claim 5 wherein the attributes of the page table entries include no_block_MAPPINGs for marking the MAPPING of the limit block_MAPPING megapages.
7. A method for implementing address mapping in a driver under a Feiteng server platform as in claim 5, wherein the attributes of the page table entries include NO_CONT_MAPPINGS for marking restriction mapping consecutive physical pages.
8. A method for implementing address mapping in a driver under a Feiteng server platform according to claim 1 or 5, wherein the virtual memory address space applied in the step S4 is 1GB.
9. A method for implementing address mapping in a driver under a Feiteng server platform according to claim 1 or 5, wherein in the step S5, when the reserved physical address space is obtained, the kernel function module of the Linux operating system further obtains the physical address head address and the physical address size of the reserved physical address space.
10. A method for implementing address mapping in a driver under a Feiteng server platform according to claim 1 or 5, wherein the step S7 further configures a page table when mapping the applied virtual address space and the reserved physical address space implemented by using the interface of the address mapping function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110563969.XA CN113326213B (en) | 2021-05-24 | 2021-05-24 | Method for realizing address mapping in driver under Feiteng server platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110563969.XA CN113326213B (en) | 2021-05-24 | 2021-05-24 | Method for realizing address mapping in driver under Feiteng server platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113326213A CN113326213A (en) | 2021-08-31 |
| CN113326213B true CN113326213B (en) | 2023-07-28 |
Family
ID=77416421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110563969.XA Active CN113326213B (en) | 2021-05-24 | 2021-05-24 | Method for realizing address mapping in driver under Feiteng server platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113326213B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102929747A (en) * | 2012-11-05 | 2013-02-13 | 中标软件有限公司 | Method for treating crash dump of Linux operation system based on loongson server |
| CN103678175A (en) * | 2013-12-05 | 2014-03-26 | 长城信息产业股份有限公司 | Address concealing method and embedded equipment based on address concealing technology |
| US8745442B1 (en) * | 2011-04-28 | 2014-06-03 | Open Invention Network, Llc | System and method for hybrid kernel- and user-space checkpointing |
| CN103984655A (en) * | 2014-04-29 | 2014-08-13 | 三星半导体(中国)研究开发有限公司 | Processing method and processing device for data access |
| CN107368384A (en) * | 2017-07-21 | 2017-11-21 | 郑州云海信息技术有限公司 | A kind of Linux server abnormal information dump system and method |
| CN108062253A (en) * | 2017-12-11 | 2018-05-22 | 北京奇虎科技有限公司 | The communication means of a kind of kernel state and User space, device and terminal |
| CN108334413A (en) * | 2017-12-22 | 2018-07-27 | 天津麒麟信息技术有限公司 | Information transmitting methods between a kind of firmware and operating system based on platform of soaring |
| CN109710544A (en) * | 2017-10-26 | 2019-05-03 | 杭州华为数字技术有限公司 | Memory pool access method, computer system and processing unit |
| CN111444119A (en) * | 2020-03-25 | 2020-07-24 | 北京计算机技术及应用研究所 | Feiteng platform nonvolatile memory registration method based on kernel parameters |
| CN112148229A (en) * | 2020-10-12 | 2020-12-29 | 北京计算机技术及应用研究所 | Method for accelerating nonvolatile memory read-write efficiency under Feiteng server platform |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017023773A1 (en) * | 2015-07-31 | 2017-02-09 | Digital Guardian, Inc. | Systems and methods of protecting data from injected malware |
-
2021
- 2021-05-24 CN CN202110563969.XA patent/CN113326213B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8745442B1 (en) * | 2011-04-28 | 2014-06-03 | Open Invention Network, Llc | System and method for hybrid kernel- and user-space checkpointing |
| CN102929747A (en) * | 2012-11-05 | 2013-02-13 | 中标软件有限公司 | Method for treating crash dump of Linux operation system based on loongson server |
| CN103678175A (en) * | 2013-12-05 | 2014-03-26 | 长城信息产业股份有限公司 | Address concealing method and embedded equipment based on address concealing technology |
| CN103984655A (en) * | 2014-04-29 | 2014-08-13 | 三星半导体(中国)研究开发有限公司 | Processing method and processing device for data access |
| CN107368384A (en) * | 2017-07-21 | 2017-11-21 | 郑州云海信息技术有限公司 | A kind of Linux server abnormal information dump system and method |
| CN109710544A (en) * | 2017-10-26 | 2019-05-03 | 杭州华为数字技术有限公司 | Memory pool access method, computer system and processing unit |
| CN108062253A (en) * | 2017-12-11 | 2018-05-22 | 北京奇虎科技有限公司 | The communication means of a kind of kernel state and User space, device and terminal |
| CN108334413A (en) * | 2017-12-22 | 2018-07-27 | 天津麒麟信息技术有限公司 | Information transmitting methods between a kind of firmware and operating system based on platform of soaring |
| CN111444119A (en) * | 2020-03-25 | 2020-07-24 | 北京计算机技术及应用研究所 | Feiteng platform nonvolatile memory registration method based on kernel parameters |
| CN112148229A (en) * | 2020-10-12 | 2020-12-29 | 北京计算机技术及应用研究所 | Method for accelerating nonvolatile memory read-write efficiency under Feiteng server platform |
Non-Patent Citations (1)
| Title |
|---|
| 基于飞腾平台的GPU图形加速技术研究及应用;李荣振;《中国优秀硕士学位论文全文数据库 (信息科技辑)》;I138-5735 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113326213A (en) | 2021-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11567803B2 (en) | Inter-server memory pooling | |
| US9185034B2 (en) | Low latency communication via memory windows | |
| US8484307B2 (en) | Host fabric interface (HFI) to perform global shared memory (GSM) operations | |
| CN113868155B (en) | Memory space expansion method and device, electronic equipment and storage medium | |
| US9584628B2 (en) | Zero-copy data transmission system | |
| US8868883B1 (en) | Virtual memory management for real-time embedded devices | |
| TW201423422A (en) | System and method for sharing device having PCIe interface | |
| US11656779B2 (en) | Computing system and method for sharing device memories of different computing devices | |
| KR20210001886A (en) | Data accessing method and apparatus, device and medium | |
| EP4407470A1 (en) | Request processing method, apparatus and system | |
| CN104794069A (en) | User state allocation method and system for cache in CPU | |
| CN115309511B (en) | Xen-based data interaction method and device, storage medium and electronic equipment | |
| EP4293523A1 (en) | Technologies for memory tagging | |
| CN113326213B (en) | Method for realizing address mapping in driver under Feiteng server platform | |
| CN111290829A (en) | Access control module, virtual machine monitor and access control method | |
| WO2021238594A1 (en) | Storage medium management method and apparatus, and device and computer-readable storage medium | |
| CN111666579B (en) | Computer device, access control method thereof and computer readable medium | |
| US11334245B1 (en) | Native memory semantic remote memory access system | |
| US20240176640A1 (en) | Method and apparatus for managing memory | |
| CN115858422A (en) | Page table processing method, electronic device and storage medium | |
| CN117591248B (en) | Terminal system processing method based on containerized virtual machine and electronic equipment | |
| CN117857253A (en) | Inter-domain communication method based on jailhouse smmu | |
| CN115576971A (en) | Data query method, device, equipment and storage medium based on federal learning | |
| JP2022141463A (en) | Information processing device, information processing method, and program | |
| CN116915770A (en) | Service deployment method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |