Method and device for deadlock detection

Abstract

An embodiment of the invention discloses a method and device for deadlock detection and is applied to the technical field of computer. According to the deadlock detection method, use information of resource locks used in the running process by threads / processes of every application program is stored in the deadlock detection device, so that when the stored resource locks are mutually exclusive locks and operation time of the resource locks is longer than or equal to preset time, the deadlock detection device can obtain occupation information and / or waiting information of the resource locks and judge whether the resource locks are deadlocks, and if the resource locks are the deadlocks, information for showing that the resource locks are the deadlocks is output. As long as the operation time of the resource locks meets certain time, the possibility that the resource locks are the deadlocks is high, whether the resource locks are the deadlocks is further determined, people do not need to search the resource locks which are the deadlocks out of a resource lock map corresponding to the application program, deadlock detection time can be saved, resource consumption can be reduced, and the detection performance is high.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a deadlock detection method and device. Background technique [0002] Multi-process / thread deadlock in computer programs refers to an abnormal phenomenon in which two or more processes / threads are waiting for each other to use shared resources due to competition for shared resources during the execution process, usually causing the entire computer to The program's service is paralyzed (hanged). Generally, the occurrence of deadlock mainly has the following types: mutual exclusion, circular waiting, partial allocation, and lack of priority. [0003] In the process of computer program development and design, in order to avoid deadlock, it is necessary to improve from the prevention and detection links. In the prior art, the detection of deadlock is mainly to monitor the process / thread to obtain resource lock information, and then establish a resource lock map , judging whether ...

AI Technical Summary

Problems solved by technology

However, for many large-scale computer programs, the amount of code is large, and there are many processes and threads involved in the operation, so the demand for resource locks is also large, making it difficult to generate resource lock graphs and search for deadlocks from resource lock graphs. The operation takes a lot of time and system resources, and the performance is not very good

Images