199 results about "Doubly linked list" patented technology

A concurrent FIFO queue is implemented as an “optimistic” doubly-linked list. Nodes of the optimistic doubly-linked list are allocated dynamically and links between the nodes are updated optimistically, i.e., assuming that threads concurrently accessing the FIFO queue will not interfere with each other, using a simple store operation. Concurrent, linearizable, and non-blocking enqueue and dequeue operations on the two ends of the doubly-linked list proceed independently, i.e., disjointly. These operations require only one successful single-word synchronization operation on the tail pointer and the head pointer of the doubly-linked list. If a bad ordering of operations on the optimistic FIFO queue by concurrently executing threads creates inconsistencies in the links between the nodes of the doubly-linked list, a fix-up process is invoked to correct the inconsistencies.

The invention relates to a management method for a computer Nand-Flash storage system. A plurality of physical blocks form a logic block chain; the information of the block numbers of the physical blocks and the block number of a forward physical block thereof are stored in an extended area of a first physical page of each physical block; when the system is initialized, a bidirectional chain table of the logic block chain is reconstructed by the stored information. When in read operation, the bidirectional chain table is utilized to research forwards and read a target physical block; the number of a read physical page is the same as the number of a logic page. When in write operation, the bidirectional chain table is utilized to research backwards; when a first physical block with the number of used pages on the logic block chain smaller than the number of a logic page requested to write, data is directly written in the logic page requested to write by the physical block; if the numbers of used pages of all physical blocks are larger than the number of the number of the logic page requested to write, a free physical block is added into the logic block chain; the data of effective pages is copied to the new current free physical block, and the data of the current page is written.

One embodiment of the present invention provides a system that supports inserting or deleting nodes at any location within a doubly-linked list which is lock-free, wherein lock-free means that the doubly-linked list can be simultaneously accessed by multiple processes without requiring the processes to perform locking operations (non-blocking) and furthermore that a finite number of steps performed by a process will guarantee progress by at least one process (lock-free). During operation, the system receives a reference to a target node to be deleted from the doubly-linked list. Next, the system atomically marks a forward pointer in the target node to indicate that the target node is deleted, wherein the forward pointer contains the address of an immediately following node in the doubly-linked list, and wherein the marking operation does not destroy the address of the immediately following node. Additional cleanup steps are then done by this or any other process. The system may also receive a new node which is accessible by only the requesting thread and may then insert the new node into the doubly linked list after a reference node. The system accomplishes this by setting the new node's backward pointer to the reference node and forward pointer to the successor of the reference node. Next, the system atomically changes the forward pointer of the reference node from the successor node to the new node. Additional cleanup steps are then done by this or any other process. An update operation that atomically performs a delete of an old node and an insert of its replacement node is also described.

The invention discloses an optimization method for creating an FAT file system on an NAND Flash storage. A plurality of physical blocks which are mapped to the same logical block in the NAND Flash storage file system are represented by adopting a doubly linked list mode; wherein, the head node of the doubly linked list represents one logical block; the node of the physical block represents one physical block; a plurality of physical blocks are mapped to the same logical block; the physical block nodes corresponding to the physical blocks are linked to the nodes of the logical block so as to form the doubly linked list; a plurality of nodes of the logical block are also linked in the doubly linked list mode so as to form a two-level doubly linked list; and a data structure brokering object with the doubly linked list is created in a memory. As the plurality of physical blocks which are mapped to the same logical block in the NAND Flash storage file system are represented by adopting the doubly linked list mode, only a plurality of pages needing to be written need to be written for each write request and not all the pages of the whole block need to be written, thus improving the speed of file writing.

The invention discloses a method and a device for adding, searching and deleting a hash table. The method for adding a hash table comprises the following steps of: extracting keywords of data to be added; performing hash operation on the keywords to obtain a hash address and a hash fingerprint; judging whether a corresponding target storage position is occupied according to the hash address; when the target storage position is occupied, searching for an unoccupied idle storage position from the hash table, and judging whether an occupied node at the target storage position is a first node of a bidirectional linked list; if so, adding a target node at the idle storage position according to the hash fingerprint, and linking the target node to the tail of the bidirectional linked list where the occupied node is positioned; and otherwise, moving the occupied node to the idle storage position, linking the occupied node to the tail of the bidirectional linked list where the occupied node is positioned, and adding a target node at the target storage position to constitute a single-node bidirectional linked list according to the hash fingerprint,. According to the method and the device, the utilization efficiency of the hash table and the utilization ratio of a storage space can be increased.

The invention relates to a management method of shared memory and a device, wherein the method comprises a memory double linked list is built, the memory double linked list comprises node information, data information and edge boundary marker information which correspond to each memory block, the node information is located in a node area, the data information is located in a data area, and the edge boundary marker information is located in a marker area. The node information at least comprises the size of the data area, a state marker, an address of a forward memory clock and an address of a backward memory block, wherein the state marker is used to express whether the data area has the data information, corresponding node information is read, and corresponding memory blocks are deleted according to the node information. The invention overcomes the management defect in the prior art that the shared memory can not be effectively managed after building a memory linked list, provides an effective protection mechanism of the shared memory, which leads an application program to effectively manage the space of the shared memory.

An editor or software engineering tool may be configured to represent program code as a doubly-linked list of lexical tokens and to maintain, coincident with an operation that modifies contents of the list, an undo object that identifies opposing end nodes of a sublist of one or more lexical tokens corresponding to a substring that is either inserted into or removed from the list by the operation. In this way, lexical tokens corresponding to an inserted substring can be readily and efficiently excised to restore a pre-insertion tokenized list state. Similarly, lexical tokens corresponding to a removed substring can be readily and efficiently reinstated to restore a pre-deletion tokenized list state. Advantageously, undo support once employed to restore a prior tokenized list state is symmetrically available to support redo operations. In some embodiments in accordance with the present invention, undo-redo entries are maintained in an operation ordered set that is traversed to support one or more operations in either the undo or redo directions. In some realizations, such an ordered set of undo-redo entries is maintained by, or in conjunction with, an undo-redo manager.

The invention relates to a service ordering method and a service ordering platform. The service ordering method comprises the steps of transmitting preset service request information by a service requiring party to a service ordering platform; after receiving preset service request information by a service ordering platform, transmitting preset service request information to a service provider; querying a preset task by the service provider, and filtering the service requiring party and filtering the preset task of the service requiring party; paying advance payment of a preset task by the service requiring party to the service provider in preset time period, confirming successful order establishment and establishing a bidirectional chain table through the service ordering platform; reading information in the bidirectional chain table by the service provider, establishing an identity verification system; keeping the service requiring party and the service provider operate in an allowed or preset time period until time is up, and after transaction, paying cost by the service requiring party to the service provider through the service ordering platform. The service ordering platform comprises a terminal logging-on module, a service searching module, a service provider filtering module, a payment module, an identity verification module, an evaluation module and a platform processing module.

The invention discloses a NAT session management method based on a distributed system. The NAT session management method based on a distributed system comprises the following steps: (1) creating a NAT session; (2) maintaining updates to the created NAT session; (3) performing an aging operation on the updated NAT session; (4) querying the NAT session performed the aging operation. According to the NAT session management method based on a distributed system, the hash bucket, the hash linked list and other data structures are used to realize the creation and the update maintenance management of the NAT session, and the session key values can be hashed evenly into each hash bucket by setting the hash bucket parameters to proper values, therefore, the session key search speed can be effectively improved; meanwhile, the sessions are linked by means of the separate double linked list to complete the aging and query management of the created sessions, therefore, frequent locking of the hash buckets is avoided, and the aging and query efficiency of the sessions is improved.

The invention relates to a system for picking up IPv6 data at high speed and quickly storing, and a relative method. Wherein, it comprises an initialization module; an IPv6 data picking up module for real-time picking up the IPv4/v6 dataflow through detecting interface, while the picked data can be used in following protocol analysis; IPv6 protocol analyze module for processing the picked IPc4/v6 data to match the picked data from outer to inner with variable protocols, to analyze the protocols of all layers, to be stored in memory unit set for different protocols. The invention uses bidirectional linkage list technique to quickly write and discompose the picked IPv6/v4 data pack in memory queue. And it uses multi-process technique to pick and record high-speed data at IPv6/v4 network, to be analyzed and stored.

The method comprises: applying a memory area having a fixed length as a buffer area; receiving packet and extracting the relevant information of the packet; ordering the packet in the buffer area according to relevant information; if passing through a minimal time delay value, then checking the packets in the buffer area to decide if it confirms with the referring condition; if yes, then referring the packets in the buffer area to the application program; if not, then continuing to receive the packets, until the packets in buffer area being accordance with the refereeing condition, the packet in the buffer area just be referred out; if passing through a maximal time delay value, then packets in the buffer area are forced to be referred out.

The invention relates to a periodic wear balancing method and a memory management method of a phase change memory. The method comprises the steps of: S1, maintaining a counter for overall writing times for a physical memory of the phase change memory; S2, increasing used-list and threshold-list doubly linked lists to each sublist, and simultaneously increasing an overall bad-list to maintain the memory region which can not be allocated; S3, when the data is written to a physical memory page, synchronously updating the counter for the overall writing times and a counter for accumulative writing times of a memory block during current allocation; S4, setting a writing threshold, after the writing times of the memory block during the current allocution exceeds the writing threshold, releasing the memory block and placing the memory block in the threshold-list of the corresponding sublist, then applying to allocate a memory block with identical capacity, and copying the data content from an old memory block to the new allocated memory block; and S5, when the overall writing times of the physical memory page exceeds the life limit of a phase change storage unit, and placing the memory page in the bad-list so as not to allocate for ever.

The invention discloses a method for distributing and reclaiming idle blocks of a file. The distributing method comprises the following steps: dividing the file into N file blocks according to the preset size, and numbering the file blocks; using K file blocks in the N file blocks as index blocks, and connecting the index blocks by adopting a bidirectional link table; storing a previous index block sequence number of the index block, a rear index block sequence number and the number of the idle blocks managed by the index block into the index blocks; and distributing the idle blocks managed by the index block in turn according to the order of the index blocks in the bidirectional link table, and correspondingly modifying information stored in the index blocks. The reclaiming method comprises the following steps: reclaiming the idle blocks in turn by the index blocks according to the order of the index blocks in the bidirectional link table, and correspondingly modifying the information stored in the index blocks. The method is particularly suitable for application situations with high real-time requirements. The method can save a large amount of storage space in the storage of mass historical data of the real-time database, and meanwhile can ensure the real-time response of the real-time database to the data due to extremely high idle block distribution speed.

A directory structure for a sparsely filled data container comprises a linked list, doubly linked list, skip list, or other non-fully populated list technique. One or more hierarchical levels of such lists may be used for sparsely filled directories. The directory structure may be converted to a conventional look-up table directory by reconstructing the directory when the directory becomes populated to certain point.

The embodiment of the invention provides a memory leak detecting method and a memory leak detecting device. A memory comprises a doubly linked list of an allocated memory; and the doubly linked list comprises memory header information of one or more allocated memory blocks. The method comprises the steps of: determining a detection time interval; traversing the header information in the detection time interval in the doubly linked list; and in accordance with the memory header information, determining whether the corresponding one or more allocated memory blocks are subject to memory leak. By the memory leak detecting method and the memory leak detecting device, memory leak can be visually detected in real time, the software performance is improved, and manual operation is reduced or avoided, so that the method and the device are labor-saving and time-saving, the operation difficulty is reduced, the working efficiency is improved, the cost is reduced, the accuracy of the detection of memory leak is improved, and the leaked memory can be recycled in the entity life cycle.

The invention relates to a real-time event management method for intelligent electronic equipment. The method comprises the following steps of: 1) separately storing and processing a data entity and a logical relationship of a real-time event; 2) managing an idle index node by using a chain pool mechanism; 3) adopting a doubly linked list structure with one write pointer and a plurality of read pointers; 4) processing the logical relationship among real-time events by using two keywords, namely event occurrence time and event priority; 5) fully and correctly reestablishing the whole doubly linked list in the process of initialization according to a generation time keyword stored in an SRAM event data entity; and 6) establishing a process list and event management routine interface for connecting an operation system. The method has the advantages of capacity of managing various real-time events in a converting station, adapting to various real-time operation systems, simplifying the complexity of managing the real-time events by multiple kinds of intelligent electronic equipment, particularly analyzing, calculating and processing the real-time events.

The invention provides a method for forwarding Linux kernel protocol stack network bridge data, when an LINUX kernel protocol stack is established, a double lined list which stores a forwarding table is initialized, when the data message enters a kernel protocol stack network bridge, an HASH value is computed through a virtual local area network (VLAN) and a multi-access computer (MAC), and is taken as an index to establish the forwarding table to be inserted into the double lined list which stores the forwarding table, and for the MAC addresses from one source, the forwarding tables which correspond to a plurality of different output equipment can be learnt. The method has the beneficial effects that the whole process is simple to realize and easy to carry out, ensures the stability and the expansibility of an LINUX kernel, and is very suitable for the requirement on network bridge data forwarding of embedded devices which are based on an LINUX system.

A method for traversing a multiplexed data packet stream including choosing a starting location in a multiplexed data packet stream, making a bi-directional search from the starting location to locate a nearest system stream object; and traversing the system stream as a doubly-linked list of objects. Traversing the system stream preferably includes traversing object links of the doubly-linked list of objects. The objects are further preferably cached with an LRU process. Multiple disjoint locations in the system stream can be accessed with a virtual linked list. The virtual linked list can be used to search forwardly, backwardly, or both forwardly and backwardly.