The embodiment of the invention provides a
data processing method and device.The method comprises the steps that a target application program is analyzed, a
parameter type and a function calling relation corresponding to a target function in the target application program are obtained, and the target function comprises a to-be-processed function and a
correlation function corresponding to the to-be-processed function; generating a
data processing program corresponding to the target function by utilizing a target program generation template according to the
parameter type corresponding to the target function and the function calling relationship; inserting a detection point at a function entry of the target function, and establishing an association relationship between the detection point and a
data processing program; under the condition that it is determined that the target function is called, a data
processing program is triggered to execute the function event in the execution process of the target function through the detection point, an event execution result is obtained from the kernel mode through the data
processing program, and the event execution result comprises performance data corresponding to the target function in the execution process of the kernel mode; fine and timely-response
troubleshooting and performance optimization are realized.