42 results about "Mmap" patented technology

The embodiment of the invention provides a method and a device for processing a client log. The method comprises the following steps of adopting a double buffering mechanism, previously setting two MMAP (map files or devices into memory) internal memories to be respectively used as a current buffering area and a standby buffering area, switching the standby buffering area to be the current buffering area when the current buffering area is filled, and transmitting a log stored in the filled buffering area to a distributed file system to be stored, so that the problem that the log processing capacity is limited because the quantity of the processed logs is limited by the limit space of the existing disk can be solved. Due to the characteristics of the MMAP technology, in the operation process, the data in the buffering area can be forcedly outputted to the disk by the system, so that even if a server is restarted or collapsed, the data stored in the buffering area can be restored through the disk without being lost, and the safety of the log stored in the buffering area can be guaranteed.

A method and product for allocating memory in a computing environment, the method comprising providing a memory allocation routine adapted to use mmap( ) with a MAP SHARED or equivalent flag specified so that the memory is allocated from a file system buffer cache. In one embodiment, the method comprises allocating the memory by using mmap( ) with a MAP_SHARED or equivalent flag specified so that the memory is allocated from a file system buffer cache.

The invention provides a method for capturing real-time network data. The method comprises the following steps: S1, a kernel mode is initialized, and register information for receiving and sending message functions by a kernel network card is deleted in the process of initializing; S2, the kernel mode provides a large patch applied by internal storage, an Mmap memory map and an interface function ioctl of a user mode operation kernel, and the starting address of physical memory and the starting address of a network card configuration space are recorded; S3, a user mode automatically detects and acquires the bus number of the network card, the number of equipment and a function number, and is in communication with the kernel mode according to the interference function ioctl; corresponding resource allocation and corresponding network card register configuration are conducted on network card receiving and sending messages, and the user mode receives and sends the messages. According to the method for capturing the real-time network data, loading and driving in the kernel mode and message receiving and sending in the user mode of the network card are achieved, and network card driving interrupt in the kernel mode and performance damage caused by data copy are removed.

The invention discloses a method for enhancing performance of WEB servers on the basis of zero copy. The method is characterized by comprising the following steps of: loading a PF-RING module in a linux kernel; capturing a data package by a network card which is loaded with the PF-RING module by adoption of an equipment polling-based NAPI manner; storing the data package in a ring-shaped cache byadoption of a DMA manner; and directly transmitting the data package to an application layer by the PF-RING module through an MMAP mechanism. According to the method, intermediate copy processes, in internal storage of hosts, of communication data, so that the data package reception and transmission speeds are improved, the data communication efficiency is improved, corresponding speeds are improved and then the performance of web servers are improved.

The invention discloses a method and device for reinforcing a dynamic link library SO file of an Android installation package. The method includes the following steps: adding a shell to an original SO file of the Android installation package to obtain a reinforced SO file; preserving a loading space for the original SO file in initialization codes of the reinforced SO file, so that when the Android system loads the original SO file, the original SO file is loaded in a memory space distributed for the preserved loading space. According to the technical scheme, the method no longer calls a mmap function alone to distribute a memory space for the original SO file, but uses a part of memory space distributed to the reinforced SO file for loading of the original SO file when the reinforced SO file is loaded, therefore, when the system releases the memory space occupied by the reinforced SO file, the memory space occupied by the original SO file can be released together, normal use of the original SO file is not affected, and memory leaking is not caused.

A device securely accesses data in a memory via an addressing unit which provides a memory interface for interfacing to a memory, a core interface for interfacing to a core processor and a first and second security interface. The device includes a security processor HSM for performing at least one security operation on the data and a remapping unit MMAP. The remapping unit enables the security processor to be accessed by the core processor via the first security interface and to access the memory device via the second security interface according to a remapping structure for making accessible processed data based on memory data. The device provides a clear view on encrypted memory data without requiring system memory for storing the clear data.

The invention provides a system and a method for handling kernel mmap call failure. The method is used for creating a private memory device in a kernel mode by the aid of a driver, and all memory mappings and data sharing are operated on the private memory device, so that sharing of data in a user mode and in the kernel mode of each kernel version is realized. The system and the method for handling kernel mmap call failure have the advantages that firstly, compatibility is greatly improved as mmap can be independent from a kernel only by writing a small-scale driver, and compatibility of the kernel versions of various systems is realized; and secondly, controllability is greatly enhanced as user mode application programs and bottom equipment can be arbitrarily accessed and controlled only by writing one driver. If the scheme is not adopted, the kernel is recompiled or third-party software is adopted, both compatibility and controllability are greatly reduced.

The invention discloses a program debugging method. The program debugging method comprises the steps of mapping data files of a business process to a memory space of a debugging process in an mmap manner; and parsing and debugging the data files according to resource files and symbol table files of the business process by utilization of a BE binary editor, wherein the resource files and the symbol table files are generated during a compilation process of the business process in advance. By mapping the data files of the business process to the debugging process, the data files can be parsed to complete debug according to the resource files and the symbol table files without influencing execution efficiency of the business process. The invention also discloses a program debugging apparatus having the above beneficial effect.

The invention provides a multi-process data sharing method and a related device, wherein the method comprises the steps: enabling an application carried by each application container on the same hostmachine to map a cache file in an MMAP sharing mode, and achieving the sharing caching based on a memory. And each application on the same host machine performs data access by utilizing the shared cache based on the memory. On one hand, the process accesses the shared cache data conveniently and quickly like accessing a stack memory or a heap memory, so that the data exchange efficiency among theprocesses is improved; on the other hand, only one copy of the shared cache data exists in the physical memory of the host machine, so that the overhead of a system protocol stack is effectively reduced, and the overall performance of the system is improved; furthermore, by means of the cache loading master control assembly, after the shared cache data of all the host machines in the distributed computing system are successfully loaded, the shared cache data in all the host machines take effect in a unified mode, and the consistency of the cache data of all the computing nodes is guaranteed.

The invention discloses an access method and device for an input/output port. The access method includes the steps that a device file of a current operation system is analyzed, and whether an interface device file exists in the device file of the operation system or not is judged; if not, device interface process information is added in a system kernel of the operation system, so that the device file of the operation system is prompted to have the interface device file; through a preset functional instruction, the interface device file is called to have access to the input/output port. When the operation system lacks the interface device file, the device interface progress information is added, so that the operation system is prompted to have the interface device file, a user can call the interface device file to have the access to the port through the preset functional instruction, the trouble that a mmap function, mapping and other operation need to be called to have the access to the input/output port by user space is avoided, and technical requirements are reduced for the user.

The invention discloses a Linux operating system file read-write transparent encryption and decryption method, comprising the following steps: 1, finding the address of a sys_close function; 2, starting from the starting address of the memory, reading the data in the memory and matching; and if the content is the address of the system call sys_close, determining that the matching is successful, and carrying out the next step; If not, determining that the matching fails, and reading the data in the next memory and performing matching; 3, after the matching is successful, calling the open, read,write, mmap and msync functions on the hook system to encrypt and decrypt the data; 4, achieving the function of file encryption protection. The method has the advantages of simple process and wide platform compatibility, and is not only suitable for Mips processors, but also suitable for Arm processors. The search method is simple; the code is simple and understandable.

The invention discloses a method and a device for improving the calling efficiency of a system command line interface, includes requesting a first memory in a kernel space of an operating system, thefirst memory is then mapped to a user space via MMAP, A first system command line interface for accessing the first memory is provided in the form of a file, so that a first application program on user space accesses the first memory on the kernel space by using the first system command line interface to realize interaction with the kernel space. As seen, that application program directly access the first memory in the kernel space using the first system command line interface, The first memory is used for data exchange between the application program and the kernel space, so that the data transmission performance between the application program and the kernel space is improved, and the invocation efficiency of the system command line interface is further improved.

The invention discloses a simulator for simulating an LCD (liquid crystal display) in a mobile terminal and an implementing method thereof. The simulator comprises a simulation drive and a simulation UI (user interface) for LCD simulation; the simulation drive is to be provided for an application program and the simulation UI; the application program and the simulation UI gain direct access to a device memory by an mmap method to read or write the simulation drive; the application program writes an action command into the simulation drive through an interface; the simulation UI reads the action command from the simulation drive through the interface and simulates the action command, and the function of LCD simulation is achieved. The simulator and the implementing method have the advantages that the application program can be developed independently of hardware, regardless of a difference of running on the simulator and the true hardware, late modification is not required, and development efficiency of software is greatly improved.

Embodiments of the invention provide a multi-machine backup method and system, so that the problem that a business process of a backup machine is not consistent with a business process of an original machine after being recovered can be avoided. The method comprises the steps that: the original machine calls a memory map creation function mmap() of the business process to map data of the business process into a memory, and marks that a mapping region of mapping data in the memory is shared; the original machine calls a refresh change function msync() to write the data of the business process in the mapping region into a backup file, wherein the backup file is positioned in a disk; and the original machine sends the data in the backup file to the backup machine. The multi-machine backup method and system are suitable for multi-machine backup of a business system.